United States
             Environmental Protection
Environmental Research
Gulf Breeze FL 32561
             Research and Development
&EPA       Workshop
April 1979
             Degradation of
             Pollutants in  Marine


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination of traditional grouping  was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
      1.  Environmental  Health Effects Research
      2.  Environmental  Protection Technology
      3.  Ecological Research
      4.  Environmental  Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

                                      April  1979

         Pensacola Beach, Florida
              9-14 April 1978
                 Edited by
     A.  W.  Bourquin and P. H. Pritchard
      Environmental Research Laboratory
        Gulf Breeze,  Florida 32561
        This workshop was conducted
            in  cooperation with
Georgia State University, Atlanta, Georgia
          Contract No. 68-03-1325
   U.S. Environmental Protection Agency
        GULF BREEZE, FLORIDA 32561

     This report has been reviewed by the Environmental Research  Laboratory,
Gulf Breeze, U.S. Environmental Protection Agency,  and approved for publica-
tion.  Approval does not signify that the contents  necessarily  reflect  the
views and policies of the U.S.  Environmental Protection Agency, nor does men-
tion of trade names or commercial products constitute endorsement or  recom-
mendation for use.

     As man continues to develop and utilize  new  organic chemicals, the poten-
tial for environmental pollution increases  accordingly.  Once a chemical
enters the environment, its eventual transfer to  the aquatic environment be-
comes a distinct possibility.   Increasing evidence indicates that a thorough
knowledge of the fate of these chemicals is required.  Inherent in any fate
study is the need for an accurate assessment  of the potential for the micro-
bial degradation of the pollutant.  However,  information on biodegradation,
specifically in aquatic environments, is sparse.  Data from laboratory and
field studies with soil microorganisms are  often  based on assumptions that
have no sound scientific basis or consensus,  regardless of the fact that
aquatic environments appear sufficiently unique to demand separate considera-

     The microbiologist has no consistent guideline regarding appropriate
techniques for the determination of recalcitrance in aquatic environments.
Further, microbiologists in general have not  provided the chemical industry
and regulatory agencies with evaluations of chemical structures that can or
cannot be degraded by aquatic microbial populations.

     The Workshop on Microbial Degradation  of Pollutants in Marine Environ-
ments was organized to offer representatives  from government, academia, and
industry a forum to examine key issues regarding  the future direction and
focus of scientific investigations in this  field.  Participants addressed a
number of questions of primary concern: What are the differences and simi-
larities between biodegradation processes in  aquatic, terrestrial, and labora-
tory environments?  Should a separate research effort determine biodegradation
potential in aquatic environments or can adequate information be obtained from
other sources?   What methodological criteria must be established to provide
interchangeable degradation information?  What constitutes a recalcitrant
molecule?  What is the potential for a particular environment or its labora-
tory simulation to naturally dispose of a polluting chemical?  We hope that
this publication of the workshop proceedings  will provide an up-to-date ref-
erence for professionals involved in the fate, regulation, and production of
potential aquatic pollutants.
     This international workshop, held April 10-14, 1978, at Pensacola Beach,
Florida, focuses on pertinent issues related to the scientific investigation
of microbial degradation of organic chemicals in aquatic environments.  Par-
ticipants discuss methodological criteria for these investigations and the
need for biodegradation studies.  Speakers and contributed papers for open
sessions explore these topics:   (1) biochemistry of microbial degradation;
(2) transformation in aquatic environments;  (3) compartmentalization in
aquatic environments; (4) biodegradation in microcosms;  (5) degradation
methodology; and  (6) persistence and extrapolation.  Discussions within each
session are presented.  These proceedings conclude with a summary report and
workshop consensus reports drafted by special task groups with recommenda-
tions concerning the research, production, and regulation of potential aquatic
pollutants.  This report is submitted in fulfillment of Contract No. 68-03-
1325 by Georgia State University under the sponsorship of the U.S. Environ-
mental Protection Agency.  Work was completed as of 30 December 1978.

Foreword	    ill
Abstract	     iv
Acknowledgment	     ix
Editorial Committee  	     x
Advisory Committee 	    xii

                                INTRODUCTION                                I

The Ultimate Sink:  Keynote Address  	     3
    Holger W. Jannasch
Introduction and Welcome to the Workshop 	     10
    Thomas W. Duke
Need for Biodegradation Studies by the EPA	     11
    Arthur Stern
Mission of the Workshop	     15
    A. W. Bourquin


Transformation of Xenobiotics by Microbial Activity  	     19
    Jean-Marc Bollag
Degradation Mechanisms 	     28
    Peter J. Chapman
Role of Cometabolism	     67
    Martin Alexander
Diauxic and Cometabolic Phenomena in Biodegradation Evaluations   ....     76
    T.-W. Chou and N. Bohonos
Metabolism of Cyclopropane Carboxylic Acid by Bacteria 	     89
    C. 0. Patterson and G. D. Hegeman

Discussion	     97


Nutrient Limitation  	    107
    G. D. Floodgate

Natural Heterotrophic Activity in Estuarine and Coastal Waters 	    H9
    Richard T. Wright

Structural Changes Caused by Environmental Pollution or Eutrophication
    in Bacterial Communities in Sea and Lake Water	    135
    Yuzaburo Ishida and Hajime Kadota

Nitrogen/Carbon Ratios and Rates of Ammonia Turnover in Anoxic
    Sediments	    148
    T. H. Blackburn

Effect of Chemical Structure and Concentration on Microbial
    Degradation in Model Ecosystems  	    154
    M. J. DiGeronimo, R. S. Boethling and Martin Alexander

Discussion	    167

The Marine Anoxic Environment   	   181
    N. J. Poole
Aquatic Environment and Pesticide Behavior:  Microorganism Binding
    Phenomena	   191
    E. Paul Lichtenstein
Role of Surface Microlayers	   201
    Birgitta Norkrans
Surface Microlayers of the North Atlantic:  Microbial Populations,
    Heterotrophic and Hydrocarbonoclastic Activities 	   214
    F. J. Passman, T. J. Novitsky and S. W. Watson
Interfaces and the Aquatic Environment 	   227
    D. G. Cooper, J. E. Zajic and D. F. Gerson

Discussion	   241
                        BIODEGRADATION IN MICROCOSMS

System Design Factors Affecting Environmental Fate Studies in
    Microcosms	    251
    P. H. Pritchard, A. W. Bourquin, H. L. Frederickson and T.  Maziarz

Biodegradation in Microcosms:  Application of Mathematical Modeling  .  .    273
    Lenore S. Clesceri
Microbial Community Structure  	    2P"
    J. H. Slater

Predicting the Fate of Organics:  Conceptual Model and
    Experimental Approaches  	   296
    P- W. Rodgers and J. V. DePinto
Transport and Fate of Anthracene in Aquatic Microcosms	   312
    J. M. Giddings, B. T. Walton, G. K. Eddlemon and K. G. Olson
Discussion	   321
                             CONTRIBUTED PAPERS

Resistance of Pollutants to Degradation in Saline Environments 	   337
    E. M. Davis, J. Bishop and R. K. Guthrie
Methanogenic Biodegradation of Aromatic Compounds  	   348
    J. B. Healy, Jr. and L. Y. Young

Some Approaches to Studies on the Degradation of Aromatic Hydrocarbons
    by Fung'i	   360
    C. E. Cerniglia, R. L. Hebert, R. H. Dodge, P. J. Szaniszlo
    and D. T. Gibson

A Novel Selective Enrichment Technique for Use in Biodegradation
    Studies	   370
    D. Liu
Distribution, Abundance, and Petroleum-Degrading Potential of Marine
    Bacteria from Middle Atlantic Continental Shelf Waters 	   380
    A. E. Maccubbin and Howard Kator
Distribution and Degradative Potential of Kepone-Resistant Bacteria in
    the James River and Upper Chesapeake Bay 	   396
    Steve A. Orndorff and Rita R. Colwell

Discussion	   408
                           DEGRADATION METHDOLOGY

Degradation Methodology:  Chemical-Physical Effects  	   423
    Jack R. Plimmer
Microbial Degradation in the Environmental Hazard Evaluation Process .  .   434
    W. E. Gledhill and V. W. Saeger
Microbial Degradation of Organochlorine Compounds in Estuarine
    Waters .and Sediments	   443
    Richard F. Lee and Charlotte Ryan
Application of a Laboratory Freshwater Lake Model in the Study of
    Linear Alkylbenzene Sulfonate  (LAS) Biodegradation 	   451
    C. R. Eggert, R. G. Kaley and W. E. Gledhill
Discussion	   462


Prediction from Laboratory Studies of Biodegradation of Pollutants
    in "Natural" Environments  .....................   479
    Arthur M. Kaplan

Use of Microcosms to Assess Microbial Degradation Processes and
    Sediment-Water Interactions in Reservoirs that Develop
    Anaerobic Hypolimnions .......................   485
    D. Gunnison, J. M. Brannon and P. L. Butler

Task Group Reports	   513

Discussion	   530

Summary of the Conference	   534
    Stanley Dagley


Attendees	   545

     We wish to express our appreciation for the tremendous  help of  the  edi-
torial committee in reviewing and editing the final report of  this workshop,
and of the advisory committee in establishing workshop goals and reviewing
abstracts of contributed papers.  We also are grateful for the encouraging
remarks and cooperation of the participants and session chairmen.  Ms. Peggy
Wells, our workshop secretary, deserves special recognition  for her  efficient
and gracious handling of all aspects of the workshop from meager tasks to the
coordination of both professional sessions and social events.

                           EDITORIAL  COMMITTEE
                        Samuel P.  Meyers,  Coordinator
                         Department of Food Science
                         Louisiana State University
                            Baton  Rouge, LA 70803
Donald G. Ahearn
Department of Biology
Georgia State University
Atlanta, GA 30303

Warren Cook
Department of Biology
Georgia State University
Atlanta, GA 30303
Joseph Cooney
Chesapeake Biological  Lab
University of Maryland
Solomons, MD 20688

Sidney Crow
Department of Biology
Georgia State University
Atlanta, GA 20303

                           ADVISORY  COMMITTEE
 Al W. Bourquin and P.  H.  Pritchard
  Environmental Research Laboratory
U.S. Environmental Protection Agency
     Gulf Breeze,  FL 32561

Donald G. Ahearn
Georgia State University
Atlanta,  Ga.

Robert Belley
Eastman Kodak
Rochester, N.Y.

Jean-Marc Bollag
Pennsylvania State University
State College, Pa.

Peter Chapman
University of Minnesota
St. Paul, Minn.
                     Holger Jannasch
                     Woods Hole Oceanographic  Institution
                     Woods Hole, Mass.

                     Donald Kaufman
                     U.S.  Department of Agriculture
                     Beltsville, Md.

                     Jean Pulliam
                     U.S.  Environmental Protection Agency
                     OPP,  Washington,  D.C.

                     Richard Raymond
                     Suntech, Inc.
                     Marcus Hook, Pa.

                     Arthur Stern
                     U.S.  Environmental Protection Agency
                     OTS,  Washington,  D.C.


                            THE ULTIMATE SINK

                               KEYNOTE ADDRESS

                             Holger W. Jannasch
                    Woods Hole Oceanographic Institution
                            Woods Hole, MA 02543

                The discussion of the marine environment as a site
           for degradation processes is introduced by reviewing some
           fundamental concepts of microbial ecology.  The effective
           environment for bacterial and fungal activities is the
           micro-habitat of the individual cell.  Characteristics of
           the aquatic environment are closely related to the physi-
           cal and chemical properties of water in general, and of
           seawater in particular.  On the macro-scale, the marine
           environment is divided, more for logistic than for micro-
           biological reasons, into estuaries, littoral zones, water
           column, sediment, etc.  For open ocean waters, the extreme
           dilution of metabolizable organic materials and pollutants
           is characteristic, and poses specific problems of micro-
           bial uptake and degradation.  As an example, a microbial
           degradation study in the deep sea is discussed with empha-
           sis on hydrostatic pressure and low temperatures as en-
           vironmental factors.  Microbial activities were reduced at
           deep sea conditions, and barophilic responses of natural
           microbial populations were not found.  The consequences
           for offshore and deep sea dumping of organic waste mate-
           rials are discussed.

     The ocean had to endure this not so romantic slogan, the "ultimate sink,"
ever since decomposition processes of organic and inorganic materials have
been considered on the global scale.  Slogans, however, are at least as often
wrong as they appear to be downright.  Inorganic marine deposits and solute
complexes indeed represent accumulative evidence of global geological pro-
cesses, but does this concept hold in the biological realm?  In setting the
stage for this workshop on Microbial Degradation of Pollutants in Marine En-
vironments, it appears useful to attempt an answer to this question and to
discuss briefly the environmental characteristics of marine microbial habitats

with respect to potential degradation processes.

     As a general principle, organic materials are produced as well as de-
graded by metabolic processes of organisms.  If this recycling was complete in
a given time period, and if the above principle was not riddled with excep-
tions, the term "ultimate sink" for biodegradation makes little sense for any
particular environment.  Considering, however, that a substantial number of
organic materials are, or become during breakdown, recalcitrant to further
biodegradation  (a topic discussed in detail by Alexander [1]), and that more
and more man-made and non-metabolizable products become part of our environ-
ment, many of these compounds will ultimately reach the ocean.  Above all,
however, it is the mere size of the marine portion of the bisophere which con-
stitutes its significance as a site of biodegradation.  Considering the aver-
age depth of the world oceans at 3800 m, as compared to 2 or  3 orders of mag-
nitude less for the terrestrial and freshwater environment, more than 99% of
the biosphere consists of seawater  (13) .  Certainly,- a large  portion of this
immense volume compares to a marine desert, while some small  fractions, es-
tuaries and littoral zones, are similar in their biological turnover rates to
highly productive terrestrial and freshwater environments.
                              THE MICRO-HABITAT

     Before we look into the characteristics of the various marine environ-
ments on the macro-scale, however,- we have to remind ourselves that the typi-
cal habitat of any individual organism has to be considered in the perspective
of its own dimension.  The vegetative microbial cell with its relatively large
and metabolically reactive surface responds quickly to types and concentra-
tions of utilizable substrates and to changes of physico-chemical conditions
of its immediate surroundings.  As a consequence,  "the effective habitat of a
microorganism is its micro-habitat, the immediate  surrounding of the cell in a
compatible scale of space and time in relation to  the radius of its metabolic
action and interaction"  (4).  As a corollary, it is hazardous to describe mi-
crobial isolates by macro-environmental characteristics, e.g., estuarine or
deep sea bacteria, before typically adaptive traits have actually been found.
We know too well the ubiquity of microorganisms and their ability to survive
unfavorable conditions for a long time in resistant, non-vegetative stages.
In addition, most of our means to look at the size and composition of natural
microbial populations are indirect and selective.

     A classical case is that of Nathanson's (12) abortive attempt,  in the
wake of systematic monograph-writing on marine plants and animals in the late
1800s, to describe the bacterial flora of the Gulf of Naples.   After years of
hard work he gave up when he realized that the number of isolates kept in-
creasing without yielding the slightest clue whether they were typical for
this particular environment,  the Gulf of Naples.   His vain attempt has been
repeated many times since with respect to marine  environments  of various de-
scriptions .

     More recently, the predominant significance  of micro-habitats  succeeded
in eliminating the sharp dividing line between aquatic,  terrestrial  and even
medical microbiology.   In ecological  research the distinction  between  these

academic disciplines is now more and more de-emphasized under the label of
basic and-applied environmental or biogeochemical microbiology.
                           THE MARINE ENVIRONMENT

     There are, of course, certain microorganisms specifically adapted to live
in aquatic habitats.  Characteristically, many of them are photo- or chemo-
lithotrophic, or have certain morphological traits, such as being sheathed,
stalked or prostecate forms.  Although these organisms will be indirectly in-
volved in biodegradation processes, our primary concern has to be with those
heterotrophic bacteria which play the main role in the breakdown of pollutants
of non-marine origin.  Such bacteria are usually highly substrate-specific,
but they themselves may or may not be of marine origin.  For example, the de-
composition of phenolic compounds in seawater may be determined a)  by the spe-
cific activity of a marine pseudomonad or b) by the constraints of the marine
environmental conditions on the metabolism of a non-marine pseudomonad.  In
many cases, we will not be able to separate between a) and b).

     What, then, are the basic properties that characterize the aquatic, and
specifically the marine, in contrast to the terrestrial environment?  The
basic feature of the predominantly aqueous environment, in contrast to more or
less dry soils, "are the physico-chemical properties of water, especially its
ability to carry those organic and inorganic chemical species which are essen-
tial and readily available for microbial metabolism in dissolved and ionic
form.  The mobility of nutrients in water by diffusion and currents produces
the characteristic constancy of environmental conditions upon which many of
the typically aquatic microorganisms depend" (4).  In principle, the only
additional and specific characteristics of marine habitats appear to be the
salt complex of seawater  (more its constancy than its particular composition)
and hydrostatic pressure in deeper waters.  The lingering doubt in the exis-
tence of truly marine bacteria, not satisfied by growth response on seawater
agar or obligate sodium requirement, is only presently being more and more re-
moved by accumulating evidence in Baumann's systematic assessment of marine
heterotrophic bacteria (e.g., Baumann et al. [2, 3]).   Although many bacteria
of non-marine origin grow well in seawater, the majority of the marine iso-
lates clearly exhibit a number of well defined metabolic differences.  At this
stage it is too early to speculate on whether and how these differences have
an effect on the characteristic patterns of biodegradation in marine versus
terrestrial environments.

     After having pointed out the significance of the micro-environment in mi-
crobial ecology, any subdivision of the marine environment on the macro-scale
may appear to be of little use.  However, there are a number of practical rea-
sons for doing so.   The mere logistics of field work is one.  Another is the
technical differences in working with rich, shallow water sediments or oligo-
trophic pelagic seawater,  etc.  Summarizing from the literature, the following
six marine habitats have been designated as such and individually studied:
Esturaires; Littoral Zones (beaches, reefs); Shelf:  water column and sediment;
Open Ocean:  water column and sediment.

     Possibilities and needs for further subdivisions are obvious.  Of special

 microbiological  interest  are  interfaces  as habitats.  The water/air interface
 has  physical  and chemical properties which give rise to the highly concen-
 trated  "neuston" populations.  They are  important  in the inoculation and deg-
 radation  of oil  slicks.   Solid/water interfaces, recently treated comprehen-
 sively  by Marshall  (11),  provide  the habitat  for "Aufwuchs" populations.  Mi-
 crobial growth takes place  not only on the surface of utilizable organic sub-
 strates but also on inert suspended particulate matter, a topic summarized by
 Jannasch  and  Pritchard  (6).   Interfaces  between soluble materials are of simi-
 lar  importance as demonstrated by the rich microbial populations surrounding
 oxic/anoxic layers of waters  in estuaries and other highly productive marine
 environments.  With the exception of some special  areas as the well studied
 Black Sea and Cariaco Trench, where stagnant  anoxic water has accumulated,
 offshore  waters  and deep  sediments are rarely found to be anoxic largely due
 to the  comparatively low  rates of oxygen uptake.

     At  this point,  let me diverge a bit from my proper duty to introduce this
workshop by discussing general concepts, and let me use some of my own recent
work as  an example of specific marine-oriented research on microbial degrada-
tion.  Also, I believe, no one else  at this workshop is dealing with the deep
sea as a special marine environment.  At the beginning of this talk, I men-
tioned that 99% of the biosphere by  volume comprises marine environments.  If
we go beyond the 200 m depth of the  shelf and let the deep sea deliberately
begin at a depth of  1000 m, we come  to the surprising result that still 75% of
the biosphere is located below that  line; in other words, it is subjected to
100 or more atm of pressure, temperatures in the range of 1-4°C and, of course,
receives no light.

     It  is important to realize that most of this huge area is far removed
from direct inputs of organic sources of energy.  Its character is oligo-
trophic, the deeper  the more so.  Most of the organic matter produced photo-
synthetically in surface water is decomposed in  the upper few hundred meters.
The huge volume of deep sea water appears to make the mere dilution of pollu-
tants almost indefinite.  At the same time, as we found in earlier chemostat
work, the microbial degradation of substrates is slowed down significantly if
they are diluted beyond a critical point  (5).  Thus, in spite of the large
volume of water, certain pollutants  have been found everywhere in the ocean.
It is self-evident that metabolic processes in higher organisms as well as
characteristic chemical reactions of seawater and sediments participate in
transformations and breakdown of pollutants.

     Our interest in the deep sea has been intensified by the fact that 50
million  tons of solid waste materials are dumped into the sea annually by this
country  alone.  Since shelf waters comprise over 90% of the marine fisheries,
there is a tendency to use deeper water as dumping sites for materials that
might interfere with bottom trawling.  Over decades of work on benthic deep
sea populations and from many recent experimental studies with research sub-
mersibles,  the general impression is that of a highly specialized and fragile
community.   Our question was, how will these populations be affected by the
input of extragenous materials in unusual quantities?  Will microbial enrich-

ment occur in a similar way it does in estuaries or shallow waters?  In short,
can the deep sea be considered as an active treatment system for organic waste

     Over the last seven years, we have deposited and recovered a large va-
riety of organic substrates in depths from 1800 to 5300 m.  These in-situ in-
cubation studies eliminated the potential effect of decompression which occurs
during the recovery of deep sea samples for the subsequent incubation in pres-
surized laboratory cylinders.  In some of these field experiments, media-con-
taining bottles were inoculated with the aid of the research submersible Alvin
directly on the deep sea floor, thus excluding all surface-borne microorga-
nisms.  This way we hoped to find enrichments of truly barophilic bacteria,
which are defined as metabolizing at an increased rate at pressures higher
than one atmosphere.  Such a proposed indigenous microflora might be dispro-
portionally outnumbered if surface-originated organisms were not excluded from
the experiment.  The substrates used in these quantitative studies included a
number of organic acids, amino acids, carbohydrates (sugars and polysaccha-
rides) and some complex materials such as starch, gelatin and agar.  Semi-
quantitative recoveries were made with samples of wood, refined cellulose,
chitin, food stuffs and seaweed.  Incubation times ranged from several days to
15 months.  Since these studies aimed at the general question of bacterial me-
tabolism under deep sea conditions, pollutants were not yet included.  The
specific techniques are discussed elsewhere (7, 8, 14).

     As a general result, deep sea-adapted microbial activities were not
found.  The effect of increased pressure and low temperature appeared to be
the reason for the slowdown of the measured activities as compared to labora-
tory controls incubated at equal temperature.   This effect was the same in mi-
crobial populations exclusively collected at the deep sea floor and in those
originated at the sea surface.  In both cases decompression resulted in in-
creased activity.  In other words, no truly barophilic response could be re-
corded.  These results were substantiated by growth and uptake studies using
pressure-retaining samplers which permitted measurements on deep sea popula-
tions that were never subjected to changes of their indigenous pressure and
temperature (9, 10) .

     It is very likely that these results will also apply to the microbial
degradation of pollutants wherever they reach the deep sea.  They also indi-
cate that a substantial accumulation of certain materials might occur in deep
waters.  In short, the deep sea has definite limitations as a dumping site for
organic wastes.  Several observations made with the aid of baited cameras de-
posited on the deep sea floor led our attention to the role of higher orga-
nisms in the removal of organic waste materials.  It will be difficult to
separate these effects in the deep sea entirely from microbial activities
since much of it appears to take place in the intestinal tracts of inverte-
brates and fishes.  Our work in this area is continuing.

     Where does this leave us with respect to the "ultimate sink"?  If ''sink"
or "dump" refers to places where discarded materials do persist, at least par-
tially, then the open ocean and the deep sea,  as by far the largest portion of
the biosphere with its relatively limited degradation potential, might indeed
be considered an ultimate sink.  Development of needed monitoring technology

and smooth cooperation between producers of new potential pollutants, govern-
ment agencies, and independent research laboratories are paramount require-
ments .

                                THE WORKSHOP

     The program of this workshop is based on the present-day state of the art
in our knowledge of microbial breakdown processes, modeling of systems, the
types of organisms involved, techniques of monitoring and the variety of pol-
lutants existing at this time.  Oil, herbicides and pesticides, DDT and PCBs,
domestic and industrial sewage, are those pollutants foremost on the public's
mind.  Their slow disappearance from the polluted environment is generally un-
derstood to be caused by biological and chemical activities.  Bacteria, who
generally suffer from their bad reputation as one of man's worst enemies, have
gained considerable favor by being mentioned in connection with environmental
cleanups.  To this day, the microbial ecologists, since Winogradski, are try-
ing to put the image of microorganisms, tarnished by dominating medical view-
points, in the right perspective.  The present workshop, brought about ulti-
mately by public interest, is a veritable success in these efforts.

     The scene is set by the steering committee of this workshop to include a
wide variety of topics in a number of well organized sessions.  Let us discuss
more than just deliver; conduct an actual workshop, not a mere convention.
The outcome should be stimulating for the further development of research in
microbial degradation, a field that will be of more and more importance as
long as variety and quantity of pollutants in the environment are on the in-

                              LITERATURE CITED

1.  Alexander, M.  1973.  Non-biodegradable and other recalcitrant molecules.
     Biotech. Bioeng. 15:611-647.

2.  Baumann, L., P. Baumann, M. Mandel, and R. D. Allen.  1972.  Taxonomy of
     aerobic marine eubacteria.  J. Bacteriol. 110:402-429.

3.  Baumann, P., and L. Baumann.  1977.  Biology of the marine enterobacteria:
     genera Beneckea and Photobacterium.  Ann. Rev. Microbiol. 31:39-61.

4.  Jannasch, H. W.  1978.  Microorganisms and their aquatic habitat.  In
     W. B. Krumbein  (ed.), Environmental biogeochemistry.  Ann Arbor Sci.
     Publ., Ann Arbor, Mich.

5.  Jannasch, H. W., and R. L. Mateles.  1974.  Experimental bacterial ecology
     studied in continuous culture.  Adv. Microb. Physiol. 11:165-212.

6.  Jannasch, H. W., and P. H. Pritchard.  1972.  The role of inert particu-
     late matter in the activity of aquatic microorganisms.  Mem. 1st. Ital.
     Idrobiol. 29 Suppl.:289-308.

 7.   Jarmasch, H. W., and C. O. Wirsen.  1973.  Deep sea microorganisms:  in
      situ response to nutrient enrichment.  Science 180:641-643.

 8.   Jannasch, H. W., and C. O. Wirsen.  1977.  Microbial life in the deep sea.
      Sci. Amer. 236:42-52.

 9.   Jannasch, H. W., C. 0. Wirsen, and C. L. Winget.  1973.   A bacteriological,
      pressure-retaining, deep sea sampler and culture vessel.  Deep-Sea Res.

10.   Jannasch, H. W., C. 0. Wirsen, and C. D. Taylor.  1976.   Undecompressed
      microbial populations from the deep sea.  Appl. Environ. Microbiol.

11.   Marshall, K. C.  1976.  Interfaces in microbiology ecology.  Harvard Uni-
      versity Press, Cambridge, Mass.

12.   Nathanson, A.  1902.  Unpublished correspondence with Anton Dohrn.
      Archives of the Zoological Station, Naples.

13.   Sverdrup, H. W., M. W. Johnson, and R. H. Fleming.  1942.  The oceans.
      Prentice Hall, Inc., Englewood Cliffs, N.J.

14.   Wirsen, C. O., and H. W. Jannasch.  1976.  The decomposition of solid or-
      ganic materials in the deep sea.  Environ. Sci. Technol. 10:880-887.


                               Thomas W.  Duke
                 Director,  Environmental  Research Laboratory
                    U.S.  Environmental Protection Agency
                            Gulf Breeze,  FL 32561

     It is my pleasure to welcome you to  Santa Rosa Island and to this Work-
shop on behalf of the U.S.  Environmental  Protection Agency and Georgia State
University.  As Director of the Environmental Research Laboratory,  your host
this week, I would like to invite each of you to  tour the laboratory during
your stay here.  At the laboratory we study the impact of toxic organics on
the marine environment.  The primary emphasis in  our research has been the ef-
fects of pesticides on marine organisms.   Early in our history we concluded
that in order to study effects of toxic organics, you certainly must have some
knowledge of the fate of these materials.  When we study fate of pollutants,
we cover a variety of disciplines which play an active part in determining
what happens to an organic compound when  it enters the marine environment.

     In the past we have dealt primarily  with organochlorine compounds which
were quite resistant to biodegradation.  Therefore, fate research focused on
other processes such as photolysis, hydrolysis and sorption kinetics.   How-
ever, more recently our research program  has become concerned with the "bio-
degradable" compounds such as organophosphorus and carbamate pesticides.  This
evolution has, of course, brought more emphasis to our studies on biological
transformations.  For this reason, we have scientists on our staff such as Al
Bourquin, Hap Pritchard, and Dick Garnas  who are  involved in studies to deter-
mine biological and physio-chemical fate  of organics in the marine environ-
ment.  The importance of good scientific  research in environmental studies is
evident to all of us.  Dr.  Stern will be  telling  you about the pressing needs
of the Office of Toxic Substances to develop good techniques to evaluate en-
vironmental fate information to make regulatory decisions.  Even a cursory
glance at the program shows we have some  of the leading experts in the field
present in this Workshop.  We are extremely pleased to have this opportunity
to host this Workshop and I want to second a remark of Dr. Jannasch in his
keynote address.  That is,  the success of this meeting will be that it is not
just one of lecturing and responding but  one of a true workshop.  I leave you
with this challenge that there will be good intensive discussions on fate pro-
cesses so that we can reach the goals of  the workshop.  Your participation is
needed to reach these goals.


                                Arthur Stern
                         Office of Toxic Substances
                    U.S. Environmental Protection Agency
                               Washington, DC

     In 1976, the Toxic Substances Control Act  (TSCA) was passed by Congress.
The Act created the Office of Toxic Subtances (OTS) and charged that office
with the responsibility of developing procedures whereby all chemicals brought
into commercial use, other than pesticides, foods, drugs and cosmetics, would
be evaluated with regard to their effect on the public health and the environ-
ment.  One of the requirements imposed on OTS by the Act was the need to de-
velop test standards, protocols and/or guidelines for use by industry.  These
would be employed to generate the data which would be used by the agency to
make risk assessments.  Tacit is the understanding that only state-of-the-art
technology be incorporated into test protocols.   Another important point is
that the Act stipulates that testing requirements be reviewed annually for
adequacy, since new developments often lead to better, more relevant and
cheaper techniques.

     At the present time, it is our intent to spell out test protocols in some
detail.  This would expedite the risk assessment process considerably and
should ultimately result in faster adjudication of applications.  Furthermore,
advance detailing of the criteria upon which assessments will be made by the
agency will allow potential applicants to make preliminary judgments of their
own.  This does not mean that published testing requirements will be totally
inflexible.  It is hoped that enough flexibility will be incorporated into
such requirements to take into consideration the large variation in chemical
structures, properties, end-uses and distribution.  In short, our objective is
the development of the minimal number of cost-effective testing requirements
the data from which should effectively provide answers essential in making
risk assessments.   To this end,  our present thinking is to arrange tests in a
series of tiers.  The evaluation of data generated by tests carried out in
each tier will lead to a decision as to whether or not the next higher tier of
tests will be required.  The lowest tier will consist of relatively inexpen-
sive short-term tests, while subsequent tiers would be characterized by more
extensive testing requirements.   All of this represents our current thinking
and is a somewhat idealized concept.  Since we are only starting out in the
rule-making process, it is quite possible some elements of this concept may
prove to be impractical and have to be modified accordingly.

                            ENVIRONMENTAL TESTING

     Environmental testing includes the areas of ecological effects and chemi-
cal fate.  There are many problems that are associated with the development of
ecological-effects testing which I will not dwell on since our purpose this
week is to consider one aspect of chemical fate.  However, I would like to
mention a few of these problems before moving on to the subject of persistence
testing.  The environment consists of several compartments, many ecosystems
within each compartment and diverse phyla, genera and species of plants and
animals within given ecosystems.  The question of whether testing of single
species in a laboratory bears any relevance to what happens in the real world
must be answered ultimately.  While this correlation is far from clear, we
have to start somewhere and judgment on such correlations has to be reserved
until more information is acquired.  It is our present feeling that microcosm
and field studies will be eventually needed to validate predictions made on
the basis of individual species testing.  These problems also happen to be
germane to the area of biodegradation.

                                CHEMICAL FATE

     Chemical fate, which is a part of the environmental picture, really con-
sists of the two separate, but intimately related, areas of chemical transport
and persistence.  Conceptually, it is impossible to differentiate these areas
but, from the standpoint of developing testing requirements, we have found it
convenient to attack each separately.  When it comes to interpreting data de-
rived from test protocols, both areas must certainly be evaluated in concert.


     Some generalized predictions with regard to chemical transport can be
made on the basis of physical/chemical characteristics of test materials.  In
other words, some concept may be derived from an analysis of a chemical's key
physical/chemical characteristics regarding the environmental compartments
which may serve as its major deposition site or sites.  Persistence data give
us some indication of the expected residence time of a chemical after deposi-
tion.  Finally, some idea of the exposure potential of a chemical pollutant
(within a few orders of magnitude) can then be derived by integrating trans-
port and persistence data with information acquired from responses to Section
8 of the Act, i.e., production volume, production distribution, etc.

     This analysis has several important applications.  First, as stated be-
fore, it provides a rough idea with regard to environmental exposure.  This,
in turn, can serve as a guide in determining the dosage levels to use in both
health and environmental effects testing.  Second, certain environmental com-
partments can be ruled out as primary deposition sites and, thus, the magni-
tude of ecological testing required can be reduced.


     As stated earlier, persistence data allow some predictions to be made
with regard either to the duration of a chemical's residence after deposition,
if its application is intermittent or occurs only once, or to its turnover rate
if its application is continuous.  Chemicals which are found to be innocuous
at low dosages can, if persistent, accumulate to hazardous environmental levels
over a period of time.  Further, chemicals which are degraded cannot be consid-
ered exempt from further concern as potential hazards.  There are quite a few
examples of degradation products posing a greater threat than the parent com-
pound from which they originated.  Thus, persistence testing must include some
provisions for the examination of degradation products of non-persistent


     There are three general mechanisms by which chemicals can be degraded.
These are photochemical, chemical and biological.  Since our major concern this
week is the latter, let me point out some of the problems we face in developing
test standards/protocols for biodegradation.

     First of all, literature on biodegradation testing is rather diffuse.
While it is true that a good deal of work has been published on the degrada-
tion of organic compounds, much of it has been directed at the determination
of metabolic pathways by which certain types of chemically pure materials are
transformed by pure cultures in a well defined medium.  Other studies which
had more applied objectives  (such as the stabilization of organics)  usually
only dealt with a restricted number of chemical types.  This doesn't imply
that the techniques used were necessarily limited to these types of chemicals
but it does mean that a good deal of validation work would have to be carried
out to determine the chemical limits of method applicability.  There have been
methods described which show some promise of general utility, such as the
river die-away test, but these have certain inherent deficiencies that make
their adoption as a sole means of biodegradation testing impractical.  It is
quite likely that ultimately we will have to require the use of a battery of
test methods to accommodate the spectrum of variables that influence the ex-
tent and rate of biodegradation.

     I would now like to list those issues which are particularly important  in
devising meaningful tests for biodegradation and which I hope will be addressed
by the speakers and workshops during the coming week:

     • Mixed vs pure cultures
     • Sources of mixed and/or pure cultures
     • Laboratory vs natural media
     • Adapted vs unadapted cultures (continuous vs single slug or intermit-
       tent doses)
     • Aerobic vs anaerobic testing
     • Concentration(s) of toxicant to use in tests, i.e., differential re-
       sponses to low and high concentrations of many chemicals run the gamut
       from inhibition to stimulation of growth

     • Role of cometabolites
     • Synergistic, commensal, symbiotic and antagonistic relationships be-
       tween species in a natural environment.

Some other problems which are shared by all methods of degradation can be sum-
marized as follows:

     • Correlation of tests on isolated reactions in the laboratory with mul-
       tiple reactions in nature
     • Design of experimental conditions to meet those most frequently encoun-
       tered in nature (pH; temperature; protective/concentrating influence of
       particulates such as soil, dust particles and suspended matter in
       aquatic environments)
     • Additive vs non-additive effects of multiple routes of degradation
     • Definition of "significant degradation'1
     • Analysis and identification of "most important" degradation products
     • Testing requirements for identified significant degradation products.

     I would like to conclude with the hope that you will:  address the issues
that must be resolved, those I have mentioned and others that you are aware of;
identify methodology in the area of biodegradation that is almost, but not
quite, state-of-the-art;  and identify short- and long-range method development
programs whose successful completion could improve our current capabilities in
this area.

                         MISSION OF THE  WORKSHOP

                               A. W. Bourquin
                      Environmental Research Laboratory
                     U.S. Environmental Protection Agency
                            Gulf Breeze, FL 32561

     The remarks of Drs. Stern and Duke emphasize the timeliness of this Work-
shop.  With the legislative passing of TOSCA, creation of the Office of Toxic
Substances, the proposed pesticide registration guideline in the Office of
Pesticide Programs, and the new effluent guidelines, the need for more empha-
sis on biodegradation studies and a better understanding and interpretation of
biodegradation results becomes increasingly important.  That is the mission of
this Workshop, to develop guidelines for better understanding and interpreta-
tion of biodegradation studies in aquatic environments.

     In order to make a regulatory decision on any chemical entering the en-
vironment, we need to know what will happen to that chemical when it enters a
particular environment.  What is the potential for biodegradation in a given
environment?  If it is low, will the compound persist and cause the environ-
ment harm?  If not persistent, what are the possible degradation products?
What is the toxicity of these metabolites to non-target organisms in relation
to the parent compounds?

     With the increased usage of the new era of pesticides, the organophos-
phates as opposed to the organochlorines, information on the biodegradation of
these compounds is even more important.  Currently, much information on the
fate of a chemical in aquatic environments is determined by techniques devel-
oped for soil or terrestrial studies.  For example, in the proposed pesticide
registration guidelines, section on aquatic metabolism, I quote:  "The pre-
ferred substrate is sediment, but use of flooded soils may be inadequate."  If
this proposed method is adopted, do we have all the needed information for
such a substitution?  Can we really use soils in place of sediment when study-
ing aquatic metabolism?  These are the type questions I hope this group will
attack this week.

     The aquatic environment is unique.  It should be studied with the incor-
poration of certain characteristic factors.  The aquatic environment is an
open system, encompassing many time-independent processes.  At the same time,
it is a dynamic system with continuously changing inputs—pollutants or nu-
trients .  These substances may be put into the system in pulses, such that one
does not have a continuous system.

     The aquatic environment is a highly dilute system and pollutants entering
this environment are diluted out rapidly to low levels.  The diluted environ-


merit of aquatic microorganisms makes their survival capacities or physiology
different from other microbes surviving in high nutrient environments.  Micro-
organisms are continuously exposed to very low nutrient concentrations; there-
fore, the microbes are probably capable of growth at extremely slow rates.  In
fact, it is quite probable that aquatic microorganisms may have an entirely
different physiology from microorganisms living in a soil environment.  On the
other hand, pollutants and nutrients are sometimes concentrated in specific
interfaces of the aquatic environment, such as surface slicks and sediments.
These highly concentrated areas offer a special ecological niche in an other-
wise dilute environment.  This is an area of concern to us.  What role do
these special environments play in the overall metabolism of aquatic micro-
organisms?  All these factors seem to dictate a specific methodology for
aquatic studies.

     The aquatic environment is characterized by water continuously passing
over sediments, as opposed to a soil environment where the water to soil ra-
tios are such that water, most of the time, soaks into the soil with only oc-
casional runoff.  This movement of water over sediment causes changes in sorp-
tion/desorption kinetics of pollutants into and from sediments.  The occur-
rence of suspended particulates is a unique characteristic of the aquatic en-
vironment and may affect the biodegradation rate of chemicals.  Biodegradation
in the sediments is a relatively unexplored area in aquatic ecology, particu-
larly the role of sediments to initiate degradation prior to the involvement
of planktonic bacteria.  As in soil systems, pollutants are irreversibly bound
in aquatic sediments.  We must yet determine if these bound residues are bio-
logically recyclible or inert.  These are all factors that should be consid-
ered when designing studies or attempting to design standard tests for regula-
tory actions to determine fate in the aquatic environment.

     In addition to the physical characteristics of the aquatic environments,
the chemical characteristics of higher salinities and higher pH added to the
complexities of biodegradation studies in marine environments.  How these fac-
tors affect biodegradation is of primary importance to us at ERL-GB.  The fate
of a chemical in aquatic environments is an interaction of all these various
processes on the chemical.  None of the current methodologies are applicable
to assess all of these processes in the aquatic environment.  How other degra-
dative processes affect biodegradation is an important factor in assessing
fate in the environment.  For example, environmental parameters such as sali-
nity, pH, and other physical factors may alter the microbial community or af-
fect its innate biodegradation potential.  In the design of laboratory systems
to assess biodegradation potential, type of water, sediments, ratio of sedi-
ment to water volume are factors for consideration to properly design systems
to simulate environmental conditions.

     I hope some of the topics I have touched on here will help to orient the
discussion on the differences and similarities in biodegradation studies in
terrestrial and aquatic environments.  We have scheduled six task groups which
will meet outside the formal sessions to intensify discussion in the areas of:
1) biochemistry, 2)  transformations in aquatic environments, 3)  compartmentali-
zation in aquatic environments,  4)  microcosm development, 5) methodologies,
and 6)  extrapolation and persistence.


        Chairperson, David T. Gibson


                              Jean-Marc Bollag
                       Laboratory of Soil Microbiology
                           Department of Agronomy
                      The Pennsylvania State University
                          University Park, PA 16802

                The fate of xenobiotic compounds in the environment
           represents an area of intense contemporary interest, since
           disappearance, persistence, or partial transformation of
           such a compound determines its usefulness or its potential
           hazardous effect.  There may be biological, chemical and
           physical factors that influence the fate of a xenobiotic,
           but the least predictable transformation is usually caused
           by microorganisms.

                Mechanisms of microbial intereference with xenobiotics
           include the use of a compound as the source of both carbon
           and energy, cometabolic transformation, conjugation and
           polymerization reactions, or the mere accumulation of a
           chemical within a microbe.  It is difficult to predict
           which molecular change can be expected by microbes, since
           each group of organisms, even various strains of one
           genus, can alter a selected molecule differently.  Conse-
           quently, for numerous xenobiotics tens of metabolites can
           be formed.  However, a certain mode of biological trans-
           formation can be anticipated in a specific environment if
           metabolic reactions or pathways with various microorga-
           nisms are determined, their formed intermediates and pro-
           ducts identified, and the biochemical mechanisms estab-
           lished.  To illustrate this concept, a study performed
           with the insecticide carbaryl is presented in which the
           metabolic activity of various organisms was followed.

     The introduction into the environment of chemicals, both naturally occur-
ring and man-made xenobiotics, causes changes in the various ecosystems.
These can be beneficial or harmful.  For millions of years, nature adapted to
various natural organic products, but only in the last 30 years have numerous
man-made organic compounds been used in the environment.  Therefore, it is a
recent task to investigate how nature or the environment can cope with the


newly synthesized chemicals.  It is obvious that chemical and physical factors
influence the fate of xenobiotics, but most of the degradative activities are
initiated by microorganisms.  Furthermore, it is merely academic whether mi-
crobes are infallible in transforming a chemical, but it is an urgent environ-
mental problem to determine the time span of recalcitrance and the chemical
transformations that take place.  A long resistance time of a chemical can re-
sult in its accumulation to toxic levels for different organisms.  Further-
more, microbial metabolic processes can create new chemicals with undesirable
or harmful effects.  Metabolic changes of xenobiotic compounds in soil, fresh-
water, estuarine or marine environments appear to be caused primarily by bac-
teria, actinomycetes and fungi.  However, there are considerable differences
in the composition of the microbial populations in the various ecosystems and,
consequently, different transformation reactions can be expected.

     If a xenobiotic compound is exposed to a microbial species there are sev-
eral possibilities for its transformation or inactivation:  (a) the compound
can serve as a substrate for growth and energy;  (b) xenobiotic compounds can
undergo cometabolism, i.e., microbes transform it but cannot derive energy for
growth from it; (c) a xenobiotic molecule or an intermediate of it can be con-
jugated or polymerized with naturally occurring compounds; and (d) the xeno-
biotic is incorporated and accumulates within an organism.

     The complete microbial degradation of an organic molecule to its inor-
ganic components is desirable if one is interested in avoiding the persis-
tence of a potentially hazardous compound in the environment.   If a xeno-
biotic can be used in such a way, it is metabolized and fragmented to com-
pounds that can be channeled into known oxidative cycles; thus, the organism
can derive all the necessary energy.  The phenomenon of complete biodegrada-
tion can occur often in nature, and it can be anticipated that usually several
microbial species participate in the metabolic transformations.

     Cometabolism, the transformation of a chemical by a microbe without de-
riving energy to support its growth, has been observed and studied only in re-
cent years.  From an environmental viewpoint, the phenomenon of cometabolism
of xenobiotics has to be considered carefully, since this reaction generally
does not result in extensive degradation of a molecule.  Consequently, it is
difficult to foresee if such a product reduces or increases toxic effects in
the environment.

     Higher organisms use conjugation systems to convert xenobiotics into an
excretable form.  Such a mechanism is usually not of importance for microbial
metabolism, but microbes possess enzymes which cause such reactions.  Examples
can be found in soil or sediments where oxidative coupling enzymes can combine
xenobiotic compounds, such as halogenated phenols or anilines, with humus com-
pounds .

     In many instances it was shown that xenobiotic compounds are incorporated
into microorganisms by an active or passive accumulation mechanism.   Not only
live bacterial cells but also autoclaved ones show a similar uptake.  This ap-
pears to indicate that a natural metabolic factor is not always involved in
the accumulation process.  Since aquatic microorganisms and plankton in fresh-
water and marine environments are an important nutrient source for a broad


spectrum of aquatic filter-feeding organisms, their accumulation of xenobiotic
chemicals can constitute a hazardous link in the food chain to fish and higher
vertebrates.  Therefore, the findings of extensive biomagnification by these
organisms has to provoke considerable concern.

     In the study of drug metabolism, most compounds are metabolized in higher
organisms by non-specific enzymes that catalyze relatively few general reac-
tions.  Similar findings have also been made for microorganisms.  One can as-
sume that the important enzymatic reactions taking place have been described,
and, consequently, it appears possible to anticipate the biotransformation of
a certain chemical molecule.  Although the aforementioned is true, many trans-
formation mechanisms exist, and the tremendous variation of organisms with
different enzyme systems can result in formation of numerous different pro-
ducts.  Even a similar kind of mechanism can produce slightly different chemi-
cal products which themselves may require different transformation reactions.
Some products may be easily degradable while others may be quite resistant to
biological attack.

     From these general remarks it can be concluded that it is a very complex
problem to establish the metabolic fate of a certain xenobiotic substance.
The major metabolic pathways of a compound have to be elaborated, but one must
always be aware that specific circumstances can cause a deviation from such a
pathway.  Under various conditions or in different ecosystems a chemical can
be transformed differently and the resulting product can vary considerably.
However, knowledge of enzymatic reactions in the metabolism of chemicals, or
their identified intermediates, should contribute to understanding transfor-
mation possibilities.  Essentially, the biochemical processes have been in-
vestigated in model systems using isolated microbial cultures or enzyme sys-
tems.  This appears, with presently available techniques, to be the only
feasible approach, and is a prerequisite for establishing and predicting the
possibility of microbial xenobiotic transformations in a natural environment.

     To illustrate this concept, a study with the insecticide carbaryl (1-
naphthyl-N-methylcarbamate), in which the metabolic activity of various orga-
nisms was followed, was performed in our laboratory.  Carbaryl was selected as
a pesticide since it is the most used carbamate insecticide in the United
States, being registered for use in controlling more than 160 different insect
pests and on 85 different food and fiber crops.  Carbaryl was first synthe-
sized in 1953 and introduced as a commercial insecticide in 1958, but did not
receive widespread attention until it was recommended as a safe substitute for
DDT.  Although carbaryl is a relatively simple organic molecule, its metabo-
lism is very complex and there is still much to be learned about the bio-
chemistry of this compound (9).  Chemical degradation of carbaryl by hydroly-
sis occurs with relative ease (especially in alkaline media) and it was first
assumed that the predominant metabolic pathway of carbaryl is hydrolysis of
the carbamate ester linkage.

     It was possible to isolate several microorganisms capable of hydrolyzing
carbaryl to 1-naphthol (1).  In these studies it became evident that 1-naphthol
is a more toxic compound to microorganisms that the original pesticide.  This
observation was confirmed with other organisms and, therefore, we concentrated
our later studies on the metabolic fate of 1-naphthol.  The existence of a


non-hydrolytic pathway was first found with mammals, plants and insects  (8),
but studies with the soil fungus Gliocladium roseum also showed the formation
of 4-hydroxy-l-naphthyl- and 5-hydroxy-l-naphthyl-N-methylcarbamate (10) .  Hy-
droxylation of the side chain also occurred and yielded 1-naphthyl-N-hydroxy
methylcarbamate.  The ability to hydroxylate carbaryl at different positions
occurred with several other soil fungi, but the products differed qualita-
tively as well as quantitatively with the various fungi (2).   The metabolism
of xenobiotic compounds is often initiated by hydroxylation,  but these results
show that it is difficult to predict which exact molecular change, in this
case hydroxylation, can be expected by a specific microbe.  Each group of or-
ganisms, even various strains of one genus, can alter a certain molecule dif-
ferently.  For instance, Aspergillus terreus caused the formation of 1-naphthyl-
N-hydroxymethylcarbamate as the major metabolic product while A. flavus had a
stronger tendency to hydroxylate carbaryl in the ring position  (Table 1) .


                           1-Naphthyl         4-Hydroxy-         5-Hydroxy-
                           N-hydroxy-         1-naphthyl         1-naphthyl
       Fungus            methylcarbamate    methylcarbamate    methylcarbamate

Aspergillus flavus            4352*               126                228
Aspergillus fumigatus            000
Aspergillus terreus           4854                 59                262
Gliocladium roseum             829                254                408
Mucor racemosus                134               1275                834

Mucor sp.                       65                406                666
Penicillium sp.                266               3342               1003

  *Amounts of radioactivity  (d.p.m.) detected.
     It is obvious to assume that the subsequent metabolic transformation of
various hydroxylated products will differ considerably.  Further microbial al-
terations of the ring-hydroxylated carbaryl was not followed, but we investi-
gated the metabolic transformation of 1-naphthyl-N-hydroxymethyl carbamate by
Aspergillus terreus (11).   1-Naphthyl-carbamate was established as the next
intermediate and subsequently 1-naphthol was formed (Fig. 1).

     As previously outlined, 1-naphthol appears to be the major intermediate
in the degradation of carbaryl, and therefore we studied the bacterial and
fungal metabolism of this compound.  Several bacteria isolated from river
water were capable of degrading 1-naphthol  (3).  After 60 h of incubation with
1-naphthol-1- ^C, it was possible to trap 44% as Cll*02, and 22% was recovered
in the bacteria  (Table 2).  The release of C1''-labeled C02 clearly indicated
that complete biodegradation of the chemical had taken place.  In addition, it
was found that 15-20% of radioactivity from 1-naphthol remained in the growth


            IN  BACTERIA





                                                          9 ^
                                                        0-C-N -CH3

                   C AR B ARYL
                                                           0 H
                                                                            0 H
                                                                            i i
0- C- N-CH,
                                           0  H

                                                                            0 H

                                                                         0-C-N -H

                                  TRIMER -> TETRAMER  ->   HIGHER POLYMERS
Figure  1.   Metabolic transformation of the  insecticide carbaryl  (1-naphthyl-N-methylcarbamate)  by

            various microorganisms.

medium, apparently without further change.  This suggests that at least two
different pathways are involved in degradation of 1-naphthol by the investi-
gated bacteria.  The radioactivity remaining in the growth medium was parti-
tioned by ether extraction into an organic and an aqueous phase at a ratio of
35 to 65%, respectively.  The ether extract was analyzed by thin-layer chroma-
tography, but no attempt was made to characterize or identify the composition
of the aqueous phase.  After 48 h or incubation no further change of the
formed metabolites occurred, and the dominant product was isolated and identi-
fied as 4-hydroxy-l-tetralone  (Fig. 1).  This product suggests that one alter-
nate pathway involves hydroxylation of the naphthyl ring in the 4 position,
and conversion of an aromatic ring to an aliphatic cyclic compound.  The  exis-
tence of  such an apparent shunt pathway has been confirmed by Walker and  co-
workers  (13) .
                    CONTAINING 20 ppm OF 1-NAPHTHOL-l-1kC

                 Growth    Ether    Aqueous    Trapped       In         Total
                 medium    phase     phase       l4CO2     bacteria    recovery

 Bacterium         16.5%     5.5%     11.0%      43.5%      22.0%        82.0%

   control        100.0%    97.5%      3.0%        0.0%       —          100.5%
      Knowledge  related  to transformation of 1-naphthol by fungi is even more
 sparse  than  bacterial naphthol metabolism.  In experiments with various Rhi-
 zoctonia  species  it was  found that the fungi were able to transform 1-naphthol
 from an ether-extractable to a water-soluble product  (4).  It was also ob-
 served  that  the growth medium, after removal of the fungal cells, possessed
 the  ability  to  transform 1-naphthol, indicating activity of an extracellular
 enzyme.   Attempts to analyze the radio-labeled material in the aqueous phase
 gave indications  that the radioactivity was associated with a relatively high-
 molecular weight  compound.  Therefore, the mixture of a culture filtrate which
 had  been  incubated with  Cll+-labeled naphthol was analyzed on a column of Sepha-
 dex  G-75,  and most of the radioactivity was recovered in the first fraction
 which coincided with the distribution of protein.

      In a different experiment the protein fraction of the culture filtrate
 from Rhizoctonia  praticola was chromatographed on a column of Sephadex G-200
 and  fractions obtained that were able to transform 1-naphthol as a substrate
 (12).   The benzene extract of the reaction mixture from such an assay was sub-
 jected  to mass  spectrometric analysis, and major m/e peaks at 144, 286, 428,
 570,  and  712 were observed.  A hypothetical model presenting 1-naphthol poly-
 merization is shown in Figure 2.  The compound has a mass of 144 and the de-
 tected  peaks represent oligomers of the substrate.  In subsequent experiments
 it was  possible to separate the individual compounds  (2 dimers, 1 trimer and 1
 tetramer)  by thin-layer  chromatography and to establish that they are not
 fragment  ions from higher molecular weight polymers.  The extracellular enzyme
 responsible  for the polymerizing activity was isolated, and according to its


 characteristics it was identified as  a laccase  (6).
          OH                OH                  OH    OH
                                 —>   ^^
     M.W.  144
 M.W. 286
 M.W. 428
          Figure 2.
                         OH    OH
                         M.W. 570

Scheme for polymerization products  of  1-naphthol
    formed by a fungal laccase.
     In nature, biological oxidation and coupling of phenols are key reactions
that result in formation of products such as lignins, melanins,  tannins,  alka-
loids, and especially humus compounds (7).  Therefore, it is not surprising
that enzymes capable of these reactions also interact with xenobiotic phenols
and it can be assumed that they are incorporated into soil organic matter or
sediments in aquatic environments.

     The phenol oxidase isolated from the fungus Rhizoctonia praticola was
also capable of polymerizing other phenolic and naphtholic compounds (Table 3).
It could also be shown that the ring-hydroxylated carbaryl metabolites were
oxidatively coupled by the fungal enzyme  (5).   It appeared that  these com-
pounds were hydrolyzed to the free naphthol prior to the coupling reaction.
The mass spectra of these products indicated that the observed dimers or  tri-
mers had undergone further oxidation or dehydration corresponding to loss of
Ha or water (Table 4).

     Figure 1 summarizes the various metabolic transformations which were ob-
served in our laboratory with carbaryl.   The scheme does not include all  the
transformations of carbaryl by microorganisms, but it indicates  that the  meta-
bolic fate of a certain xenobiotic presents a very complex problem.  One  has
to establish all the possible transformation reactions of a chemical in order
to foresee the possible formation of potentially hazardous compounds in nature,

The isolation and identification of xenobiotic products from natural environ-
ments is a very difficult undertaking, but the knowledge acquired from  indi-
vidual metabolic studies can help immensely in this task.  Studies with iso-
lated microorganisms and enzymes are essential and can contribute basic infor-
mation for determining the fate of a xenobiotic compound in nature.
                         FROM Rhizoctonia praticola
     Phenolic compound
                    Enzymatic products as
                detected by mass spectrometry
2 , 6-Dimethoxyphenol
2 , 4-Dichlorophenol
Dimer ,
Dimer ,
Dimer ,
Dimer ,
Dimer ,
Dimer ,
trimer ,

trimer ,

trimer ,
trimer ,


tetramer, pentamer, hexamer
tetramer , pentamer
               Tetramer  Pentamer

 me thy1carbamate * *
    1,4-Naphthalene diol

    1,5-Naphthalene diol



472 (-2H2)
456 (-H2;H2O)
160    300 (-H20)
    *Isolated by TLC.

   **Hydrolysis to  free naphthol during oligomerization.

                               LITERATURE CITED

 1.   Bollag,  J.-M.,  and S.-Y. Liu.  1971.  Degradation of Sevin by soil micro-
      organisms.  Soil Biol. Biochem. 3:337-345.

 2.   Bollag, J.-M., and S.-Y. Liu.  1972.  Hydroxylations of carbaryl by soil
      fungi.  Nature  236:177-178.

 3.   Bollag, J.-M., E. J. Czaplicki, and R. D. Minard.  1975.  Bacterial meta-
      bolism of 1-naphthol.  J. Agr. Food Chem. 23:85-90.

 4.   Bollag, J.-M., R. D. Sjoblad, E. J. Czaplicki, and R. E. Hoeppel.   1976.
      Transformation  of 1-naphthol by the culture filtrate of Rhizoctonia
      praticola.  Soil Biol. Biochem. 8:7-11.

 5.   Bollag, J.-M., R. D. Sjoblad, and R. D. Minard.   1977.  Polymerization of
      phenolic intermediates of pesticides by a fungal enzyme.  Experientia

 6.   Bollag, J.-M., R. D. Sjoblad, and S.-Y. Liu.  1978.  Characterization of a
      polymerizing enzyme from the soil fungus Rhizoctonia praticola.   Submit-
      ted for publication.

 7.   Brown, B. R.  1967.  Biochemical aspects of oxidative coupling of phenols,
      p. 167-201.  In W. I. Taylor and A. R. Battersby (eds.), Oxidative coup-
      ling of phenols.  Marcel Dekker, Inc., New York.

 8.   Dorough, H. W.,  N. C. Leeling, and J. E. Casida.  1963.  Nonhydrolytic
      pathway in metabolism of N-methylcarbamate insecticides.  Science 140:

 9.   Kuhr, R. J., and H. W. Dorough.  1976.  Carbamate insecticides:   chemistry,
      biochemistry, and toxicology.  CRC Press, Inc., Cleveland, Ohio.

10.   Liu, S.-Y., and J.-M. Bollag.  1971.  Metabolism of carbaryl by a soil
      fungus.  J. Agr. Food Chem. 19:487-490.

11.   Liu, S.-Y., and J.-M. Bollag.  1971.  Carbaryl decomposition to 1-naphthyl
      carbamate by Aspergillus terreus.  Pestic. Biochem. Physiol. 1:366-372.

12.   Sjoblad, R. D.,  R. D. Minard, and J.-M. Bollag.   1976.  Polymerization of
      1-naphthol and related phenolic compounds by an extracellular fungal en-
      zyme.  Pestic.  Biochem. Physiol. 6:457-463.

13.   Walker, N., N. F. Janes, J. R. Spokes, and P. Van Berkum.  1975.  Degrada-
      tion of 1-naphthol by a soil Pseudomonad.  J. Appl. Bact. 39:281-286.

                        DEGRADATION  MECHANISMS

                              Peter J.  Chapman
          Department of Biochemistry and Department of Microbiology
                           University of Minnesota
                             St.  Paul,  MN 55108

                The role of fungi  and bacteria in accomplishing the
           biodegradation of organic compounds is stressed by con-
           sidering the variety of reactions which these microorga-
           nisms can employ to initiate  attack on different classes
           of organic molecules.   Reference is made to the value of
           this type of information, obtained by pure culture
           studies, in understanding the chemical events which may
           occur in natural environments,  yet at the same time serv-
           ing only to suggest possibilities rather than to predict
           specific fates.  Among  the various classes of compounds
           considered are linear,  branched and cyclic hydrocarbons,
           aromatic acids, and selected  heterocycles.  Mechanisms
           used for the enzymic modification and displacement of
           different substituents  such as  sulfonic acids,  nitro,
           alkoxyl, and halide groups are  also discussed.   Finally,
           mention is made of the  difficulties of assessing which
           of various degradative  reactions may occur with a given
           substrate by reference  to unexpected oxygen-incorporating
           reactions, alternative  degradative routes for the same
           compounds found in different  microbial groups,  and the
           influence which environments  can exert on the strategies
           available for microbial attack.
     Bacteria and fungi are the principal agents which accomplish the biodeg-
radation of organic compounds,  an important process  in the carbon cycle re-
turning carbon to the mineral state for its use in synthesis.   Our knowledge
of the mechanisms employed by these microorganisms for the dissimilation of
different organic compounds derives largely from studies  of pure cultures able
to grow at the expense of a selected substrate usually at high concentrations
and with other conditions optimized principally for  the convenience of the in-
vestigator.  Through such studies mechanisms of degradation have been defined
by the characterization of intermediates and enzymes which comprise the over-
all degradative pathways.  In many instances,  extensive studies of individual
enzymes have revealed the precise details of the mechanisms of the reactions


catalyzed.  In this account the overall strategies of microbial degradation as
revealed by the pathways employed will be emphasized.  It might be argued that
investigations which yield this information are conducted under conditions
which bear little relationship to conditions prevailing in the environment by
failure to recognize, for example, that biodegradation occurs naturally in mi-
crobial communities at low substrate concentrations.  However, it should be
pointed out that the objectives of much of this work are frequently biochemi-
cal and microbiological, not environmental.  Insofar as they reveal the bio-
chemical mechanisms by which different classes of organic compounds are de-
graded and identify the reactions characteristic of different groups of micro-
organisms, they provide a valuable basis of information from which to suggest
the fates of various environmental pollutants.  Only from knowledge of the
role of individual microbial species in degrading single organic substrates
can one begin to explain the complex relationships which are involved when mi-
crobial communities accomplish biodegradation of multiple substrates.

     Our understanding of the processes by which microorganisms acquire the
ability to degrade novel compounds depends in large part on a background of
information about mechanisms of biodegradation.  Acclimation, a term used to
describe the lag periods during which organisms in water treatment plants
gradually "learn" how to degrade some novel chemical, is a phenomenon familiar
to most civil engineers.  One explanation of this phenomenon is that particu-
lar organisms are evolving new degradative abilities.  Microortanisms able to
utilize organic compounds for growth do so by converting a portion of that com-
pound's structure into compounds which are central to the processes of inter-
mediary metabolism before synthesis of cell constituents can occur.  Suffi-
cient energy for synthesis is derived by complete oxidation of the remainder
to carbon dioxide and water.  For utilization of compounds such as sugars,
fatty acids, and amino acids, relatively short pathways of degradation are in-
volved since these compounds bear close structural similarity to intermediary
metabolites and are readily converted to such products by enzyme-catalyzed re-
actions of the type readily found in a standard biochemistry text.  Reactions
such as retroaldol cleavage, thiolytic cleavage of g-keto.acyl thio-esters and
oxidative decarboxylation of a-keto acids are employed to cleave the carbon-
carbon bonds of these substrates just as these reactions are characteristic of
the metabolism of the same classes of compounds normally present inside the
cell.  The structures of many other types of compounds such as the natural
products of plant biosynthesis are not immediately suitable substrates for re-
actions of this type and often require extensive structural modification be-
fore they can enter central metabolic schemes.  Much interesting biochemistry
has been discovered from studies of these so-called "peripheral pathways" by
which these types of compounds are degraded because appropriate alterations of
their structures may involve reactions very different from those of central
metabolism.  Microorganisms capable of elaborating the necessary enzymes for
such sequences are numerous and widespread as seen by their ready isolation
from soil and water by selective enrichment procedures.  Many such organisms
are versatile in this respect, having the ability to utilize many different
organic compounds for growth.  Others may be unable to derive carbon or energy
for growth from a given compound yet may modify its structure, for example, by

     A recent compilation of different microbial transformations of cyclic com-
pounds illustrates the wide array of reactions of which microorganisms are


capable (90).   Clearly such organisms have acquired the ability to act on na-
tural products and have had a considerable period of time to evolve the neces-
sary enzymes.   For those novel chemicals more recently introduced into the en-
vironment, and sometimes referred to as xenobiotics, microorganisms have had
much shorter periods of time to acquire degradative enzymes.  Nevertheless, it
is possible to isolate bacteria able to grow with such compounds as the herbi-
cide 2,4-dichlorophenoxyacetic acid, first introduced as an agricultural chemi-
cal in about 1945.  Evidently the requisite enzymes have been acquired by these
bacteria and,  according to current views  (71), enzymes possessing the appro-
priate mechanisms have been recruited from pre-existing degradative pathways.
Whether these enzymes were previously encoded by the genes of the evolved
strain or acquired by an exchange of genetic material their effective function-
ing in a new setting requires alterations in their specificity, and in the
regulatory mechanisms which control their synthesis.  As more is learned about
how these processes can occur, directed evolution in the laboratory of strains
better equipped to degrade specific pollutants becomes a realistic possibility.
Attempts to construct such strains will depend on an adequate knowledge of the
degradative pathways available in different microorganisms, of the specifici-
ties of participating enzymes, and of the mechanisms which regulate enzyme

     The following is an attempt to identify some of the distinctive features
associated with the microbial degradation of different classes of organic com-
pounds and to describe the general nature of certain types of reactions used
to effect their catabolism to common intermediary metabolites.
                         I.  ALIPHATIC HYDROCARBONS

     The most commonly encountered mechanism employed by bacteria, yeasts and
 fungi  for  initiating attack on n-alkanes which serve as growth substrates is
 terminal methyl hydroxylation.  For the simplest member of the series, meth-
 ane, cell-free preparations have been obtained from methylotrophic bacteria
 which  form methanol  (27,129).   For the longer chain hydrocarbons, two differ-
 ent hydroxylating  systems have been described, one from species of Pseudomonas
 which  involves three different proteins for activity (118).  The other from a
 species of Corynebacterium grown with n-octane differs in that cytochrome Pit 50
 appears to be involved in one of the two fractions required for activity  (15).
 The primary  alcohols which result from their action  (Fig. 1A) are oxidized via
 the corresponding  aldehydes to fatty acids which can undergo 3-oxidation as
 their  coenzyme A thioesters.  In some instances, diterminal oxidation can oc-
 cur to give  dicarboxylic acids, apparently by w-oxidation of fatty acids  (96).
 It has been  suggested that 1-alkenes may be intermediates in the process of
 hydroxylation and  the principal evidence for their participation is from an-
 aerobic experiments with a denitrifying organism, Pseudomonas aeruginosa  (26).
 A recent report has provided evidence that a Pseudomonas growing anaerobically
 with n-decane and  nitrate as electron acceptor forms 1-decene as an intermedi-
 ate en route to 1-decanol  (114).  There are conflicting opinions, however,
 about  the  role of  1-alkenes in the aerobic formation of primary alcohols.  They
 can be excluded as intermediates in the hydroxylation of n-alkenes by the hy-
 droxylating  systems from Pseudomonas and Corynebacterium since their action on
 1-alkenes  leads only to the formation of epoxides  (107, 84).  On the other
 hand,  evidence for 1-alkene formation from alkanes has also been reported by


other groups working with Nocardia species  (148), Candida rugosa  (80), and
with Candida tropicalis  (101) suggesting the possibility of an alternative,
possibly  anaerobic, route.

     Mechanisms  of n-alkane  oxidation other than by terminal  methyl  hydroxyla-
tion have been described.  A number  of organisms are known  to form methylke-
tones from  alkanes and  further  metabolism  occurs via a Baeyer-Villiger type
reaction  to give the  acetate ester of the  alcohol having one  carbon  less  than
the substrate  (105)  (Fig.  1A) .   The  properties  of an ester-forming monooxygen-
ase have  been described recently (15).  Hydrolysis  then furnishes the primary
alcohol for conversion  to  a  fatty acid and 3-oxidation. While formation  of
ketones is  frequently shown  as  proceeding  via the corresponding  secondary al-
cohols, these compounds could be formed readily by ketone reduction  rather
than by direct hydroxylation of the  subterminal methylene group.   In the
eucaryotic  yeasts formation  of  1,2-diols from 1-alkenes is  logically accounted
for by hydration of epoxide  intermediates  (139) .  Degradation of  the diols in
these and other  organisms  can proceed by oxidation to 2-hydroxy  acids, fol-
lowed by  oxidative decarboxylation to give fatty acids one  carbon shorter than
the parent  hydrocarbon  (92).
  Figure 1.  A.
Pathways of n-alkane degradation in bacteria and yeasts involv-
ing (i) terminal methyl hydroxylation, ii) subterminal oxidation
to form a methylketone and an acetate ester, iii) formation of
1-alkene and its conversion to a 1,2 diol.
Pristane catabolism by a Brevibacterium sp. to give mono- and
dicarboxylic acids.

     Both mono- and diterminal hydroxylation are found as initial reactions in
the degradation of branched alkanes such as 2-methylundecane and the diterpene-
derived, pristane (2,6,10,14-tetramethylpentadecane), by Brevibacterium ery-
throgenes (120) as shown by the series of mono- and dicarboxylic acids isolated
from culture medium.  Their structures also reveal that 3~°xidation has oc-
curred, removing alternately acetate and propionate residues (Fig. IB).  Mc-
Kenna and Kallio reached similar conclusions working with strains of Mycobac-
teria  (109).  Evidently the presence of alpha-methyl substituents in fatty
acids does not preclude the same general mechanism of g-oxidation established
for unsubstituted mono- and dicarboxylic acids (22, 73).  By contrast, the
presence of methyl substituents beta to the carboxyl group of acids such as
geranoic and citronellic acids does impede (3-oxidation as can be seen from a
consideration of the route employed by a species of Pseudomonas for degradation
of the monoterpenes, geraniol and citronellol (132) (Fig. 2A) .   A modification
of the 3-oxidation route involves an ATP-dependent carboxylation of the (3-
methyl a3-unsaturated fatty acid as its coenzyme A thioester before hydration.
This is, in effect, a vinylogous (3-carboxylation as distinct from $-carboxyla-
tion and is directly analogous to the reaction used in the bacterial degrada-
tion of leucine where (3-methylcrotonyl CoA is carboxylated (106) .  The reac-
tions are catalyzed by different enzymes, however  (70).  After hydration, a
substrate for a retroaldol cleavage is formed and its cleavage releases acetic
acid, in effect replacing the methyl substituent with a carbonyl group and
thereby forming a substrate for a cycle of conventional 3-oxidation until an-
other 3-methylsubstituted acyl CoA is formed  (131) .  A recent review (119)
summarizes these and related topics.

     With alkanes possessing a high degree of methyl substitution, as for ex-
ample where quaternary carbon atoms are present in a terminal neopentyl group,
available evidence suggests that incomplete degradation occurs.  For example,
when an Achromobacter grows with 2,2-dimethylheptane attack is  initiated at
the linear terminus and pivalic acid (2,2-dimethylpropionic acid) accumulates
(18).  Similar findings were reported by the same group for degradation of
tert-butylbenzene.  Here initial degradation of the aromatic ring leads to the
formation of pivalic acid.  These patterns of attack at other sites may repre-
sent the only available routes for the degradation of compounds containing the
neopentyl group possibly because the degree of steric hindrance imposed by the
presence of a cluster of three methyl groups prevents terminal  methyl hydroxy-
lation.  McKenna had shown earlier that organisms capable of growth with n-
hexadecane could not utilize alkanes possessing diterminal neopentyl groups
(108).   Somewhat different considerations apply to those alkanes possessing
gem-dimethyl substituents forming quaternary carbon atoms at internal regions
of the chain (e.g., 4,4-dimethylheptane), where it can be predicted that al-
though limited degradation can occur from one terminus, a point will be reached
where 3-oxidation is prevented by dimethyl substitution because a3~dehydrogena-
tion cannot occur.  Such is true for certain methyl-substituted fatty acids
(66).-  In this regard it is of interest that gem-dimethyl compounds such as
pantothenic acid (103) and 2,2-dimethylsuccinic acid (135) are readily biode-
gradable, suggesting that diterminal attack may be an obligatory process for
complete degradation of this class of alkanes.  The problems of biodegradation
of compounds in the environment containing branched alkane functions can best
be illustrated by reference to the present use of linear alkylbenzene sulfon-
ates (LAS)  detergents in place of the alkylbenzene sulfonates used prior to
1965.   This latter group, manufactured using tetrapropylene, contained a


significant proportion of compounds having extensive branching in their alkyl
sidechains which persisted in the environment causing widespread foam on natu-
ral bodies of"water.  Doubtless the presence in the alkyl sidechains of these
detergents of neopentyl and gem-dimethyl groups limited their biodegradation
for the reasons outlined above providing a rational explanation for their en-
vironmental persistence.  It must not be assumed, however, that all compounds
possessing this type of alkyl branching are similarly persistent.  This is
manifestly not so as can be shown by noting that the bicyclic monoterpene,
camphor, which possesses two quaternary carbon atoms, is a growth substrate
for a number of microorganisms  (63).  Care must be exercised, therefore, to
avoid drawing generalized, untenable conclusions about the resistance to bio-
degradation of a particular structure (77) without reference to the entire
molecule in which it is present and to the biodegradation mechanisms available
for that class of compounds.
Figure 2.  A. Pathway of geranoic acid catabolism by Pseudomonas citronellolis
              involving ATP-dependent carboxylation and cleavage of acetate.
           B. Squalene degradation by an Arthrobacter sp. showing how cleavage
              of internal double bonds can form geranylacetone.
     In situations where a hydrocarbon is both branched and possesses a number
of double bonds, as with the naturally occurring triterpene, sg^ialene, a dif-
ferent type of attack can occur as shown by studies with a strain of Arthrobac-
ter able to utilize this compound  (155).  In place of terminal hydroxylation
(or any other of the degradative routes for n-alkanes mentioned earlier) this
compound undergoes cleavage at its internal double bonds to generate two mole-
cules of geranyl-acetone as shown by isolation of this product from cultures
(Fig. 2B).  This cleavage is reminiscent of what is apparently a dioxygenase-


catalyzed cleavage of 3~car°tene to vitamin A aldehyde observed with rat liver
and intestinal enzymes (61, 112).  Other compounds recovered from the Arthro-
bacter cultures included isovaleric, geranoic, citronellic and {3,3-dimethyl-
acrylic acids.

     The latter conversions while still obscure appear relevant only to that
class of hydrocarbons which possesses internal olefinic groups, possibly lim-
ited, therefore, to those compounds formed by the pathways of terpene biosyn-
thesis.  There is evidence, however, that in some instances double bonds may
be introduced at internal positions in n-alkanes as shown by the formation of
a family of octadecene isomers by a species of Nocardia (1).

     To illustrate some of the alternative reaction mechanisms employed by bac-
teria for the conversion of alicyclic compounds to aliphatic structures refer-
ence will be made initially to such simple compounds as cyclohexane and cyclo-
hexane carboxylic acid.  The degradative mechanisms known to be used for such
compounds include the oxidative conversion of cyclic ketones to lactones fol-
lowed by hydrolysis, the formation of 1,3-diketones and $-ketoacyl coenzyme A
derivatives for hydrolytic or thiolytic attack, respectively, and the aromati-
zation of 6-membered alicyclic rings before ring cleavage by dioxygenases
characteristic of aromatic catabolism.  This topic has recently been reviewed
at length by Perry  (115) and for more complete discussion of this subject the
reader should consult this source.

     The isolation of a biotin-requiring Nocardia strain able to use both cy-
clohexane and methylcyclohexane was reported recently  (140) and represents one
of the first accounts of the complete degradation of an alicyclic hydrocarbon
by pure bacterial cultures.  The pathway proposed for its degradation possesses
features previously shown to apply in the bacterial dissimilation of cyclo-
hexanol and related alicyclic alcohols by Trudgill and colleagues  (39, 62).
This is not surprising since cyclohexanol is apparently the first formed in-
termediate.  Of particular note is the formation of cyclohexanone and its con-
version to a seven-membered lactone, e-caprolactone, by a monooxygenase, simi-
lar to that purified by Trudgill et al.  (38)  (Fig. 3).  After hydrolytic ring
opening of the lactone the product, e-hydroxycaproic acid, is converted by
successive dehydrogenase reactions to adipic acid, ultimately to enter the
pool of intermediary metabolites via the pathways used for dicarboxylic acid
catabolism  (22, 73).  Mono-oxygenase reactions leading to the formation of
lactones from cyclic ketones appear to be frequently encountered in this area
of metabolism as a means of accomplishing ring opening as can be seen from
their involvement in the catabolism of cyclopentanol  (62) , camphor  (12), and
in the transformations of fenchone  (23) , progesterone and 17-a-hydroxyproges-
terone  (51, 117), testosterone  (122) and eburicoic acid  (100) .  Such reactions
are the biological counterpart of a chemical reaction, the Baeyer-Villiger re-
action already referred to in the previous section in connection with methyl-
ketone catabolism.

     While pure cultures of cyclohexane-utilizing organisms obviously can be
isolated  (with due attention paid to growth-factor requirements), there is
evidence from both laboratory and field studies that degradation of cyclo-


hexane may  occur  more  widely through  the agency of  microbial  communities  also
provided with  with  other hydrocarbons.   Thus,  one organism utilizing a second
hydrocarbon provides the ability to hydroxylate cyclohexane to cyclohexanol,
the  further metabolism of which  is  assumed by  other organisms (8).
Figure  3.  Pathways of bacterial degradation of alicyclic compounds,  i) cyclo-
           hexane conversion via cyclohexanone and e-caprolactone to e-hydroxy-
           caproic acid, ii) cyclohexane carboxylic acid conversion to cyclo-
           hexanol-2-carboxylic acid and pimelic acid, probably as coenzyme A
           thioesters, iii) cyclohexane carboxylic acid aromatization to 4-
           hydroxybenzoic acid before ring cleavage of protocatechuic acid,
           iv) aromatization of ring A of 9a-hydroxy-3,17-androstadienedione
           by aldol cleavage of ring B.
     Not shown in Figure 3 is an alternative hydrolytic mechanism of alicyclic
ring cleavage which involves hydrolysis of a cyclic (3-diketone.  By a suitable
substitution reaction a hydroxyl group is introduced into the camphor molecule
by a soil diphtheroid  (now recognized as a strain of Mycobacterium rhodochrous)
to generate a carbonyl group beta to that already present, thereby permitting
hydrolysis of the six-membered ring system (24).  A 1,3-diketone is also impli-
cated, in the catabolism of myo-inositol by Klebsiella species formed apparently
by dehydrogenation and dehydration reactions allowing hydrolysis of the six-
membered ring to give 2-deoxy-5-keto-D-gluconic acid  (2) before phosphoryla-
tion and aldolase cleavage (3).

     For cyclohexane carboxylic acid degradation a related mechanism is em-
ployed by a number of bacteria which involves a hydrolytic (or possibly thio-
lytic) cleavage of a $-keto acyl derivative of coenzyme A.  This and the

 reactions leading to this point are formally those of 3-oxidation where the
 cyclic structure can be viewed as an 8-alkyl substituted fatty acid (Fig. 3).
 This initial formation of a coenzyme A thioester allows a$-dehydrogenation,
 hydration and dehydrogenation of the resultant cyclohexanol-2-carboxylic acid
 derivative to the corresponding ketone.   Ring opening then yields pimelic acid
 as its coenzyme A derivative, for further degradation by 3-oxidation or decar-
 boxylation gives cyclohexanone and adipic acid by the sequence outlined ear-
 lier (126).   Work in this laboratory (Chatterjee and Chapman,  unpublished re-
 sults)  suggests that it is the trans form of cyclohexanol-2-carboxylic acid
 which is involved in this pathway in a strain of Mycobacterium rhodochrous.
 It should be pointed out that this sequence of reactions may represent a more
 general one for the degradation of the related cyclopentane- and cycloheptane-
 carboxylic acids since these are usually found to serve as carbon sources
 (either immediately or after a spontaneous mutation)  by organisms utilizing
 cyclohexane-carboxylic acid by the above route (Chapman,  unpublished observa-
 tions; 142).  Cyclobutane- and cyclopropane-carboxylic acid are not utilized,
 however; the bacterial catabolism of the latter compound is a topic to be dis-
 cussed in another communication in this  workshop.

      Another route for cyclohexane carboxylic acid degradation studied by
 three research groups, those of Blakley  (10),  Kaneda (85)  and Trudgill (142),
 illustrates  how aromatization of a six-membered alicyclic ring can occur
 (Fig. 3).  This route therefore cannot be a useful one for degradation of
 five- or seven-membered ring-containing  carboxylic acids.   Initial trans hy-
 droxylation  at Ci> followed by dehydrogenation yields cyclohexanone-4-carboxylic
 acid and this can be converted to 4-hydroxybenzoate by dehydrogenation reac-
 tions.   Thereafter,  the pathway is that  characterized for bacteria utilizing
 4-hydroxybenzoate (see next section),  namely 3-hydroxylation to protocatechuic
 acid and either ortho cleavage by the organisms studied by Blakley and Kaneda
 or meta-cleavage in that studied by Trudgill and in one organism described by
 Smith and Callely (134).

      The last scheme in Figure 3 shows an aromatization step which occurs when
 9a-hydroxyandrostadienedione,  an intermediate in steroid degradation,  is acted
 on by strains of Nocardia  and Pseudomonas.   In fact,  the aromatization occurs
 as a result  of a cleavage  of the alicyclic ring B accomplished by another type
 of ring opening mechanism,  an aldolase reaction,  the substrate in this case
 being a vinyl-interrupted  aldol  (37).  The phenol  so formed is subsequently
 hydroxylated to a 3,4-diphenol before  its ring is  cleaved by a dioxygenase
 (59) .
                          III.  BFJSIZENOID COMPOUNDS

     In the previous section it was noted that microbial degradation of cer-
tain six-membered alicyclic rings may occur by conversion to benzenoid ring
systems.  One of the most distinctive features of the reaction sequences used
by aerobic microorganisms for catabolism of benzenoid compounds is the utili-
zation Of molecular oxygen as a cosubstrate in the cleavage of such rings.
For cleavage to occur generally two hydroxyl groups must be present in either
an ortho or a para substitution pattern and for all examples studied this far
both atoms of molecular oxygen undergo incorporation into product (68) .  Such
reactions are catalyzed by enzymes known as dioxygenases.  Examples of para-


substituted diphenols known to be cleaved by enzymes of  this  type  include hy-
droquinol  (99) , gentisic  (29) and homogentisic acids  (21).  The sequences of
reactions which are initiated by such cleavages, while often  mechanistically
similar, are generally accomplished by specific enzymes.  For ortho-substituted
diphenols two types of ring cleavage dioxygenases are known.  The  first group,
so-called ortho or intra-diol-cleaving dioxygenases, catalyze cleavage of the
bond between carbon atoms carrying the hydroxyl substituents.  For catechol and
protocatechuic acid (3,4-dihydroxybenzoic acid) ortho cleavage reactions ini-
tiate parallel reaction sequences which converge on a common  intermediate, $-
ketoadipate enol lactone, and both routes involve lactonization and isomeriza-
tion steps (Fig. 4).  An alternative type of cleavage of 1,2-diphenols  (often
referred to as meta- or extradiol-cleavage, terms which  seldom define the pre-
cise bond cleaved) is that encountered when the bond adjacent to only one of
the hydroxyl groups is ruptured.  A wide variety of dioxygenase enzymes cata-
lyzing such reactions has been described and includes those which  have as sub-
strates, catechol, 2,3-dihydroxybenzoic acid, 3,4-dihydroxybenzoic acid, 3,4-
dihydroxyphenylacetic acid, 2,3-dihydroxy-3-phenyl propionic  acid, 1,2-dihy-
droxynaphthalene and others  (19).  For catechol and its  3- and 4-methyl homo-
logues the reaction sequences initiated by these enzymes  are  very  similar and
may even be catalyzed by single  enzymes at least at certain steps  (6, 127).
The similarity of these reactions prompted an early recognition of their func-
tion in terms of a general reaction sequence  (32).  It can be seen from Figure
5 that generation of 2-keto-4-alkenoic acids either by hydrolysis  of ring fis-
sion products or by dehydrogenation and decarboxylation  of the aldehyde-
containing products, allows generation of aldols by hydration and  their subse-
quent cleavage by aldolase reactions.  It does not appear to  be widely appre-
ciated that the general mechanistic features of both the  dehydrogenase and
hydrolase routes of this scheme  are also found to apply  in a  number of other
reaction sequences initiated by meta-cleaving dioxygenases as for  example in
the bacterial catabolism of homoprotocatechuic acid (136) and of biphenyl (17).


Figure 4.  The convergent pathways of ortho-fission of catechol  and protocate
           chuic acid leading to 3-ketoadipate  formation.


                         OXIDATIVE ROUTE                 Keto--  (CHj)
                                           CHJ/COOH    isomerase /COOH
                                            *<   COOH 	> f^
                       OH         X^^OH
                          Catechol           --^_^^  Hydrolase   ICHj)
                        2,3-oxygenase           ^^V-\     II   COOH

                                              ^^   H  1
                              HYDROLYTIC ROUTE (CH3> HCOOH     ^^0
                                                   2-Keto- 4-alkenoic acid

                             CH3      COOH    Aldo|ase      CH3   COOH
                             CHO +    C = 0   "           ^C^1 /C*o

                                     •CH3            H      2
Figure 5.  Oxidative and hydrolytic  routes  which follow meta-fission of cate-
           chol.  The methyl groups  in  parentheses indicate the principal
           routes used for  3-methyl- and 4-methylcatechol degradation.

     In order to generate suitable substrates for these ring-cleaving dioxy-
genases, it is frequently necessary  for organisms to alter or remove certain
substituents and to insert  hydroxyl  groups  in appropriate positions.  When
these are introduced singly, as usually occurs when an aromatic compound al-
ready possesses at least one hydroxyl group,  the oxygen for such reactions
again derives from molecular oxygen  and reactions catalyzed by monooxygenases
bring about the introduction of one  atom of oxygen into substrate with the
other atom undergoing reduction to water.  Many enzymes of this type are flavo-
proteins or require a flavin coenzyme (46).  For aromatic compounds not pos-
sessing phenolic hydroxyl substituents,  hydroxyl groups may be introduced
singly, as for example the  conversion of L-phenylalanine to L-tyrosine  (64) or,
as appears more usually the  case for  aromatic hydrocarbons, two hydroxyl sub-
stituents can be introduced by a  dioxygenase-initiated mechanism  (Fig. 6)
which involves the intermediacy of cis-dihydrodiols (56).  Reactions of this
latter type have been described in different bacteria which utilize the aro-
matic hydrocarbons, benzene, toluene, ethylbenzene, biphenyl and naphthalene,
for growth.  A similar dioxygenation of benzoic across the Ci~C2 bond has been
shown to operate when Alcaligenes eutrophus (125) and other bacteria  (124)
utilize this substrate for  growth; similar  dioxygenation mechanisms are thought
to function in the degradation of benzene sulfonic acid and various of its
alkyl-substituted homologues  (14), naphthalene-2-sulfonic acid  (93), and ani-
line  (4).  By this means dioxygen addition  can generate unstable intermediates
which lose the sulfonic acid substituent as sulfite, the aromatic amine group
as ammonia or facilitate loss of  carboxyl as carbon dioxide.

     With the knowledge that bacteria usually degrade aromatic hydrocarbons by
introducing dioxygen to form 1,2-diphenols  and that such diphenols are usually
degraded by sequences of reactions of the meta fission category referred to


earlier, one possesses a framework of information within which a limited num-
ber of reasonable predictions can be made about pathways of degradation.  A
number of examples are available to illustrate this.  The formation of benzoic,
phenylacetic, atropic and hydratropic acids as intermediates in the degradation
of biphenyl  (17), diphenylmethane  (47), 1,1-diphenylethylene (48), and 1,1-
diphenylethane  (49) , respectively, are compounds all of which could have been
predicted as logical intermediates on the basis of previous information.  It
should be stressed, however.- that the predictive value of this information
lies in its use in limiting the number of working hypotheses for investiga-
tional purposes.  The existence of systems which constitute exceptions to the
above generalizations points to the need for exercising caution in making dog-
matic predictions about how degradation will proceed.  For example, the pres-
ence of different functional groups in a given compound provides alternative
sites for microbial attack, each of which may be characteristic of certain
groups of organisms.
Figure 6.  The conversion of toluene  to  3-methyl-catechol by addition of di-
           oxygen  forming an intermediate cis-l,2-dihydrodiol compared with
           its conversion to benzoic  acid by methyl group oxidation.

     A number of aromatic hydrocarbons contain alkyl substituents and, as dis-
cussed earlier, when  such sidechains  are extended, linear structures, it is
generally  the sidechain which  is  the  preferred site of initial attack by No-
cardia species, thereby forming as  intermediates a series of phenylalkanoic
acids  (35).  Compounds with short sidechains may either be attacked in a simi-
lar manner or may  undergo ring dioxygenation to give 3-alkyl substituted cate-
chols, depending on the microorganism chosen for study  (5).  Even alkyl side-
chains as  short as one carbon  atom, as for example in toluene, m- and p-xylene,
may be the preferred  site of oxidative attack to give benzoic acid and its
methyl homologues  (Fig. 6).  In pseudomonads, degradation of toluene, m- and
p-xylene by  this route is a property  frequently dependent on the presence of a
degradative  plasmid designated TOL  or XYL depending on its conjugative proper-
ties  (52,  153).  In another example,  Kiyohara  (91) has shown that phenanthrene
is initially attacked by a 1,2-dioxygenation as anticipated but a subsequently
formed intermediate,  l-hydroxy-2-naphthoic acid, is itself a substrate for a
ring cleavage dioxygenase rather  than the anticipated 1,2-dihydroxynaphthalene.
Another aromatic compound, 3-hydroxyanthranilic acid, is cleaved by a mamma-
lian dioxygenase despite the fact that it possesses only one hydroxyl group

(Ill).   Alternative meta cleavage pathways of alkyl-substituted catechols also
have been described.  Horvath  (75) reported that extracts of a benzoate-grown
Achromobacter species catalyzed a novel 1,6-cleavage of 4-methylcatechol and
other alkyl-substituted catechols.  Such reactions imply the existence of meta
fission pathways different from those outlined in the earlier scheme  (Fig. 5),
but at this time no further details of this pathway are available.

     For more complete accounts of the topic of aromatic catabolism by bac-
teria the reader is referred to a number of extensive reviews  (19, 31, 57).
                         IV.  HETEROCYCLIC COMPOUNDS


     By comparison with the extensive body of information available on the mi-
crobial degradation of aromatic compounds, comparatively few studies have been
reported which deal with different categories of heterocyclic compounds.  From
those which are available, however, certain mechanistic features can be dis-
cerned which appear to be characteristic of their degradation.

     Studies by Trudgill and his associates of the bacterial catabolism of un-
saturated cyclic compounds containing an oxygen atom, typified by such com-
pounds as furan-2-carboxylic acid  (146), have uncovered a number of general
degradative features (Fig. 7).   For the named compound its initial conversion
to a coenzyme A thioester is required before its hydroxylation to give the enol
tautomer-of a lactone.  Ring opening of the latter, presumably in the keto
form, by hydrolysis yields the coenzeyme A derivative of a-ketoglutaric acid
 (initially formed as its enol).  It should be noted that these reactions are
in a sense the complement of those described earlier for such compounds as
cyclohexanone and cyclopentanone, i.e., with oxygen already present in the
ring the strategy is to make a lactone by introduction of a hydroxyl group.
In the earlier examples, the strategy was to make a lactone from a cyclic ke-
tone by insertion of oxygen into the ring.  The enzyme system responsible for
introducing the hydroxyl group uses water, not oxygen, and can be visualized
as catalyzing water addition followed by dehydrogenation.  Consistent with
this view are the findings that the reaction requires a suitable electron ac-
ceptor and proceeds in the absence of oxygen.  An essentially similar sequence
of reactions is employed by organisms growing with thiophene 2-carboxylate as
carbon source  (30)  (Fig. 7).  Here again hydroxylation of the ring is a reac-
tion using the oxygen of water.  Reactions of this type are also encountered
in organisms able to hydroxylate certain pyridine ring systems (see next sec-
tion) .  When the heterocyclic ring is part of a more complex multi-ring sys-
tem, as for example in dibenzothiophen, then degradation of the other ring
systems may represent the first if not the only type of attack.  The compounds
isolated by Laborde and Gibson (97), when a biphenyl-grown Beijerinckia was
incubated with dibenzothiophene, include several of those reported by others
studying organisms growing at the expense of dibenzothiophene  (94).  As shown
in Figure 7, the most extensive alterations are those in the benzene ring
which is dioxygenated and cleaved by the reactions typical of a meta fission
pathway (Fig. 5).  Formation of the sulfoxide is the only indication of a mi-
crobial attack on the thiophen ring.


                                                COSCoA   HOOC  O^COSCoA
Figure 7.  Bacterial degradation of oxygen- and sulfur-containing heterocyclic
           compounds, i) furoic acid conversion to the coenzyme A derivative
           of ot-ketoglutarate by its hydroxylation at Cs.  Shown in parenthe-
           ses is the position of the  sulfur  in thiophene-2-carboxylic acid
           and its metabolites, ii) pathways  of dibenzothiophen oxidation by a
           Beijerinckia sp. showing aromatic  ring cleavage and sulfur oxida-
     Oxygen-containing rings, which are part of more complex multi-ring sys-
tems, may be the initial site of attack as is seen in the degradation of the
3-hydroxyflavone, quercetin, by a strain of Aspergillus flavus and other fungi
(95)  (Fig., 8) .  The formation of carbon monoxide and the depside shown are con-
sistent with a dioxygenase attack on the pyrone-ring, either via the indicated
hydroperoxide or by concerted cyclic peroxide formation.  Photosensitized oxy-
genation of quercetin also forms carbon monoxide apparently via the hydroperox-
ide route (104).  By contrast, the same 3-hydroxyflavone is degraded by Pseudo-
monas putida by initial hydroxylation of the adjacent aromatic ring at Ce
(Fig. 9).  The resultant product can now serve as a substrate for a dioxygenase
which effects incorporation of an oxygen function in the pyrone ring and which
is then believed to undergo a dioxygenase cleavage across C2-Cs (130).  The
related taxifolin, which is an intermediate in the degradation of (+)-catechin
by a Pseudomonas species, is similarly hydroxylated at Cs to give 2,3-dihydro-
gossypetin (82), a substrate for a ring cleavage dioxygenase; however, the
structures of later intermediates  (81) suggest that the absence of a C2,Ca
double bond in this series of compounds requires that a degradative sequence
differing from that suggested by Schulz et al. (130) is involved.


     For saturated ring systems containing nitrogen the most usually encoun-
tered cleavage mechanism is hydrolysis of the cyclic imine formed by dehydro-
genation of the parent compound.  The bacterial conversion of L-proline to
glutamic semialdehyde (50)  and of piperidine-2-carboxylic acid to Ct-aminoadipic

        OH  0
       8.  The alternative routes of quercetin degradation by Aspergillus and
           Pseudomonas sp.
                                               COOH   HC02H
                     H  H
                        CH2-C-COOH   Q2
Figure 9.  Examples of bacterial degradation of nitrogen-containing hetero-
           cycles.  i) Nicotinic acid conversion to maleic acid by Pseudomonas
           fluorescens via hydroxylation at Ce.  ii) Pyridine degradation by a
           Bacillus sp. involving 1,4-reduction and ring cleavage without
           prior hydroxylation.  iii) Oxygenase cleavage of the pyrrole ring
           of tryptophan.

semialdehyde  (7) are examples of  this mechanism.  For unsaturated heterocyclic
acids such as the pyridine carboxylic acids, nicotinic and picolinic acids,
the initial step in their microbial degradation usually  involves hydroxylation
of the ring in  the 6-position with the  introduced oxygen function derived  from
water, not oxygen, in reactions analogous  to those  found for  furan-2-carboxy-
late catabolism.  This initial reaction is found in organisms which degrade
nicotinic acid  both aerobically and anaerobically from such genera as Pseudo-
raonas, Arthrobacter and Clostridium.  The  routes taken after  this reaction may
vary with the organism and growth conditions.  The  pathway of nicotinic acid
degradation elucidated for Ps. fluorescens (9,78) is shown  (Fig. 9).  It can
be seen that a  second hydroxyl group is introduced, this time by a monooxygen-
ase mechanism displacing the carboxyl before ring cleavage yields an amide de-
rivative of maleic acid and ultimately  fumaric acid.  A  number of organisms
able to utilize nicotine employ a similar  water-dependent hydroxylation mecha-
nism to form 6-hydroxynicotine  (72).  Once substituted with one hydroxyl group,
however, pyridine compounds generally undergo further hydroxylation via mono-
oxygenase-type  mechanisms to provide suitably substituted substrates for ring
cleavage  (46).  With pyridine itself  (150)  and also of an N-methylpyridinium
derivative, 4-carboxy-N-methylpyridine  hydrochloride, formed photochemically
from Paraquat,  no hydroxylated intermediates have been detected.  Instead, an
initial 1,4-reduction reminiscent of pyridine nucleotide reduction appears to
be necessary, yielding 1,4-dihydro-pyridine derivatives.  Evidence supports a
cleavage mechanism by direct dioxygenation of one of the double bonds.  In
Figure 9 is shown the pathway proposed  for pyridine degradation by a species
of Bacillus.

     There is a marked similarity between  this type of ring opening mechanism
and that catalyzed by the dioxygenase,  L-tryptophan pyrrolase (Fig. 9).  Later
steps in pyridine breakdown by this organism yield  ammonia, formate and suc-
cinate.  A different type of ring opening  has been  described for indole degra-
dation in which a Gram-positive coccus  appears to form 2,3-dihydroxy indole
before its intradiol cleavage.  Evidence to suggest the  participation of a di-
hydrodiol intermediate is provided by the  indentification of 2-keto-3-hydroxy-
skatole from skatole, the 3-methyl-substituted homologue of indole (53).

     Another way of evaluating the degradative reactions which an organic com-
pound may undergo is to consider the different functional groups in the com-
pound in relation to what is already known of microbial action toward such
groups.  Selected functional groups will be considered here with a number of
examples chosen to illustrate their degradation.


     The presence of an O-alkyl ether derivative of a phenyl in a given com-
pound represents one site where microbial action may be initiated.  From
studies with a number of O-methyl ethers of aromatic compounds, many having
structures related to lignin monomers, it is now recognized that aerobic mi-
croorganisms can effect demethylation by monooxygenation to form what is prob-
ably the unstable hemiacetal of the parent phenol and formaldehyde.  Such an
intermediate would then yield spontaneously the phenolic and formaldehyde pro-
ducts observed (128,144) (Fig. 10).

                                   HO  OCH2COOH
             CH?OH  „
             I  2    0
Figure 10.  The bacterial cleavage of phenolic ethers.  i)  The conversion of
            p-anisic acid to formaldehyde and p-hydroxybenzoic acid by forma-
            tion of the intermediate hemiacetal.   ii)  Formation of glyoxylate
            from p-chlorophenoxyacetic acid by a monooxygenative formation of
            a hemiacetal.  iii)  Formation of glycollic acid and p-benzoquinone
            from p-hydroxyphenoxyacetic acid via the intermediate hemiketal.

     Such reactions also explain the cleavage of the ether sidechain of phen-
oxyacetic acids as with the degradation of chlorinated herbicides such as 4-
chlorophenoxyacetic (43) and 4-chloro-2-methylphenoxyacetic acids (55) .  In
this instance, glyoxylate is formed together with the parent phenol (Fig. 10).
Fortuitous monooxygenation of 4-hydroxyphenoxyacetic acid at C-l by an enzyme
from Pseudomonas acidovorans, which has 4-hydroxyphenylacetic acid as its nor-
mal substrate, can also accomplish cleavage of the ether bond to yield glycol-
lic acid and 1,4-benzoquinone (67)  (Fig. 10).  A number of examples have been
described, however, which show that aromatic O-methyl substituents are not al-
ways subject to this type of attack but may keep this structure intact until
aliphatic products are formed where they may now be present as methylesters
(as with syringate and 3-0-methylgallate degradation by Pseudomonas putida
[141]) or as ethers of enols (as is the case for degradation of methoxyl-
substituted gentisates and their precursors)  (102; Chapman, Abstr. Annu. Meet.
Am. Soc. Microbiol. 1977, Q90,  p. 276).  Hydrolysis and hydration reactions
can then liberate the methyl moiety as methanol.

     Studies of the microbial degradation of strictly aliphatic ethers have
not been extensive; the mechanisms for cleavage of such ether bonds are not
yet clear.  In one study, however, investigation of the microbial degradation
of carboxymethoxy-succinate (CMOS), a potential replacement of phosphate as a
detergent builder, has revealed a mechanism which is formally a (3-trans-

elimination of glycollic acid  forming  fumaric acid  (116)  (Fig.  11).  Other  3-
eliminations of this type are  known  for the microbial depolymerization of $-
1,4-polymers of uronic acids,  such as  alginic acid  (123).  Microbial degrada-
tion of a related compound, carboxymethoxymalonic acid  (Builder M) occurs ap-
parently via a reductive type  of cleavage  to yield  glycollic and malonic acids
(Gledhill, Abstr. Annu. Meet.  Am. Soc. Microbiol. 1976, Q65, p. 201; personal
communication).  Bacteria capable of utilizing ethylene glycol  and its various
polymers have been described  (83,87),  but  there  is  presently no clear indica-
tion as to how the ether bonds of the  polymers are  cleaved.

     ~°^I\OH HO/ °-
J||).\  ,/  °\°H      /   "\
   >  °v(\OH HO/   +   \OH HO/
        YL^       Y-Y
Figure 11.  Cleavage of aliphatic ethers by a ^-elimination mechanism.
            i) Elimination of glycollic acid from carboxymethoxysuccinate
            (CMOS).  ii) Depolymerization of alginic acid by a 3-eliminase.

     In the case of thio-ethers, microorganisms are known which oxidize the
sulfur of such functional groups to the sulfoxide.  This is found both for
methyl thio-ethers  (98) as well as with compounds containing sulfur in hetero-
cyclic rings such as d-biotin (156) and the earlier mentioned dibenzothiophen
 (97) .


     The subject of microbial dehalogenation has been dealt with in recent re-
views (20,60).  Since many of the chemicals, which persist in the environment
and which may undergo biomagnification to toxic levels, are organohalides, a
general view has developed that all halogen-containing compounds are of indus-
trial origin and are potentially harmful.  Currently at least three hundred
different halogen-containing compounds of naturall occurrence are known (133)
and many of these are of marine origin.  Admittedly this represents only a
small fraction of the total range of naturally synthesized organic compounds.
Nonetheless microorganisms can be isolated which modify or completely degrade
both natural and synthetic organohalides.  Table 1 shows some of the types of
enzyme reactions known by which organohalides can lose halide.  For additional
details the reader is referred to the excellent review by Goldman  (60).  Reac-
tions such as these have been studied usually in bacteria capable of growth
with a particular halogenated substrate.  Other reactions are known, however,

in which dehalogenation occurs fortuitously because halogenated compounds  can
serve as substrates for enzymes normally acting on non-halogenated  substrates
and halide is labilized as a consequence of the reaction mechanism.  One ex-
ample of microbial utilization of an organohalide will be discussed to  show
how dehydrohalogenation can account for halide mineralization.  When Pseudo-
monas and Arthrobacter strains degrade 2,4-dichlorophenoxyacetic  acid,  3,5-
dichlorocatechol is an intermediate which undergoes an ortho-dioxygenative
ring cleavage (Fig. 12).  Lactonization then yields the doubly unsaturated
lactone whose formation is most reasonably formulated as proceeding via the
lactone in square brackets.  Loss of the elements of HCl probably occurs spon-
taneously  (44,143).  The enol lactone is then hydrolyzed to a chlorine-substi-
tuted maleylacetic acid.  Recent work from these laboratories  (Keenan and
Chapman, unpublished observations) has shown that the next step is  a reductive
one requiring NADH2, but interestingly, two moles of reduced pyridine nucleo-
tide are consumed per mole of chloromaleylacetic acid; the product  is 3-keto-
adipate and chloride is formed quantitatively.  Since a similar action  is  ob-
served by maleylacetate reductase, an NADH2-dependent enzyme reducing maleyl-
acetate to 3-ketoadipate which is found in bacteria degrading resorcinol  (25),
it appears that such an enzyme may have been recruited for use in dehalogena-
tion.  A possible explanation for the experimental observations is  suggested
in Figure  12 where the first formed reduction product is shown as $-eliminating
HCl to generate the related maleylacetate which can serve as substrate  in  a
second reduction to give 3-ketoadipate.
                  I  Substitution   a) Hydrolytic
                                    b) Neighboring group attack
                  II  Elimination    Dehydrohalogenation
                 III  Reduction      Reductive Dehalogenation
                          Cl         CHO  Cl
                             .OCH..COOH  /  ^^vOH
                             COOH  cr
  a- r.-*^ >
NADH? fXl''
Figure  12.  Pathway of 2,4-dichlorophenoxyacetic acid degradation  showing how
            lactone formation can eliminate chloride and how  reduction can
            lead to loss of second chloride substituent.


     A number of industrial chemicals contain nitro- and sulfonyl-functional
groups.  Reactions for the removal of these groups have been described in mi-
croorganisms utilizing compounds containing these functional groups where it
appears that these groups must be removed in early steps before the remainder
of the molecule can be utilized.

     Recent work with organisms able to assimilate nitroalkanes has revealed
the presence of a dioxygenase which converts two molecules of 2-nitropropane
to acetone and nitrite (89).  This mechanism appears to be one specific for
nitroalkanes and is distinct from the oxidase type of attack observed when D-
amino acid oxidase acts on nitromethane or nitroethane  (121) or the release of
nitrite effected by the action of glutathione S-transferase (65).  Many aro-
matic nitro compounds can undergo microbiological reduction to amino and hy-
droxylamino products, but reduction does not appear to be the principal route
by which aromatic nitro compounds such as p-nitrobenzoic acid are degraded by
Nocardia species (16).  The first reaction is removal of the nitro group
usually as nitrite and its replacement by a hydroxyl group  (Fig. 13A).  Whether
this occurs by a hydrolytic or, as appears more likely, an oxygenase-type re-
action is not established because of difficulties in obtaining active cell-
free enzyme preparations.  An oxygen-dependent enzyme system capable of con-
verting p-nitrophenol to hydroquinone and nitrite has recently been obtained
in Dr. D. T. Gibson's laboratory (Spain and Gibson, personal communication),
suggesting that oxygenative removal of aromatic nitro groups is likely  to be
encountered elsewhere.

     The linear alkyl benzene sulfonate (LAS) detergents represent one  impor-
tant class of aryl sulfonates used on a large scale.  Studies of their  micro-
bial dissimilation have tended to concentrate largely on the degradation of
the alkyl sidechains since it is this process which rapidly alters the  deter-
gent property of these molecules.  Loss of the sulfonyl group evidently takes
place as can be shown by formation of inorganic sulfur-containing products and
disappearance of the UV-absorbing aromatic ring in pure microbial cultures and
in microbial communities.  One way in which this can occur is evidently by the
displacement of the sulfonic acid group by a single hydroxyl group as found in
the degradation of alkylbenzene sulfonates by a species of Bacillus (152)
(Fig. 13B).  It is not yet clear whether this is a monooxygenase-type reaction
as appears likely.  Fungi apparently can desulfonate sulfophenylalkanoic acids
in a reductive manner (151).  An alternative mechanism for sulfonic acid group
removal has been identified in studies on the degradation of benzene-,  toluene-
and naphthalene-sulfonic acids (14,93).  Here, a dioxygenase mechanism  appears
to be involved, similar to that noted earlier for the dioxygenation of  benzoic
acid.  Addition of dioxygen to form a 1,2-diol where the sulfonic acid  group
is also attached to one of the oxygenated carbon atoms  (Fig. 13C)  constitutes
a feasible mechanism for the observed loss of sulfite and formation of  a 1,2-
diphenol.  Unlike the pathway of benzoic acid degradation referred to,  no de-
hydrogenase step is required here;  spontaneous elimination accomplishes aro-
ma tization directly.


Figure 13.  Elimination of nitro and sulfonic acid substituents.  i) Displace-
            ment of a nitro group from p-nitrobenzoic acid to form nitrite.
            ii) Removal of a sulfonic acid group from a sulfophenyl alkanoic
            acid metabolite of a linear alkylbenzene sulfonate.  iii) Sulfite
            formation by dioxygenative attack of p-toluene sulfonic acid.

     A wide variety of aniline derivatives are synthesized for use as herbi-
cides.  The derivatives may be variously substituted in the aromatic ring with
alkyl, alkoxyl and halide groups and the aniline nitrogen derivatized either
as an acyl anilide, a carbamate or a phenylurea.  Three examples of named her-
bicides based on 3,4-dichloroaniline are shown in Figure 14.  Bacteria and
fungi capable of modifying these structures have been described  (86).  One
general mechanism which may be employed is that whereby cleavage of the carbon-
nitrogen bond liberates the parent aniline together with aliphatic products.
An acylamidase enzyme system possessing a broad substrate specificity and cap-
able of hydrolyzing examples of all three classes of these herbicides has been
purified from a strain of Bacillus sphaericus (41).  For N-alkyl-substituted
phenylureas, an alternative mechanism of degradation involves one or more suc-
cessive demethylation or demethoxylation reactions as necessary to yield the
parent phenylurea  (147).

                                              C02 + CH3OH
                  -a      Propanil
               ci       { Acylanilide)

                  "Ci     Linuron
               CI       (Pbenylurea)
                                              CH30-N-CH3 + C02
Figure 14.  Degradation of aniline-based herbicides.  Cleavage of carbamate-,
            anilide-,  and phenylurea-herbicides to yield the parent aniline by
            an  amidase-type reaction.

     Having outlined  some  of the mechanisms employed by microorganisms for deg-
radation of different classes of organic compounds and for their attack on a
selected range of  functional groups, some specific examples will be considered
here to illustrate  a  few factors which may influence or modify the degradative
routes which are used.   These examples should further serve to illustrate the
difficulties of  predicting routes of degradation.

     It was stated  earlier that aromatic hydrocarbons may be degraded by micro-
organisms which  either oxidize alkyl sidechains or introduce dioxygen into the
ring to form cis-diols.  Consideration of the degradation route of the aromatic
hydrocarbon, styrene,  by bacteria which grow with this compound reveals that
the former possibility is  realized in an unexpected manner.  Recent results
from this laboratory  (Battles and Chapman, unpublished observations)  implicate
styrene oxide as the  first formed intermediate (Fig. 15)  which subsequently
undergoes isomerization to phenylacetaldehyde; thereafter its dehydrogenation
furnishes phenylacetic acid.  While this conversion is equivalent to sidechain
oxidation, it illustrates  that bacteria, like fungi and higher organisms (34),
are also capable of forming epoxides from certain hydrocarbons.  Isomerization
(rather than hydration as  occurs with arene oxides in animal tissues [110]) ap-
pears to be the  reaction used here for further degradation.  A number of cyclic
ketones known to be formed from such chemicals as  the cyclodiene insecticides
appear to be similarly formed by isomerization of precursor epoxides.

                     = CH2       HC—-CH2       CH2CHO        CH2COOH
Figure 15.  Conversion of styrene to phenylacetic acid by formation of styrene
            oxide and its isomerization to phenylacetaldehyde.

     When naphthalene is utilized as a carbon source by pseudomonads and other
bacteria the first reaction is one which forms a 1,2-cis-dihydrodiol  (Fig. 16)
 (56).  When such organisms are confronted by the related 1,5-dimethyl-naphtha-
lene it is a methyl substituent which is oxygenated to the primary alcohol and
thereafter converted to l-methyl-5-naphthoic acid (36); neither of the aromatic
rings is oxidized.  In a similar situation under study in this laboratory  (Tarn
and Chapman, unpublished observations), the tricyclic hydrocarbon, acenaph-
thene, is converted by a naphthalene-grown Pseudomonas to 1-acenaphthenol and
then to 1-acenaphthenone (Fig. 16).  Here, the site of oxygenation is the
methylenic carbon of the five-membered ring rather than the aromatic ring sys-
tem probably because it is the site of highest electron density and preferred
by an electrophilic species of oxygen.  The related acenaphthylene (not shown
in Fig. 16) possesses a double bond in the five-membered ring and it is this
group which is also the site of attack by naphthalene-grown cells, forming the
1,2-cis-dihydrodiol presumably by a dioxygen addition.  It seems clear that
the naphthalene dioxygenase is responsible for this unexpected attack on the
five-membered ring, giving either mono- or dioxygenated products depending
upon the structure involved.  These selected examples illustrate how a dioxy-
genation reaction can form oxygenated products which on purely steric grounds
would not be anticipated.  Others are known such as the 2,3- or meta-cleavage
of 3-methyl- and 3-methoxy-catechols by catechol 1,2-dioxygenase  (54,76), the
formation of o-phthalide from 2-methyIbenzoic acid by a benzoate-l,2-dioxy-
genase  (79) and the formation of small amounts of 1-phenylethanol from ethyl-
benzene apparently through the action of the enzyme system which forms 3-
ethylcatechol as the principal product (58).  There are also examples to show
that monooxygenase enzymes may also act to form products which would not be
expected from purely steric considerations.  The formation of 4-hydroxymethyl-
phenylalanine, of 3-hydroxy-4-methyl-phenylalanine and of 3-methyltyrosine by
the action of L-phenylalanine-4-hydroxylase on 4-methylphenylalanine  (33) can
be rationalized only from consideration of the nature of the reaction mecha-
nism and the form of "active" oxygen involved.  Substituents other than methyl,
such as chlorine or bromine, can also undergo migration as is found when cer-
tain halogenated aromatic compounds are acted upon by those hydroxylating en-
zumes which act by forming arene oxide intermediates  (34).  It is the rear-
rangement of such epoxides which underlies this migration of substituents and
which is known as the "NIH shift."

     It should not be assumed that the introduction of hydroxyl groups into
non-heterocyclic compounds is invariably an oxygen-dependent process.  The
conversion of p-cresol to 4-hydroxybenzylalcohol by species of Pseudomonas is
a reaction which uses the oxygen of water as the source of the hydroxyl group


 (74).  The mechanism appears to involve formation of a quinone methide inter-
mediate by a dehydrogenation reaction followed by hydration.  This mechanism
also accounts for the strict specificity of this hydroxylation in acting only
on 4-methyl-substituted phenols.  A similar reaction involving a quinone meth-
ide intermediate has been described in the degradation of a-conidendrin by a
Pseudomonas species  (145).
Figure 16.  Conversion of naphthalene and related compounds to oxygenated pro-
            ducts,  i) Formation of 1,2-dihydroxynaphthalene from naphthalene
            via cis-naphthalene 1,2-dihydrodiol by a naphthalene-utilizing
            Pseudomonas sp.   ii) Conversion of 1,5-dimethyl naphthalene to 1-
            methyl-5-naphthoic acid via 5-hydroxymethyl-l-methylnaphthalene.
            iii) Formation of 1-acenaphthenol and 1-acenaphthenone from
     The realization that frequently there are a number of alternative routes
by which a given chemical may be degraded by bacteria and fungi has prompted
examination of the possibility that such differences may be uniquely character-
istic of different microbial groups.  This has proved to be true for procary-
otic and eucaryotic groups within the microbial world, for different genera
and even for different species within the same genus.

     Within the genus Pseudomonas two different mechanisms for the cleavage of
protocatechuic (3,4-dihydroxybenzoic) acid are known and are useful in taxo-
nomic work (138); the ortho or 3,4-cleavage appears characteristic of fluor-
escent pseudomonads while the 4,5-(meta)-cleavage is found in the non-flour-
escent or acidovorans group (Fig. 17A).  Curiously, the subsequent reactions
of the degradative sequences initiated by these dioxygenase cleavages appear
to be present in both groups of organisms (113,137).  A third dioxygenase
cleavage of protocatechuic acid, across the 2,3-bond, has been described in
certain species of Bacillus (28).

     Another feature which distinguishes the fluorescent and non-fluorescent

pseudomonads is the^route of degradation of 4-hydroxyphenyiacetic acid.  Fluor-
escent pseudomonads form homprotochatechuic (3,4-dihydroxyphenylacetic) acid,
whereas non-fluorescent members of this group convert it to homogentisic acid
by a novel hydroxylation which effects migration of the carboxymethyl  side-
chain  (Fig. 17B)  (67,136).
 Figure  17.  Alternative degradative routes for aromatic acids,  i) Ortho
            cleavage of protocatechuate in fluorescent pseudomonads  compared
            with meta  cleavage by non-fluorescent pseudomonads. ii)  Conversion
            of p-hydroxyphenylacetic acid to homoprotocatechuate by  fluor-
            escent pseudomonads compared with its conversion  to homogentisate
            by non-fluorescent pseudomonads.

      To illustrate generic  differences, reference will be made to  the  degrada-
 tion of 4-hydroxybenzoic  acid and its  3-methoxy-substituted derivative (vanil-
 lic acid).  Members of the  genus Pseudomonas convert these compounds to proto-
 catechuic  acid either  by  direct hydroxylation or demethoxylation respectively.
 Cleavage of protocatechuic  acid will occur either 3,4- or 4,5- depending  upon
 the pseudomonas group  involved  (Fig. 18).  While certain of the above  reac-
 tions may  also be found in  members of  the genus Bacillus, other members of
 this group can effect  a novel hydroxylation-facilitated carboxyl migration of
 4-hydroxybenzoic acid  and vanillic acid not yet found in other groups  (88) .  A
 consequence of this is that gentisic acid and its 4-methoxyl-derivative are
 now the respective products; they are  subsequently degraded by pathways com-
 pletely different from the  protocatechuic acid routes  (Fig. 18).

      One example will  suffice to show  how fungi and bacteria  act in  very  dif-
 ferent  ways on the same chemical.  The herbicide, 2,4-dichlorophenoxyacetic
 acid (2,4-D),  is utilized for growth by some bacteria and the principal reac-
 tion sequence  used appears  to be that  shown  (Fig. 19), namely,  loss  of the
 sidechain  and  hydroxylation at C& before ring cleavage.  The  sequence  described
 earlier (44)  (Fig. 12) shows the continuation of the pathway.  By  contrast, the
 fungus, Aspergillus niger,  which does  not utilize 2,4-D for growth,  introduces
 hydroxyl groups both at C5  and C^, the latter hydroxylation causing  a  migra-
 tion of chlorine to an adjacent position  (Fig. 19)  (45).

                                                HOOC  0 COOH

Figure 18.  Degradation of 4-hydroxybenzoic  and  vanillic  acids by  species of
            Pseudomonas and Bacillus.   i) Formation  of protocatechuate by hy-
            droxylation and demethoxylation  reactions before  ring  cleavage  in
            pseudomonads.  ii) Formation of  gentisate and 4-methoxygentisate
            by carboxyl migration  facilitated by hydroxylation as  found in
            species of Bacillus.

Figure 19.  Action of bacteria and  fungi on  2,4-dichlorophenoxyacetic acid
             (2,4-D).  Formation of  3,5-dichlorocatechol by sidechain loss and
            ring hydroxylation by bacteria compared with fungal hydroxylation
            at Ci*  (with chlorine migration)  and at Cs.
     Lastly, mention should be made of the fact that while much of the fore-
going discussion has described degradative sequences which use molecular oxy-
gen at one or more steps, many of the environments into which natural and syn-
thetic compounds are introduced are anaerobic.  Degradation must occur there-
fore by catabolic routes which are independent of oxygen, i.e., by anaerobic
photometabolism, by fermentation and by anaerobic respiration.  Few detailed
studies of such types of degradation have been reported.  The anaerobic degra-
dation of benzoate, however, has been the subject of extensive studies, prin-
cipally by the research group of Evans  (42) who has studied its dissimilation

by both photosynthetic and nitrate-respiring bacteria.  The general features
of its photometabolism (40) are compared with one known route for its aerobic
catabolism in Figure 20.   It can be seen that anaerobic degradation of this
aromatic compound joins the pathways of alicyclic degradation outlined earlier
(Fig. 3).  Reduction and hydration of the coenzyme A thioester of benzoate oc-
curs to give the alicyclic cyclohexanol-2-carboxylic acid.  The strategy is to
generate a ring system which can be opened by a reaction not requiring oxygen.
This is accomplished by formation of the cyclohexanol-2-carboxylic acid deriva-
tive which is cleaved either by a hydrolytic or thiolytic mechanism.  The re-
sulting pimelic acid or its thioester can then be oxidized by using electron
acceptors other than oxygen.
              COz + 3ACETYL CoA
               |    COOH
 Figure  20.   Comparison of benzoate degradation by aerobic and anaerobic bac-
             teria,   i) Conversion of benzoate to cis,cis-muconate by oxygen-
             dependent reactions by aerobic bacteria,  ii) Conversion of ben-
             zoate by anaerobic bacteria to alicyclic compounds by reduction
             and hydration permitting ring cleavage in the absence of oxygen.
      Recent reports  indicate that aromatic compounds such as phenol and cate-
 chol  are  degraded by anaerobic communities of methane-forming organisms  (69).
 The reactions  by which  these conversions occur have not yet been elucidated.

      Much of our knowledge of the mechanisms employed by microorganisms for
 the dissimilation of organic compounds derives from studies with organisms
 isolated  from  terrestrial and freshwater environments.  Few studies have been
 reported  of the mechanisms of degradation of organic chemicals by marine orga-
 nisms in  marine environments.  It remains to be established whether those
 groups of microorganisms currently recognized as playing important roles in
 degradative processes in soil and freshwater are such important agents in
 brackish,  estuarine  and marine waters.  There is some evidence to suggest that
 fresh water organisms may not be able to effect degradation of such compounds
 as  nitrilotriacetic  acid  (11) or may be inhibited  in their degradation of hy-
 drocarbons (149) in  saline environments.  It is evident, therefore, that

 continuing research on biodegradation of organic compounds should address
 these and other questions so as to amplify our present knowledge of these
 processes in marine environments.


      The author's research is supported by NSF Grant No.  PCM76-18793.

                               LITERATURE CITED

 1.  Abbott, B. J., and L. E. Casida.  1968.  Oxidation of alkanes to internal
      monoalkenes by a Nocardia.  J. Bacteriol. 96:925-930.

 2.  Anderson, W. A., and B. Magasanik.  1971.  The pathway of myo-inositol
      degradation in Aerobacter aerogenes.  Identification of the intermediate
      2-deoxy-5-keto-D-gluconic acid.  J. Biol. Chem. 246:5653-5661.

 3.  Anderson, W. A., and B. Magasanik.  1971.  The pathway of myo-inositol
      degradation in Aerobacter aerogenes.  Conversion of 2-deoxy-5-ketogluconic
      acid to glycolytic intermediates.  J. Biol. Chem. 246:5662-5675.

 4.  Bachofer, R., F. Lingens, and W. Schafer.  1975.  Conversion of aniline
      into pyrocatechol by a Nocardia sp.; incorporation of oxygen-18.   FEES
      Letters 50:288-290.

 5.  Baggi, G., D. Catelani, E. Galli, and V.  Treccani.  1972.  The microbial
      degradation of phenylalkanes:  2-phenylbutane, 3-phenylpentane,  3-phenyl-
      dodecane and 4-phenylheptane.  Biochem.  J. 126:1091-1097.

 6.  Bayly, R. C., and G. J. Wigmore.  1973.  Metabolism of phenol and  cresols
      by mutants of Pseudomonas putida.  J. Bacteriol. 113:1112-1120.

 7.  Basso, L. V., D. R. Rao, and V. W. Rodwell.  1962.  Metabolism of  pipe-
      colic acid in a Pseudomonas species.  II. A'-Piperideine-6-carboxylic
      acid and a-aminoadipic acid-6-semialdehyde.  J. Biol. Chem. 237:2239-2245.

 8.  Beam, H. W., and J. J. Perry.  1974.  Microbial degradation of cycloparaf-
      finic hydrocarbons via co-metabolism and commensalism.  J. Gen.
      Microbiol. 82:163-169.

 9.  Behrman, E. J., and R. Y. Stanier.  1957.  The bacterial oxidation of
      nicotinic acid.  J. Biol. Chem. 228:923-945.

10.  Blakley, E. R.  1974.  The microbial degradation of cyclohexanecarboxylic
      acid:  a pathway involving aromatization to form p-hydroxybenzoic acid.
      Can. J. Microbiol. 20:1297-1306.

11.  Bourquin, A. W., and V. A. Przybyszewski.  1977.  Distribution of  bacteria
      with nitrilotriacetate-degrading potential in an estuarine environment.
      Appl. Environ. Microbiol. 34:411-418.

12.  Bradshaw, W. H. ,  H. E. Conrad, E. J. Corey, I. C. Gunsalus, and D. Led-
      nicer.  1959.  Microbial degradation of (+)-camphor.  J. Am. Chem. Soc .

13.  Britton, L. N. , and A. J. Markovetz.  1977.  A novel ketone monooxygenase
      from Pseudomonas cepacia.  Purification and properties.  J. Biol. Chem.

14.  Cain, R. B. , and D. R. Farr.  1968.  Metabolism of aryl sulphonates by
      microorganisms.   Biochem. J. 106:859-877.
15.  Cardini, G. , and P. Jurtshuk.  1968.  Cytochrome Pi^so involvement in the
      oxidation of n-octane by cell-free extracts of Corynebacterium sp.
      strain 7E1C.  J. Biol. Chem. 243:6070-6072.

16.  Cartwright, N. J. ,  and R. B. Cain.  1959.  Bacterial degradation of the
      nitrobenzoic acids.  Biochem. J. 71:248-261.

17.  Catelani, D., A. Colombi, C. Sorlini, and V. Treccani.  1973.  Metabolism
      of biphenyl.  2-Hydroxy-6-oxo-6-phenylhexa-2 ,4-dienoate :   the meta-
      cleavage product from 2, 3-dihydroxybiphenyl by Pseudomonas putida.
      Biochem. J. 134:1063-1066.

18.  Catelani, D. , A. Colombi, C. Sorlini, and V. Treccani.  1977.  Metabolism
      of quaternary carbon compounds:  2,2-dimethylheptane and tertbutylben-
      zene.  Appl. Environ. Microbiol. 34:351-354.

19.  Chapman, P. J.  1972.  An outline of reaction sequences used for the bac-
      terial degradation of phenolic compounds, p. 17-55.  In Degradation of
      synthetic organic molecules in the biosphere .   National Academy of
      Sciences, Washington, D.C.

20.  Chapman, P. J.  1976.  Microbial degradation of halogenated compounds.
      Biochem. Soc. Trans. 4:463-466.

21.  Chapman, P. J., and S. Dagley.  1962.  Oxidation of homogentisic acid by
      cell-free extracts of a vibrio.  J. Gen. Microbiol. 28:251-256.

22.  Chapman, P. J. , and R. G. Duggleby.  1967.  Dicarboxylic acid catabolism
      by bacteria.  Biochem. J. 103 :7C.

23.  Chapman, P. J., G.  Meerman, and I. C. Gunsalus.  1965.  The microbiologi-
      cal transformation of fenchone.  Biochem. Biophys. Res. Commun.

24.  Chapman, P. J., G.  Meerman, I. C. Gunsalus, R.  Srinivasan, and K. L.
      Rhinehart.  1966.   A new acyclic acid metabolite in camphor oxidation.
      J. Am. Chem. Soc.  88:618-619.

25.  Chapman, P. J. , and D. W. Ribbons.  1976.  Metabolism of resorcinylic
      compounds by bacteria:  alternative pathways for resorcinol catabolism
      in Pseudomonas putida.  J. Bacteriol. 125:985-998.

26.  Chouteau, J., E. Azoulay, and J. C. Senez.  1962.  Anaerobic formation of
      n-hept-1-ene from n-heptane by resting cells of Pseudomonas aeruginosa.
      Nature 194:576-578.

27.  Colby, J., and H. Dalton.  1976.  Some properties of a soluble methane
      monooxygenase from Methylococcus capsulatus strain Bath.  Biochem. J.

28.  Crawford, R. L.  1975.  Novel pathway for degradation of protocatechuic
      acid in Bacillus species.  J. Bacteriol. 121:531-536.

29.  Crawford, R. L., S. W. Hutton, and P. J. Chapman.  1975.  Purification
      and properties of gentisate 1,2-dioxygenase from Moraxella osloensis.
      J. Bacteriol. 121:794-799.

30.  Cripps, R. E.  1973.  The microbial metabolism of thiophen-2-carboxylate.
      Biochem. J. 134:353-366.

31.  Dagley, S.  1971.  Catabolism of aromatic compounds by microorganisms.
      Adv. Microb. Physiol. 6:1-46.

32.  Dagley, S., P. J. Chapman, D. T. Gibson, and J. M. Wood.  1964.  Degrada-
      tion of the benzene nucleus by bacteria.  Nature 202:775-778.

33.  Daly, J., and G. Guroff.  1968.  Production of m-methyltyrosine and p-
      hydroxymethylphenylalanine from p-methylphenylalanine by phenylalanine
      hydroxylase.  Arch. Biochem. Biophys. 125:136-141.

34.  Daly, J. W., D. M. Jerina, and B. Witkop.  1972.  Arene oxides and the
      NIH shift:  the metabolism, toxicity and carcinogenicity of aromatic
      compounds.  Experientia 28:1129-1149.

35.  Davis, J. B., and R. L. Raymond.  1961.  Oxidation of alkyl-substituted
      cyclic hydrocarbons by a Nocardia during growth on n-alkanes.  Appl.
      Microbiol. 9:383-388.

36.  Dean-Raymond, D., and R. Bartha.  1975.  Biodegradation of some poly-
      nuclear aromatic petroleum components by marine bacteria.  Dev. Ind.
      Microbiol. 16:97-110.

37.  Dodson, R. M., and R. D. Muir.  1961.  Microbiological transformations.
      VI. The microbiological aromatization of steroids.  J. Am. Chem. Soc.

38.  Donoghue, N. A., D. B. Norris, and P. W. Trudgill.  1976.  Purification
      and properties of cyclohexanone oxygenase from Nocardia globerula CL-1
      and an Acinetobacter NCIB 9871.  Eur. J. Biochem. 63:175-192.

39.  Donoghue, N. A., and P. W. Trudgill.  1975.  The metabolism of cyclo-
      hexanol by Acinetobacter NCIB 9871.  Eur. J. Biochem. 60:1-7.

40.  Dutton, P. L., and W. C. Evans.  1969.  The metabolism of aromatic com-
      pounds by Rhodopseudomonas palustris.  Biochem. J. 113:525-536.


41.  Engelhardt,  G.,  P.  R.  Wallnofer,  and R.  Plapp.   1973.   Purification and
      properties  of an aryl acylamidase of Bacillus  sphaericus,  catalyzing the
      hydrolysis  of various phenylamide herbicides and fungicides.   Appl.
      Microbiol.  26:709-718.

42.  Evans, W. C.  1977.  Biochemistry of the bacterial catabolism  of aromatic
      compounds in anaerobic environments. Nature 270:17-22.

43.  Evans, W. C., B. S. W. Smith,  P.  Moss, and H. N. Fernley.   1971.  Bacte-
      rial metabolism of 4-chlorophenoxyacetate.  Biochem.  J.  122:509-517.

44.  Evans, W. C., B. S. W. Smith,  H.  N. Fernley,  and J. I. Davies.  1971.
      Bacterial metabolism of 2,4-dichlorophenoxyacetate.   Biochem. J.

45.  Faulkner, J. K., and D. Woodcock.  1967.  Fungal detoxication.  VII. Me-
      tabolism of 2,4-dichlorophenoxyacetic and 4-chloro-2-methylphenoxy-
      acetic acids by Aspergillus niger.  J.  Chem. Soc., p. 1187-1191.

46.  Flashner, M. S., and V. Massey.  1974.  Flavoprotein oxygenases, p. 245-
      283.  In O. Hayaishi  (ed.), Molecular mechanisms of oxygen activation.
      Academic Press, New York.

47.  Focht, D. D., and M. Alexander.  1970.  Bacterial degradation  of diphenyl-
      methane, a DDT model substrate.   Appl.  Microbiol. 20:608-611.

48.  Focht, D. D., and H. Joseph.  1974.  Degradation of 1,1-diphenylethylene
      by mixed cultures.  Can. J. Microbiol.  20:631-635.

49.  Francis, A.  J., R. J.  Spanggord,  G. I. Ouchi, R. Bramhall,  and N.  Bohonos.
      1976.  Metabolism of DDT analogues by a Pseudomonas sp.  Appl.  Environ.
      Microbiol.  32:213-216.

50.  Frank, L., and B. Ranhand.  1964.  Proline metabolism in  Escherichia coli.
      III. The proline catabolic pathway.  Arch. Biochem.  Biophys.  107:325-331.

51.  Fried, J., W. Thoma, and A. Klingsberg.   1953.   Oxidation of steroids by
      microorganisms.  III. Side chain degradation,  ring D-cleavage and dehy-
      drogenation in ring A.  J. Am. Chem. Soc. 75:5764-5765.

52.  Friello, D.  A., J. R.  Mylroie, D. T. Gibson,  J. E. Rogers,  and A.  M.
      Chakrabarty.  1976.  XYL, a non conjugative xylene-degradative plasmid
      in Pseudomonas pxy.  J. Bacteriol. 127:1217-1224.

53.  Fujioka, M., and H. Wada.  1968.   The bacterial oxidation of indole.
      Biochim. Biophys. Acta 158:70-78.

54.  Fujiwara, M., L. A. Golovleva, Y. Saeki, M. Nozaki, and O.  Hayaishi.
      1975.  Extradiol cleavage of 3-substituted catechols by  an intradiol
      dioxygenase, pyrocatechase, from a Pseudomonad.  J.  Biol.  Chem.

55.  Gamar, Y., and J. K. Gaunt.  1971.  Bacterial metabolism of 4-chloro-2-
      methylphenoxyacetate.  Formation of glyoxylate by side-chain cleavage.
      Biochem. J. 122:527-531.

56.  Gibson, D. T.  1971.  The microbial oxidation of aromatic hydrocarbons.
      Grit. Rev. Microbiol. 1:199-223.

57.  Gibson, D. T.  1968.  Microbial degradation of aromatic compounds.
      Science 161:1093-1097.

58.  Gibson, D. T., B. Gschwendt, W. K. Yeh, and V. M. Kobal.  1973.   Initial
      reactions in the oxidation of ethylbenzene by Pseudomonas putida.
      Biochemistry 12:1520-1528.

59.  Gibson, D. T., K. C. Wang, C. J. Sih, and H. Whitlock.  1966.  Mechanisms
      of steroid oxidation by microorganisms.  IX. On the mechanism of ring A
      cleavage in the degradation of 9,10-seco steroids by microorganisms.
      J. Biol. Chem. 241:551-559.

60.  Goldman, P.  1972.  Eazymology of carbon-halogen bonds, p. 147-165.
      In Degradation of synthetic organic molecules in the biosphere.
      National Academy of Sciences, Washington, D.C.

61.  Goodman, D. S., and H. S. Huang.  1965.  Biosynthesis of vitamin  A with
      rat intestinal enzymes.  Science 149:879-880.

62.  Griffin, M., and P. W. Trudgill.  1972.  The metabolism of cyclopentanol
      by Pseudomonas NCIB 9872.  Biochem. J. 129:595-603.

63.  Gunsalus, I. C., and V. P. Marshall.  1971.  Monoterpene dissimilation:
      chemical and genetic models.  Grit. Rev. Microbiol. 1:291-310.

64.  Guroff, G., andC. A. Rhoads.  1967.  Phenylalanine hydroxylase from
      Pseudomonas species  (ATCC 11299a).  Purification of the enzyme and  acti-
      vation by various metal ions.  J. Biol. Chem. 242:3641-3645.

65.  Habig, W. H., M. J. Pobjst, and W. B. Jakoby.  1974.  Glutathione S-
      transferase.  J. Biol. Chem. 249:7130-7139.

66.  Hammond, M. W., and M. Alexander.  1972.  Effect of chemical structure on
      microbial degradation of methyl-substituted aliphatic acids.  Environ.
      Sci. Technol. 6:732-735.

67.  Hareland, W. A., R. L. Crawford, P. J. Chapman, and S. Dagley.  1975.
      Metabolic function and properties of 4-hydroxyphenylacetic acid 1-
      hydroxylase from Pseudomonas acidovorans.  J. Bacteriol. 121:272-285.

68.  Hayaishi, O., and M. Nozaki.  1969.  Nature and mechanisms of oxygenases.
      Science 164:389-396.

69.  Healy, J. B., and L. Y. Young.  1978.  Catechol and phenol degradation by
      a methanogenic population of bacteria.  Appl. Environ. Microbiol.


70.  Hector,  M.  L.,  and R.  R.  Fall.   1976.   Evidence for distinct 3-methyl-
      crotonyl-CoA and geranyl-CoA carboxylases in Pseudomonas citronellolis.
      Biochem.  Biophys. Res.  Commun.  71:746-753.

71.  Hegeman, G. D.   1972.   The evoluation  of metabolic pathways in bacteria,
      p. 56-72.  In Degradation of synthetic organic molecules in the biosphere.
      National Academy of Sciences,  Washington, D.C.

72.  Hochstein,  L. I., and B.  P. Dalton.   1965.  The hydroxylation of nicotine:
      the origin of the hydroxyl oxygen.  Biochem. Biophys.  Res. Commun.

73.  Hoet, P. P., and R. Y. Stanier.   1970.  The dissimilation of higher di-
      carboxylic acids by Pseudomonas fluorescens.  Eur. J.  Biochem. 13:65-70.

74.  Hopper,  D.  J.,  and D.  G.  Taylor.  1977.   The purification and properties
      of p-cresol-(acceptor)  oxidoreductase (hydroxylating)  a flavocytochrome
      from Pseudomonas putida.  Biochem.  J. 167:155-162.

75.  Horvath, R. S.   1970.   Co-metabolism of methyl- and chloro-substituted
      catechols by an Achromobacter sp. possessing a new meta-cleaving oxy-
      genase.  Biochem. J.  119:871-876.

76.  Hou, C.  T., R.  Patel,  and M. O.  Lillard.   1977.  Extradiol cleavage of 3-
      methylcatechol by catechol 1,2-dioxygenase from various microorganisms.
      Appl. Environ. Microbiol. 33:725-727.

77.  Howard,  P.  H.,  J. Saxena, and H. Sikka.   1978.   Determining'the fate of
      chemicals.  Environ.  Sci. Technol.  12:398-407.

78.  Hunt, A. L., D. E. Hughes, and J.  M. Lowenstein.   1958.   The hydroxyla-
      tion of nicotinic acid by Pseudomonas fluorescens.  Biochem.  J.

79.  Ichihara, A., K. Adachi,  K. Hosokawa,  and Y. Takeda. 1962.  The  enzy-
      matic hydroxylation of aromatic carboxylic acids;  substrate specifici-
      ties of anthranilate and benzoate oxidases.  J.  Biol.  Chem. 237:2296-2302,

80.  lizuka,  H., M.  lida, and S. Fujita.  1969.  Formation of n-decene-1-1
      from n-decane by resting cells of Candida rugosa.   Z.  Allg. Mikrobiol.

81.  Jeffrey, A. M., D. M.  Jerina, R. Self, and W. C.  Evans.   1972. The bac-
      terial  degradation of flavonoids.   Oxidative fission of the A-ring of
      dihydrogossypetin by a Pseudomonas  sp.   Biochem. J. 130:383-390.

82.  Jeffrey, A. M., M. Knight, and W.  C. Evans.   1972.   The bacterial degra-
      dation  of flavonoids.  Hydroxylation  of the A-ring of  taxifolin  by a
      soil pseudomonad.  Biochem. J.  130:373-381.

83.  Jones, N.,  and G. K. Watson.  1976.  Ethylene glycol and polyethylene
      glycol  catabolism by a sewage  bacterium.   Biochem. Soc. Trans. 4:891-892.

84.  Jurtshuk, P., and G. E. Cardini.  1971.  The mechanism of hydrocarbon
      oxidation by a Corynebacterium species.  Grit. Rev. Microbiol.  1:239-289.

85.  Kaneda, T.  1974.  Enzymatic aromatization of 4-ketocyclohexane  carboxylic
      acid to p-hydroxybenzoic acid.  Biochem. Biophys. Res. Commun.

86.  Kaufman, D. D., and J. Blake.  1973.  Microbial degradation of several
      acetamide, acylanilide, carbamate toluidine, and urea pesticides.   Soil
      Biol. Biochem. 5:297-308.

87.  Kawai, F., T. Kimura, M. Fukaya, Y. Tani, K. Ogata, T. Ueno, and H.
      Fukami.  1978.  Bacterial oxidation of polyethylene glycol.  Appl.
      Environ. Microbiol. 35:679-684.

88.  Keenan, S. L., and P. J. Chapman.  1978.  Carboxyl migration facilitated
      by bacterial hydroxylation of 4-hydroxybenzoate.  Chem. Commun. (in

89.  Kido, T., K.  Soda, T. Suzuki, and K. Asada.  1976.  A new oxygenase,  2-
      nitropropane dioxygenase of Hansenula mrakii.  Enzymologic and  spectro-
      photometric properties.  J. Biol. Chem. 251:6994-7000.

90.  Kieslich, K.  1976.  Microbial transformations of non-steroid cyclic com-
      pounds .  Georg Thieme Pub., Stuttgart.

91.  Kiyohara, H., and K. Nagao.  1978.  The catabolism of phenanthrene  and
      naphthalene by bacteria.  J. Gen. Microbiol. 105:69-75.

92.  Klug, M. J., and A. J. Markovetz.  1968.  Degradation of hydrocarbons by
      members of the genus Candida.  III. Oxidative intermediates from 1-
      hexadecene and 1-heptadecene by Candida lipolytica.  J. Bacteriol.

93.  Knackmuss, H.-J., W. Beckmann, E. Dorn, and W. Reineke.  1976.  On  the
      mechanism of the biological persistence of halogenated and sulfonated
      aromatic hydrocarbons.  Zentralbl. Bakteriol. Hyg. B162:127-137.

94.  Kodama, K., K. Umehara, K. Shimizu, S. Nakatani, Y. Miroda, and  K.  Yamada.
      1973.  Identification of microbial products from dibenzothiophene  and
      its proposed oxidation pathway.  Agric. Biol. Chem. 37:45-50.

95.  Krishnamurty, H. G., and F. J. Simpson.  1970.  Degradation of rutin by
      Aspergillus flavus.  Studies with oxygen 18 on the action of a  dioxy-
      genase on quercetin.  J. Biol. Chem.  245:1467-1471.

96.  Kusunose, M., E. Kusunose, and M. J. Coon.  1964.  Enzymatic co-oxidation
      of fatty acids.  I. Products of octanoate, decanoate and laurate oxida-
      tion.  J. Biol. Chem. 239:1374-1380.

97.  Laborde, A. L., and D. T. Gibson.  1977.  Metabolism of dibenzothiophene
      by a Beijerinckia species.  Appl. Environ. Microbiol. 34:783-790.


 98.   Lanzilot'ta,  R.  P.,  D.  G.  Bradley,  and M.  L.  Maddox.   1977-   Stereospeci-
       fic sulfur  oxidation  of  7-methylthioxanthone-2-carboxylic  acid by Calo-
       nectria decora.  Appl. Environ. Microbiol.  34:56-59.

 99.   Larway,  P.,  and W.  C.  Evans.   1965.   Metabolism of quinol and resorcinol
       by soil pseudomonads.  Biochem. J.  95:52P.

100.   Laskin,  A.  I.,  P. Grabowich,  C. D. Meyers, and J.  Fried.  1964.   Trans-
       formations  of  eburicoic  acid.  V. Cleavage  of ring A by  the fungus Glo-
       merella fusarioides.  J.  Med. Chem.  7:406-409.

101.   Lebeault, J. M.,  B.  Roche,  Z.  Duvnjak, and E.  Azoulay.  1969.   General
       scheme  for  the metabolisation of  hydrocarbons by  Candida tropicalis.
       Antonie van Leeuwenhoek  J.  Serol. Microbiol.  35:F34-F34.

102.   Lee, Y.-L.  T.,  V. L. Sparnins, and S.  Dagley.   1978.   Catabolism of 2,4,
       5-trimethoxybenzoic acid and 3-methoxy-crotonic acid.  Appl.  Environ.
       Microbiol.  35:817-819.

103.   Magee, P. T., and E. E. Snell.  1966.   The bacterial  degradation of pan-
       tothenic acid.  IV. Enzymatic conversion of aldopantoate to a-ketoiso-
       valerate.   Biochemistry  5:409-416.

104.   Matsuura, T., H.  Matsushima,  and H.  Sakamoto.   1967-   Photosensitized
       oxygenation of 3-hydroxyflavones.   A possible model  for  biological oxy-
       genation.   J.  Am.  Chem.  Soc.  89:6370-6371.

105.   Markovetz,  A. J.  1971.   Subterminal oxidation of  aliphatic hydrocarbons
       by microorganisms.  Crit.  Rev. Microbiol. 1:225-237.

106.   Massey,  L.  K.,  J. R. Sokatch,  and  R.  S. Conrad.  1976.  Branched-chain
       amino acid catabolism in bacteria.   Bacteriol.  Rev.  40:42-54.

107.   May, S.  W.,  and B.  J.  Abbott.  1973.   Enzymatic  epoxidation.   II.  Com-
       parison between  the epoxidation and hydroxylation reactions catalyzed
       by the  w-hydroxylation system of  Pseudomonas  oleovorans.   J.  Biol.
       Chem. 248:1725-1730.

108.   McKenna, E.  J.   1972.  Microbial metabolism  of normal  and branched chain
       alkanes, p. 73-97.  In Degradation  of synthetic organic  molecules in  the
       biosphere.   National  Academy of Sciences, Washington, D.C.

109.   McKenna, E.  J., and R. E.  Kallio.  1972.  Microbial metabolism of  the
       isoprenoid  alkane  pristane.   Proc.  Nat.  Acad.  Sci. U.S.A.  68:1552-1554.

110.   Oesch, F.  1973.  Mammalian epoxide  hydrases.   Inducible  enzymes catalyz-
       ing the inactivation  of  carcinogenic  and cytotoxic metabolites  derived
       from aromatic  and  olefinic compounds. Xenobiotica  3:305-340.

111.   Ogasawara, N.,  J. E. Gander,  and L.  M. Henderson.   1965.  Purification
       and properties of  3-hydroxyanthranilate  oxygenase from beef kidney.
       J. Biol. Chem. 241:613-619.

112.  Olson, J. A., and O. Hayaishi.  1965.  The enzymatic cleavage of (3-caro-
       tene into vitamin A by soluble enzymes of rat liver and intestine.
       Proc. Soc. Nat. Acad. Sci. 54:1364-1370.

113.  Ornston, M. K., and L. N. Ornston.  1972.  The regulation of the 3-keto-
       adipate pathway in Pseudomonas acidovorans and Pseudomonas testosteroni.
       J. Gen. Microbiol. 73:455-464.

114.  Parekh, V. R., R. W. Traxler, and J. M. Sobek.  1977.  n-Alkane oxidation
       enzymes of a pseudomonad.  Appl. Environ. Microbiol. 33:881-884.

115.  Perry, J. J.  1977.  Microbial metabolism of cyclic hydrocarbons and re-
       lated compounds.  Grit. Rev. Microbiol. 5:387-412.

116.  Peterson, D., and J. Llaneza.  1974.  Identification of a carbon-oxygen
       lyase activity cleaving the ether linkage in carboxymethyloxysuccinic
       acid.  Arch. Biochem. Biophys. 162:135-146.

117.  Peterson, D. H., S. H. Eppstein, P. D. Meister, H. C. Murray, H. M.
       Leigh, A. Weintraub, and L. M. Reineke.  1953.  Microbiological trans-
       formations of steroids.  IX.  Degradation of Cai steroids to Gig ke-
       tones and to testololactone.  J. Am. Chem. Soc. 75:5768-5769.

118.  Peterson, J. A., D. Basu, and M. J. Coon.  1966.  Enzymatic OJ-oxidation.
       I. Electron carriers in fatty acid and hydrocarbon hydroxylation.
       J. Biol. Chem. 241:5162-5164.

119.  Pirnik, M. P.  1977.  Microbial oxidation of methylbranched alkanes.
       Crit. Rev. Microbiol. 5:413-422.

120.  Pirnik, M. P., R. M. Atlas, and R. Bartha.  1974.  Hydrocarbon metabo-
       lism by Brevibacterium erythrogenes:  normal and branched alkanes.
       J. Bacteriol. 119:868-878.

121.  Porter, D. J. T., J. G. Voet, and H. J. Bright.  1972.  Nitromethane:  a
       novel substrate for D-amino acid oxidase.  J. Biol. Chem. 247:1951-1953.

122.  Prairie, R. L. , and P. Talalay.  1963.  Enzymatic formation of testolo-
       lactone.  Biochemistry 2:203-207-

123.  Preiss, I., and G. Ashwell.  1962.  Alginic acid metabolism in bacteria.
       I. Enzymatic formation of unsaturated oligosaccharides and 4-deoxy-L-
       erythro-5-hexoseulose uronic acid.  J. Biol. Chem. 237:309-316.

124.  Reiner, A. M.  1971.  Metabolism of benzole acid by bacteria:  3,5-
       cyclohexadiene-l,2-diol-l-carboxylic acid as an intermediate in the
       formation of catechol.  J. Bacteriol. 108:89-94.

125.  Reiner, A. M., and G. D. Hegeman.  1971.  Metabolism of benzoic acid by
       bacteria.  Accumulation of  (-)-3,5-cyclohexadiene-l,2-diol-l-carboxylic
       acid by a mutant strain of Alcaligenes eutrophus.  Biochemistry  10:

126.  Rho,  E.  M., and W. C.  Evans.  1975.  The aerobic metabolism of cyclo-
       hexane  carboxylic acid by Acinetobacter anitratum.  Biochem. J.

127.  Ribbons, D. W.   1970.   Specificity of monohydric phenol oxidations by
       meta cleavage pathways in Pseudomonas aeruginosa TI.   Archiv. Mikrobiol,

128.  Ribbons, D. W.   1972.   Metabolism of methoxy- and methylenedioxyphenyl
       compounds by bacteria, p. 98-115.  In Degradation of synthetic organic
       molecules in the biosphere.  National Academy of Sciences, Washington,

129.  Ribbons, D. W.   1975.   Oxidation of C-l compounds by particulate frac-
       tions from Methylococcus capsulatus:  distribution and properties of
       methane-dependent reduced nicotinamide adenine dinucleotide oxidase
       (methane hydroxylase).  J. Bacteriol. 122:1351-1363.

130.  Schultz, E., F. E. Engle, and J. M. Wood.  1974.  New oxygenases in the
       degradation of flavones and flavanones.  Biochemistry 13:1768-1776.

131.  Seubert, W., and E. Fass.  1964.  Untersuchungen uber den bakteriellen
       Abbau von Isoprenoiden.  V. Der mechanisms des isoprenoidabbaues.
       Biochem. Z. 341:35-44.

132.  Seubert, W., and U. Remberger.  1963.  Untersuchungen u'ber den bak-
       teriellen Abbau con Isoprenoiden.  II. Die Rolle der Kohlensaure.
       Biochem. Z. 338:245-264.

133.  Siuda, J. F., and J. F. DeBernardis.  1973.  Naturally occurring haloge-
       nated organic compounds.  Lloydia 36:107-143.

134.  Smith, D. I., and A. G. Callely.  1975.  The microbial degradation of
       cyclohexane carboxylic acid.  J. Gen. Microbiol. 91:210-212.

135.  Sorlini, C., and V. Treccani.  1969.  Microbial degradation of aliphatic
       branched compounds:  isobutyric, 2,2-dimethylmalonic  and 2,2-dimethyl-
       succinic acids.  Experientia 25:1032.

136.  Sparnins, V. L., P. J.  Chapman, and S. Dagley.   1974.   Bacterial degra-
       dation of 4-hydroxyphenylacetic acid and homoprotocatechuic acid.
       J. Bacteriol.  120:159-167.

137.  Sparnins, V. L., and S. Dagley.  1975.  Alternative routes of aromatic
       catabolism in Pseudomonas acidovorans and Pseudomonas putida:  gallic
       acid as a substrate and inhibitor of dioxygenases.  J. Bacteriol.

138.  Stanier, R. Y., N. J.  Palleroni, and M. Doudoroff.  1966.  The aerobic
       pseudomonads:   a taxonomic study.  J. Gen. Microbiol. 43:159-271.

139.  Stewart, J. E., W. R.  Finnerty, R. E. Kallio, and D. P. Stevenson.  1960.
       Esters  from bacterial  oxidation of olefins.  Science 132:1254-1255.


140.  Stirling, L. A., R. J. Watkinson, and I. J. Higgins.  1977.  Microbial
       metabolism of alicyclic hydrocarbons:   isolation and properties of a
       cyclohexane-degrading bacterium.  J. Gen. Microbiol. 99:119-125.

141.  Tack, B. F., P. J. Chapman, and S. Dagley.  1972.  Metabolism of gallic
       acid and syringic acid by Pseudomonas putida.  J. Biol. Chem.

142.  Taylor, D. G. , and P. W. Trudgill.  1978.  Metabolism of cyclohexane
       carboxylic acid by Alcaligenes strain Wl.  J. Bacteriol. 134:401-411.

143.  Tiedje, J. M. , J. M. Duxbury, and M. Alexander.  1969.  2,4-D-metabolism:
       pathway of degradation of chlorocatechols by Arthrobacter sp.  J.  Agr.
       Food Chem. 17:1021-1026.

144.  Toms, A., and J. M. Wood.  1970.  The degradation of trans-ferulic  acid
       by Pseudomonas acidovorans.  Biochemistry 9:337-343.

145.  Toms, A., and J. M. Wood.  1970.  Early intermediates in the degradation
       of a-conidendrin by a Pseudomonas multivorans.  Biochemistry 9:733-740.

146.  Trudgill, P. W.  1969.  The metabolism of 2-furoic acid by Pseudomonas
       F2.  Biochem. J. 113:577-587.

147.  Tweedy, B. G., C. Loeppky, and J. A. Ross.  1970.  Metabolism of 3- (p-
       bromo-phenyl)-l-methoxy-l-methylurea by selected soil microorganisms.
       J. Agric. Food Chem. 18:851-853.

148.  Wagner, F., W. Zahn, and U. Buhring.  1967.  1-Hexadecene, an interme-
       diate in the microbial oxidation of n-hexadecane in vivo and in vitro.
       Angew. Chem. Internat. Ed. 6:359-360.

149.  Ward, D. M., and T. D. Brock.  1978.  Hydrocarbon biodegradation in hy-
       persaline environments.  Appl. Environ. Microbiol. 35:353-359.

150.  Watson, G. K., and R. B. Cain.  1975.  Microbial metabolism of the pyri-
       dine ring.  Metabolic pathways of pyridine biodegradation by soil bac-
       teria.  Biochem. J. 146:157-172.

151.  Willetts, A. J.  1973.  Microbial metabolism of alkylbenzene sulphonates.
       Fungal metabolism of 1-phenylundecane-p-sulphonate and 1-phenyldodecane-
       p-sulphonate.  Antonie van Leeuwenhoek J. Microbiol. Serol. 38:585-597.

152.  Willetts, A. J., and R. B. Cain.  1972.  Microbial metabolism of alkyl-
       benzene sulphonates.  Bacterial metabolism of undecylbenzene-p-
       sulphonate and dodecylbenzene-p-sulphonate.   Biochem. J. 129:389-402.

153.  Williams, P. A., and M. J. Worsey.  1976.  Ubiquity of plasmids  in cod-
       ing for toluene and xylene metabolism in soil bacteria:  evidence for
       the existence of new TOL plasmids.  J. Bacteriol. 125:818-828.

154.  Wright,  K.  A.,  and R.  B.  Cain.   1972.   Microbial metabolism of pyridi-
       nium compounds.   Metabolism of 4-carboxy-l-methylpyridinium chloride,
       a protolytic product  of  Paraquat.   Biochem.  J.  128:543-559.

155.  Yamada,  Y., H.  Motoi,  S.  Kinoshita,  N.  Takada,  and H.  Okada.   1975.
       Oxidative  degradation of squalene  by Arthrobacter sp.   Appl.  Microbiol.

156.  Yang, H.-C., M. Kusumoto, S. Iwahara, T.  Tochikura, and K.  Ogata.   1969.
       Studies on degradation of d-biotin by  microorganisms.   I.  d-Biotin
       degradation by yeasts and molds.   Agric. Biol.  Chem.  33:1730-1736.

                          ROLE OF  COMETABOLISM

                              Martin Alexander
                       Laboratory of Soil Microbiology
                           Department of Agronomy
                             Cornell University
                              Ithaca, NY 14853

                Cometabolism is the metabolism by a microbial popu-
           lation of a chemical which that species cannot use as a
           nutrient or an energy source.  The cometabolizing popu-
           lation does not increase in abundance in response to in-
           troduction of that chemical into natural ecosystems.  Be-
           cause the rate of transformation of many chemicals paral-
           lels the rate of population growth, a common consequence
           of cometabolism is a slow microbial transformation.  Fur-
           thermore, products typically accumulate in a cometabolic
           process, the accumulated metabolite being structurally
           similar to the original compound.  Examples of cometabo-
           lism are found among pesticides, chemicals used in the
           home, and products of importance to the manufacturing
           industry.  A probable explanation of cometabolism is
           the existence in the active population of an enzyme of
           broad substrate specificity, the initial enzymatic re-
           action yielding products that do not serve as substrates
           for other enzymes present in the responsible cells.  The
           possible role of cometabolism in nature is discussed.

     Microbiologists and biochemists have traditionally limited their atten-
tion in investigations of catabolism to a small number of compounds.  Although
it is widely believed that the number is large, the range includes but a few
sugars, amino acids and nucleotides and their polymers, fatty acids and a mod-
est assortment of simple aliphatic and aromatic hydrocarbons.  These studies
have resulted in an enormous and valuable literature, but much of the research
has little relevance to modern environmental problems.  The reason is that the
types of molecules of current environmental importance, and of concern to regu-
latory agencies, include several hundred pesticides as well as tens of thou-
sands of chemicals of industrial importance, few of which are structurally
similar to the common laboratory substrates.  The reason for the attention to
the number of nonpesticidal compounds is the enactment of the Toxic Substances
Control Act, which brings under regulatory view an array of molecules


heretofore not considered by biologists or environmental scientists.

     Environmental monitoring has clearly discredited the view that all or-
ganic molecules are biodegradable, and this fanciful notion was brought to a
sudden demise under the weight of the many new chemicals of potential environ-
mental significance.  Furthermore, the generally held view that all molecules
which microorganisms can metabolize will also provide them with a source of
carbon and energy has also fallen beneath the weight of new chemicals.

     Compounds in the ever-increasing assortment that are of potential regula-
tory concern can be divided into three broad groups.  First are those that can
be used as a source of carbon, some other nutrient element or energy.  When
these compounds are introduced into a natural ecosystem, the heterotrophs able
to grow on them proliferate, and the small initial population active on these
substrates increases in size with time.  Parallel to increase in numbers or
biomass of active cells is an ever-increasing rate of destruction of the mole-
cule.  Essentially all of the chemical will disappear in reasonably short pe-
riods of time, and the products are either carbon dioxide or molecules unre-
lated to the parent substance.  Second are those chemicals that are totally
recalcitrant.  These are not transformed microbiologically because no enzymes
can modify them or because they do not penetrate into microbial cells that
contain the requisite enzymes.  These molecules thus frequently persist, un-
less there is some nonbiological mechanism for their modification or downward

     The third group of compounds includes those subject to cometabolism.  Co-
metabolism refers to the metabolism by a microorganism of a chemical which
that orgaMsm cannot use as a nutrient or an energy source.  Although there is
some confusion and overlap with the term "cooxidation," the former term is
more desirable because many of the processes involved do not entail an oxida-

     Although most investigations of cometabolism have centered on compounds
that do not contain nutrient elements other than carbon, hydrogen and oxygen,
the substrate acted on by the cometabolizing population also might not serve
as a source of any nutrient which may be limiting in a particular environment;
thus, the organism may not use the molecule as a source of nitrogen, sulfur or
phosphorus, if these elements exist in the structure.

     Several outcomes of cometabolism are almost self-evident.   For example,
the compound that serves as a nutrient source provides the active populations
with a selective advantage, and they become more abundant in time.  However,
a chemical that does not serve as a nutrient does not give the organism acting
on it a selective advantage; hence, its cell density or biomass does not rise
with time.  This lack of population growth during cometabolism has rarely been
shown in nature or even in culture, but it would seem to be a natural conse-
quence of the phenomenon.  Furthermore, inasmuch as biochemical activity in a
natural ecosystem is roughly proportional to the number or biomass of cells
carrying out the specific transformation, the absence of an increase in popu-
lation size would be reflected in the lack of a rapid rise in the rate of bio-
degradation, as is typical of those substrates that support microbial growth.

     As an example, consider a synthetic organic compound that no population
in a particular environment can use as a carbon source.  Inasmuch as growth on
organic-molecules requires their conversion to precursors for biosynthesis of
amino acids, sugars, nucleic acids and fatty acids incorporated into cell con-
stituents, organisms failing to grow on the compounds in question must be un-
able to form these intermediates.  Were they able to do so, the populations
would proliferate.  If precursors required for biosynthetic purposes are not
generated, a product that the organism forms from the parent molecule would
likely accumulate.  This would be the end of the metabolic sequence in trans-
formation of the synthetic molecule.  Hence, a consequence of cometabolism is
the formation of one or more products that are structurally related to the
original substrate, rather than the typical products of complete degradation.
Such accumulations have been observed in cultures of many bacteria and fungi
provided with a wide range of organic compounds.

     In. some of our studies, for example, DDT and its analogues were found to
be transformed in cell suspensions or samples of water into a series of pro-
ducts similar to those reported by other investigators.  On the other hand, it
was observed that one of the two rings of the insecticide molecule could be
cleaved by a cometabolizing population to give a product with a single benzene
ring.  This compound, however, would not sustain growth of the original come-
tabolizimg population  (5, 6, 14, 15, 17).  In other of our studies, it was ob-
served that nitro aromatic compounds  (16), certain phenoxy pesticides (11),
and benzoates  (8, 9) were cometabolized by individual cultures, and in the pro-
cess, were converted to products quite similar in structure to the original
molecule  (Fig. 1, 2, 3).  Our studies in this connection are not unique,  and
many other microorganisms have been observed to cometabolize a multitude of
chemicals of dissimilar structures to yield products that the active species
could not further degrade  (3, 4, 7, 10, 12, 13, 21, 22) .  Chemicals acted on in
this manner include insecticides, herbicides, surfactants, aliphatic and aro-
matic hydrocarbons, as well as chemicals that are widely used in industry as
solvents or precursors for manufactured products.  The reactions involve de-
halogenation, introduction of hydroxyl groups, ring cleavage, epoxidation and
oxidation of methyl groups to the corresponding alcohols or aldehydes.  Hence,
many microorganisms have the capacity for cometabolism, the rangeof substrates
is truly impressive and a variety of biochemical reactions can be effected by
the responsible heterotrophs.

     How does one explain cometabolism?  For the teleologist, it is comforting
to believe that every enzymatic reaction that microorganisms catalyze does the
organism some good, but this is obviously sheer teleology and has no basis in
fact.  Many enzymatically catalyzed conversions are fortuitous, and the orga-
nism obtains no value from the reaction that it effects.  It is our view  that
the likeliest explanation is simply that the molecule undergoing cometabolism
is acted on by an enzyme that normally serves a physiological function; how-
ever, the products of the reaction with the substrate for cometabolism are not
further transformed by enzymes of the active populations.  Thus, should the
initial enzyme acting on a physiological substrate have a broad specificity,
it may  transform not only the normal substrate but related molecules.  In the
former  instance, the product of the reaction will be further utilized, but in
the latter situation, the product may not be transformed further because  the
succeeding enzyme has a more narrow substrate specificity and is unable  to


transform the product made  available  to it.  Some evidence exists to support
this view;  for example,  the enzyme cleaving 2,4-D acts on a number of related
phenoxy compounds to yield  the  corresponding phenols, but the enzymes further
degrading the phenols by the widely understood routes of aromatic metabolism
act on some,  but not all, of the phenols that are generated (2, 11).   Those
that are not substrates  for the phenol hydroxylating enzymes will accumulate,
and thus the preceding phenoxy  compound will be cometabolized and not support
«, '-0

                                     .	onp__
                                      	n	'_'	i..y
                        I      234567

Figure 2.  Conversion of DDT to ODD, DDE  and DBP  in waste water.  The three
           products are all chlorine-containing diphenylmethane derivatives.
           The samples received  (A) no addition,  (B) glucose or  (C) nonchlori-
           nated diphenylmethane.

     As one moves from the purity  of the  axenic culture  to  the marine environ-
ment, obvious shortcomings of the  in vitro  view of cometabolism become  evi-
dent.  Thus, products of cometabolism of  one species may be excellent sub-
strates for a second species, and  indeed  the second species may grow on the
product and thus lead to its rapid destruction.   Under these circumstances,
the product of cometabolism will not reach  appreciable or possibly  even detec-
table levels.  Thus, we have observed that  DDT cometabolism in culture  gives
rise to p-chlorophenylacetic acid.  The latter is not further degraded  by the
original population, but other organisms  have been isolated which can grow on
the product of cleavage of one of  the two rings of the DDT  molecule (14).
Still, the first phase of this biphasic process will probably be  very slow as
a rule because the organisms are not deriving energy from the conversion,

although the product of the first phase will be destroyed as quickly  as  it be-
comes available.  Not only insecticides like DDT but also other chemicals  may
be subject to this two-population effected degradation; thus,  2,3,6-trichloro-
benzoate is cometabolized to yield 3,5-dichlorocatechol (8).   The  latter com-
pound is itself a substrate for microorganisms that grow on related chlori-
nated aromatics (1), and thus presumably is destroyed totally  by  species able
to grow on it.
                             p-Chlorophenylglycol aldehyde
 Figure  3.   Products and proposed pathway for the cometabolism of  analogues  of
            DDT by Pseudomonas putida.
      What is  the  evidence  that cometabolism actually  functions  in  natural eco-
 systems?   Microbiolegists, with their propensity  for  wishful  thinking,  assume
 that  anything their pet  organisms do in  culture,  they will  also do in nature.
 Such  wishful  thinking may  in  fact reflect processes in natural  waters,  but
 meaningful evidence is clearly needed.   The best  evidence,  albeit  indirect,
 comes from studies in soil rather than water.  A  large number of chemicals,
 particularly  among the pesticides, when  added to  soil are converted to  pro-
 ducts similar to  the parent molecules.   The processes appear  to be microbial
 because the reaction is  abolished or markedly reduced in rate upon sterilizing
 the soil.   Inasmuch as the process is the result  of microbial processes or is
 enhanced  by microorganisms, the microbiologist endeavors to isolate species
 carrying  out  the  conversion.  In some cases he is successful; in other  instan-
 ces he is not.  The failure may be attributable to the absence  from the isola-
 tion  medium of an essential nutrient for the organisms with the correct array
 of enzymes; thus, most investigators working with synethetic  molecules  use en-
 richment  media containing  the test chemical as the sole carbon  source.   Yet,
 heterotrophs  that require  growth factors will not develop in  these media.  If

 growth factors are included in a  synthetic medium together with a somewhat re-
 sistant organic chemical as the major  carbon source,  the organisms that typi-
 cally develop in the enrichment culture  are those that use the growth factors
 as  carbon sources rather than synthetic  molecule.  Granting that the inability
 to  isolate an organism from a natural  environment in  which a process is occur-
 ring may thus frequently result from the absence of growth factors from the
 isolation medium, it is plausible that often the process in the natural eco-
 system arises not from the activities  of fastidious populations but rather
 from cometabolizing ones.  A case in point is the herbicide 2,4,5-T, which is
 converted to the corresponding phenol  (Fig.  4);  organisms growing on 2,4,5-T
 are rare or do not exist.  In addition,  many reasonably resistant organic com-
 pounds are degraded more rapidly  in  samples of natural environments to which
 are added readily utilizable energy  sources,  and it is possible that these
 energy sources are providing carbon  and  energy for the cometabolizing popula-
 tion so it can transform the subject chemical.   These lines of evidence are
 admittedly not strong, but they are  suggestive that cometabolism is responsi-
 ble for the transformation of many synthetic chemicals.
                 CM" 20
                          *    *-   •	*—«
-0.3 i
                                                          -0.2 C
                                                          -0.3 .
                                                           0.2 i
Figure 4.  Disappearance of 2,4,5-trichlorophenoxyacetate  (2,4,5-T)  and  forma-
           tion  of  the corresponding phenol in samples of  freshwater and soil
            (K. W. Sharpee and M.  Alexander, unpublished data).

      Many topics for investigation remain for the microbiologist, biochemist
 and environmental scientist.  For example, it is important to establish what
 chemicals and products are involved in this type of conversion, and which mi-
 crooorganisms are responsible.  Since cometabolic reactions probably are gen-
 erally slow  (unless the responsible populations are initially large and their
 enzymes are particularly active),  then concern with environmental quality dic-
 tates that means be sought to increase the rates.  Such an enhancement may
 come about because of an understanding of the process, or it may arise because
 of a technique discovered in an entirely fortuitous fashion.  Finally, because
 this is a phenomenon that appears to be widespread and of practical signifi-
 cance, information should be obtained on how and why it occurs.  With even
 some information in these various areas of inquiry, it is hoped we shall be in
 a better position to understand the processes concerned with microbial degra-
 dation of pollutants and in maintenance of the quality of marine environments.
                               LITERATURE CITED

 1.  Bollag, J.-M., G. G. Briggs, J.  E.  Dawaon, and M.  Alexander.   1968.  2,4-D
      metabolism.  Enzymatic degradation of chlorocatechols.   J.  Agric. Food
      Chem. 16:829-833.

 2.  Bollag, J.-M., C. S. Helling, and M. Alexander.  1968.   2,4-D metabolism.
      Enzymatic hydroxylation of chlorinated phenols.   J.  Agric.  Food Chem.

 3.  Bourquin, A. W.,  S. K.  Alexander, H. K. Speidel,  J.  E.  Mann,  and J. F.
      Fair.  1972.  Microbial interactions with cyclodiene pesticides.
      Develop. Ind. Microbiol. 12:264-276.

 4.  Flashinski, S. J., and E. P. Lichtenstein.  1974.   Degradation of Dyfonate
      in soil inoculated with Rhizopus arrhizus.   Can.  J.  Microbiol.  20:

 5.  Focht, D. D., and M. Alexander.   1970.  DDT metabolites and  analogs:   ring
      fission by Hydrogenomonas.  Science 170:91-92.

 6.  Focht, D. D., and M. Alexander.   1971.  Aerobic cometabolism of DDT ana-
      logs by Hydrogenomonas.  J. Agric. Food Chem.  19:20-22.

 7.  Hashimoto, K.  1973.  Oxidation of phenols by yeast.   II.  Oxidation of
      cresols by Candida tropicalis.   J. Gen.  Appl.  Microbiol.  19:171-187.

 8.  Horvath, R. S., and M.  Alexander.  1970.   Cometabolism:   a technique for
      the accumulation of biochemical products.  Can.  J.  Microbiol. 16:

 9.  Horvath, R. S., and M.  Alexander.  1970.   Cometabolism of m-chlorobenzoate
      by an Arthrobacter.  Appl. Microbiol. 20:254-258.

10.  Jensen, H. L.  1963.  Carbon nutrition of some microorganisms decomposing
      halogen-substituted aliphatic acids.  Acta Agric.  Scand.  13:404-412.


11.  Loos, M. A., R. N. Roberts, and M. Alexander.  1967.  Phenols as inter-
      mediates in the decomposition of phenoxyacetates by an Arthrobacter
      species.  Can. J. Microbiol. 13:679-690.

12.  Matsumura, F., V. G. Khanvilkar, K. C. Patil, and G. M. Boush.  1971.
      Metabolism of endrin by certain soil microorganisms.  J. Agric. Food
      Chem. 19:27-31.

13.  Patel, R. N., and D. S. Hoare.  1971.  Physiological studies of methane
      and methanol-oxidizing bacteria.  J. Bacteriol. 107:187-192.

14.  Pfaender, F. K., and M. Alexander.  1972.  Extensive microbial degrada-
      tion of DDT in vitro and DDT metabolism by natural microbial communi-
      ties.  J. Agric. Food Chem. 20:842-846.

15.  Pfaender, F. K., and M. Alexander.  1973.  Effect of nutrient additions
      on the apparent cometabolism of DDT.  J. Agric. Food Chem. 21:397-399.

16.  Raymond, D. G. M., and M. Alexander.  1971.  Microbial metabolism and co-
      metabolism of nitrophenols.  Pestic. Biochem. Physiol. 1:123-130.

17.  Subba-Rao, R. V., and M. Alexander.  1977.  Cometabolism of products of
      1,1,l-trichloro-2,2-bis(p-chlorophenyl)ethane (DDT) by Pseudomonas pu-
      tida.  J. Agric. Food Chem. 25:855-888.

18.  Suggs, J. , R. Hawk, A. Curley, E. Boozer, and J. McKinney.  1970.   DDT
      metabolism:  oxidation of the metabolite 2,2-bis(p-chlorophenyl)ethanol
      by alcohol dehydrogenase.  Science 168:582.

19.  Tober, C. L., P. Nicholls, and J. D. Brodie.  1970.  Metabolism and enzy-
      mology of fluorosuccinic acids.  II. Substrate and inhibitor effects
      with soluble succinate dehydrogenase.  Arch. Biochem. Biophys. 138:

20.  Tranter, E. K., and R. B. Cain.  1967.  The degradation of fluoro aro-
      matic compounds to fluorocitrate and fluoroacetate by bacteria.
      Biochem. J. 103:22P-23P.

21.  Walker, J. D., and J. J. Cooney.  1973.  Oxidation of n-alkanes by Clado-
      sporium resinae.  Can. J. Microbiol. 19:1325-1330.

22.  Willetts, A. J.  1974.  Microbial metabolism of alkylbenzene sulfonates.
      Oxidation of key aromatic compounds by a Bacillus.  Antonie van
      Leeuwenhoek 40:547-559.


                          T.-W.  Chou and N.  Bohonos
                            Life Sciences Division
                              SRI International
                            Menlo Park,  CA 94025

                Methyl parathion,  benzo[b]thiophene,  dibenzothiophene,
           quinoline,  benzo[f]quinoline,  and isoquinoline  are  bio-
           degraded by mixed culture  systems obtained from organisms
           in oligotrophic or eutrophic waters,  sewage effluents, or
           polluted estuarine waters.  Induction of degradative  en-
           zymes and/or cometabolism  with structurally-related
           chemicals is necessary  to  degrade some of  these compounds.
           The biodegradability of these  substances can be affected
           when nutrients that are assimilated more readily are
           present in the medium.   These  facts should be considered
           in the development of ecosystems  to be used for assessing
           whether pollutants are  biodegraded.

     Previous papers presented in this workshop described degradative pathways
and cometabolism involving pure culture systems.   Presently, we  are  reporting
on studies primarily concerned with mixed culture  biodegradation during the
first stages of relatively large-scale enrichment  processes.   These  latter pro-
cedures may be more representative of environmental  conditions than  pure cul-
tures and small flask processes.   We also studied  mixed  enrichment cultures
that were serially propagated on inorganic salts/substrate media,  following
which commonly used microbial nutrients were added.   In  this way,  we could ob-
tain an insight into the effects of simultaneous pollution by  other  nutrients.
The substrates used were methyl parathion (MP), benzo[b]thiophene  (BT), di-
benzothiophene (DBT),  quinoline (Q),  isoquinoline  (IQ),  and benzo[f]quinoline
(BQ).  Their structures are presented in Figure 1.   Naphthalene  (N)  and anthra-
cene (A) were used as  inducers and/or cometabolic  substrates.

     In previous studies on the microbial degradation of parathion,  both pure
cultures (25, 26) and mixed cultures (1, 18) readily degraded the  insecticide.
The metabolic pathways (17, 19) and the properties  of parathion hydrolase (16)
have been defined.  Several reports also indicated that  MP was biodegraded in


waters or soils  (2, 5, 14, 15, 21, 31).   Our prior studies (1, 4, 27) with mixed
culture systems  suggested  that  not all organisms degraded MP.   In the current
study, we have demonstrated  that problems  may be encountered if other readily
assimilable nutrients  are  present  in the media.
Figure 1.  Compounds used  for  the development of biodegrading cultures by en-
           richment techniques.
     Aerobic degradation  of  DBT has been investigated extensively (7,8, 10-13,
29, 30) but most os  these  studies were conducted with pure cultures.   Aerobic
degradation of BT was  reported less frequently (1, 24, 27-29)  and only under
mixed substrate conditions.   These may have involved cometabolic phenomena but
this was not definitely established.   Suggestive evidence for biodegradation
of Q in soils was obtained in 1916 (23)  and 1917 (6).   However, before our in-
investigations  (1, 27)  no  reports were published on  aerobic microbial degrada-
tion of IQ or BQ.   Q,  IQ, and BQ as well as BT and  BDT coexist with many types
of polycyclic aromatic compounds in coal tars and crude petroleum.  Conse-
quently, studies on the biodegradation of these heterocyclic aromatic compounds
are of environmental importance.
                            MATERIALS  AND METHODS
     MP was obtained from Chem Service (West Chester, Pennsylvania);  BT (thia-
naphthene) and DBT were obtained from Aldrich Chemical (Milwaukee, Wisconsin).
All were recrystallized from methanol.  For enrichment studies, Q  (Aldrich)
was used without further  purification, but was purified by gas chromatography
(gc) to produce reference material  for analysis.   BQ (Eastman, Rochester, New
York), IQ  (Aldrich), N  (Mallinckrodt,  St.  Louis,  Missouri), and A  (Sigma, St.

Louis, Missouri) were used without further purification.


     Eutrophic water samples were obtained from two sources:  Coyote Creek in
San Jose, California, and a pond near Searsville Lake in Woodside, California.
Sewage effluents were obtained from treatment plants with different feed char-
acteristics  (Palo Alto, California; South San Francisco, California; Shell Oil
Refinery, Martinez, California; and Monsanto Chemical Company, Anniston, Ala-
bama) .  Oligotrophic water samples were obtained from Lake Tahoe, California.
Polluted estuarine water was collected from the vicinity of the San Mateo
Bridge that crosses San Francisco Bay.


     Water samples with sediment present were placed in large sterile collec-
tion  reservoirs for about one h to allow the sediment to settle by gravity.
The supernatants were siphoned and filtered through fine-mesh polyester cloth
to remove insects and particles that did not settle.  In preliminary studies
we had higher success rates with large volumes, as compared with shaker flask
scale enrichment.  Consequently, 4-liter volumes of filtered water samples
were  added to sterile 9-liter glass fermentors that contained one liter of a
solution of 0.5 g of  (NHiJ 2SOi,.  [NH^NOa was substituted in BT and DBT enrich-
ments] and 10 g of Kt^POit-K^HPOit buffer to maintain the solution at pH 7.  Tes,t
chemicals for enrichment studies were added as concentrated aqueous solutions
(Q, IQ, BQ hydrochloride, and sterile microbial nutrients as indicated) or as
concentrated solutions in dimethyl sulfoxide (MP, BT, DBT, N and A) and mixed
thoroughly in the diluted water samples.  The initial concentrations of test
compounds was 10 yg ml""1 .  The 9-liter fermentors were fitted for sterile
aeration, air exhaust, sampling, and addition of other nutrients as needed.
Incubations were carried out at 25°C except for the Lake Tahoe water samples
that were maintained at 5-10°C while being transported to the laboratory and
then were incubated at 15°C.  Cultures were continuously aerated, and samples
were  taken for analyses at frequencies based on our past experiences with the
test  compounds.  Sampling intervals varied from 2 to 48 h.  Cultures were in-
cubated for up to 6 weeks.  When BT and/or N were used, they were added to the
fermentors at 2-day intervals because control tests indicated that during that
time  these compounds were removed from the culture media by aeration.

     When 50-70% of a test compound was degraded, enrichment cultures were
transferred to shaker flasks containing basal inorganic salts media with dif-
ferent concentrations of the test compound.  Each liter of basal salts medium
contained:  1.4 g of K2HPO4, 0.6 g of KH2POi,, 0.5 g of (NHit)2SOit, 0.1 g of
NaCl, 0.1 g of MgSOit-7 H2O, 0.02 g of CaCl2«2 H20, 0.005 g of FeSO4.7 H20, and
1 ml of trace elements solution.  The latter contained 0.1 g of H3BO3, 0.05 g
each of CuSO^-5 H20, MnSO4'H2O, ZnS04'7 H2O, Na^oO^ and CoCl2«6 H2O liter"1.
In studies in which BT or DBT were substrates,  (NHI+)2S01+ was replaced by
NHi»N03 and other sulfate salts were replaced with the corresponding chloride
salts.  Shaker flasks containing BT and N were capped with Teflon-lined screw-
on caps.

     Separate rotary shaker incubators were used for each individual substrate
to prevent contamination of shaker flask media by the vapors of readily


metabolized substrates  (e.g., N, BT) or by metabolites from transformations.


     Absorption measurements of broths or extracts at critical differential
(peak and lower value) uv wavelengths were particularly useful for rapid analy-
ses of the many degradations being conducted.  Combinations of wavelengths used
were:  270 and 300 nm for MP; 296 and 300 nm for BT; 290 and 300 nm for DBT;
312 and 320 nm for Q; 345 and 355 nm for BQ; and 316 and 324 nm for IQ.  Sam-
ples from critical points in the degradation curves were always analyzed by gc
or high-performance liquid chromatography (hplc).

     Samples for analyses were extracted twice with equal volumes of ethyl ace-
tate or hexane.  The extracts were adjusted to fixed volumes and dried with
NaaSOi,.  Aliquots of the extracts were injected, directly or after being evapo-
rated under nitrogen, into gc or hplc analyzers.  MP analyses were made with a
Varian Associates Model 5711 with a APIS detector  (Varian Model 2740); the
column was 5 ft x 0.25 in. packed with 4% SD-30 and 6% OV-210 on 80/160 mesh
Gas Chrom Q; column temperature, 200°C; Na carrier gas at 33 ml, min"1; reten-
tion time, 3.73 min.  Ethyl parathion was used as the internal standard.  A
Spectra Physics Model 3500 analyzer was used for hplc analyses of heterocyclic
hydrocarbons.  For BT, a \i Bondapack Cie (30 cm x 4 mm) (Waters Associates)
column was used with a mobile phase consisting of 60% methanol and 40% water,
and a Schoeffel Variable uv detector set at 228 nm (sensitivity, 500 pg).  For
DBT, the column was Spherisorb ODS  (24 cm x 6 mm)  (Spectra Physics) with a mo-
bile phase containing 55% methanol and 45% water, and a Spectraphysics 230 uv
Detector set at 254 nm  (sensitivity, 20 ng).  For Q and IQ, the column was y
Bondapack Cie  (30 cm x 4 mm) (Water Associates) with a mobile phase composed
of 60% methanol and 40% water.  A Schoeffel Variable uv Dectector was set at
220 nm (sensitivity, 100 pg).  For BQ, the column was Spherisorb ODS (24 cm x
5 mm)  (Spectraphysics) with a mobile phase consisting of 55% methanol and 45%
H2O.  A Schoeffel Variable Fluorescence Detector was used at 360 nm  (sensi-
tivity, 4 ng) .

                           RESULTS AND DISCUSSION


     MP-degrading organisms were present in all eutrophic water bodies and
sewage effluents (1, 4, 27).  To obtain an adequate supply of cells for deter-
mining the kinetic constants defining the degradation of malathion, Paris et
al.  (22)  grew degrading organisms in a medium containing malathion and 1:10
diluted nutrient broth.  Because one of our assignments was to obtain rate
constants with mixed enrichment cultures from natural water bodies, we also
required adequate concentrations of cells.  Therefore, we investigated the ef-
fect of nutrient broth on a MP-degrading mixed culture system transferred more
than 10 times in an inorganic salts/MP medium after the original enrichment
degradation.  Table 1 shows MP degradations when cultures were grown:   (I) in
shaker flasks one time in an inorganic salts medium with MP (50 yg ml" ),
(II) in nutrient broth  (Difco)  diluted 1:10 and with MP (50 yg ml"1), and
(III) when the mixed culture was transferred daily three times in  the diluted
nutrient broth/MP medium used in (II) and then the fourth transfer was


incubated for 86 h in a similar medium.


                                      Serial     Incubation time            .
                                                _ ...  ,  .      ,-    Degradation
            Medium                  transfers  of final transfer    3  .
                                    (numbers)         (h)
Inorganic salts + MP
(50 yg ml'1)
Nutrient broth (0.1 cone.) + MP
(50 yg ml"1)
Same as (II)
 *The inoculum used at the initiation of this experiment had already been
transferred more than 10 times on inorganic salts/MP medium.

**The first 3 transfers for (III) were made daily.
     After one transfer in the diluted nutrient broth/MP medium (II) ,  MP deg-
radation increased slightly in the 21-h incubation compared with the control
degradation in inorganic salts/MP medium (I) .   The nutrient broth supported
growth of a greater number of cells.  However, when the organisms were trans-
ferred four times in the diluted nutrient broth/MP medium (III), MP-degrading
capacity was markedly diminished.  This was interpreted as being due to ingre-
dients in the nutrient broth that favored growth of organisms other than those
involved in degradation, leaving few of the primary MP-degrading organisms and/
or exoenzymes involved in modifying the structure of MP.


     Table 2 presents the data from enrichment studies conducted with BT.  Be-
cause of the volatility of BT, degradations were considered to be positive
when transfers made to screw-capped flasks containing inorganic salts medium
with BT, or BT plus additive, indicated BT degradation.  When BT was the sole
substrate, no degradation was observed in six water samples during 6 weeks of
incubation.  Naphthalene (N) was selected as a cosubstrate because it coexists
with BT in coal tars and petroleum products and has a similar chemical struc-
ture.  BT degradation was observed in four of five water samples to which N
was added with BT at 10 yg ml"1 levels every 2 days.  The one sample that did
not develop a BT-degrading culture within 6 weeks under this mixed substrate
condition was the oligotrophic sample from Lake Tahoe.  When BT, glucose, and
yeast extract were added to one water sample in a 9-liter fermentor, BT degra-
dation was not observed.  In other studies (27) , we found that the breakdown
was a cometabolic process in which N could not be replaced by quinoline  (Q)
using cultures that had been transferred many times on salts/N/BT medium.
Similarly, if such a culture mixture was transferred to media in which N was
replaced by tryptone or glucose plus yeast extract, BT degradation did not
occur (27).


   TVJJ-J--   j_            ,             Time for development         BT
   Addition to water samples                 _   , .            ,  .  ,     ,  . .  4.
                        ^                   of cultures       biodegradation*
BT alone
BT + naphthalene (N)
BT + glucose + yeast extract
— t
5-8 days
— t
*Number of water samples that developed BT-biodegrading cultures/number of
different water samples tested.
  id not develop within 6 weeks of incubation.

     Naphthalene  (N) or anthracene  (A) were added to DBT because they coexist
with DBT in wastes  from the coal and petroleum industries and because they
have similar structural characteristics.  The selection of these two aromatic
hydrocarbons was  arbitrary for many other compounds could have been selected.
The data in Table 3 indicate that even though DBT-degrading organisms could be
obtained by adding  DBT as the sole substrate, N was an inducer in development
of DBT-degrading  organisms.  The presence of N caused more rapid degradation
of DBT.  Although only one enrichment was initiated with the DBT/A combination,
the data suggest  that A may also be an inducer of enzymes used in DBT degrada-
tion.  All five mixed cultures developed in the DBT plus N substrate enrich-
ments were able to  degrade DBT on subculturing in media with decreasing con-
centrations of N.

                                                  Time for DBT
       Additions to water samples                 degradation             .  ^
                                                     (days)        Sgra

Eutrophic waters and sewage effluents
  DBT alone                                           17-22           3/4
  DBT + naphthalene  (N)                                2-5            3/5
  DBT + anthracene  (A)                                  8            1/1
Oligotrophic waters
  DBT alone                                            —t           0/1
  DBT + N                                              32            1/1

*Number of water samples that developed DBT-biodegrading cultures/number  of
different water samples tested.

     not develop within 6 weeks.



     Cultures  that  could degrade Q were readily obtained.  Figure 2 illus-
trates the time course  of degradations using procedures designated as "Cascade
Batch Fermentations"  (27).  Two concentrations of Q were used.  During the
first 2 days of incubation, no degradation of Q occurred, following which Q
was totally degraded  in both cultures.  The latter were inoculated (0.2% v/v)
into shaker flasks  containing freshly collected water samples mixed with one-
fifth volumes  of (NHi,) 2 SO i,-potassium phosphate buffer and the respective ini-
tial concentrations of  Q.  These processes were repeated four times.   In all
instances, short lag  periods occurred followed by rapid degradations  of Q.
These Cascade  Batch Fermentations suggest that, under natural conditions, a
lag period would occur  in which Q would be degraded downstream from the source
where it was first  introduced, but gradually degradation would occur  increas-
ingly closer to the source of pollution.  This was demonstrated in a  multi-
chamber fermentation  unit with p-cresol as the substrate (3).
           9-liter BOTTLE
  - 8
  01 c
  a. 6
  O 2
                                     TIME (days)
Figure 2.  Cascade  Batch Fermentations with a freshly developed degrading sys-
           tem transferred  in fresh pond water media.

     The results with BQ  and quinoline  (Q) as pollutants in Table 4 are essen-
tially similar to those in Table 3.  Q was the inducer of enzymes that degraded
BQ.  However,  in contrast to the data obtained with DBT (Table 3),  even after
more than 6 weeks of induction and sub-transfer, cometabolism was still in-
volved in rapid BQ degradation by three of the six culture mixtures that ini-
tially degraded BQ readily in the presence of Q.  Again, in the presence of
microbial nutrients such  as glucose plus yeast extract, no degradation of BQ

was observed after 6 weeks of incubation of the one water sample tested.  The
data in Figure 3 were obtained with one mixed culture transferred 30 times in
shaker flasks containing basal inorganic salts medium plus BQ and Q.  The
presence of Q accelerated degradation of BQ, but with this culture mixture,
only 30% of the BQ was degraded in the absence of Q.  Addition of glucose plus
yeast extract to the medium completely eliminated BQ degradation.  One trans-
fer in trypticase soy broth plus Q medium still resulted in BQ degradation in
the absence of Q, but with several transfers, results similar to those with MP
(Table 1) probably would have been obtained.
         Addition to water samples
Time for BQ
Eutrophic waters and sewage effluents
  BQ alone
  BQ + quinoline (Q)
  BQ + glucose + yeast extract

Oligotrophic waters
  BQ alone
  BQ + Q

*Number of water samples that developed BQ-biodegrading cultures/number of
different water samples tested.
'Did not develop within 6 weeks.
                                      2      3
                                      TIME (days)
Figure 3.  Biodegradation of benzoff]quinoline inoculated with a previously
           isolated benzo[f]quinoline and quinoline degrading mixed culture.
           Supplements to inorganic salts:  BQ + Q (•); BQ alone (•); BQ +
           glucose + yeast extract (o), and BQ + trypticase soy broth (A).

     The experiments in Figure 4 were conducted with  estuarine water.  Q was
degraded after a lag period and BQ did not  interfere  with Q degradation.  How-
ever, Q was necessary to induce enzymes for BQ degradation, functioning as a
cometabolic substrate.  Initially, BQ degradation began  at about 12 d (point
A) and ceased when Q was completely degraded.  The  addition of more Q on day
15 initiated a second degradation of BQ, commencing at about 17 d (point B).
Naphthalene (N) also functioned as an inducer and cometabolic substrate in BQ
degradation in estuarine water (Fig. 5).
                  10 —i
                                    TIME (days)              LSJJ|

Figure 4.  Biodegradation of guinoline and benzo[f]quinoline in estuarine
           water.  Q in waters supplemented with Q alone  (•) or BQ plus  Q (D)
           BQ in waters supplemented with BQ  (A) alone, BQ plus peptone  (A),
           or BQ plus Q (o).  Additional Q was added to BQ + Q supplemented
         ,,  waters at 15 days.
                   5 5
                                                  J	L
                                 5          10
                                    TIME (days)

Figure 5.  Biodegradation of benzo[f]quinoline in estuarine water with  naphtha
           lene as a cometabolic substrate.  N (10 yg ml"1) was  added at 0,
           2.5, 5, and 7 days (i).

     As indicated in Figure 6, development of isoquinoline(IQ)-degrading orga-
nisms in saline estuarine water appeared to be considerably slower than devel-
opment of Q-degrading cultures with Q as the added substrate.  IQ degradation
apparently was facilitated by the presence of BQ, but this may be an artifact
in that particular fermentor.  BQ was not degraded after 56 days in the pres-
ence of IQ.  In this instance, IQ was not a cometabolic substrate for BQ deg-
Figure 6.  Biodegradation of isoquinoline and benzoff]quinoline in est    le
           water.  IQ in waters supplemented with IQ alone  (•) or BQ plus IQ
            (O).  BQ in waters supplemented with BQ plus Q  (o).

     Data are presented here that demonstrate the problems of determining the
biodegradability or recalcitrance of compounds in natural waters.  These ob-
servations could be particularly significant in some ecosystems.  Compounds
with somewhat similar structural features can be important as inducers and co-
metabolic substrates in developing enrichment cultures and possibly for isola-
tion of pure strains of organisms for studies of metabolic pathways.  Naphtha-
lene has been an excellent inducer and cometabolic substrate in degradation of
the aromatic heterocyclic compounds studied in this report.

     The presence of readily assimilated microbial nutrients inhibits forma-
tion and maintenance of mixed culture-degrading systems.  These observations
may have to be considered in assessing the effectiveness of sewage treatment
plants that process domestic and industrial feeds and of ecosystems that leach
organic nutrients into aqueous phases.

     We  thank  the  applied microbiologists and analytical chemists at SRI Inter-
national who participated in these studies.  This work was supported in part by
the U.S. Environmental Protection Agency under Contract No. EPA-600/7-77-13 and
by internal SRI  International funds.


                               LITERATURE CITED

 1.  Bohonos,  N.,  T.-W.  Chou,  and R.  J.  Spanggord.   1977.   Some observations on
      biodegradation of pollutants in aquatic systems.   Japan. J.  Antibiot.

 2.  Bourquin, A.  W.,  M.  A.  Hood,  and R.  L.  Garnas.   1977.   An artificial mi-
      crobial  ecosystem for  determining  effects and  fate of toxicants in a
      salt-marsh  environment.   Dev.  Ind.  Microbiol.  18:185-191.

 3.  Chou,  T.-W.,  S. S.  Lee, R. J. Spanggord, and N.  Bohonos.   1976.   Con-
      tinuous-flow compartmentalized fermentation units in  environmental deg-
      radation studies.   172nd National  Meeting,  American Chemical Society.
      Abstract MICR 16.

 4.  Chou,  T.-W.,  R. Spanggord, E. Shingai,  and N. Bohonos.   1976.  The fate of
      methyl parathion in fresh water systems.   III.  Biodegradation studies.
      172nd National Meeting,  American Chemical Society.

 5.  Eichelberger, J.  W., and J.  J.  Lichtenberg.   1971.  Persistence  of pesti-
      cides in river water.   Environ. Sci. Technol.  5:541-544.

 6.  Funchess, M.  J.  1917.   The nitrification of pyridine,  quinoline, quani-
      dine  carbonate,  etc.,  in soils. Alabama Agric.  Expt.  Sta.,  Bull. 196;
      Tech. Bull.  3:65-82; Chem.  Abstra.  11:3362  (1917).

 7.  Gibson, D. T.  1975. The microbial  degradation  of aromatic petroleum pro-
      ducts, Annual Report No. 2.   Prepared  for Office  of Naval Research under
      Contract No. NOOO14-67-A-0126-0020,  University  of Texas  at Austin (30
      June  1975).

 8.  Hou, C. T.,  and A.  I. Laskin.  1975.  Microbial  conversion of dibenzothio-
      phene.  Dev. Ind.  Microbiol. 17:351-362.

 9.  Hsieh, D. P.  H.,  and M. Munnecke.   1972.  Accelerated microbial  degrada-
      tion  of  concentrated parathion. Proc.  IV Ferm.  Symp.  Ferment.  Technol.,
      pp. 551-554.

10.  Knecht, A. T., Jr.   1962.  Microbial oxidation of  dibenzothiophene and its
      possible application in the desulfurization of  coal and  petroleum.   Diss.
      Abst. 22:3350.

11.  Kodama, K.  1977.  Co-metabolism of dibenzothiophene by Pseudomonas
      jianii.   Agric.  Biol.  Chem.  41:1305-1306.

12.  Kodama, K.,  S. Nakatani,  K.  Umehara,  K.  Shimizu, Y. Minoda, and  K. Yamada.
      1970. Microbial conversion of petro-sulfur compounds.   III. Isolation
      and identification of  products from dibenzothiophane.  Agric. Biol.  Chem.

13.  Kodama, K., K. Umehara, K. Shimizu, S. Nakatani, Y. Minoda, and K. Yamada.
      1973.  Identification of microbial products from dibenzothiophene and
      its proposed oxidation pathway.  Agric. Biol. Chem. 37:45-50.

14.  Lichtenstein, E. P., and K. R. Schulz.  1964.  The effects of moisture
      and microorganisms on the persistence and metabolism of some organophos-
      phorous insecticides in soils with special emphasis on parathion.
      J. Econ. Entomol. 57:618-627.

15.  Maleszewska, J.  1974.  Degradation of methyl parathion by microorganisms
      occurring in surface water and sewage.  Pol. Arch. Hydrobiol. 21:163-171.

16.  Munnecke, D. M.  1976.  Enzymatic hydrolysis of organophosphate insecti-
      cides, a possible pesticide disposal method.  Appl. Environ. Microbiol.

17.  Munnecke, D. M., and D. P. H. Hsieh.  1974.  Microbial decontamination of
      parathion and p-nitrophenol in aqueous media.  Appl. Microbiol.

18.  Munnecke, D. M., and D. P. H. Hsieh.  1975.  Microbial metabolism of a
      parathion-xylene pesticide formulation.  Appl. Microbiol. 30:575-580.

19.  Munnecke, D. M., and D. P. H. Hsieh.  1976.  Pathways of microbial metabo-
      lism of parathion.  Appl. Environ. Microbiol. 31:63-69.

20.  Nakatani, S., T. Akasaki, K. Kodama, Y. Minoda, and K. Yamada.  1968.
      Microbial conversion of petro-sulfur compounds.  II. Culture conditions
      of dibenzothiophene-utilizing bacteria.  Agric. Biol. Chem.  32:1205-1211.

21.  Naumann, K.  1967.  Decomposition of parathion by soil bacteria.
      Phytopathol. Z. 60:343-357.

22.  Paris, D. F., D. L. Lewis, and N. L. Wolfe.  1975.  Rates of degradation
      of malathion by bacteria isolated from aquatic systems.  Environ. Sci.
      Technol. 9:135-138.

23.  Robbins, W. J.  1916.  The cause of the disappearance of coumarin, vanil-
      lin, pyridine and quinoline in the soil.  Science 44:894-895.

24.  Sagardia, F., J. J. Rigau, A. Martinez-Lahoz, F. Fuentes, C.  Lopez, and
      W. Flores.  1975.  Degradation of benzothiophene and related compounds
      by a soil pseudomonas in an oil aqueous environment.  Appl.  Microbiol.

25.  Sethunathan, N., and T. Yoshida.  1973.  Parathion degradation in sub-
      merged rice soils in the Philippines.  J. Agric. Food Chem.  21:504-506.

26.  Sidderamappa, R.,  K. P. Rajaram, and N. Sethunathan.  1973.  Degradation
      of parathion by bacteria isolated from flooded soil.  Appl.  Microbiol.

27.  Smith,  J.  H.,  W.  R.  Mabey,  N.  Bohonos,  B.  R.  Holt,  S.  S.  Lee,  T.-W.  Chou,
      D. C.  Bomberger, and T.  Mill.   1977.   Environmental pathways  of selected
      chemicals in freshwater  systems.   Part I:   Background and experimental
      procedures.   EPA Publ. No.  EPA-600/7-77-113.   Part II:   Laboratory
      studies.   EPA Publ.  No.  EPA-600/7-78-074.

28.  Walker, J. D., and R.  R.  Colwell.   1974.   Microbial petroleum  degradation:
      use of mixed hydrocarbon substrates.   Appl.  Microbiol.  27:1053-1060.

29.  Walker, J. D., R. R.  Colwell,  and  L. Petrakis.   1975.  Microbial petro-
      leum degradation:  application of computerized mass spectrometry.   Can.
      J. Microbiol. 21:1760-1767.

30.  Yamada, K., Y. Minoda, K.  Kodama,  S. Nakatani,  and  T.  Akasaki.   1968.
      Microbial conversion of  petroleum compounds.   I. Isolation  and  identifi-
      cation of dibenzothiophene utilizing bacteria.  Agric. Biol.  Chem.

31.  Yasuno, M., S. Hirakoso,  M.  Sasa,  and M. Uchida.  1965.   Inactivation of
      some organophosphorous insecticides by bacteria in polluted water.
      Japan. J. Exp. Med.  35:545-563.


                      C. 0. Patterson  and G. D. Hegeman
                          Department of Microbiology
                              Indiana University
                             Bloomington, IN 47401

                 A Corynebacterium strain was isolated from soil by
            enrichment at the expense of cyclopropanecarboxylate (CPC) .
            Substrate utilization and oxygen uptake patterns indicate
            that the pathway for use of CPC by this bacterium is in-
            ducible and differs from that previously described for
            the fungus, Fusarium, and liver.  Properties of mutants
            that have lost the ability to grow at the expense of CPC
            and their revertants implicate in the oxidation of CPC
            portions of a pathway previously described as functional
            in the degradation of valine by bacteria.  Measurement of
            enzyme activity in crude extracts confirms this conclu-
            sion.  The mode of initial attack upon the cyclopropane
            ring remains obscure.

      Alicyclic  (cycloparaffin and naphthene) compound occur widely in orga-
 nisms and in organic deposits, such as petroleum, derived from them.  Some  of
 the most interesting of such compounds are cyclopropane and its derivatives.
 The three-membered ring structure present in these compounds is highly
 strained, and displays certain chemical similarities to a simple carbon-carbon
 double bond; for instance, it is readily reactive with bromine.  Further,  the
 unsaturated analog of cyclopropane, namely cyclopropene, has been shown to
 serve as substrate for bacterial nitrogenase, an enzyme capable of reducing
 triple bonds (6).

      Cyclopropane carboxylic acid  (CPC) is of interest as a model compound  and
 as a substance, the metabolism of which is of interest in its own right.   Com-
 pounds that are derivatives of, contain, or may yield CPC occur in petroleum,
 pyrethrins, hypoglycins, bacterial fatty acids, a miticide now being developed
 for commercial use (cycloprate:  the hexadecyl ester of CPC), and other
       Present address:  Department of Biological Sciences,  Tucker Hall,  Uni-
versity of Missouri, Columbia, MO 65201.

compounds proposed for use or in use.  Examples are a coccidiostat  (cyproquin-
ate),  an herbicide (cypromid),  a tranquilizer (prazepam), and several narcotic
antagonists (naltrexone, cyclazocine, and oxilorphan) (11, 13).

     CPC can serve as sole carbon and energy source for growth of some micro-
organisms with complete degradation of the cyclopropane ring.  Two mechanisms
for this degradation have been identified.  Tipton and co-workers (5, 14)
found that Ochromonas danica and Tetrahymena pyriformis mineralize cyclopro-
pane derivatives by a simple modification of |3-oxidation, in which the ring is
treated as a double bond.  Quite a different mechanism was identified in the
fungus Fusarium oxysporum.  Chung and co-workers (3, 9, 10) showed that cleav-
age of the ring occurred by direct addition of the elements of water across
one of the bonds adjacent to the a-carbon.  This attack yields y-hydroxybuty-
ric acid, which is then degraded by known routes.  The present paper reports
still another route for degradation of the cyclopropane ring in bacteria.
                            MATERIALS AND METHODS


     A Corynebacterium sp., designated strain CPC-1,  was isolated by liquid
enrichment using a mineral medium (7) with CPC as sole source of carbon and
energy.  For other purposes this medium was modified by inclusion of 1% (w/v)
lonagar No. 2  (Colab, Inc.) and by replacement of 0.2% Na-CPC by other carbon
sources.  Growth was measured at 30°C in side arm-fitted Erlenmeyer flasks
using a Klett colorimeter (No. 66 filter).


     Mutants were isolated from independent cultures following treatment with
various concentrations of l-methyl-3-nitro-l-nitrosoguanidine (Aldrich Chemi-
cal Co.) in growth medium, several generations of growth in mutagen-free me-
dium for expression, and plating on solid medium containing 0.2% Na-CPC and
0.01% Na-succinate.  Small colonies were picked and screened for loss of the
ability to grow at the expense of CPC.


     All organic chemicals used as substrates were obtained from Sigma Chemi-
cal Co., with the exception of iso-butyric acid (Fisher Scientific Co.), CPC,
3-hydroxybutyric acid, methacrylic acid  (Aldrich Chemical Co.),  1-hydrocybuty-
ric acid (K & K Laboratories), cyclopentane carboxylic acid (PCR Incorporated)
and cyclobutane carboxylic acid (ICN Pharmaceuticals, Inc.).  Inorganic chemi-
cals were all of reagent grade.  Jso-butyryl-CoA was synthesized by the mixed
anhydride method (1) using freshly distilled reagents.


     Gas chromatographic measurement of CPC and other volatile acids in acidi-
fied (HsPOiJ aqueous media was performed with a Hewlett-Packard Model 402


instrument fitted with a 2 m x 2 mm "U" tube column of 10% SP-1200, 1%
on 80-100 mesh Chromosorb WAW (Supelco, Inc.) at 105°C.

     Oxygen uptake was determined polarographically with resting cell suspen-
sions at 30°C in the buffer portion of the mineral growth medium (above)  using
a Yellow Springs Instrument Co.  Model 53 oxygen monitor connected to a Hewlett-
Packard Model 7128A recorder.

     Cell extracts for enzyme assay were prepared by sonic extraction at 0°C
of a 10% wet weight suspension of cells in 0.1 M phosphate buffer  (pH 7.2).
Following extraction, unbroken cells and debris were removed by centrifugation
at 10,000 g for 10 min at 0°C.  Jso-butyryl-CoA dehydrogenase was measured ac-
cording to Green et al. (2).  Protein in extracts was measured using Pardee's
modification of the biuret reagent  (4); bovine  serum albumin (Sigma Chemical
Co.) served as standard.

                           RESULTS AND DISCUSSION

     CPC is metabolized by strain CPC-1 with the accumulation in the culture
fluid of compounds detectable by gas chromatography using a flame ionization
detector.  Accordingly, it was decided to search for evidence of the pathway
being used by means of a study of blocked mutants, substrate utilization, and
oxygen uptake patterns.  Before it was possible to undertake either of the
latter two approaches, it was necessary to consider which compounds might be
intermediates in the mineralization of the cyclopropane ring.

     Cleavage of the cyclopropane ring of CPC could yield various products,
depending on the site and mode of attack.  These are summarized in Figure 1.
Hydrogenation of either of the bonds adjacent to the a-carbon would give n-
butyrate, while hydrolysis of those bonds would give either a-hydroxybutyrate
or Y~hydroxybutyrate  (Chung  [10] found that F. oxysporum produces yhydroxy-
butyrate).  On the other hand, attack upon the bond opposite the a-carbon
yields iso-butyrate or a related compound such as methacrylate or 3-hydroxy-
isobutyrate.  The formation of metacrylate from CPC would be an internal ar-
rangement, in which the strained ring configuration is resolved into a true
double bond.  Such a dismutation could also involve one of the adjacent bonds,
yielding 2-butenoate or 3-butenoate.  Still another possibility is the oxyge-
nation of the cyclopropane ring to the corresponding lactone (butyrolactone).

     The results of growth experiments presented in Table 1 indicate that deg-
radation of CPC proceeds neither by attack at a bond adjacent to the a-carbon,
or by lactonization of the ring.  Neither n-butyrate or its hydroxy deriva-
tives supported growth of the bacterium.  Butyrolactone was not attacked.  In
contrast, iso-butyrate, metacrylate, valine and 2-oxoisovalerate all supported
growth of the organism at rates equal to or slightly greater than the growth
rate on CPC.  The unsaturated acid, 2-butenoate, also supported growth, though
only after a lengthy lag and at sluggish rates.  Growth on 3-butenoate was at-
tended by no lag, but again growth was extremely slow.

     To distinguish between intermediates involved in degradation of CPC and
fortuitous use of putative intermediate compounds, measurement of oxygen up-
take in suspensions of resting cells grown on these compounds was examined.


                                                  iso -butyric
Figure 1.  Possible modes  of attack upon cyclopropanecarboxylic acid  (CPCO,
           and the resulting products.   The arrows indicate the bond  attacked.
           The unlabeled arrows  indicate ring cleavage via rearrangement.
           Fusarium oxysporum engages in a hydrolytic fission of the  cyclo-
           propane ring with y~hydroxybutyric acid as product  (10).
         Cyclopropane  carboxylate   ++
         Glutarate                    +
         Succinate                   ++
         Jso-butyrate                ++
         Methacrylate                ++
         n-butyrate                   -
                                             2-oxoisovalerate        ++
                                             Valineb                 ++
                                             a-hydroxybutyrate        -
                                             2-butenoate              +
                                             3-butenoate              +
                                             Butyrolactone            -
aGrowth was measured turbidimetrically in liquid mineral medium  as  indicated
in Materials and Methods.   Substrates of organic acids were supplied at a con-
 Supplied as DL-valine.
centration of 0.2%
as the sodium salts.

The results are summarized in Table 2.  Several points deserve special atten-
tion.  First, utilization of CPC, iso-butyrate or methacrylate is strictly in-
ducible-.  When each of these substrates is offered to starved cells grown on
succinate, Oa consumption is not stimulated, but remains at background level.
However, when each of these substrates is offered to starved cells previously
grown on CPC or on iso-butyrate, Oa consumption is immediately stimulated.  In
contrast, when cells are grown on either 2-butenoate or 3-butenoate, supply of
CPC, iso-butyrate or methacrylate to starved cells produces no stimulation of
Oa consumption, i.e., neither induces oxidation of CPC, iso-butyrate or meth-
acrylate.  Likewise, growth of cells on CPC or on iso-butyrate show a slight
stimulation of Oa consumption when given 3-butenoate, but never reach Oa con-
sumption rates comparable to those elicited by succinate.
Cells grown at the expense of:
aMicroliters of Oa taken up by a turbidimetrically standardized suspension per
unit time at 30°C.  Cells were harvested by centrifugation at room temperature
and starved by incubation in growth medium to reduce the endogenous level of
Oa uptake for a standard time.  Measurements were performed as indicated in
Materials and Methods.
 Supplied as the sodium salts at a final concentration of 0.2% (w/v).

cNot done.
     Since patterns of induction appear to eliminate 2-butenoate and 3-buteno-
ate as intermediates in CPC degradation, attention is focused on iso-butyrate
and methacrylate.  Sokatch  (12) and others have shown that in several bacte-
rial strains, degradation of valine proceeds via the pathway shown in Figure
2, in which 2-oxoisovalerate, iso-butyrate and methacrylate are intermediates.

     A mutant of strain CPC-1 was isolated, following mutagenesis, that had
lost the ability to grow at the expense of CPC.  It was found to have simul-
taneously lost the ability to grow with iso-butyrate, methacrylate and valine,
although growth on succinate was unimpaired.  When large numbers of succinate-
grown mutant cells are plated onto agar with either CPC or valine, a small
number of presumed revertant colonies appear following incubation.  Forty-
seven revertant colonies were picked from CPC and transferred to fresh valine
plates; 100 revertant colonies were picked from the valine plates and trans-
ferred to fresh CPC medium.  When growth had occurred, the plates were exam-
ined:  100% of the CPC revertants had regained the ability to grow on valine,

and 100% of the revertants  selected for regain of ability to grow on valine
had simultaneously regained the  ability to grow with CPC.

                 |-2H,  +CoASH
            tso-butyryl-CoA  + C02
                 T                 +CQASH	P^>-COOH
            Methacrylyl-CoA      -*                          L-^

3-Hydroxyi sobutyryl-CoA


3-Hydroxyisobutyrate + CoASH


 semi aldehyde
              Propionaldehyde + C02   '2H> +CoASH     Propionyl-CoA
Figure 2.  The convergence  of valine and CPC metabolism proposed to occur  in
           Corynebacterium  strain  CPC-1 (modified from Sokatch  [12]).   It  is
           not known whether  activation by coenzyme A precedes ring fission,
           or which of  the  three possible ring fission mechanisms occur.

     Direct enzymatic evidence for participation of iso-butyrate as an  inter-
mediate in CPC degradation  was sought by assay of iso-butyryl-CoA dehydrogen-
ase in extracts of succinate-, CPC-, and iso-butyrate-grown cells.  The re-
sults are given in Table  3.   It is clear that iso-butyryl-CoA dehydrogenase is
elevated in CPC-grown cells as in  iso-butyrate-grown cells.
          Corynebacterium CPC-la (nmoles/hr/mg protein)
                  Cells  grown on CPC                  11.5
                  Cells  grown on iso-butyrate         13
                  Cells  grown on succinate             2

aCells were harvested  during exponential growth, extracted, and enzyme ac-
tivity and protein measured as indicated in Materials and Methods.

     In light of the results described above, we conclude that Corynebacterium
CPC-1 degrades CPC by attack on the cyclopropane ring opposite the a-carbon.
Subsequent metabolism occurs via a scheme previously described for valine deg-
radation by bacteria (12; Fig. 2).

     The bacterial metabolism of higher alicyclic hydrocarbons and their de-
rivatives generally entails initial conversion of the compound to the corre-
sponding alicyclic ketone and subsequent introduction of an oxygen atom adja-
cent to the oxidized carbon in a mixed-function oxidase-mediated reaction.
The lactones formed in this way hydrolyze spotaneously or in an enzyme-cata-
lyzed process to yield aliphatic hydroxy acids susceptible to further metabo-
lism by conventional routes (8).  CPC is an apparent exception to this pattern
either because of the presence of the carboxyl function, which may prevent the
occurrence of some step of the usual process, or the size and/or configuration
of CPC may cause it to be handled initially as an aliphatic compound.  The oc-
currence of CPC activation either as the carnitine  (Fusarium [3])  or CoA
(Corynebacterium CPC-1) thioester, if it precedes ring cleavage,  suggests that
this latter explanation may be true.  The higher cycloalkanes are metabolized
by bacteria without initial activation as CoA thioesters.

     However, the proposed pathway does have certain general features common
to other pathways for the degradation of alicyclic hydrocarbons.   First, the
pathway is inducible.  Second, there is a high degree of specificity as to
substrates.  Corynebacterium CPC-1 is totally unable to grow on any other ali-
cyclic compound tested in the Ci» to Cy range.  Third, the portion of the pro-
posed degradative pathway unique to CPC metabolism is very short,  in this case
only two steps  (activation and ring cleavage) long.  It is generally true, as
here, that degradation of alicyclic compounds proceeds by conversion of the
alicyclic compound to a form that can feed into some commonplace central meta-
bolic pathway.  All of these points emphasize the key role of the initial
steps involving primary attack upon the substrate molecule, as well as the
regulatory mechanisms that govern their synthesis and the synthesis of enzymes
functional in later steps.


     The authors are grateful for the collaboration of Mr. Bruce Bromley in
some early aspects of this work.  Acknowledgment is made to the Donors of the
Petroleum Research Fund, administered by the American Chemical Society, for
support of this research.

                              LITERATURE CITED

1.  Goldman, P., and P. R. Vagelos.  1961.  The specificity of triglyceride
     synthesis from diglycerides in chicken adipose tissue.  J. Biol. Chem.

2.  Green, D. E., S. Mii, and H. R. Mahler.  1954.  Studies on the fatty acid
     oxidizing system of animal tissues.  III. Butyryl coenzyme A dehydrogen-
     ase.  J. Biol. Chem. 206:1-12.

 3.  Guilbert,  C.  C.,  and A.  E.  Chung.   1974.   Metabolism of cyclopropane car-
      boxylic acid.   A new role  for carnitine.   J.  Biol.  Chem.  249:1026-1030.

 4.  Layne, E.   1957.   Spectrophotometric and  turbidimetric  methods  for measur-
      ing proteins.   III. Biuret method,  p.  450-451.   In  S.  P.  Colwick and
      N. 0. Kaplan (eds.), Methods  in enzymology, Vol.  3. Academic  Press, New

 5.  Magat, W.  J., and C. L.  Tipton.   1970.  Oxidation of cyclopropane fatty
      acids by Ochromonas danica.   Arch.  Biochem. Biophys. 141:353-355.

 6.  McKenna, C. E.,  M.-C. McKenna,  and M. T.  Higa.   1976.  Organic  probes of
      nitrogenase.  J. Amer.  Chem.  Soc.  98:4657-4659.

 7.  Palleroni, N.,  and M. Doudoroff.   1956.   Mannose  isomerase of Pseudomonas
      saccharophila.   J.  Biol. Chem.  218:535-548.

 8.  Perry, J.  J.   1977.   Microbial metabolism of cyclic  hydrocarbons  and re-
      lated compounds.  CRC Grit. Rev.  Microbiol. 5:387-412.

 9.  Schiller,  J.  G.,  and A.  E.  Chung.   1970.   The metabolism of cyclopropane
      carboxylic acid.  J. Biol.  Chem.  245:5857-5864.

10.  Schiller,  J.  G.,  and A.  E.  Chung.   1970.   Mechanism  of  cyclopropane ring
      cleavage  in cyclopropane carboxylic acid.  J. Biol.  Chem.  245:6553-6557.

11.  Schooley,  D.  A.,  G.  B. Quistad,  and  L. E.  Staiger.   1978.   Cyclopropane
      carboxylic acid:  chain elongation  to oo-cyclopropyl fatty acids  by mam-
      mals and  plants.  Science  199:544-545.

12.  Sokatch, J. R.   1968. Bacterial physiology and metabolism,  p.  180-182.
      Academic  Press,  New York.

13.  Tanaka,  K. E.,  A. Kean,  and B.  Johnson.   1976.  Jamaican vomiting sick-
      ness.  New England  J. Med.  295:461-467-

14.  Tipton,  C. J.,  and N. M. Al-Shatir.   1974.  The metabolism of cyclopro-
      pane fatty acids by Tetrahymena pyriformis.  J. Biol. Chem.  249:886-889.



TIEDJE:  Dr. Chapman, do you see any reason why marine organisms or a marine
environment would effect a different type of metabolism?

CHAPMAN:  Recent studies have drawn attention to the fact that hydrocarbons
are degraded with increasing difficulty at increasing levels of salinity.
Furthermore, studies by Dr. Bourquin on the degradation of the compound nitri-
lotriacetic acid (NTA) by a freshwater organism have demonstrated that there
is a serious impairment to the degradation of NTA when freshwater organisms
are present in estuarine environments.  Based on those two examples, I would
merely say that we know very little about the constraint of salinity upon the
degradation of chemicals by freshwater and soil microorganisms, and I feel
that there is a great deal to be learned.  Possibly even more complex alterna-
tives are going to turn up from a study of organisms which have the marine en-
vironment as their specific habitat and are not responsible to growth under
freshwater conditions.

CHAPMAN:  Dr. Bollag, fungi are notorious for making secondary metabolites and
they are known formers of coupled products when they are the symbionts in
lichens.  A number of the products that are formed from carbaryl are evidently
hydroxylation products.  Furthermore, they are coupled products, oxidative
coupling products.  Do you think there is a relationship that characterizes
these fungi or relates these conversions to their normal process of secondary

BOLLAG:  Well, they are not mainly hydrolytic compounds, they are phenolic-
type compounds.  My opinion is  that the nature of these phenolic compounds
represents a special group of metabolic products.  We assume that they are ac-
tive in secondary metabolic activity and that the fungi have a major function
in this kind of polymerization process.

GIBSON:  You said the enzyme was an extracellular enzyme, Dr. Bollag.  It seems
to me those products are very similar to what you get in phenol couplings with
enzymes like horseradish peroxidase and laccases; I think you mentioned yours
was a laccase.  Do you have any evidence to support this?

BOLLAG:  That's right, we did find it to be a laccase.  That means we purified
and identified it as a laccase, but it is also possible that a peroxidase can
catalyze this reaction.

CHAPMAN:  Dr. Alexander, do you think that the non-growth substrates can also
serve as the inducers of the enzymes which may or may not be responsible for


 their degradation?  I think much of the biochemical and microbiological basis
 for cometabolism resides, as you've quite correctly stated,  between the struc-
 tural relationships of growth substrates and non-growth substrates.  You men-
 tioned, quite correctly I believe,  the importance of evaluating enzyme speci-
 ficity in this regard, but there's  a second very important barrier that has to
 be overcome and that is the de-repression of appropriate enzymes by such mole-
 cules.  Would you care to discuss this?

 ALEXANDER:  The problem of the induction or de-repression is a very real one
 in the understanding of the biochemistry/microbiology of cometabolism.  I
 agree that the work ought to be done,  and we have little information on this
 matter.  We have tried to estimate  whether induction was significant by adding
 biodegradable analogs to a natural  environment.   This work was done in fresh-
 water, sewage, and soils, using a number of different compounds—benzoates,
 diphenylmethanes,  and others.  In no case did we find much of a change.  Prob-
 ably the cometabolizing organisms are not benefitted greatly by the substrates
 we think they ought to respond to.   The organisms that respond to such addi-
 tions are probably pseudomonads, flavobacteria or other organisms which quickly
 utilize the added compound, which is then all gone by the time the cometaboliz-
 ing populations begin to metabolize the chlorinated molecules to a significant

 CHAPMAN:   Undoubtedly, it is particularly important to determine what sort of
 things are going on in these processes, and if one can achieve conditions
 where one can direct or facilitate  degradation by a suitable choice of a growth
 substrate, then I  think it's important that we understand these relationships
 so that we can direct or facilitate degradation in this fashion.

 ALEXANDER:  It is  also important to know what substrates do to natural popula-
 tions,  as well as  obtaining the biochemical information.  These population
 studies will not be easy to do well.

 TIEDJE:   Another consideration here should be the phenomenon of catabolite re-
 pression,  because  I think a lot of  substrates that we might add in nature could
 often  have a negative effect on degradation of materials.

 ALEXANDER:   In dealing with a microbial community of many populations,  we tend
 to  extrapolate too readily from pure cultures to natural environments.  The
 organism  which is  affected by catabolite repression may not  be  the  organism
 involved  in cometabolism.  Again, let me stress  that it is important to get
 the information on the biochemistry and microbiology,  and similarly on the
 ecology of the responsible species.

 TIEDJE:   You made  a statement that  was  a little  surprising to me.  You said
 that the  evidence  for cometabolism  in nature  is  shaky, and I want to know what
 kind of evidence you're  looking for  that would be more solid.  What kind of

ALEXANDER:   Something other than our inability to isolate the responsible  or-
ganisms.   It would  be  satisfying to show that the disappearance  of the  mole-
cules was parallel  to  no  increase in population of the active organzsms.   Al-
ternatively, if C1" organic  compounds subject to  cometaboHsm are not con-


verted to Cllf cellular products in the environment, then likely cometabolism
is occurring.

RAYMOND:  Dr. Alexander, I can't quite fathom the comment you made that co-
metabolism will not lead to C02.  Did I misinterpret that?

ALEXANDER:  I should retract somewhat.  Cometabolism will not lead to a major
conversion of the substrate, but a cometabolic decarboxylation reaction would
lead to CO2.  But cometabolism will not give, as with glucose in a natural en-
vironment, 6 moles of CC>2 for one mole of glucose, or 5, considering carbon-
assimilatory processes.

ZAJIC:  My questions really have to do with the carbon monoxide data.  One, I
wonder whether the reaction was run under aerobic or anaerobic conditions, and
two, whether any gaseous hydrogen was produced.  Also, I don't know what hap-
pened to Carboxymonas in the recent edition of Bergey's Manual; maybe they
don't exist any more, but the older data by A. J. Kluyver in 1945-46 indicated
that oxidation of carbon monoxide occurred under anaerobic conditions.  There
was a reductive cleavage of water, and some hydrogen was produced.  I wondered
if this might not be what you're seeing.

ALEXANDER:  Carbon monoxide was oxidized under anaerobic and aerobic conditions.
The generic name you cited does not appear in the current volume.  We have not
determined whether hydrogen is evolved.

ZAJIC:  Many times with carbon monoxide oxidation to C02 and with hydrogen re-
lease you^see methane as the end product.  That's just a comment; .1 want to
ask something else.  I wonder if we have indirect evidence of cometabolism.
Natural products like lignin are very difficult to oxidize with pure cultures
in the laboratory.  Yet in nature, we find turnover is slow but of a far
greater rate than  predictions  from anything we have been able to accomplish
in the laboratory to date.  In making another example, the alphaltene mate-
rials from crude oil, if put in the soil, are oxidized to only a limited ex-
tent.  It is almost impossible to find a single culture to metabolize it,  yet
a long-term study in the soil shows something happening that we've never been
able to do in the laboratory using more exacting controlled conditions.   These
are somewhat indirect examples, but things are happening in the environment
that we haven't been able to approach or handle in the laboratory under con-
trolled conditions.  What is your reaction to an indirect type of evidence?

ALEXANDER:  Such arguments are quite tenuous, although the facts may be cor-
rect.  In the case of depolymerization of lignin in pure culture, it's quite
possible that the monomeric fragments build up, and these may be quite toxic;
but in a mixed culture in a natural ecosystem, there would be populations
which would destroy these phenolic or methoxy compounds.  Also, the initial
populations depolymerizing lignin could be growth-factor requiring, but in
nature, organisms produce the growth factors for the depolymerizing species.

MATSUMURA:  I am interested in your definition of cometabolism.  I understand
that certainly in related chemicals we can talk about cometabolism.  How about
metabolism by quite general enzymes, such as hydrolytic enzymes, excreted out-
side of the cell, hydrolysis on something like carbaryl by exo-enzymes—do you
call this degradation or cometabolism?


ALEXANDER:  You're trying to nail me to the wall on semantics, but I have a
very low semantic sensitivity.  For the sake of the definition, I would say
the organism is not cometabolizing the compound.  On the other hand, a product
is accumulating.  Such accumulations occur with compounds which contain qua-
ternary carbons.  The organism keeps attacking the molecule until it gets
close to the quaternary carbon and then it stops.  The consequence is the im-
portant point, whatever the phenomenon is called.

GIBSON:  In the batch system, Dr. Chou, did you look for accumulation of pro-
ducts, or did you just measure substrate disappearance?

CHOU:  In most of these experiments, I measured disappearance of substrate;
however, in the mixed culture, in our experiments it seems to me the products
are degraded further, maybe as a function of the mixed organisms.

GIBSON:  You may get extensive degradation?

CHOU:  Yes.

RAYMOND:  Dr. Gibson, I can't pass up the opportunity to explore further into
this business of products of cometabolism accumulating in nature, which Dr.
Alexander has alluded to.  Our evidence, certainly with hydrocarbons, is that
products do not accumulate in nature from cometabolism, or rather from the co-
oxidation.  Admittedly, we have not isolated all the possible cultures from
that particular aquatic environment.  Because we, like you, have had a lot of
negative studies using a pure compound and attempting to find a microorganism
in the environment which utilizes it for growth and energy.  Most of the time
we don't find them.  What we do find, in nature, is, if we first add a sub-
strate that can be used for growth and energy, then introduce the second (non-
utilizable) substrate, the second is oxidized.  Furthermore, in freshwater and
soil environments, we don't see end products accumulating.  I don't know that
this is proof, but certainly we have looked exhaustively for end products in
these systems.  Perhaps some of them are tied up or conjugated so that we don't
see them, but with our extractive methods we don't see them.  I think this is
part of what we have to establish this week:  whether the methods we use to
determine biodegradation are satisfactory.

ALEXANDER:  I think there are two possible explanations.  First, if an orga-
nism can't be isolated it does not mean that the reaction is necessarily co-
metabolic.  Second, as in the case of p-chlorophenylacetic acid in a two-
culture system, possibly the initial reaction is cometabolic and the product
of the initial reaction will support growth; therefore, there is no accumula-

RAYMOND:  Are you still insisting, though, that if you cannot find microorga-
nisms, at least by identifiable names, to utilize certain materials,  or if we
can't find in the laboratory that these materials are cometabolized,  that
therefore they are not biodegradable.  Therefore, they are dangerous materials
because they will accumulate in nature.  That cannot be the gist of this re-
moval from nature, the way I understand it.

ALEXANDER:  I focus on the environment first and on the microbiology second.
If there is no accumulation in nature, one would not explain the disappearance
by cometabolism.  It may be that cometabolism is occurring but the product
would not be found.  I suggest cometabolism may occur if the reaction is slow
and there is an accumulation.  That does not mean that accumulation is the
only consequence of cometabolism in nature.  My microbiology begins with the
observation in nature and then we work with cultures whereas many of my col-
leagues prefer to begin with cultures and hope their data are relevant in

GIBSON:  I think if we understand what is going on, it doesn't really matter
what term we ascribe to the phenomenon.

ALEXANDER:  I agree, Dave, we have enough problems without semantics.  There
is nothing particularly unique in the question of cometabolism.  It is inap-
propriate to worry about terminology in important environmental issues.  If we
understand what is going on and we can use biochemical information to predict
what will happen in nature, that should be the focus of a microbiologist.

GIBSON:  You suggested that there were possible hazards of carcinogens produced
in the environment, for example, nitrosamines.  Aren't nitrosamines reactive
alkylating agents?  Are they going to hang around long enough to be dangerous?

ALEXANDER:  We have put dimethyl, dipropyl, diethyl nitrosamines into soils,
water and sewage.  In fresh water, there was no disappearance at all over four
months.  In the other environments about half of the concentration disappeared
and the rest stayed in the system.  So far as we know, they are quite persis-
tent.  Debra Raymonds found that nitrosamines can be taken up by plants (no
one eats the soil as a rule).  They can also move through soil into ground
water.  So, at least with the dialkyl nature of the amines and the N-alkyl di-
ethanolamine, I'd be very concerned.  They do seem to be reasonably persistent
molecules.  There are animal metabolism studies which suggest you can get a
dealkylation of molecules enzymatically, but not at natural population levels
of microorganisms.

GIBSON:  They must be metabolized into active carcinogens by animals that in-
gest them.  Is that correct?

ALEXANDER:  I would assume so.  They can then become alkylating agents after

GIBSON:  Have these nitrosamines that you mentioned been tested in the Ames

ALEXANDER:  Well, I am sure they have been tested in Ames test.  They have
been tested in animals and they are carcinogens.

IVANOVICI:  In cometabolism studies, it seems useful to assess the growth
state of the microorganisms you are working on as to whether or not they are
growing.  A lot of microorganisms won't grow even though they remain viable.
So you don't really know what's happening to the energy.  Have you ever con-
sidered measuring ATP as an indicator of microbial growth state, by an

adenylate charge ratio?  And if you had,  what sorts of results did you obtain?
If you haven't measured it,  might you consider measuring it,  because it could
give you lots of very useful information  about the growth state of the orga-

ALEXANDER:   Well, we have considered it,  but we haven't done  the experiment.
It would be interesting to see if there is any such benefit to the organism.
We have, in looking at heterotrophic oxidation of nitrogen compounds,  calcu-
lated what the energy (end kilocalories)  would be if the organism was  getting
all the energy by converting ammonium to  nitrate.  It would be trivial by com-
parison with the cell yield of the organism.   But we haven't  done the  experi-

GIBSON:  You said that the list of chemicals under TOSCA was  almost bizarre,
but you never mentioned any of them.  What kind of things are we faced with

ALEXANDER:   I think that I shall leave that matter to Dr. Stern.  However,  the
list is enormous, and if one considers the number of precursors used for the
major biosynthetic  reactions and the solvents involved, one  obtains some idea
of the possibilities.  The focus of attention should not be with compounds
that are of interest biochemically,  but those on large quantities produced  by
industries.  The ten most wanted list is  public information.   That would in-
clude cresols, xylenes, nitrobenzenes, octachlorobutadiene, and pthalates.

STERN:  The list was compiled on at least three criteria:  1)  production vol-
ume, 2) distribution of material, and 3)  structural relationships to materials
which have been shown to have some hazardous effects.   Although the list is
long, it doesn't mean everything is subject to regulation.  It means merely
that they are going to be given a closer  scrutiny.

GIBSON:  In terms of structure, then there is nothing unusual.   I got  the im-
pression we are going to be looking at some bizarre chemicals and I was get-
ting nervous about our lack of information.

ALEXANDER:   No, if you consider the precursors for major polymers,  and the
major solvents, it is evident that TOSCA  is concerned with chemicals we usually
do not study biochemically.

GIBSON:  You did follow it up by saying we must look into new areas of bio-
chemical processes.  I would be interested to know what they  are.

ALEXANDER:   At the moment we should consider the general approaches for scien-
tists to follow to get reasonable information both from purely laboratory
studies with pure cultures and from model ecosystem studies.

COONEY:  I  have a perhaps naive view of many aquatic microorganisms living  for-
ever on the edge of starvation in a dilute nutrient environment.  I wonder  if
you would comment on induction in this environment.   It is difficult to envi-
sion induction as a widespread phenomenon, since it would require.the  synthe-
sis of new enzymes.

ALEXANDER:  This topic is outside my field of competence, although we have
been concerned with problems of very low concentrations of synthetic chemicals.
Problems of induction and of whether diffusion of substrates to the cell is
sufficiently rapid to provide for maintenance energy and for the cell to mul-
tiply are real.  It is possible the chemical concentration is so low in natu-
ral environments that there is no appreciable biodegradation.  There could be
bioaccumulation and consequent protection from biodegradation, because the
chemical is biomagnified in the tissues; it may never be exposed to microorga-
nisms outside.

PRITCHARD:  I would like to make two comments relative to what has just been
said.  One, there is a report by Horvichi et al.  (1962) which showed that by
growing E. coli in a chemostat at very low nutrient concentrations and at very
low dilution rates over long periods of incubation with lactose as the sole
source of carbon and energy, a mutant strain of E. coli could be selected that
was constitutive for lactose metabolism.  At these low dilution rates and low
substrate concentrations the constitutive mutant outcompeted the inducible
parent.  I think this points to the possibility that the physiology of the so-
called "brink of starvation type bacteria" may be unique, and the more we can
find out about their physiology, the more capable we will be at asking ques-
tions about the role of these bacterial types in natural transformation pro-
cesses.  The second point I'd like to make is directed toward Peter Chapman.
Considering the staggering number of compounds that the Office of Toxic Sub-
stances is handling, are we in a position to begin making significant generali-
zations about the ability of microorganisms to oxidatize or metabolize these
compounds?  If we can categorize these compounds into potentially biodegradable
or non-biodegradable, should we begin to emphasize the ecosystem or environmen-
tal effects on degradation processes rather than on the mechanistic aspects?

CHAPMAN:  Hap, I think there is a significant kicker in your question—you
asked whether one can make significant predictions?  I have no doubt, from
the vast accumulated information we have on biodegradation, that we can iden-
tify functional groups in natural and synthetic molecules which we can pre-
dict what reactions they will undergo.  The next step is to say whether those
reactions do, in fact, take place in certain environments.  Clearly,  if such a
reaction is seen, say, to be catalyzed by a mono- or a dioxygenase, it will
happen in aerobic environments, but not in anaerobic environments.  So much of
this prediction has to be conditioned by relationships within prevailing en-
vironments, habitats and their constituent microflora.  One can say,  yes,
these reactions could occur, but I don't think one can say at this time,  they
will occur.

COLWELL:  I am compelled to make a brief statement concerning low nutrient
concentrations in seawater.  It is obvious that suspended particulate matter
in seawater results in different concentrations of nutrients in the water col-
umn.  Large concentrations can occur in microniches, an important considera-
tion.  For toxic chemicals and their effect on the environment, we must be
concerned that we may be disrupting the community structure of the microbial
populations.  For example, fuel oil, or refined oil, will selectively inhabit
or perhaps kill organisms that are capable of carrying out certain ecologi-
cally important activities such as cellulose decomposition or starch hydroly-
sis, whatever that effect may be on the total ecosystem.  Microorganisms in

the environment carry out ecological functions.   Effects of toxic chemicals
considered to date have not been thought about in this respect.   What effects
are there on the natural populations and,  therefore,  on the ecosystem?  I
think that is an aspect of microbial ecology we  must  consider.


        Chairperson, Carol Litchfield

                           NUTRIENT LIMITATION

                               G. D. Floodgate
                         Marine Science Laboratories
                      University College of North Wales
                      Menai Bridge, Gwynedd, Wales, U.K.

                Substances which are degradable in the laboratory can
           become apparently recalcitrant in the natural environment
           because of the lack of a necessary nutrient.  The physical
           state of the nutrient, such as dissolved,  biphasic or par-
           ticulate, can govern the rate of biologically induced
           change in the system.  The commonest nutrient deficiency in
           the oceans is a lack of nitrogen and less  often phosphorus.
           Hence, to degrade organic material, microorganisms must be
           able to mobilize the nutrients efficiently, and this will
           depend upon the biochemical constants of the enzymes in-
           volved.  These points are illustrated using crude oil as
           an example, both of molecules which are recalcitrant be-
           cause of their chemical structure and also because of the
           low concentrations of nitrogen in sea water.
     In discussing the problem of pollutants in the marine environment,  and
the factors that govern their disappearance by biological action,  we are at-
tempting to explore the growing point between some aspects of marine micro-
biology and marine chemistry.  The area is hardly overstocked with facts.
Certainly, there are an increasing number of papers about marine organic chem-
istry  (7), and considerable work was done a few years ago on the effects of
substrate limitation on the growth of pure cultures in a chemostat.   The ef-
fect of starving bacteria also has been studied by many workers and their
findings have been reviewed by Dawes (5).   Unfortunately, most of this work is
only obliquely related to the task of this workshop.  As so often in the past,
it is not the production of bacterial biomass that concerns us primarily,
though this is clearly relevant, but the removal of foreign chemicals which
have been placed in the marine environment by human activity, and whose con-
centration and chemistry can be changed by nonbiological phenomena such as
wind and wave mixing, dilution, and photochemical reaction.

     In fact, there is still comparatively little information about the rates
of destruction of organic substances in the sea, and most of the available in-
formation deals with the fate of small, but ecologically significant,


molecules.  Unhappily, too many of the papers dealing with the fate of pollu-
tants describe experiments carried out under conditions very different from
those found in the sea.  Generally, the tendency has been to maximize the
rates of decay, leading to over-optimistic forecasts of the duration of the
pollutant in the oceans.  However, we do know that some organics, both natural
and human artifacts, are only slowly broken down.  Whatever may be the final
definition of persistence or recalcitrance to emerge from this workshop, it
seems that there are at least three possible reasons why a carbon substrate
may not be degraded in the marine environment.  These are:

     1) There are some substances whose molecules are so strongly constructed
that an excessive amount of energy is required to change them into a form that
can enter the biochemical mechanisms of the microbial cell.  Because of this
large expenditure of energy, the breakdown rate is necessarily slow.

     2) There is another set of circumstances in which the microbial popula-
tion does not have the ability to destroy the molecule.  In other words, the
information as to how to make the appropriate enzymes does not exist within
the gene pool of the population.  Hence, attack is impossible.

     3) Finally, there are those instances where, although breakdown is ener-
getically feasible and degradation pathways are available, oxidation is pre-
vented by environmental conditions.  Such conditions include unsuitable tem-
peratures, lack of suitable electron donors, or a paucity of nutrients.  All
three reasons for recalcitrance can co-exist at one time, but in this paper we
are concerned only with the third.  In certain instances, the missing moiety
may be a growth factor or a metal.  For this reason, yeast extract is often
added to media to isolate heterotrophs.  Occasionally, growth factor depen-
dency leads to mutualism (22).  Again, Litchfield and Floodgate (14) found
sediment organisms which required amino acids while Dibble and Bartha  (6)
found that soluble iron stimulated the removal of crude oil.  Nevertheless
since most natural populations, as a whole, are not likely to be nutritionally
exacting, the commonest reason for the nondevelopment of microorganisms in an
otherwise generally favorable environment is most likely to be an insuffi-
ciency of a major nutrient.  In the sea, nitrogen and phosphorus are the most
probable candidates; indeed, the commonest constraint on bacterial growth in
the sea is a dearth of available nitrogen.  Phosphorus is less likely to be
the limiting factor since it appears to be rapidly recycled (18) .  However, Le
Petit and N'Guyen (13) have observed that the optimum concentration for bac-
terial growth on hydrocarbons in natural waters depends on the ionic strength.
The concentrations for maximum growth ranged between 2 x 10" ^ M to 8 x 10" "* M
phosphate in waters of salinity close to that of oceanic water and between
1.5 x 10"  M to 3 x 10"3 M phosphate where the natural water receives a large
input of low salinity water.  Excessive amount of phosphate also inhibited
bacterial growth.

     Perhaps the best known example of nutritionally induced recalcitrance in
the sea is that of oil degradation.  This has been demonstrated by many
workers but is shown here (Fig. 1) in the well known example provided by Atlas
and Bartha (1).   Presumably, nitrogen limitation acts by either preventing the
growth of sufficient biomass to deal with the substrate, or by preventing the
organisms that are there from being maximally active.  It is generally


accepted that a high bacterial count does not necessarily imply a high degree
of activity and vice versa.  Further, it must be assumed that the organic mat-
ter is not able itself to supply nitrogen, so that this form of limitation  is
confined to those substrates that are low in available nitrogen or from which
nitrogen is absent.  Therefore, the substrates involved are largely carbohy-
drates and their polymers such as cellulose and starch, lipids  (including
steroids), hydrocarbons  (including spilled oil), some phenolics, and a number
of synthetic pollutants.
X IO"2]" I X IO'2! I X IO"21 5 X IO"3 I X IO"3   00

Figure 1.  Conversion of petroleum  in  sea water  supplemented by various con-
           centrations of NaaHPOit and  KNOs during  18 days of incubation.
           Solid portions of  the histograms represent mineralization percen-
           tage as measured by C02  evolution.  The upper limits of the cross
           hatched and empty  portions  of the histograms represent biodegrada-
           tion percentages as measured by gas chromatography and by residual
           weight, respectively  (from  Atlas and  Bartha  [1]).  Normal sea water
           concentrations of  nitrogen  and phosphorus are below any of the
           additions by several orders of magnitude.

     It is well established that various bacteria, especially certain species
of Desulfovibrio and Clostridia, are able to fix nitrogen,  the ATP required
coming from the oxidation of  organic matter  (2).   Hexadecane has been reported
as providing a carbon and energy source for nitrogen fixing (19) soil bac-
teria.  However, it is generally concluded that  bacterial nitrogen fixation is
an insignificant part of the  annual nitrogen input to the seas  (2).

     It should also be remembered that the available organic matter will be
very dilute and variable in concentration.  Hence, there are advantages in
thinking, not only in terms of the  concentration of the nitrogen and carbon,
but also of the carbon to nitrogen  ratio.  However, the significance of this

ratio is complicated by the fact that the dissolved organic matter is not the
only degradable form of organic carbon in the marine environment.   There is
the particulate matter as well, and whereas the greater part of the dissolved
organic matter is known to be in a recalcitrant form such as "gelbstoff," much
of the particulates are composed of polymers such as starch, cellulose, etc.,
that are intermediate in their degradation rates.   Clearly, in the example of
such particulate organic carbon, or of oil, either floating or on the surface
or in finely divided form, we are dealing with a biphasic system in which the
bacteria are either attached, or closely associated with a separate carbon-
rich phase.  It follows that there is a series of possible systems in which
the C/N ratio varies with the degree of homogeneity.  This is illustrated in
Figure 2.
Figure 2.  Effect of homogeneity of environment on carbon/nitrogen ratio.
           • indicates available nitrogen;  gj indicates available carbon
           source.  For discussion see text.
Organism A is an authochthanous bacterium,  that is,  its resources are scarce
but its ability to harvest these resources  is high.   In contrast, the orga-
nisms at B have plentiful supplies in terms of available carbon and nitrogen.
This has led to overgrowth,  with consequent deterioration of the culture.  The
organisms on particulate matter at C have solid phase organic carbon on which
to adhere; inorganic nitrogen has sorbed onto the detritus,  setting up a fa-
vorable C/N ratio.  The hydrocarbonoclastic bacterium at D is closely associ-
ated with the oil phase but like the organism at A,  has to scavenge nitrogen


from the aqueous phase.  A rather different view of the relationship between
the oil phase and hydrocarbonoclastic bacteria has been proposed by Velankar
and his colleagues  (22).  They suggested that surface active material,  possibly
formed by the bacteria themselves, remove miscelles from an oil droplet, the
miscelles being about one tenth of the size of a bacterium.  The miscelles
"recruit" nutrient ions around themselves and the whole is transported across
to the bacterium which receives a packet of carbon, nitrogen and phosphorus.
If this model is true, then the ability of the miscelle to scavenge nitrogen
could be a limiting factor.

     The variations in concentration of nutrients between geographical areas
and with time will have a bearing on the biodegradation rate, particularly
when it is remembered that the most pronounced changes occur in the shallow
coastal waters, which are the most productive as well as the most likely to
receive polluting wastes.  Detailed information of the amounts of nitrogen in
these environments can be found in any major marine chemistry text (e.g., 18
and 20), but for the sake of convenience it is worth recalling a few salient
facts.  The concentration of organic nitrogen, both dissolved and particulate,
varies from 5 to 10 yM-N.  The inorganic nitrogen is found in several oxida-
tion states, of which nitrate is the most important.  Concentrations of up to
600 yg nitrate-N I"1 are found, while nitrite and ammonia are present at up to
approximately one tenth of this amount.  Temperate surface oceanic water con-
tains around 10-15 yg nitrate-N I"1.  In tropical water this might fall below
1 yg nitrate-N 1~ .  The amount increases with depth of the water column up to
about 1000 m where it reaches a value between 300 and 600 yg nitrate-N"1.  The
concentration remains at about this value to the bottom.  From the degradative
point of view, the comparatively high concentrations in estuaries and coastal
waters will assist decomposition but once the recalcitrant material has es-
caped to the oceanic water, the lack of nitrogen is likely to further delay
oxidation, particularly if the pollutant stays near the surface.  There are
also important seasonal variations to consider.  Much of this is due to the
competitive effects of the algae which lock up large amounts of nitrogen in
the spring and autumn blooms.  There are other variations caused by disconti-
nuities such as thermoclines and frontal systems, and sometimes a patchiness
of no apparent origin.  A recalcitrant material therefore might meet a variety
of nitrogen regimes before it finally disappears.

     Although there is very little evidence as to the fate and function of nu-
trients in marine microbial degradation, our work on oil in sea water using a
fixed amount of oil and varying amounts of added nutrients led to development
of the idea of "nitrogen demand."  This is defined as the amount of nitrogen
required in the breakdown of 1 mg of oil, at least insofar as it was changed
over an experimental period of nine to twelve months (8, 9, 10).  The idea dif-
fers in several important aspects from the Biochemical Oxygen Demand, the most
important difference being that the oxygen of the BOD is directly involved in
the chemistry of the breakdown whereas the nitrogen is not.  Hence, a stereo-
metric relationship exists at least theoretically between BOD and the oxida-
tion of the substrate.  No such relationship exists with the nitrogen demand
which is related only to the degradative ability of the system under consid-
eration.   (There is, of course, an exception in the cast of nitrate dissimila-
tion where the nitrate is part of the electron acceptor mechanisms.)  Using a
variety of nitrogen regimes that fell within the variations that are known to


occur in the Irish Sea,  Gibbs (8)  showed a linear relationship between the
rate of oxygen uptake and the rate of nitrogen uptake.   He found a regression
equation of
       02 uptake mg week"1 = 0.72  + 0.83 (nitrogen uptake ymol week"1)

If it is assumed that the theoretical oxygen demand for crude oil is 3.5 g oil
oxidized per gram oxygen, it is possible to calculate the amount of nitrogen
assimilated in association with the disappearance of 1  mg oil.  Under the ex-
perimental conditions used, at 14°C and with Kuwait crude oil as the substrate,
the value works out at, 4 y moles N mg"1 oil.  If crude  oil is assumed to be
90% carbon, the ratio of carbon oxidized to nitrogen assimilated, calculated
on a molar basis, is 18:8 for summer temperature and 6:8 in winter.  Further-
more, using this data and making some assumptions as to the rate of nitrogen
regeneration, Gibbs  (8,  9) calculated that one cubic meter of Irish sea water
would provide sufficient nitrogen to degrade 30 g oil in summer year"  and 11
g oil year"1 at winter temperature.

     A second set of data was obtained using a continuous flow method in which
a stream of water was passed through a tube whose inner surface was coated
with oil (10).  Again assuming that oxygen remained in  excess, nitrogen uptake
rates were the limiting factor.  For the apparatus running at 14°C the regres-
sion analysis generated the equation
         Oxygen uptake mg hr"1 = 0.005 +0.33 (N supply y mole hr~ )
This was felt to be satisfactorily close to that for the respirometry experi-

     In a third experimental design (11) ,  a gravel column was used as a matrix
for the standing oil phase, and sea water containing different concentrations
of nutrient at several percolation rates was passed down it.  Again,  providing
low concentrations of oxygen were avoided by a high flow rate, nitrate addi-
tion rates were the limiting factor.  The regression equation under these con-
ditions was
        Oxygen uptake mg hr"1 = 0.12 + 0.23 (nitrate supply y moles hr"1)

     Unfortunately, none of these experimental designs  allow the worker to
monitor the biomass, numbers or types of microorganism  without terminating the
experiment.  Nevertheless, it was possible to obtain some information by ex-
amining the effluent water.  These data suggest that the bacteria developed
rapidly in the early days to a value of around 105 bacteria ml"1 as shown on
normal heterotrophic media, and subsequently stayed more or less steady.  This
would indicate that the nitrogen that was being continuously added was not
being used to increase bacterial numbers or biomass, but only to replace dead
cells or possibly was stored in the amino acid pool.

     The inorganic nitrogen uptake was determined by difference from the known
added amounts and that found in the effluent water.  This is illustrated in
Figure 3 from the respirometry experiments using data from a flask receiving
2 ymol N/week~  as sodium nitrate and 0.2 ymol P/week"1 as sodium dihydrogen
phosphate (8).  Although the oxidation proceeded at a more or less steady rate
for several months, the amount of nitrogen in the flask was close to zero
whatever the rate of nitrate addition.  The disappearance of the nitrogen


clearly poses an interesting series of questions.  Figure 4 shows the change
in the nitrogen concentration during one week (No. 17) of the flask receiving
the highest rate of nutrient addition.  It is clear that the inorganic nitro-
gen is almost completely assimilated by the microbial population within three
                                     15    !0
Figure  3.  From Gibbs  (8).  Oil added at week 12.  A= gross rate of oxygen
           uptake; T = rate of oxygen uptake less control; 	 = nutrient
           added; 	 = nutrient utilized; • = nitrate concentration;
           • = phosphate concentration.

     Unfortunately,  the result was not exactly reproducible.  When the experi-
ment was repeated using another batch of aged sea water, the nitrate concen-
trations were rapidly reduced in a few days but not to the same degree as in
the first instance.  The difference is believed to be due to a change in the
microbial population in the area of the Irish Sea from which both samples were


Figure 4.  From Gibbs (8).  Week 17 in detail for respirometer receiving
           2 ymol N week"1 and 0.2 ymol week"1.   A= rate of oxygen uptake;
           • = nitrate concentration; • = phosphate concentration.
     In an interesting, and to some extent parallel, series of papers Parks
noted a relationship between utilization of nitrate and the oxidation of cel-
lulose by several species of fungi (15, 16, 17).  It was found that the rate of
cellulose removal is affected by the nitrogen concentration, and that differ-
ent species show various patterns of removal.  The ratio of carbon removed
•from the substrate to nitrogen assimilated varied from below 10 to around 48
depending upon the chemical and biological conditions (15).  Park (17) also
emphasized the ambiguity of C/N ratios and suggested that the carbon/mineral
ratios should be expressed in three forms.  The provisional ratio is the pro-
portion of the nutrient elements that are available in any particular environ-
ment.  It is suggested that this ratio will be a factor in determining which
of several possible decomposers in the vicinity of the substrate will become
the dominant organism.  The second ratio suggested is the assimilation ratio,
which is the amount of carbon in the decomposer biomass per unit of nutrient
assimilated, and thirdly, the decomposition ratio which is the amount of car-
bon oxidized per unit mineral element assimilated.  This third ratio is clearly
close to the nitrogen demand which we have suggested for oil decomposition in
the sea.  It should be remembered that although nitrogen demand or decomposer
ratio is useful for the purpose for which it was originally intended, i.e., as
a means of estimating the acceptable loading of a recalcitrant pollutant in a
particular sea area, it is limited in that there is no certainty that the or-
ganic material is mineralized to carbon dioxide.  If the carbon is only par-
tially oxidized, then the nitrogen demand will increase before complete min-
eralization is achieved.

     The question of what happens to the nitrogen and how it is assimilated by
the organisms is obviously of interest at this point.  Nitrate, the form in
greatest supply in the sea, first has to be reduced to ammonia by nitrate and
nitrite reductases.  Syrett (21) has calculated that the reduction via nitrite
to ammonia involves a free energy change of over 70 kcal mol"1 so it is not
surprising that ammonia is the preferred source.  For one pseudomonad isolate,
Brown and his colleagues (3) found that the uptake of nitrate ceased if the
ammonia concentration went above 1 mM, but at lower concentrations both sources
were used simultaneously.  The necessity to reduce nitrate means that the rate
of uptake and reduction of nitrate is a physiological limitation on the rate
of nitrogen availability even though the supply of inorganic nitrogen may be
adequate.  However, the saturation constants for nitrate and ammonia uptake
were 0.26 mM, and for nitrate reductase 0.29 mM.  These figures represent a
fair degree of scavenging ability for low concentrations of nitrate.

     The mechanism for ammonia assimilation also has been extensively studied
(2, 4).   There are two known major routes.  First there is the classical glu-
tamate dehydrogenase route whereby 2-oxo-glutarate reacts with ammonia to give
glutamic acid.  This system has a high K^, value of around 10-22 mM so that
this is not a good mechanism for scavenging ammonia and will only be important
when the ammonia concentration is his; thus, it is rarely important in the sea.
However, the mechanism is independent of the presence of ATP.  The second
route  (Fig. 5) requires ATP.  Here assimilation is by way of glutamine synthe-
tase (GS) and glutamate synthase (GOGAT*glutamine (amide); 2-oxo-glutamate
amino transferase).  The Km of glutamine synthetase is about 0.25 mM so that
it will mobilize ammonia at low concentrations.  Therefore, both major path-
ways for the reduction of nitrate and the assimilation of ammonia in the mar-
ine environment necessitate expenditure of energy.  It is possible that a bac-
terium which has lived under the starvation conditions, which exist over the
greater part of the world's oceans, may not be able to readily assimilate the
nitrogen that it requires and thus be unable to attack the polluting material.
Hence, the previous history of the organism, as well as its genetic composi-
tion, may determine whether or not it will develop on a substrate.
       NH3 -f ATP -^.     ^—    Glutamote     ^*~^    ^r- Glutamafe

                                                              NADPH + H+

        ADP-f Pj •""*"      ^^   Glufomine       """'      ^~~ 2 oxo-glutarate

              Glutamate synthesis in ammonia-limited Aerobacter aerogcues.

Figure 5.  The glutamine synthetase-glutamate synthase  (GS GOGAT) pathway of
           nitrogen assimilation.

     Once  the ammonia has been assimilated into glutamate, it may form part of
the amino  acid pool of the bacterium and is then available for protein


synthesis among other functions.   However,  a number of environmental factors
affect the size of the amino acid pool.   It is known (12)  that the pool size
of nitrogen-limited pseudomonads  in a chemostat increased three-fold as sali-
nity increased from 0.2 M to 0.5  M.  The pool size also depended upon whether
the culture was carbon or nitrogen-limited.  With carbon limitation, the pool
size was 3.9 mg liter"1 on 0.2 M  sodium chloride and 14 mg liter"1 at 0.5 M
sodium chloride.  When the conditions were  nitrogen-limiting the figures were
much smaller at 0.4 mg liter"1 at low salinity, to 2.4 mg liter"1 at the higher
salinity value.

     In summary, the limitation of nutrients on degradation of pollutants in
the marine environment appear to  have a number of general effects.  First, be-
cause the organisms which can degrade the substrate are in competition, the
concentration of nutrients may determine which species or strain is likely to
become dominant.  The strain which is most  effective in mobilizing the nitro-
gen or other nutrient in short supply will  have an advantage.   Second, the
rate at which the nutrients are regenerated will govern the rate of degrada-
tion as this will regulate their  concentration and so limit the growth of bio-
mass, or the synthesis of enzyme, or both.   Third, the characteristics of the
assimilation pathways of the nutrients and  the availability of energy within
the cell can be the limiting factor even when the environmental concentration
of the nutrient is not.  Finally, the fate  of the nutrient within the cell may
determine whether it is used in the degradation of the pollutant in question
or whether it is diverted to some other function.  In all these areas, further
research is needed, particularly  directed to what happens in the natural en-
vironment rather than in the laboratory alone.

     I wish to thank the Environmental Protection Agency and Georgia State
University for making it possible for me to attend this workshop.   I am also
grateful to various authors for permission to reproduce their diagrams and to
my colleague, Carol Turley, and my daughter Catherine for help with Figure 2.
                              LITERATURE CITED

1.  Atlas, R. M.,  and R.  Bartha.   1972.   Degradation and mineralization of pe-
     troleum in sea water;  limitation by nitrogen and phosphorus.   Biotech.
     Bioeng. 14:309-318.

2.  Brown, C. M.,  D.  S.  McDonald-Brown,  and J.  L. Meers.   1974.   Physiological
     aspects of microbial inorganic nitrogen metabolism.   Adv.  Microbial
     Physiol. 11:1-52.

3.  Brown, C. M.,  D.  S.  McDonald-Brown,  and S.  0. Stanley.   1975.   The inor-
     ganic nitrogen metabolism in marine bacteria; nitrate uptake and reduc-
     tion in a marine pseudomonad.   Mar.  Biol.  31:7-13.

4.  Brown, C. M.,  and B.  Johnson.  1976.   Inorganic nitrogen assimilation in
     aquatic microorganisms.   Adv.  Aquatic Microbiol. 1:1-113.


 5.   Dawes, E. A.  1976.  Endogenous metabolism and survival of starved pro-
      karyotes.  Symp. Soc. Gen. Microbiol. 26:19-53.

 6.   Dibble, J. T., and R. Bartha.  1976.  Effects of iron on the biodegrada-
      tion of petroleum in seawater.  Appl. Environ. Microbiol. 31:544-550.

 7.   Ehrhardt, M.  1977.  Organic substances in the sea.  Mar. Chem. 5:307-316.

 8.   Gibbs, G. F.  1975.  Quantitative studies in marine biodegradation of oil.
      I. Nutrient limitation at 14°C.  Proc. Roy. Soc. Lond. Ser. B. 188:61-82.

 9.   Gibbs, C. F., K. B. Pugh, and A. R. Andrews.  1975.  Quantitative studies
      in marine biodegradation of oil.  II. Effects of temperature.  Proc. Roy.
      Soc. Lond. Ser. B. 188:83-94.

10.   Gibbs, C. F.  1977.  Rate measurements and rate limiting factors in oil
      biodegradation in the marine environment.  Rapp. P-v. Reun. Cons. Int.
      Explor. Mer. 171:129-138.

11.   Gibbs, C. F., and S. J. Davis.  1976.  The rate of microbial degradation
      of oil in a beach gravel column.  Microbial Ecol. 3:55-64.

12.   Herbert, R. A., C. M. Brown, and S. 0. Stanley.  1977.  Nitrogen assimila-
      tion in marine environments, pp. 161-177.  In F. A. Skinner and J. M.
      Shewan, eds., Aquatic microbiology.  Appl. Bact. Symp. Ser. No. 6, Aca-
      demic Press, New York.

13.   Le Petit, J., and M.-H. N'Guyen.  1976.  Besoin en phosphore des bacteries
      metabolisant les hydrocarbures en mer.  Can. J. Microbiol. 22:1364-1373.

14.   Litchfield, C. D., and G. D. Floodgate.  1975.  Biochemistry and micro-
      bioloty of some Irish Sea sediments.  II. Bacteriological analysis.  Mar.
      Biol. 30:97-103.

15.   Park, D.  1976.  Cellulose decomposition by Pythaceous fungus.  Trans.
      Brit. Mycol. Soc. 66:65-70.

16.   Park, D.  1976.  Nitrogen level and cellulose decomposition by fungi.
      Int. Biodeterioration Bull. 12:95-99.

17.   Park, D.  1976.  Carbon and nitrogen levels as factors influencing fungal
      decomposers, pp. 41-59.  In J. M. Anderson and A. Macfayden, eds., 17th
      Symp. Brit. Ecol. Soc.  Blackwell Scientific Publ., Oxford.

18.   Riley, J. P., and R. Chester.  1971.  Introduction to marine chemistry.
      Academic Press, London.

19.   Riviere, J., J. Oudot, J. Jonqueres, and G. Gatelier.  1974.  Fixation
      d1azote atmospherique, par des bacteries utilisant 1'hexadecane comme
      source de carbon et d'energie.  Annals Agron. 25:633-644.

20.  Spender,  C.  F.   1975.   The micro-nutrient elements, pp. 245-295.  In J. P.
      Riley and G.  Skirrow,  eds.,  Chemical oceanography, 2nd ed.,  vol. 2.  Aca-
      demic Press,  London.

21.  Syrett, P. J.   1954.   Ammonia and nitrate assimilation by green algae.
      Symp. Soc.  Gen.  Microbiol.  4:355-427.

22.  Velankar,  S. K.,  S.  M.  Barnett,  C.  W.  Houston,  and H.  R.  Thompson.  1975.
      Microbial growth in hydrocarbons;  some experimental results.  Biotech.
      Bioeng.  17:241-251.

23.  Zavarin,  G.  A.,  and  A.  N.  Nozhevnikova.   1977.   Aerobic carboxydobacteria.
      Microbial Ecol.  3:305-326.

                AND  COASTAL WATERS

                   Richard T. Wright
                 Department of Biology
                    Gordon College
                   Wenham, MA 01984

     Previous studies on natural heterotrophic activity
have contributed substantially to our understanding of
the role of the bacteria in natural waters.   The research
has demonstrated that low molecular weight organic  com-
pounds are present at concentrations of several micro-
grams per liter or less, and the existence of natural
bacteria populations capable of uptake at these low sub-
strate concentrations.  These microbes account for  rapid
turnover and substantial mineralization of substrates.
Some specificity is shown in the response of natural
microbes to organic solutes, along with a considerable
range in rates of uptake.  This range is responsive to
the degree of organic pollution, differences in tempera-
ture, and other factors less easy to assess, such as the
physiological state of the bacteria.

     Recent establishment of epifluorescence direct
counting using Acridine Orange has provided a method
that gives total counts of bacteria in natural waters.
This method can be combined with radioisotope-based
measures of heterotrophic activity to give a quantita-
tive expression of the relationship between activity
and numbers of heterotrophic bacteria, the "specific
activity index."  Major differences in specific activity
were found in coastal and estuarine waters of north-
eastern Massachusetts.  These differences follow changes
in habitats; e.g., a decline with depth in the water
column and with increasing distance from the coast.
This approach, applied to enrichment studies, has re-
vealed an interesting pattern of bacterial response to
enrichment.  Cell numbers increase linearly, while  ac-
tivity increases exponentially, as does the plate count.
Based on the results of these studies and the work  of
others, an adaptive strategy of the natural heterotrophic

           bacteria can be postulated that involves reversible
           changes from functional dormancy to a high level of
           activity.  Additional responses to enrichment, or pol-
           lution, include growth and multiplication, and cer-
           tainly selection.  The sequence of responses is dis-
           cussed and related to the problems of measuring the
           response of a natural population of bacteria to any
           added substance.

     For the past fifteen years freshwater and marine microbial ecologists
have been using radiolabelled organic solutes to probe the activities of het-
erotrophic bacteria in natural waters and sediments.  These workers have em-
ployed methods designed to deal with microbial populations under conditions as
close as possible to those in nature.  The basic goal of this work has been to
understand the activity of the heterotrophic bacteria in relation to the fun-
damental processes of energy flow and material cycling in aquatic ecosystems.
Obviously, this information is essential for understanding of the potential
role microbes might play in breakdown of organic pollutants.  The first part
of this paper is a presentation of the most significant results of previous
studies of heterotrophic activity as they relate to the goal of this confer-
ence.  Following this is presented some recent work on the relationships be-
tween microbial numbers and activity in estuaries and coastal waters.


     Research on heterotrophic bacterial activity has confirmed the informa-
tion obtained from the better chemical analyses concerning the normal concen-
tration range of low molecular weight organic compounds,  i.e., below 200 mo-
lecular weight.  This confirmation derives from a kinetic analysis  of uptake
at substrate concentrations in the range of 1-100 lag/liter (35,39).  The an-
alysis yields a parameter (Kt + Sn),  which is a combined  value including natu-
ral substrate concentration (Sn) , and a transport constant (Kt) indicative of
the affinity of the microbial transport systems for the substrate in question.
Measured (Kt + Sn)  values are useful  in setting an upper  limit on Sn, cer-
tainly giving an order of magnitude indicative of the natural substrate con-
centration.  (Kt + Sn) values have been reported for a number of substrates
from a variety of aquatic systems, and have been compared with Sn measured
chemically in a number of studies (3,6,9,20,41).

     Essentially, the low molecular weight compounds are  singly present at
concentrations of several micrograms  per liter or less, results that indicate
higher concentrations are highly suspect.  For example, the concentration of
total dissolved free amino acids in estuarine and coastal waters, measured
chemically, shows a range of 25 to 100 micrograms per liter (3,9).   Even if
the waters are polluted, the range is at most only double the normal level.
Offshore and deeper waters indicate lower levels, from 2  to 10 micrograms per


liter  (3,33).  This has led to the important understanding that in order to
work with the natural bacteria, one must measure their activities relative to
such low substrate concentrations.


     Using simple radiolabelled organic compounds like glucose, glutamate and
acetate, many workers have shown that heterotrophic microbes in eutrophic and
mesotrophic waters are active in taking up and metabolizing substrates at con-
centrations of a few micrograms per liter.  The uptake can be characterized as

     (1) Uptake shows a saturation type response to substrate concentration
which can usually be analyzed with Michaelis-Menten kinetics and is best mea-
sured in a concentration range of one to 50 micrograms per liter.  This clearly
indicates that a most important adaptation for heterotrophic microbes is the
possession of transport systems functional in this concentration range.  Mea-
surements of transport constants have confirmed this (14,28,39).

     (2) The turnover of many low molecular weight compounds is rapid, from a
few hours or less in summer conditions to several days or more in colder waters
or in winter.  Turnover is much more rapid in polluted and eutrophic waters
than in oligotrophic waters.  Obviously, to accurately assess the importance
of a given compound to heterotrophs it is necessary to measure more than just
its concentration.  Two of the standard methods employed for measuring hetero-
trophic activity yield the natural turnover time (5,12,32,39), defined as the
ratio of substrate concentration to rate of use.

     (3) A substantial proportion of substrate taken up is respired or miner-
alized.  The percent varies, depending on the substrate and other factors,
from 1 or 2% to as much as 80% for some substrates like glycolic acid (12,19,
40).  Most commonly, respiration is in the range of 25-50% of uptake.  After
the first few minutes, this percentage is usually quite constant.  This pro-
cess indicates the major pathway for nutrient regeneration from dissolved or-
ganic matter in natural waters (3,36).

     (4) There has been some confusion in the past as to which organisms are
responsible for the measured heterotrophic activity in natural waters.  Wright
and Hobbie (39) used a kinetic analysis of natural and cultured populations of
bacteria and algae to conclude that most of the uptake detected at close to
natural substrate concentrations was due to the bacteria.  Algal uptake of or-
ganic compounds appeared to be measurable only at relatively high and unnatu-
ral substrate concentrations.  Two recent lines of research have established
beyond reasonable doubt that the bacteria are the primary agents of natural
heterotrophic use of dissolved organic solutes.  The use of Nuclepore filters,
with their superior properties of acting like a true sieve, has shown that in
coastal and open ocean waters, 90% of the measurable heterotrophic uptake is
in the fraaction less than one micron in size  (4).   Where significant hetero-
trophic activity is associated with larger particles, there is evidence that
bacteria on detritus are responsible for most of the activity (16).  Auto-
radiography has confirmed these results, giving a qualitative indication that
most of the heterotrophic activity of a natural plankton community is


associated with the smallest particles, that is, the bacteria (22).

     (5) Some substrates are actively taken up and metabolized by a given mi-
crobial population, while others are not.  Those substrates most frequently
used by active heterotrophic populations are the simpler compounds, like glu-
cose, acetate, lactate and most of the amino acids, with obvious entry into
the major biochemical pathways (7,15,20,40).  In nutrient-poor waters, how-
ever, little heterotrophic activity can be measured for any substrate, and the
uptake usually does not follow Michaelis-Menten kinetics (29) .  Thus, speci-
ficity occurs in terms of preferences for a given compound, and it is probably
linked to the specificity of transport systems possessed by the major bacteria
species in a population.  The lack of measurable activity toward even the simp-
ler substrates in some waters is a matter that will be addressed subsequently.

     (6) Rates of heterotrophic uptake and/or mineralization of specific sub-
strates at close to natural substrate concentrations are expressed as either
Vmax (heterotrophic potential) values or actual flux rates (39,41).  The lat-
ter rate requires an independent measurement of substrate concentration and
hence is less commonly performed.  The Vmax values are derived from a kinetic
analysis of uptake, and are an expression of the instantaneous flux of sub-
strate use by a given population if their use is saturated with substrate (see
37 for a discussion on this method).  Measured either way, the rates show a
broad range of several orders of magnitude in heterotrophic activity.  Vmax
values have been reported as high as several yg'liter"1 (2,7,9.41), and they
can be found at least four orders of magnitude lower (1,15).  The factors con-
trolling this broad range include temperature, size and physiological state of
the existing microbial population, rate of input of usable dissolved organic
matter, grazing by protozoa, and probably others.

     (7) Studies of heterotrophic activity have uncovered some important eco-
logical relationships.  For example, there is evidence of a close energy coup-
ling between planktonic algae and bacteria, with the algae nourishing the bac-
teria by excretion of dissolved organic carbon (11,18,31).  It seems certain
that this is the major autochthonous source of energy for the natural bacteria.
Most of the excreted organic carbon is readily taken up and mineralized, while
a fraction (ca 25%) is relatively recalcitrant (18).

     Some workers have applied heterotrophic activity measurements to the
sediments (8,17).  The results have revealed very high heterotrophic activity
concentrated at the surface of the sediments.  In shallow waters, this may ex-
ceed the activity of the entire water column.

     Several workers have tested the response of natural microbial populations
to short-term enrichment with low molecular weight organic solutes (30,34).
These tests have involved concentrations of 25 yg liter"1 up to several mg
liter"1, and have been measured in bottles for 24 to 48 h.  More detailed in-
formation will be given subsequently, but some general findings of these ex-
periments include:  (a) a fairly constant rate of use for several hours,
higher for higher applied concentrations but definitely not proportional to


applied concentration;  (b) a marked, often exponential increase in uptake and
mineralization of the test substrate by the end of the first day;  (c) exhaus-
tion of the substrate in two days, if it was added at concentrations up to 500
ug/liter.  Unfortunately, most of these experiments have not involved simulta-
neous measurements of microbial population  changes.  They do, however, indi-
cate that natural microbial populations can rapidly accommodate a 10- to 100-
fold increase in substrate.  This certainly corroborates the low ambient con-
centrations of organic solutes in natural waters.  The response has been found
to be more rapid in onshore and estuarine waters than in the open ocean.

     A closely related observation concerns the response of natural microbial
activity to organic pollution.  Where natural waters are obviously polluted,
heterotrophic activity is typically higher than similar unpolluted waters by
one or two orders of magnitude (2,7).  Furthermore, significant rates of me-
tabolism of more complex pollutants can be demonstrated.  For example, sig-
nificant   C-hexadecane metabolism was found in Tokyo Bay, where oil pollution
is obvious  (26).
                            POPULATION PARAMETER

     For some time workers have been measuring heterotrophic activity without
measuring the microbial population responsible for that activity, the main ex-
cuse being the lack of suitable methodology.  With the recent establishment of
the Acridine Orange Direct Count  (AODC) using epifluorescence and Nuclepore
filters, that excuse is no longer valid.  Although there are still some alter-
natives in methodology, the AODC is well enough understood, and simple enough,
so that any laboratory equipped to do heterotrophic activity measurements
should also be measuring the bacteria with AODC (21).  To quote Ralph Daley in
a recent methods symposium:  "With a reasonable amount of preliminary testing
and attention to procedural detail, AO epifluorescence can provide rapid, pre-
cise quantitative data on bacterial numbers in all marine and freshwater en-
vironments.  The use of alternative methods for this specific purpose is,
therefore, no longer justifiable"  (10).

     Given this counting technique and the already established methods for
measuring heterotrophic activity with radioisotopes,  the next logical step is
to relate heterotrophic activity to the bacterial population.  I have proposed
the term "specific activity index" to give a quantitative expression of the
relationship between activity and numbers  (38).  Because there are three basic
approaches to measuring heterotrophic activity with labelled organic solutes,
three indices will be required  (37).

     Details of the three approaches are given by Wright (37) , while the ap-
plication of these approaches to specific activity measurements is found in
Wright  (38).  Very briefly, the numerator of the given specific activity index
is a measurable function of heterotrophic bacterial activity (e.g., turnover
rate, Vmax/ direct uptake) and the denominator is the direct count of bacteria
from the sample measured.

     There are several current problems that this approach brings particularly
well into focus.  One of these is the discrepancy between the normal range of


bacterial  numbers  and the range in heterotrophic activity.  In my studies of
the  seasonal, horizontal  and vertical variation in bacteria in estuaries and
coastal water in northeastern Massachusetts,  I have found a range of 0.3 to
6 x  10 6 bacteria/ml,  slightly over one order  of magnitude.  Heterotrophic ac-
tivity, however, can  vary several  orders of magnitude.  In Figure 1 is pre-
sented the Vmax  specific  activity  index for glucose uptake at a depth of one
meter  from well  up the Essex estuary along a  horizontal transect to a station
14 km offshore in  the Gulf of Maine.   Figure  2 presents the same kind of data
for  a vertical profile at the 14 km offshore  station.   Specific activity de-
clined two orders  of  magnitude from a peak at the estuary inlet to the 40 m
depth waters.  Clearly, no direct  correlation between numbers and heterotro-
phic activity can  be  expected; this finding holds for other substrates and for
different  seasons  in  these temperate marine waters.  However, the changes in
specific activity  do  show a pattern,  one that corresponds with a profound
change in  marine habitats,  from a  shallow,  salt-marsh  bordered tidal river to
the  offshore station  46 m in depth.
                                                                 100 =
                4.0km 3.0km 1.5km |  1.5km
                   Upriver    ~~
4.0km   7.0km
14.0 km
Figure 1.  Bacteria numbers  (clear) and Vmax specific activity  index (yg«
           cell  «hr  '10 12) for glucose uptake  (stippled)  at  one  meter depth
           along a transect from within the Essex estuary  to a  station 14  km
           offshore, August 4, 1976.

                            Bacteria  10 -ml
                          0.5       1.0
                             5      ,     10
                            Specific Activity Index
Figure 2.  Bacteria numbers  (clear) and Vmax  specific  activity  index  (yg»
           cell^'hr"1 -10"12)  for glucose  uptake  (stippled)  from a  vertical
           profile at  a station 14 km  offshore  in  the  Gulf  of Maine,  August 13,
     Another current problem  is the  spectrum in  sizes  and shapes  of the bac-
teria and the possibility that some  of  this  spectrum may be  indicative of
changes in the physiological  state of the  bacteria,  where the smallest bacte-
ria are viewed as being starved forms with much  lower  physiological activity
(27).  Results from the Merrimack River, which is  quite  polluted  in the fresh-
water part of the estuary,  are shown in Figure 3.  These data represent the
separation of bacteria numbers and activity  using  Nuclepore  filters.   The up-
river stations are in fresh water; the  station 5 km  from the inlet had a sali-
nity of 7.3 ppt and the inlet station a. salinity of  23.9 ppt.   There  is a pro-
nounced shift in the size spectrum of bacteria from  an apparently large pollu-
tion-related freshwater flora to a typical coastal marine flora dominated by
the smaller bacteria.  In the transition,  specific activity  declined abruptly.
Because of the added parameters of freshwater to marine  transition and pollu-
tion, this study does not give any clear answers to  the  question  of starvation-
related size and activity changes.   This kind of approach, however, may be
useful in solving the above question.

                 Specific Activity Index
        Bacteria • Ifl'-ml"1
         £ 8
                              O 4* K) o
                              b » *.
                              -a -a -B
Figure 3.  Bacteria numbers and turnover rate specific activity  index (cell"1*
           hr~1*l«10~6) from several stations along  the Merrimack  River es-
           tuary, one meter depth, November 21, 1977.  Numbers and hetero-
           trophic activity fractioned using Nuclepore filters of  different
           pore sizes.

     Some interesting things happen when a seawater  sample  is enclosed in a
bottle and numbers and activity are followed over time.  Results from a mid-
winter sample enriched with glucose and incubated at two temperatures (Fig.  4)
illustrate two of the basic responses to enrichment  mentioned earlier,  i.e.,
an exponential increase in activity (even specific activity), and  a basically
linear increase in cell numbers.  Similar enrichment experiments have been
conducted using an additional indicator of bacterial activity.   This  is the
method of Iturriaga and Kheinheimer (23) , which gives a count of bacteria
cells with active electron transport systems via the reduction of  a tetrazo-
lium salt (INT) to a red colored formazan.  Figures  5 and 6  show the  results
of an experiment where 500 ug-liter"1  glutamate was  added to the enriched sam-
ple.  Specific activity is plotted against INT counts in Figure  7.  These fig-
ures indicate that the incubation and enrichment conditions  stimulate a sub-
stantial increase in bacterial activity accompanying a much  lower  increase in
bacterial numbers.  Some of the increase in specific activity is probably
caused by an increase in size which occurs (Fig. 8).











— -


















on *
u.u •—
1.2 E
16       28       40
Time  (hours)
Figure 4.  Bacteria  numbers (clear)  and direct
  uptake specific  activity index (yg cell"1"hr"1•
  10~12) glucose uptake  (stippled)  in Essex es-
  tuary water enriched with 100 yg-l"1 (2C incu-
  bation) or 300 yg-1"1  (15C incubation)  glucose.
                                                                                                        100 .^
                                                                             12      24
                                                                               Time (hours)
                                        Figure 5.  Essex  estuary water (salinity 25.9
                                          ppt, temperature  -2°C)  from January 31, 1978,
                                          incubated at  2°C  for  48 h,  unenriched.  A =
                                          direct counts  (AODC);  B = bacteria with active
                                          electron transport (INT counts);  C = glutamate
                                          turnover rate specific activity index  (cell"1-
                                          hr^-l-lO-6) .

                                                                             Direct Uptake Activity Index   (B)
                                                                 O.8 -
                                                                 0.6 -
                                                               .™ O.4 -
                                                                 0.2 -
                                                                               SO           1OO
                                                                              Turnover Rate  Activity Index   (A)
                         Time (hours)
    Figure  6.   Essex estuary water  from January 31,
      1978,  enriched with 500 Ug-l"1  glutamate.
      A = direct counts  (AODC); B =  INT counts;
      C = glutamate direct uptake specific activity
                                                           Figure  7.   Plot of INT counts of bacteria vs
                                                             specific activity  indices for unenriched
                                                             sample  (A)  and enriched sample  (B)  from
                                                             January  31, 1978,  sample from Essex estuary.

             _  O
              £ i.o
                              12     24
                                 Time  (hours)
Figure 8.  Mean bacterial volume  (u3) measured microscopically fron enrichment
           experiment of January  31, 1978.  A = control, B = enriched.
           Brackets indicate limits of mean ± 2 x S.E.-.
     It is instructive to summarize the bacterial responses induced by enclo-
sure of a natural sample in a bottle, with or without specific enrichment.
These are:   (1) a linear, often delayed increase in direct counts; (2) an in-
crease in specific  (meaning per cell) heterotrophic activity, delayed at first
but then exponential;  (3) an increase in the percent bacteria showing active
electron transport systems, at first rapid but eventually leveling off; (4)  not
shown by this experiment but repeatedly obtained when measured, an exponential
increase in colony-forming units as measured by plate counts; (5)  a slow in-
crease in the mean volume of the bacteria.  Based on these responses, a tenta-
tive analysis would suggest a basic adaptive strategy of the heterotrophic bac-
teria in natural waters.  This strategy has been suggested in part by others
(27).  A natural heterotrophic population is made up of a spectrum of cells
with a range of sizes and activity.  Most of them are probably on the "down"
side of functional activity, in a more or less dormant state (27) .  Given fav-
orable conditions—enrichment, enclosure in a bottle—many, if not most, come
out of dormancy.  This may mean a number of things not well understood for
natural bacteria—the energizing of membranes, the synthesis of permeases and
internal enzymes.  Whatever the causes, the changes can account for remarkable
increases in heterotrophic activity without significant increases in numbers

or selection of specific types adapted to enrichment conditions.  This strategy
is made complete by being reversible,  the evidence for the reversal of the
changes being less easy to show,  however.  The movement in the direction of
functional dormancy is presumably caused by nutrient starvation.  Figures 1
and 2, showing the marked decrease in  specific activity with distance offshore
and depth, give evidence of this  movement,  although only by inference.  The
important research by Novitsky and Morita (24,25)  has established a similar
reversible pattern for bacteria in culture conditions.

     Practically any imaginable system for measuring the response of natural
bacteria to pollutants or toxic substances will involve enclosure of a natural
sample, and thus the adaptive strategy just described.  There are at least two
other kinds of changes that will  also  occur, depending on the amount of time
of incubation.  The first one is  a function of transport kinetics, and is con-
centration dependent.  No adaptation is required.   Uptake and metabolism of a
substrate will be increased because under natural  conditions substrate concen-
tration was somewhere below saturation levels.  Uptake will approach and per-
haps exceed Vmax if other mechanisms—i.e.,  diffusion—are also present.  This
is a reversible response, entirely dependent on substrate concentrations.

     The second response is the one described above—the reversible changes
involved in movement into and out of functional dormancy.  This response be-
comes evident during the first day of  incubation.

     The third response is the growth  and multiplication of many of the bac-
teria successful in the second response.  In fact, it is difficult to separate
these two responses, because they are  outcomes of  the same physiological
changes in the bacteria.  Perhaps the  best way to  do this is to use cold tem-
perature systems to expand the responses in time.   After two or three days,
the most available naturally occurring substrate is exhausted in the enclosed
sample, as is the added substrate if it was  readily metabolizable and added in
the  <1 mg liter"1 concentration  range.  If higher concentrations were added,
and if incubation is extended, a  severe selection  of bacteria will occur and
continued growth will be a response to the high concentrations and the other
enclosure conditions.  By this time, the system is rather far removed from a
natural population, and the results have increasingly less applicability to
natural systems.  They also have  increasingly less predictive power and must
be interpreted with great caution.  Unfortunately, there may not be any rea-
sonable alternatives to enclosing and incubating samples.  When this is done,
however, it is highly advisable to measure the population as well as its re-

                               ACKNOWLE DGMENTS

     This material is based in part on research supported by the National
Science Foundation, Grants DBS 73-06650 A02  and OCE 77-19446.  I am particu-
larly grateful to Rebecca Buck for technical assistance in carrying out this

                               LITERATURE CITED

 1.   Albright, L.  J.  1977.  Heterotrophic bacterial dynamics in the Lower
      Fraser River, its estuary and Georgia Strait, British Columbia, Canada.
      Mar. Biol. 39:203-211.

 2.   Allen, H. L.   1969.  Chemo-organotrophic utilization of dissolved organic
      compounds by planktic algae and bacteria in a pond.  Int. Rev. Ges.
      Hydrobiol. 54:1-33.

 3.   Andrews, P.,  and P. J. LeB. Williams.  1971.  Heterotrophic utilization of
      dissolved organic compounds in the sea.  III. Measurement of the oxida-
      tion rates and concentrations of glucose and amino acids in sea water.
      J. Mar. Biol. Assn. U.K. 51:111-125.

 4.   Azam, F., and R. E. Hodson.  1977.  Size distribution and activity of
      marine microheterotrophs.  Limnol. Oceanogr. 22:492-501.

 5.   Azam, F., and O. Holm-Hansen.  1973.  Use of tritiated substrates in  the
      study of heterotrophy in sea water.  Mar. Biol. 23:191-196.

 6.   Burnison, B.  K., and R. Y. Morita.  1974.  Heterotrophic potential for
      amino acid uptake in a naturally eutrophic lake.  Appl. Microbiol.

 7.   Carney, J. F., and R. R. Colwell.  1976.  Heterotrophic utilization of
      glucose and glutamate in an estuary:  effect of season and nutrient  load.
      Appl. Environ. Microbiol. 31:227-233.

 8.   Christian, R. R., and J. R. Hall.  1977.  Experimental trends in sediment
      microbial heterotrophy:  Radioisotopic techniques and analysis, p. 67-88.
      In B. Coull (ed.), Ecology of marine benthos.  University of South Caro-

 9.   Crawford, C.  C., J. E. Hobbie, and K. L. Webb.  1974.  The utilization of
      dissolved free amino acids by estuarine microorganisms.  Ecology

10.   Daley, R. J.   Direct epifluorescence enumeration of native aquatic bac-
      teria:  uses, limitations and comparative accuracy.  Proc. ASTM Symposium,
      Native Aquatic Bacteria; Enumeration, Activity and Ecology, June 25,  1977,
      Minneapolis (in press).

11.   Derenbach, J. B., and P. J. LeB. Williams.  1974.  Autotrophic and bacte-
      rial production:  fractionation of plankton populations by differential
      filtration of samples from the English Channel.  Mar. Biol. 25:263-269.

12.   Gocke, K.  1976.  Respiration von gelosten organischen Verbindungen durch
      naturliche Mikroorganismen-Populationen.  Ein Vergleich zwischen ver-
      schiedenen Biotopen.  Mar. Biol. 35:375-383.

13.  Gocke,  K.   1977.   Comparison of methods for determining the turnover
      times  of dissolved organic compounds.   Mar.  Biol.  42:131-141.

14.  Hamilton,  R.  D.,  K. M.  Morgan,  and J. D.  H. Strickland.  1966.   The glu-
      cose uptake  kinetics of some marine bacteria.   Can.  J. Microbiol.

15.  Hamilton,  R.  D.,  and J. E.  Preslan. 1970.   Observations on heterotrophic
      activity in  the Eastern Tropical Pacific.   Limnol.  Oceanogr.  15:395-401.

16.  Hanson, R. B.,  and W. J. Wiebe.  1977.   Heterotrophic activity  associated
      with particulate size fractions in a Spartina  alterniflora salt-marsh
      estuary,  Sapelo Island, Georgia, USA,  and  the  continental shelf waters.
      Mar. Biol. 42:321-330.

17.  Harrison,  M.  J.,  R. T.  Wright,  and R. Y.  Morita.   1971.  Method for mea-
      suring mineralization in lake  sediments.   Appl.  Microbiol.  21:698-702.

18.  Herbland,  A.   1975.  Utilisation par la flore heterotrophe de la matiere
      organique naturelle dans 1'eau de mer.   J. Exp.  Mar.  Biol.  Ecol.

19.  Hobbie, J. E.,  and C. C. Crawford.  1969.   Respiration correction for bac-
      terial uptake of dissolved organic compounds in  natural waters.  Limnol.
      Oceanogr. 14:528-532.

20.  Hobbie, J. E.,  C. C. Crawford,  and K. L.  Webb.   1968.   Amino acid flux
      in an estuary.  Science 159:1463-1464.

21.  Hobbie, J. E.,  R. J. Daley, and S. Jasper.   1977.  Use of Nuclepore fil-
      ters for counting bacteria by  fluorescence microscopy.  Appl.  Environ.
      Microbiol. 33:1225-1228.

22.  Hoppe,  H.  G.   1976.  Determination and  properties  of  actively metaboliz-
      ing heterotrophic bacteria in  the sea,  investigated  by means of micro-
      autoradiography.  Mar. Biol. 36:291-302.

23.  Iturriaga, V. R., and G. Rheinheimer.   1975.  Eine einfache Methode zur
      Auszahlung von Bakterien mit aktivem Elektronentransportsystem in
      Wasser- und  Sedimentproben. Kieler Meeresforsch. 31:83-86.

24.  Novitsky,  J.  A.,  and R. Y.  Morita.  1976.   Morphological characterization
      of small  cells resulting from  nutrient starvation of a psychrophilic
      marine vibrio.  Appl.  Environ. Microbiol.  32:617-622.

25.  Novitsky,  J.  A.,  and R. Y.  Morita.  1977.   Survival of a psychrophilic
      marine vibrio  under long term  nutrient starvation.   Appl. Environ.
      Microbiol. 33:635-641.

26.  Seki, H.  1976.  Method for estimating  the  decomposition of hexadecane in
      the marine environment. Appl. Environ.  Microbiol.  31:439-441.

27.  Stevenson, L. H.  1978.  A case for bacterial dormancy in aquatic sys-
      tems.  Microbial Ecol. 4:127-133.

28.  Vaccaro, R. F., and H. W. Jannasch.  1966.  Studies on heterotrophic ac-
      tivity in seawater based on glucose assimilation.  Limnol. Oceanogr.

29.  Vaccaro, R. F., and H. W. Jannasch.  1967.  Variations in uptake kinetics
      for glucose by natural populations in seawater.  Limnol. Oceanogr.

30.  Vaccaro, R. F.  1969.  The response of natural microbial populations in
      seawater to organic enrichment.  Limnol. Oceanogr. 14:726-735.

31.  Wiebe, W. J., and D. F. Smith.  1977.  Direct measurement of dissolved
      organic carbon release by phytoplankton and incorporation by micro-
      heterotrophs.  Mar. Biol. 42:213-223.

32.  Williams, P. J. LeB., and C. Askew.  1968.  A method of measuring the
      mineralization by microorganisms of organic compounds in seawater.
      Deep-Sea Res. 15:365-375.

33.  Williams, P. J. LeB., T. Berman, and 0. Holm-Hansen.  1976.  Amino acid
      uptake and release by marine heterotrophs.  Mar. Biol. 35:41-47.

34.  Williams, P- J. LeB., and R. W. Gray.  1970.  Heterotrophic utilization
      of-dissolved organic compounds in the sea.  II. Observations on the re-
      sponses of heterotrophic marine populations to abrupt increases in amino
      acid concentration.  J. Mar. Biol. Ass. U.K. 50:871-88.

35.  Wright, R. T.  1973.  Some difficulties in using  ^-organic solutes to
      measure heterotrophic bacterial activity, p. 199-220.  In H. L. Steven-
      son and R. R. Colwell  (eds.), Estuarine microbial ecology.  University
      of South Carolina Press, Columbia.

36.  Wright, R. T.  1974.  Mineralization of organic solutes by heterotrophic
      bacteria, p. 546-565.  In R. R. Colwell and R. Y. Morita (eds.), Effect
      of the ocean environment on microbial activities.  University Park
      Press, Baltimore.

37.  Wright, R. T., and B. K. Burnison.  Heterotrophic activity measured with
      radiolabelled organic substrates.  Proc. ASTM Symposium, Native Aquatic
      Bacteria; Enumeration, Activity and Ecology, June 25, 1977, Minneapolis
      (in press).

38.  Wright, R. T.  1978.  On the measurement and significance of specific
      activity in the heterotrophic bacteria of natural waters.  Appl.
      Environ. Microbiol. 36:297-305.

39.  Wright, R. T., and J. E. Hobbie.  1966.  The use of glucose and acetate
      by bacteria and algae in aquatic ecosystems.  Ecology 47:447-464.

40.  Wright,  R.  T.,  and N.  M.  Shah.   1975.   The trophic role of glycolic acid
      in coastal seawater.   I.  Heterotrophic metabolism in seawater and bac-
      terial  cultures.   Mar. Biol.  33:175-183.

41.  Wright,  R.  T.,  and N.  M.  Shah.   1977.   The trophic role of glycolic acid
      in coastal seawater.   II.  Seasonal  changes in  concentration and hetero-
      trophic use in Ipswich Bay, Mass.,  USA.   Mar.  Biol.  43:257-263.

                          IN SEA AND LAKE  WATER

                      Yuzaburo Ishida and Hajime  Kadota
                         Laboratory of Microbiology
                           Department of Fisheries
                           Faculty of Agriculture
                                Kyoto, Japan

                The structure of bacterial communities  in the sea
           and lakes is changed by environmental  pollution or eutrp-
           phication.  The effect of increased concentration of or-
           ganic matter upon the structure of bacterial communities
           in natural water was observed using various  selective
           media and a continuous culture system.  A new method for
           the enumeration in natural water of extremely oligotro-
           phic bacteria which cannot be counted  using  conventional
           solid media or diluted liquid media supplemented with
           organic nutrients is presented.  It was found that ex-
           tremely oligotrophic bacteria were predominant in oligo-
           trophic lake water.

     The distribution and species composition of heterotrophic bacteria in the
sea and lakes are much influenced by environmental pollution.  When a concen-
tration of organic matter in water is increased by pollution the bacteria
usually found in the unpolluted pristine environment decrease in number or
disappear, and inversely, the bacteria which "prefer" high nutrient media in-
crease in number (1,8, 13).   However, there are few unequivocal evidences
available to document the relationship between the structure of the bacterial
community and concentration of organic matter in natural water.

     In this paper, an attempt is made to clarify the effect of an increase in
concentration of organic matter in water by pollution on the structure of bac-
terial communities in the sea and lakes by using various selective media and a
continuous culture system.

                            MATERIALS AND METHODS

     Water samples were collected at stations 3, 4 and 5 in Osaka Bay (Fig. 1)
which were at 6, 21 and 50 km, respectively,  offshore from the mouth of the
Yodo River.  The water samples from stations  3, 4 and 5 contained 5.1, 2.3 and
1.5 mg, respectively, of dissolved organic carbon per liter.  Station A was in
a heavily polluted site with 4.0 mg of dissolved organic caron.

     The Zl medium used for this field survey contained 5 g of peptone (Daigo),
0.5 g of yeats extract (Difco) and 13 g of agar (Difco)  in one liter of aged
sea water, the latter collected from an unpolluted area.  The pH value was ad-
justed to 7.7-7.8.  The Z/20 medium contained peptone and yeast extract at the
strength of 1/2 of Zl medium.
                Awaji Is
Figure 1.  Location of sampling stations in Osaka Bay,  bacterial number
           counted by using Zl and Z/20 media and bacterial  flora at each
           station.  Concentration of organic matter expressed as DOC in sur-
           face water at each station is also shown.


     The  culture vessel was  charged with 500 ml of  raw sea water from station
A and filtered through a stainless steel net  (20 ym mesh)  to remove detrital
particles and zooplankters.  The reservoir was filled with 5 liters of the
filter-sterilized sea water  collected at the same station.  A second culture
vessel was  charged with 500  ml  of unsterilized diluted sea water, which con-
sisted of one part of the polluted sea water  (DOC:   4.0 mg C/ml) and nine
parts of  clean sea water  (DOC:   0.9 mg C/ml) which  was filtered through a 20 ym
mesh stainless steel net.  The  reservoir was filled with 5 liters of the di-
luted sea water after being  sterilized through 0.22 ym filter.  The dilution
rate was  adjusted to 0.123-0.130.  The incubation  temperature was adjusted to
the in-situ water temperature  (12°C).


     Figure 2 illustrates the  scheme for counting various  groups of hetero-
trophic bacteria including extremely oligotrophic bacteria in water of Lake
     Water sample
           102  1Q3  104  10s
           A   A   A   A
TF 1
                                 Turbidity (2,3,4 weeks' incubation)
                        ,—(A)—> Colony formation test
                               2.3A . Tf ~	* O TF1 Q9Qr P'Qte
                             —(B)-» 14CO? Production from
kQlF102agar plate
2weeks' incubation
                              A weeks'
            ''C-organic subs.
            (10 nrjjc/0.1 ml/tube)
                                                        vial containing ethanolamine
                                                         scintillator for COj trapping
Figure  2.   Diagrammatic  scheme for enumeration  of extremely oligotrophic bac-
      After serial dilutions  with sterilized artificial lake water  (ALW:   0.5
mg NaCl,  0.38 mg KC1,  0.9  mg CaCl2«4H20, 0.1 mg  MgSOit-VHzO, 50 yg MnCl2-4H20,
40 yg ZnCl2«7H2O, 1 yg CuCl2-5H20, 0.5 yg CoCl2«6H20,  and 2 yg FeCls per  liter
of distilled water), one ml  of the diluted samples  were inoculated  into each
of five test tubes containing TF1 medium  (5 g  trypticase (BBL) and  0.5 g  yeast
extract (Difco) in one liter of ALW at pH 7.0).   The media TF10"1,  TF10"2,

TF10 3, and TFIO"4 were prepared to contain trypticase and yeast extract at
the concentrations of 1/1.0, 1/100, 1/1,000 and 1/10,000, respectively, of
those of the TF1 medium.  Lake water from an oligotrophic area in Lake Biwa
(Station 7) was filtered through a 0.22 ]am filter, dispensed into test tubes,
sterilized, and used as the lake water medium (LW7 medium).  The concentration
of dissolved organic matter in the latter was 1.9 mg C/liter.

     After incubating the inoculated media at 20°C for 14, 21 and 28 d, the
growth of bacteria was detected by turbidity in TFl, TF10"1 and TF10"2 media.
The most probable number of bacteria was calculated by use of Hoskins' table

     Since bacterial growth in TF10"3, TFIO"1* and LWy media could not be de-
tected by turbidity, the following procedures were used.

     a) Colony formation on agar plates:  0.1 ml of the culture at 14, 21 or
28 d incubation was spread onto each of duplicate agar plates of TFl and TF10~
media.  The plates were incubated at 20°C for 14 d.  The tube which gave more
than 10 colonies on the plate was regarded as positive.

     b) Detection of 14CC>2 production from organic compounds labelled with   C:
After incubation for 28 d, 2.5 ml of culture in each tube was aseptically
transferred to a sterilized tube containing 0.1 ml of 0.1 yCi/ml  ^C-labelled
compound as a tracer level substrate.  As the substrate, L-[U-1'tC]  glutamic
acid  (275 m Ci/m mol) ,  [U-^C] glucose  (335 m Ci/m mol) and [U-llfC]  protein
hydrolysate (57 m Ci/m atom), respectively, were used.  The tubes were incu-
bated at 20°C for 14 d.  The lltCO2 produced during incubation was expelled by
bubbling with COa-free air for 1 min after the addition of 0.1 ml of 1.2 N HC1.
The C02 was trapped with 10.5 ml of the absorbing solution composed of 1 ml of
ethanolamine, 1.5 ml of ethylcellusolve and 8 ml of scintillator which con-
tained 4 g PPO, 0.2 g POPOP, and 60 g napththalene in one liter of toluene.
Contamination of bacteria in the tube was checked by putting a drop of the
culture onto TF10"1 agar plate before acidification and incubating it at 20°C
for 14 d.  Radioactivity of the absorbing solution in vial was counted by an
Aloka LSC-502 liquid scintillation counter.  A tube which gave radioactivity
of at least 40% higher than that of the blank without added culture was re-
garded as positive.
                           RESULTS AND DISCUSSION


     Figure 3 shows the bacterial counts obtained by using Zl and Z/20 media.
Heterotrophic bacteria were most abundant in water at station 3 and most
sparse in water at station 5.  The ratio of the Z/20 bacteria counted on Z/20
medium to the Zl bacteria counted on Zl medium increased in inverse proportion
to the DOC value of water.  In general, the ratio was larger in summer than in
winter.  The data indicate that higher populations of bacteria can be obtained
from polluted waters using media rich in organic nutrients rather than that
poor in organic nutrients, and vice versa for unpolluted waters.

        Polluted seawater
1/10 diluted seawater
                           8    12    16 0    4     8    12    16
VII 50
••• Z1- bacteria 100
1=1 Z/20 -bacteria


-,il , jn , „





             8      13     0     4    8

                 Incubation time (h)
Figure 3.  A.
Viable counts of Zl bacteria and Z/20 bacteria in the culture
vessels which were fed with polluted sea water (DOC:  4.0 mg/1)
and the 1/10 diluted sea water  (DOC:  1.2 mg/1).  •= Zl bacteria;
O= Z/20 bacteria.
Response of Zl bacteria and Z/20 bacteria to different concen-
trations of organic nutrients.  The Z1£Z/20 means the number of
of bacteria which formed larger colonies on Z/20 plates than Zl
plates as second media.
     Responses to different concentrations of organic nutrients of the bac-
teria from sea water of Osaka Bay were examined on Zl and Z/20 media.  Colo-
nies formed on each of the original agar plates (Zl and Z/20 media) were trans-
ferred to the second plates of Zl and Z/20 media.  After 1 or 2 wk incubation,
it was possible to determine whether bacteria which formed colonies on the
original plates grew well on transfers.  Sizes of corresponding colonies of
bacteria which could develop on both plates were compared.  Bacteria which
grew better'/m Zl than Z/20 and those which showed superior growth on Z/20
than on Zl'was expressed as Z1>Z/20 and Z1Z/20 to the
total was 91% in the water sample from station 3, and 70% from station 4.  Of
the 49 isolates from Z/20 plate (Z/20 bacteria), the ratio of Zl^Z/20 was 82%
at station 3, 86% at station 4 and 77% at station 5  (unpolluted area).  Simi-
lar results were obtained in the other seasons.  These data indicate that most
of the bacteria enumerated on Zl mediam are ordinary heterotrophs that grow

well on nutrient-rich media, and that most of the bacteria on Z/20 medium were
relatively oligotrophic form that developed well on nutrient-poor media.

     All the strains of bacteria isolated using both media were studied taxo-
nomically according to the scheme of Shewan et al.  (10), modified by Simidu
(12).  About 40% of bacterial strains isolated from the Zl medium belonged to
the genus Acinetobacter at all the stations examined.  These strains of Acine-
tobacter grew better on Zl medium than on Z/20 medium.  At station 5, 28% of
the strains isolated, using Zl medium, belonged to the genus Vibrio.  However,
more than 50% of the bacteria isolated by use of the Z/20 medium grew neither
on Zl nor Z/20 media when transplanted to corresponding new media.  Most of
the remaining strains isolated using Z/20 medium belonged to Flavobacterium at
stations 4 and 5, and to Acinetobacter at station 3.

     Taxonomic grouping of the bacteria from sea water in Osaka Bay was made
and their responses to different concentrations of organic nutrients on the
second plates were examined.  It was found that most of the Acinetobacter
strains isolated, by using Zl medium, from stations 3 and 4 in Osaka Bay grew
better on the second Zl medium than on Z/20 medium.  On the other hand, most
of the Flavobacterium strains from stations 4 and 5 grew better on the second
Z/20 medium than on Zl medium.

     It has generally been recognized that high concentrations of peptone in
agar media allow only a limited number of aquatic bacteria to form colonies
(2, 11) .   Little work, however, has been done on physiological differences be-
tween bacteria developed on nutrient-poor media and those grown on nutrient-
rich media.  Our results indicate that bacteria which grow well on nutrient-
poor media are different from bacteria showing better development on nutrient-
rich media, not only in response to nutrient concentrations but also in the
taxonomic types.  It seems that bacteria from nutrient-poor media were tempo-
rarily suppressed by high concentrations of organic nutrients.


     To obtain further information on the effect of organic substrates on the
structure of bacterial communities in sea water we attempted an analysis with
a continuous culture system charged with sea water containing organic sub-
stances in different concentrations.

     During incubation, viable counting of Zl and Z/20 bacteria in the culture
vessels were made at intervals of 2 h; aliquots of the water samples were
spread on Zl and Z/20 media and incubated at 15°C.  Colonies formed on each of
the original agar plates were directly transplated to second Zl and Z/20
plates.  Changes in population densities of Zl and Z/20 bacteria in the cul-
ture vessels during incubation are shown in Figure 3A.  When polluted sea
water was charged,  the ratio in number of Zl bacteria to Z/20 bacteria in-
creased with incubation time, while in 1/10 diluted sea water the ratio, did
not change.  Responses of the bacteria isolated from each of the plates to
different concentrations or organic nutrients are shown in Figure 3B.  The
ratio of the Zl bacteria which grew better on the second Z/20 medium than on

the second Zl medium to the total Zl bacteria remained almost constant during
13 h incubation, regardless of the concentration of organic substances in the
sea water.  For Z/20 bacteria, however, the ratio of the bacteria which grew
better on the second Z/20 media than on the second Zl media to the total num-
ber decreased regularly during incubation when polluted sea water was charged,
but increased gradually when the diluted sea water was charged.

     The bacteria of the genus Acinetobacter  (genus Moraxella was included),
which predominated in polluted sea water, continued to be predominant in the
culture vessel during 13 h incubation during charging with the polluted sea
water.  However, with 1/10 diluted sea water, containing organic nutrients in
low concentration, density of Acinetobacter decreased during the first 8 h in-
cubation  (corresponding to the first retention time).  In the latter instance,
the genus Vibrio was predominant.  These results may explain field observa-
tions wherein bacteria which grew better on nutrient-rich medium were predomi-
nant in the polluted sea water and the bacteria which grew well on nutrient-
poor medium predominated in the unpolluted sea water.


     Apparently two different bacterial populations (Zl and Z/20 bacteria)  com-
pete for the organic substrates in sea water and the growth rates of the popu-
lations depend upon the concentration of organic substrates.   Therefore, we
examined the growth kinetics of Zl and Z/20 bacteria with a continuous culture
system.  In this study, sea water samples containing organic substances in dif-
ferent concentrations  (DOC:  1.0 mg/1, 1.1 mg/1, 1.8 mg/1 and 5.0 mg/1)  were
fed to the culture vessels.  The number of bacteria grown on Zl and Z/20 media
were plotted against the incubation time.  According to the equation of Jan-
nasch  (9), actual growth rates were calculated from the washout rates
1/t In (X/XQ)•
                             y = D + i In (x/xQ)                          [1]

where XQ = population density at zero time (t=0), y = specific growth rate
(h"1)  and D = dilution rate (h"1).

     The relationship between the growth rates and the concentration of limit-
ing substrate is illustrated in Figure 4.  The competition between populations
of Zl and Z/20 bacteria was dependent on the concentration of organic sub-
stances as nutrients.  The saturation curves of both populations crossed at
approximately 5.0 mg C/l.   When the concentration was less than 5.0 mg C/l,
the population of Z/20 bacteria predominated.  These growth rates were ob-
tained with sea water samples collected in winter.   In summer, the growth rates
of Zl bacterial populations were found to be 0.130 h"1 at 6.0 mg C/l (25°C) and
0.030 h"1 at 1.2 mg C/l (20°C) , respectively,- whereas the populations of Z/20
bacteria were 0.095 h"1 at 6.0 mg C/l and 0.090 h"1 at 1.2 mg C/l.  These re-
sults suggest that the crossing point of the curves of Zl bacteria and Z/20
bacteria shifts to lower concentration of organic substrates.  Kinetics con-
stants for the growth of both the populations (Zl and Z/20 bacteria) were
roughly estimated from the reciprocal plots for the populations of Zl and Z/20
bacteria to be as follows, respectively:  K§ = 3.0 mg C/l and y^ = 0.19 h"1
(population of Zl bacteria), and K^ = 1.2 mg C/l and y^ = 0.15 h"1  (population
                                  O        -*          ill


of Z/20 bacteria).   The data reveal that the larger ym was  associated with
larger Ks, and the  smaller ym was  associated with smaller Ks.   Accordingly,
the population of Zl bacteria exhibited lower affinity toward  organic sub-
stances at the low concentrations  and the population of Z/20 bacteria showed
higher affinity for organic substrates at the low concentrations.   It is  in-
teresting that Jannasch arrived at the same conclusions by  using mixed cul-
tures of pure strains of aquatic bacteria (8).
                    0    1.0   2.0   3.0    4.0    5.0
                 Cone of organic substrate in reservoir
                            (DOC, mg/l )
Figure 4.  The growth rate-substrate concentration relationships of  Zl  and
           Z/20 bacteria.
     When the growth constants ym,  Ks and Y (cell  yield coefficient) were
known, the steady state concentration (X)  of bacterial  population  in the cul-
ture vessels could be predicted from the dilution  rate  (D)  and  the concentra-
tion of inflowing substrate (Sr),  according to the following  equation  (2):
     When the population density of Zl bacteria is  XA and  that of  Z/20  bac-
teria is XB, the condition in which the Zl bacteria become predominant  in
place of the Z/20 bacteria was

                                  XA>XB                                  [3]

By substituting [2]  for [3]

                               yB K* - y A KB

     In this instance, the dilution rate to satisfy  [3] is more than 0.11 per
hour, when YA of the population of Zl bacteria is equal to YB of the popula-
of Z/20 bacteria.  In the case of YA>YB, the dilution rate for satisfying  [3]
will be less than 0.11.

     These experiments reveal that the structure of bacterial communities in
the sea is controlled markedly by the concentrations of organic substances.


     The Z/20 bacteria enumerated on an agar medium containing organic nutri-
ents in relatively low concentrations in all likelihood play an important role
in the biodegradation of organic matter in aquatic environments.  However,
there may exist in aquatic environments other groups of heterotrophic bacteria
which cannot be counted by using diluted media such as Z/20 medium.  Since the
concentration of organic matter in the sea and lake waters (0.2 ~ 5 mg C/l) is
usually much lower than that in Z/20 medium there may be some bacteria which
can grow only in environments poor in organic nutrients.  Such a group of or-
ganisms tentatively can be called "extremely oligotrophic" bacteria.  There-
fore, we attempted the detection and enumeration of such bacteria using water
samples from various stations in oligotrphic and eutrophic areas of Lake Biwa .

     In this study, many kinds of liquid media containing organic substrates
in low concentrations were examined by use of extinction dilution methods .
Criteria for the growth of bacteria included turbidity increase, colony forma-
tion on agar plates, and lf*CO2 production from glutamic acid, protein hydroly-
sate and glucose labelled with   C (Fig. 2).  In water samples from station O
in an eutrophic area of Lake Biwa (DOC:  4-5 mg C/l) , the bacterial number
counted by use of TF10"1 medium was very large compared with that tabulated on
other media.  Numbers of extremely oligotrophic bacteria in this water sample
were only 1.6% of the population densities obtained with TF10 1.  However, for
water from station 7, in an oligotrophic area of Lake Biwa, the ratio of ex-
tremely oligotrophic bacteria to the TF10"1 medium-counted bacteria was high
(Table 1) .

     Of three kinds of organic substrates labelled with llfC, glutamic acid and
protein hydrolysate gave higher counts than glucose.  This result might re-
flect the fact that the concentration of amino acids was higher (100 yg or
more per liter as ninhydrin-positive substances) than that of glucose in the
dissolved form (approximately 3 yg per liter) in the lake water; glucose might
not be suitable substrate for the bacteria in the lake water.  Conversely, the
concentration of glucose used may have been insufficient for the extremely oli-
gotraophic bacteria.  To confirm this presumption, the llfC02 production ac-
tivity of 12 strains of extremely oligotrophic bacteria from the TF10"2,
TF10" , TF10"  and LW? media cultures was examined by use of LW? medium, LW?
medium with 1 ppm glucose and LW? medium with 10 ppm, to which 10 myc of


Bacterial number
Criteria for growth
Station 0 (May 7, 1977)
MPN (ml"1)



(A) Colony formation on TF1 and TF10"2
(B) Respiration:
rglutamic acid
< protein hydrolysate
(A) Colony formation on TF1 and TF10"2
(B) Respiration:
rglutamic acid
/protein hydrolysate

.4 x
.4 x
.4 x
.3 x
.5 x
.3 x
.3 x
.5 x
10 5
10 5

10 3
10 3
Ratio (%)*
7 (June 9, 1977)
MPN (ml-1)
10 3

*Ratio of each count to TF10"1 count.

[U-  C]-glucose was added.  It was found by this experiment that   COz was not
produced from all the media employed.  These results suggest that, compared
with amino acids, glucose is not a suitable substrate for respiration of ex-
tremely oligotrophic bacteria in lake water.  There is, however, a possibility
that the TF media were deficient in some components required for growth of ex-
tremely oligotrophic bacteria.  Therefore, the LW7 medium prepared from fil-
tered and sterilized lake water was used for enumeration of extremely oligo-
trophic bacteria  (Table 2).  The number of heterotrophic bacteria  (B) and that
of extremely oligotrophic bacteria  (B - A) with LW? medium with l '*C-glutamic
acid were 6.7 and 5.8 times, respectively, higher than that of ordinary hetero-
trophic bacteria counted by use of TF10~  medium.  The concentration of dis-
solved organic matter in LWy medium was 1.9 mg C/l which was comparable to
that of TFlCr3 medium.
          7 IN LAKE BIWA  (JULY 7, 1977)

„  ,.                „-,_.--        ^,                   Bacterial number

v-j. j. i-ej. j.
a. J-U.L i-} j. u w ui i

"(A) Colony formation on TFl and TF1CT2
(B) Respiration:
r glutamic acid
] protein hydrolysate
L glucose


(A) Colony formation on TFl and TF10 2
(B) Respiration: ,

- glutamic acid
protein hydrolysate
- glucose
(A) Colony formation on TFl and TF10"2
(B) Respiration: ,

_ glutamic acid
protein hydrolysate
i glucose
MPN (ml"1)
1.7 x 101*
2.4 x 104
4.9 x 103
1.7 x 103
1.7 x 10 3 ^
1.7 x 101*
n \
1.3 x 103 J
5.5 x 102
4.0 x 103 -.
2.4 x 103 (
9 1
2.4 x 102> J
1.7 x 10 **
1.6 x 10 5 -,
9.2 x 101*
LL 1
1.3 x KT -)
Ratio (%)*






*Ratio of each count to TF10"1 count.
     Strains of extremely oligotrophic bacteria  from oligotrophic waters in
Lake Biwa showed a relatively high growth rate  (generation time, approximately
6 h) and a very low maximum cell yield  (less than  106 cells/ml) in a  liquid
medium containing organic matter in low concentration  (approximately  2 mg C/l).
All bacteria grown in TF10"1 medium were able to grow in LW? medium,  such or-
ganisms may be called facultatively oligotrophic bacteria.

     Data reported here suggest that some growth-promoting factors were con-
tained in natural lake water and that extremely  oligotrophic bacteria were

 predominant in water in oligotrophic  areas of Lake Biwa and probably contrib-
 ute to the biodegradation processes  in the lake.

                               LITERATURE CITED

 1.  Akagi, Y.; N.  Taga, and U.  Simidu.  1977.  Isolation and distribution of
      oligotrophic  marine bacteria.   Can. J.  Microbiol.  23:981-987.

 2.  Fonden, R. A.   1968.  Yeast-extract peptone agar for the determination of
      heterotrophic bacteria in lakes.  Vatten 2:161-166.

 3.  Hoskins, J. K.  1934.  Most probable numbers  for evaluation of  coli-
      aerogenes tests by fermentation tube method.   U.S. Public Health (Ser-
      vice) Repts.  49:393-405.

 4.  Ishida, Y., and H.  Kadota.   1974.  Ecological  studies on bacteria in the
      sea and lake  waters polluted with organic substances.   I. Responses of
      bacteria to different concentrations of organic substances. Bull.  Japan.
      Soc. Sci. Fish. 40:999-1005.

 5.  Ishida, Y., and H.  Kadota.   1975.  Ecological  studies on bacteria in the
      sea and lake  water polluted with organic substances.  II. Analysis  of
      bacterial flora by use of a chemostat.   Bull.  Japan. Soc. Sci.  Fish.

 6.  Ishida, Y., and H.  Kadota.   1975.  Ecological  studies on bacteria in the
      sea and lake  waters polluted with organic substances.   III. Growth  ki-
      netics of the bacteria populations showing different responses  to dis-
      solved organic substances.   Bull. Japan. Soc.  Sci. Fish.   41:961-964.

 7.  Ishida, Y., and H.  Kadota.   1978.  A new method for enumeration  of oligo-
      trophic bacteria in lake  water.   Arch.  Hydrobiol.  (in press).

 8.  Jannasch, H. W.  1967.  Enrichments of aquatic bacteria in continuous cul-
      ture.  Arch.  Mikrobiol. 59:165-173.

 9.  Jannasch, H. W.  1969.  Estimations of bacteria growth rates in  natural
      waters.  J. Bacteriol. 99:156-160.

10.  Shewan, J. M., G. Hobbs, and W.  Hodgkiss.  1960.  A determinative scheme
      for the identification of certain genera of gram-negative bacteria, with
      special reference  to the  Pseudomonadaceae.  J.  Appl. Bacteriol.

11.  Sieburth, J. McN.  1967.  Seasonal selection of estuarine bacteria by
      water temperature.  J. Exp.  Mar. Biol.  1:98-121.

12.  Simidu, U.  1974.  Taxonomy of marine bacteria,  p.  45-65.   In N. Taga (ed.),
      Kaiyo biseibutsu (marine  microbes).  Univ. Tokyo Press, Tokyo.

13.  Simidu, U.  1975.  In situ analysis of microbial flora changes in pol-
      luted sea, p. 50-60.  In H. Kadota, N. Taga, and M. Sakai (eds.), Eco-
      system in the sea and microorganisms.  Koseisha-Koseikaku, Tokyo.

                           IN ANOXIC  SEDIMENTS

                              T.  H.  Blackburn
                      Institute  of Ecology  and Genetics
                     University  of Aarhus,  Ny  Munkegade
                          DK-8000  Aarhus  C,  Denmark

                The net rate of ammonia production  (d-i) was measured
           in sediment sections,  at in-situ  temperatures.  Addition
           of l5N-labeled ammonia made possible  the measurement of
           total ammonia turnover (d) and of ammonia incorporation
           into cells (i).   The molar N/C ratio  (Ns) in detrital ma-
           terial undergoing breakdown, and  the  rate of carbon oxi-
           dation (Co) or sulfate reduction  (Sr) were  calculated,
           using known values for the N/C ratio  in cells  (Nc) and the
           efficiency of carbon incorporation into cells  (E), from
           the following:   Ns = E x Nc x d/i
                     2 Sr = Co = i(l-E)/(E x Nc)
                     2 Sr = Co = (d-i)(l-E)/(Ns-Nc x E)
           The calculated Ns values did not  always correspond to the
           actual N/C ratio of the organic detritus in the sediment.
           The data are consistent with the  hypothesis that  surface
           detritus of an initially high N/C ratio progressively
           loses nitrogen as it is mixed downwards and finally re-
           sults in a net uptake of ammonia  in the lower sediment

     Detritus, containing carbon and nitrogen,  is  degraded and used by bac-
teria, partially as a source of energy  and partially  for  incorporation into
new cellular material.   The net production (or  consumption) of degraded  nitro-
gen compounds depends on the N/C ratio  of  the detritus  and of the  cells,  and
on the efficiency with which detrital material  is  assimilated into cellular

     These relationships are examined and  are used to illustrate how  rates of
ammonia turnover, incorporation and net production may  be used to  calculate
N/C ratios  in the substrate and rates of carbon oxidation (or sulfate reduction)


in anoxic marine  sediments.


     The--rates  of net ammonia production (d-i) in sediment sections  were  mea-
sured at in-situ  temperatures,  where diffusional losses could not  occur.   Addi-
tion of   N-labeled  ammonia to these samples made possible the measurement of
total ammonia production (d), using net ammonia production rate  to determine
the dilution of label by the changing pool.  The difference between  the two
rates gave the  rate  of ammonia incorporation into bacterial cells  (i)  at  any
depth  (Blackburn,  unpublished results).  Organic carbon in sediments was  de-
termined by combustion in a Leco furnace, correcting for the contribution from
carbonates.  Total nitrogen was determined, following Kjeldahl digestion  (3).
Organic nitrogen  was obtained by correcting for extractable and  non-extractable
ammonia.  The sediments were sampled from the stations described by  Jjrfrgensen
(5) .

                            RESULTS AND DISCUSSION

     The fate of  C and N, following the breakdown of detritus by hydrolytic,
fermentative and  sulfate-reducing bacteria is shown in Figure 1.   Carbon  is
metabolized at  a  rate Cd, oxidized at a rate Co and incorporated into  cells at
a rate Ci  (Cd = Co + Ci; Ci/Cd = E, the efficiency of incorporation).  Simi-
larly, nitrogen is liberated at a rate d (d/Cd = Ns, the N/C ratio of  the sub-
strate) , assimilated at a rate i (i/Ci = Nc, the N/C ratio of bacterial cells)
and excess ammonia is made available at a rate d-i  (net production rate).   It
is assumed that bacteria assimilate nitrogen mostly in the form  of ammonia,
thus allowing the following relationships to be established:

                     1)  Ns = Nc x E x d/i
                     2)  Co = i(l-E)/(Nc x E)
                     3)  Co = (d-i)(l-E)/(Ns-Nc x E)
                     4)  Co = 2Sr
where Sr = the  rate  of sulfate reduction by carbon at the oxidation-reduction
level of CH20.

                            We      NIC in cells 10,16)
                            Ns      N/C in substrate
                            E       C-incorporation/C-metabolized (0.3)
                                                 ti-» Cell - C
                     Substrate -*  fermentation
                              sulfate reduction
Figure  1.   The  processes involving the production of cells,  carbon dioxide and
            ammonia from detritus in sediment.


     In order to  calculate Ns and Co from equations  1 and 2 it is necessary  to
know the values for  Nc  and E, in addition to the  ammonia rates.  The value for
Nc = 0.16  (Fig. 1) is consistent with quoted values  for bacterial cells  (1, 2).
The value E = 0.31 is based on the calculations in Table 1.  The assignment  of
approximately 1 ATP  to  the oxidation of acetate by sulfate fits with the effi-
ciency of acetate assimilation by Desulfotomaculum acetoxidans (6).

                             kcal mol
ATP available      C atoms       C atoms
 for synthesis    dissimilated   assimilated
  100 hexose C+lactate + cells   -52
  88 lactate-»• C02  + cells

  100 hexose C->C02 + cells
      It  is assumed that 1 mol ATP will form 10 g  cells or 0.42 mol of organic  C.
      Per mol substrate dissimilated.

      The data in Table 2 illustrate how d,  d-i  and i vary with depth in an
average  of four sediments from the Limfjord in  August 1976, and how these
values may be used to calculate the N/C ratio in  the substrate and the rates
of carbon  oxidation or sulfate reduction,  using the equations derived in equa-
tions 1  through 4.
                      TABLE 2.  RATES OF AMMONIA TURNOVER
Depth (cm)
Total for
12 cm



N/C molar (Ns)

N/C molar

SO?" reduction
(Sr) calculated
from eq. 2 and 4

SO? reduction
(Sr) calculated
from eq. 3 and 4

   Rates are in n mol cm * day
   Using Ns = measured N/C molar ratio in sediment.
      There was a decrease in total  ammonia production, with depth  of  sediment,
 but there was still considerable activity at 10 to 12 cm.  At this depth,  how-
 ever,  there was little net ammonia  production as ammonia incorporation was

quite high.  The latter rate showed surprisingly little variation with depth
down to 12 cm.  The calculated Ns decreased with depth and was generally lower
than the measured value of sediment N/C molar ratio.

     The calculated rate of sulfate reduction (Sr) from equations 2 and 4
showed little decrease with depth, since it was determined from the rate of
ammonia incorporation.  The values of Sr, calculated from equations 3 and 4,
decreased with depth in a manner consistent with the values measured in the
same sediments (5).  The respective values of 39.9'and 30.6 mmol m~2day~1,
down to 12 cm, are somewhat higher than the measured yearly average of 9.5
mmol m~ day"  (5).  The utilization of equation 3 has considerable advantages
as 15N tracer experiments are not necessary for the measurement of d-i, which
may even be derived from the change in flux in ammonia, between two adjacent
strata.  It may thus be calculated from an ammonia concentration profile, pro-
vided that it is an undisturbed sediment and that the diffusion coefficient is
known.  The value of Ns must be known to use equation 3.  The data in Table 2
demonstrate that the actual Ns value is not the same as the N/C ratio of the
sediment, and can be significantly different (Fig. 2).  It seems likely that,
in sediments where vertical mixing does not occur, the Ns value must equal the
sediment N/C ratio.

     The data in Figure 2 illustrate how the net rate of ammonia production
and the calculated N/C ratio of substrate  (Ns)  at various depths may be used
to explain some aspects of ammonia concentration profiles at station 5 in the

     The concave interstitial ammonia concentration profile at 6-8 cm for May
was associated with a negative net ammonia production at 4-8 cm and a low N/C
in the substrate from 4-12 cm.  This contraction in ammonia concentrations
continued into June, even though there was considerable net ammonia production
in the top 4 cm of sediment, associated with a high Ns in this region, pre-
sumably due to the deposition of fresh algal cells on the surface.  During
July and August the ammonia concentrations increased  (unpublished results) but
by September they had contracted again, due partially to the negative net am-
monia production, associated with the low Ns of the substrate undergoing de-
composition from 2-12 cm.

     These patterns fit with the hypothesis that algal cells of a high N/C
content were decomposed at the sediment surface, with preferential decomposi-
tion of protein and nucleic acids, liberating ammonia.  The partially decom-
posed cells, rich in structural materials of low N/C ratio, gradually mixed
downwards and were degraded in preference to the old and recalcitrant organic
material in the sediment, which had a higher N/C ratio  (Fig. 2, top right).
When the Ns was less than 0.05 (when i>d, E = 0.03 and Nc = 0.16) there would
be net ammonia uptake.

     The hypothesis depends on the mixing of surface detritus downwards to at
least 12 cm.  This is consistent with the observation that sulfate reduction
in these sediments was higher than could be explained by the oxidation of de-
tritus to which no new additions were made  (5).  The role of macrofauna in
sediment irrigation and mixing has recently been demonstrated  (4)  Additional
evidence for sediment mixing is derived from the fact that the measured values


                 jj mol cm'3               n moi  crr&day-l
        MAY U
         6   6
         i£   s
         Q   10
  0  2   -I*
                            •&   1-0  1-2 -20  0  20 W  60 '05 -10  '15  -20
Figure 2.  Interstitial ammonia concentrations, net ammonia production rates
           and Ns  (N/C molar ratios) at 2 cm intervals  in the  Limfjord station
           5, for May (12°C), June  (16°C) and September (12°C)  in  1977.   The
           broken line in the May values of Ns  (top right) shows the sediment
           N/C molar ratio.

for the rates of net ammonia production, particularly in the summer months,
are not consistent with the observed changes in the ammonia concentrations.
The ammonia concentration gradients are not sufficient to account for the dis-
appearance of ammonia from the lower strata, unless very high diffusion coef-
ficients are postulated for ammonia.


     This study was supported by Grant No. 521/11 from the Danish Natural
Science Council.

                              LITERATURE CITED

1.  Bardowskiy, 0. K.  1965.  Sources of organic matter in marine basins.  Mar.
     Geol. 3:5-31.

2.  Bardowskiy, O. K.  1965.  Accumulation of organic matter in bottom sedi-
     ments.  Mar. Geol. 3:33-82.

3.  Bremner, J. M.  1960.  Determination of nitrogen in soil by the Kjeldahl
     method.  J. Agr. Sci. 55:1-23.

4.  Goldhaber, M. B., R. C. Aller, J. K. Cochran, J. K. Rosenleld, C. S. Mar-
     tens, and R. A. Berner.  1977.  Sulfate reduction, diffusion, and bio-
     turbation in Long Island Sound sediments:  report of the FOAM group.
     Am. J. Sci. 277:193-237.

5.  J^rgensen, B. B.  1977.  The sulfur cycle in a coastal marine sediment
      (Limfjorden, Denmark).  Limnol. Oceanog. 22:814-832.

6.  Widdel, F., and N. Pfennig.  1977.  A new anaerobic, sporing, acetate-
     oxidizing, sulfate-reducing bacterium, Desulfotomaculum  (emend.) acetoxi-
     dans.  Arch. Microbiol. 112:119-122.


       Michael J.  DiGeronimo, Robert S. Boethling and Martin Alexander
           Laboratory of Soil Microbiology, Department of Agronomy
                            Cornell University
                             Ithaca, NY  14853

                Increasing  the  number of chlorines on benzoates in-
           creases their resistance  to attack by microorganisms in
           sewage.  Populations of bacteria able to grow on readily
           biodegradable benzoates multiply as the chemicals disap-
           pear from model  sewage ecosystems.  The major bacteria in
           sewage growing on unsubstituted benzoate will not grow on
           simple chlorinated benzoates.  The addition of benzoate
           to sewage does not lead to the loss of 2,4-di- and 2,3,6-
           trichlorobenzoates.   During the decomposition of m-chloro-
           benzoate, 4-chlorocatechol and 5-chlorosalicylate appear.
           In stream water,  parts-per-trillion concentrations of p-
           chlorobenzoate,  dimethy1amine, and diethanolamine are
           quickly converted to carbon dioxide.  In contrast, parts-
           per-million or low parts-per-billion concentrations of
           the herbicide 2,4-D  are converted to carbon dioxide very
           slowly, although higher concentrations are destroyed
           readily.   The data suggest that chemicals may be attacked
           only slowly if present in waters at very low concentra-
           tions .
     Corresponding to the continuing  rise  in production and use of synthetic
organic chemicals is an increased  discharge of these chemicals into natural
waters, either deliberately or inadvertently, as domestic and industrial
wastes, runoff from agricultural operations, or accidental spillage.  Most of
these compounds are probably rendered harmless soon after their introduction
into aquatic environments.   A surprisingly large number of organic chemicals,
however, may not be metabolized; among  those that are transformed microbio-
logically, the products may be more toxic  than the original compound, or  they
may persist for longer periods of  time.  Although a considerable amount of in-
formation has accumulated about the metabolism of these organic compounds by
pure or mixed cultures, little is  known about their fate in natural ecosystems,
both aquatic and terrestrial.

     Several factors govern the susceptibility of organic chemicals to micro-
bial destruction in nature.  In the present investigation, interest is di-
rected to the effects of chemical structure and concentration on the biodegra-
dation of organic compounds.  Ample evidence exists of a relationship between
the structure of an organic chemical and its susceptibility to microbial at-
tack (1, 11) .   Moreover, Jannasch (8) and Shehata and Marr  (12) demonstrated
that stable, axenic microbial populations could not be maintained in continu-
ous culture at low concentrations of glucose and other simple organic mole-
cules.   However, no systematic investigations have been conducted of the in-
fluence of low substrate concentration on the biodegradation of pollutants in
artificial microcosms or ecosystems.

     The present study was undertaken, therefore, to assess the effects of
chemical structure and concentration on the biodegradation of organic mole-
cules in artificial aquatic microcosms.  The experimental systems used were
designed to reflect the complex interactions in natural microbial communities.

                            MATERIALS AND METHODS

     Surface water samples were taken from Fall Creek, a tributary of Cayuga
Lake in central New York.  Sewage samples were taken from the primary settling
tank of the water treatment plant at Ithaca, New York.  All samples were pro-
cessed within 2-4 h after collection.

     An activated-sludge die-away system was used in studies of the degrada-
tion of benzoate and chloro-substituted benzoates.  Primary influents were in-
cubated at room temperature in 1.0 liter quantities in 2-liter Erlenmeyer
flasks, and the suspensions were aerated with air that had been passed through
a saturated solution of KMnO4.  Sewage was amended with the test chemical to a
final concentration of 100 mg/liter.  Cell counts were performed by a most-
probable-number method for enumerating populations of aromatic degraders (7).

     Benzoate and chlorinated benzoate were determined by gas-liquid chroma-
tography using a Perkin-Elmer gas chromatograph, model 3920B, fitted with a
flame-ionization detector.  Samples were saturated with NaCl, acidified with
HCl, and extracted with anhydrous ether.  The ether extracts were concentrated
to a volume of 2-5 ml.  Benzoate and monochlorinated benzoates were analyzed
on a teflon-lined stainless steel column  (0.6 m x 2mm, i.d.) packed with 10%
DEGA on 80/100 mesh Chromosorb G (H.P.) at a column oven temperature of 190°C.
Dichlorinated and trichlorinated benzoates were derivatized with bis(trimethyl-
silyl)  trifluoroacetamide  (Pierce Chemical Co.) in demethylformamide after con-
centration of the ether extracts, and the derivatives were analyzed on a 1.8 m
x 2 mm i.d. teflon-lined stainless steel column packed with 3% OV17 on 80/100
mesh Chromosorb G at a column oven temperature of 130°C.  Metabolic intermedi-
ates were identified using a Finnigan 3300 Quadrupole mass spectrometer  (elec-
tron impact 70 eV) after derivatization with diazomethane.  Spectrophotometrie
determinations were performed by the method of Alexander and Lustigman  (2),
and dihydroxy compounds were measured by the method of Arnow  (4).

     The degradation of low concentrations of organic compounds was studied
using a stream die-away system.  Stream water was collected aseptically and
analyzed for pH and total alkalinity by standard methods  (3).  The water was


amended with an inorganic nutrient supplement  (13) modified as follows:  NaNOs
and NaHCO3 were omitted, Zn was added at the same molarity as ZnSOit'H2O rather
than ZnCl2, and NHitCl was added at 8.0 mg/liter.

     Microbial activity was assayed by measurement of 14CO2 produced from  14C-
labeled substrates.  Stream water was incubated at 28.5°C in 100- or 200-ml
volumes in 500-ml filtering flasks modified to permit connection to a gas
train subsequent to incubation.  The flasks were incubated in the dark to  pre-
vent photodecomposition of added organic chemicals and fixation of   C02 by
algae.  Although the flasks were not shaken, measurements of dissolved oxygen
 (3) indicated that the stream water remained aerobic.

     Individual organic compounds were added to the flasks at various concen-
trations.  After specified periods of time, the contents of the flasks were
acidified with 0.12 N HaSOit, the flasks were connected to a source of N2 ,  and
the C02 in the flasks was flushed out and trapped in ethanolamine .  If neces-
sary, NaHCOa was added immediately before acidification to raise the concen-
tration of bicarbonate plus carbonate to at least 1.0 mM.  Control experiments
indicated that the recovery of * ^CC^ was quantitative, and that no degradation
of any compound to ^CC^ occurred in autoclaved stream water.  None of the
 14C-labeled compounds tested nor any impurities in their formulation were suf-
 ficiently volatile to be flushed out with N2 after acidification and trapped
 in ethanolamine.  This method for assessing biodegradation of organic chemi-
 cals does not distinguish between degradation to  kCO2 and possible conversion
 to other volatile metabolites that may be trapped in ethanolamine.

     Samples  (1.0 g) of ethanolamine containing trapped 14CO2 were added to 9
 ml of scintillation fluid  (6) , and the radioactivity was measured in a Beckman
 LS-100C liquid scintillation spectrometer.  The data were corrected for quench
 by the external standard/channels ratio method and expressed as dpm.  The
 counting efficiency of ethanolamine (1.0 g) in Bray's scintillation fluid was
 about 75%.

     The three monochlorinated benzoic acid isomers, 3,4- and 2 ,4-dichloroben-
 zoic acids, and 2 ,4-dichlorophenoxyacetic acid were obtained from Aldrich
 Chemical Co.  Scintillation grade ethanolamine and unlabeled dimethylamine hy-
 drochloride were obtained from Eastman Organic Chemicals, diethanolamine from
 Fluka AG, and sodium benzoate from Merck.  2 ,3,6-Trichlorobenzoic acid (Pfaltz
 and Bauer) was recrystallized from hot water.

     The sources and specific activity of the radiochemicals used in this
 study were as follows:  dimethylamine-1 4C hydrochloride (Amersham) ,  54 mCi/
 mmol; diethanolamine- *C (New England Nuclear), 20.8 mCi/mmol; p-chlorobenzoic
 acid-ring 2l*C (California Bionuclear) , 10.5 mCi/mmol; 2 ,4-dichlorophenoxyace-
 tic acid-acetate 2-1'*C (Amersham), 32 mCi/mmol.


     The primary degradation of the chloro-substituted benzoates in sewage was
 investigated initially.   The results are presented in Figure 1.  Primary deg-
 radation was more rapid with benzoic acid than with the chlorinated analogues.
o-Chlorobenzoate was metabolized more rapidly and disappeared sooner than did


p- and m-chlorobenzoate.  Of the polychlorinated compounds tested, 3,4-di-
chlorobenzoate disappeared in 9 days, but no loss of 2,4-dichlorobenzoate or
2,3,6-trichlorobenzoate was detected, even after an additional 18 days.
Figure 1.  Degradation of benzoate and m-, p-, o-, 2,4-di-, 3,4-di-, and
           2,3,6-trichlorobenzoates in sewage.  Sewage was amended to an ini-
           tial concentration of 100 yg/ml of each test compound.  Abbrevia-
           tions:  B.A., benzoate; o-, p-, and m-Cl B.A. are the o-, p-, and
           m-chlorobenzoates; 3,4- and 2,4-D.B.A. are 3,4- and 2,4-dichloro-
           benzoates; and 2,3,6-TBA is 2,3,6-trichlorobenzoate.
     To determine whether disappearance of the compound corresponded to an in-
crease in a population capable of utilizing aromatic compounds, sewage was
amended with 100 yg/ml of the chlorinated compounds, and cell numbers were de-
termined.  The results in Figure 2 demonstrate that the numbers of bacteria
capable of using the chlorinated compounds as carbon and energy sources for
growth rose in sewage amended with the benzoates previously shown to disappear.
The populations of chlorinated benzoate utilizers were initially less than 10
cells/ml but increased to a maximum of about 106 cells/ml.  In contrast, there
was no increase in populations capable of metabolizing 2,4-di- and 2,3,6-tri-

 Figure  2.   increase in populations of chlorinated benzoate utilizers in sewage
            amended with o-, m-, and p-chloro- and 3,4-dichlorobenzoates to an
            initial concentration of 100 yg/ml.

     To determine what effect benzoate metabolism might have on the prolifera-
 tion of the chlorobenzoate utilizers, sewage was amended with 100 yg of sodium
 benzoate/ml and monitored for changes in populations of both benzoate and
 chlorobenzoate utilizers.  The results in Figure 3 show that the number of
 benzoate utilizers increased rapidly in the first 24 h but then declined some-
 what.  In contrast, the size of the chlorobenzoate-utilizing populations was
 unaffected by the amendment and remained at less than 102 cells/ml.

     To assess whether the more refractory chlorinated compounds are modified
 in the presence of a utilizable and structurally related substrate, two flasks
 containing sewage were amended with sodium benzoate (100 yg/ml)  and either
 2,4-di- or 2,3,6-trichlorobenzoate.  The number of cells and the percent of
 the test compound remaining as a function of time are shown in Figure 4   The
 results show that there was no disappearance of either 2,4-dichlorobenzoate or
 2,3,6-tnchlorobenzoate as the benzoate was being destroyed.   No increase in
 the population capable of growing at the expense of the chloro-substituted
benzoates was evident, whereas,  concomitant with benzoate degradation was an
increase in the population capable of using this compound for growth

            • Benzole Acid Degraders (+Benzoic Acid in Sewage)

           Benzole Acid Degraders (-Benzole Acid in Sewage)
                                      m,p,+oCI Benzoic Acid Degraders (±B.A.)

                                   9       12
Figure 3.   Populations of benzoate and chlorobenzoate utilizers in sewage
             amended with 100 yg benzoate/ml  or without benzoate  addition.
    Q 4
                                     2,4-di-ond 2,^6-trichlorobenzoate
        • Benzoate
                 /2,4-di-and 2,3,6-trichlorobenzoate degraders
                -£,-_	-i-.		^-,	
                                                                             % LEFT
Figure 4.   Effect of benzoate on  the degradation  of 2,4-di-  and 2,3,6-tri-
             chlorobenzoates  in sewage.

     Studies were performed to evaluate the ability of pre-adapted  populations
to respond to new organic substrates.  Pre-adaptation was  accomplished by ini-
tially adding sodium benzoate  (100 yg/ml) to the sewage.   The  sewage  was  moni-
tored for substrate disappearance, and when benzoate could no  longer  be de-
tected, m-chlorobenzoate  (100 yg/ml) was added.  The results in  Figure 5  dem-
onstrate that sodium benzoate was degraded in approximately 12 h.   After  the
addition of m-chlorobenzoate, its concentration fell by  18% in four days, but
no further change in concentration was evident until day 8.  At  24  h  after the
addition of m-chlorobenzoate, an ortho-substituted dihydroxy compound was de-
tected, and the identity of the product was confirmed by gas chromatography-
mass spectrometry.  The methyl ester of this metabolite  showed a molecular ion
with an m/e of 172 and a  fragmentation pattern identical to that of authentic
4-chlorocatechol.  The concentration of the chlorocatechol (9.0  yg/ml) re-
mained constant until 48 h, after which it declined until  none was  detectable
at 4 d.  During this time, there was no significant increase in  the population
capable of using m-chlorobenzoate for growth.  After day 8,  the  chlorobenzoate
disappeared quickly, and  none was left at day 12.  The sewage  was again amended
with m-chlorobenzoate  (100 yg/ml),  and monitoring of the compound showed  that
the chlorobenzoate was degraded in  3 d.  During the rapid  phase  of  decomposi-
tion of m-chlorobenzoate, a second  product was detected, and it  was identified
by mass spectrometry as 5-chlorosalicylate.  Methylation of this metabolite
gave a molecular  ion with an m/e  of 186 and a fragmentation pattern identical
to that of  authentic 5-chlorosalicylate.
         w 100 ppm m-chlorobenzoic acid added
                                                         lOOppm m-chloro-
                                                          benzoic acid added
          Q I
Figure 5 .
            The degradation of zn-chlorobenzoic acid in sewage pre-adapted with
            benzoic acid.

     Studies were conducted to determine the effect of low substrate concen-
tration on the degradation of p-chlorobenzoate and other organic compounds.
The procedure involved measuring the production of 1 "*C02 from 1 "*C-labeled com-
pounds in a stream die-away system.  The production of 1^COa from p-chloroben-
zoate was measured at four initial p-chlorobenzoate concentrations  (Table 1).
The data are expressed as the percent of the initial  ll*C-labeled material re-
covered as  ltC02.  Little or no decomposition of the parent compound was ob-
served in at least 12 d at 47 ppm, the highest concentration used.  In con-
trast, at 47 ppt, p-chlorobenzoate was extensively degraded within 4 d.  At
intermediate concentrations, intermediate rates were obtained.  At an initial
concentration of 470 ppb, a lag of at least 6 d was evident before significant
degradation occurred, but the apparent lag was approximately 2 d or less at
4.7 ppb.


          	Initial p-chlorobenzoate concentration	
Days         47 ppm            470  ppb            4.7 ppb             47 ppt
          (3 x lO""* M)      (3 x 10~6 M)      (3 x 10~8 M)       (3 x ID'10  M)


% of lltC
recovered as C02

     The secondary amines dimethylamine, diethylamine, and diethanolamine were
selected for study in part because they have been shown to be precursors for
the formation of nitrosamines in model ecosystems, most of the nitrosamines
being potent carcinogens.  Typical results showing the production of   CO2
from dimethylamine-lkC added to stream water at four initial concentrations
are given in Table 2.  Extensive decomposition of dimethylamine was observed
within one day at the two lowest initial concentrations.  Longer periods were
required at the two highest initial concentrations.  Apparent decomposition to
C02 was complete within 3 d at an initial concentration of 90 ppb, whereas at
an initial level of 18 ppm, the fraction degraded did not increase signifi-
cantly for at least 3 d.  Similar results were obtained with diethylamine.

     When stream water was amended with diethanolamine, a somewhat different
picture emerged.  The results of a typical experiment are shown in Figure 6.
At the lowest initial concentration of diethanolamine  (21 ppt), degradation of
the parent compound continued at an approximately constant rate for at least 4
d; at this time, about 30% of the original compound had apparently been con-
verted to CO2.  In contrast, at an initial level of 210 ppb, an accelerating
curve for degradation of diethanolamine was obtained.  At 21 ppm, the highest
initial concentration, less than 5% of the original chemical was converted to
COa within 4 d.

Initial dimethylamine concentration
Days 18 ppm 90 ppb 230 ppt
(4 x 10-1* M) (2 x 10~6 M) (5 x 10~9 M)
% of 1'*C recovered as llfC02
1 1.6 14.8 84.9
2 4.3 47.0 96.3
3 4.2 100.0
6 90.1

< 40
o 30
3? 20


210 ppbO
(XIO2) /
/ "
/A -

21 ppt A//
(XIO6)/ /
A o
/ /
/ /
A Q'
300 §
250 S
200 •$>
150 i
100 ^
2.8 ppt
(6 x 10 M)


Figure 6.
       Formation of L ^COa from diethanolamine-1 4C added at several concen-
       trations to stream water.

     The biodegradation of 2,4-dichlorophenoxyacetate  (2,4-D) was similarly
studied at four initial concentrations of the herbicide in stream water.  The
results are shown in Figure 7.  The effect of initial substrate concentration
on the extent of degradation of the parent compound was strikingly different
from that observed with p-chlorobenzoate and dimethylamine.  At the two lowest
initial concentrations, only a small fraction of the starting material was de-
graded within 8 d, whereas the data indicate that 2,4-D was extensively de-
graded in 6 d or less when initially present at 220 ppb and 22 ppm.
                                                         -60 ~
Figure 7.  Formation of   CO2 from 2,4-dichlorophenoxyacetate  (acetate-2-  C)
           added at several concentrations to stream water.
     The 2,4-D data were not the result of oxygen depletion in the sealed in-
cubation vessels; the opposite effect of substrate concentration would be ex-
pected.  The apparent absence of significant decomposition of 2,4-D at low ini-
tial concentration could not be explained on the basis of high efficiency of
assimilation by the microbial population.  At most, several percent of the
starting material was assimilated after 7 d of incubation at an initial 2,4-D
level of 2.2 ppb, as indicated by the dpm retained when particulate matter
from stream water was collected on 0.45 urn cellulose-acetate  (Millipore) fil-
ters and the filters counted by liquid scintillation.

     Control experiments indicated that the data in Figure 7 were not the re-
sults of sorption of 2,4-D to the glass incubation vessel or to particulate

matter in stream water.  No radioactivity above background was detected on the
filters when autoclaved stream water was incubated with 2.2 ppb of 2,4-D for 5
d and the particulate fraction collected on 0.45 ym filters.  In addition, no
loss of 2,4-D from solution was detected when autoclaved stream water was in-
cubated with 2.2 ppb of 2,4-D for 36 h and the water samples centrifuged at
110,000 X g for 3 h to remove particulate matter.  Dimethylamine, diethanol-
amine, and p-chlorobenzoate at similar concentrations were not detectably
sorbed to particulate matter or glass in this test.  Only 0.1% of the initial
radioactivity was found in the fraction collected on 0.45 ym filters when non-
sterile stream was was incubated for 2 d with 2.2 ppb of 2,4-D.


     The data suggest that the chemical structure of chlorobenzoates affects
their persistence in sewage.  These findings are in agreement with published
reports, in that the position and number of substituents on the aromatic ring
affect susceptibility of the compound to microbial attack (1,9).   In the
present data, with the exception of 3,4-dichlorobenzoate, the chlorobenzoates
were more refractory if two or more chlorine atoms were present.

     The data further suggest that members of the microbial community of sew-
age used the monochlorinated benzoates and 3,4-dichlorobenzoate as carbon and
energy sources for growth.  The populations active in the degradation of the
mono- and dichlorinated benzoates were initially small, but the active popula-
tions increased in the presence of the usable chlorobenzoate.  The apparent
maximum population size was well below 108 cells/ml, the approximate number of
bacteria expected for a substrate concentration of 100 yg/ml, but this dis-
crepancy may be a result of the feeding by protozoa on the chlorobenzoate uti-

     Since there was only partial disappearance of m-chlorobenzoate at 48 h
and no detectable increase in the population capable of using this compound as
sole source of carbon and energy, the results suggest that the initial disap-
pearance of m-chlorobenzoate was the result of cometabolic activity by ben-
zoate degraders.  Although little is known about the role of cometabolism in
degradation of chlorobenzoates under natural conditions, the ability of ben-
zoate-degrading bacteria from various genera to cometabolize chlorobenzoates
is well documented (14).

     The results from studies of the effect of concentration of p-chloroben-
zoate and dimethylamine on their biodegradation indicate that these compounds
are extensively decomposed within a few days at the lowest initial concentra-
tions and thus would not persist in freshwater at these levels.  In contrast,
all of the compounds studied at ppm levels, except for 2,4-D, were degraded
only after significantly longer lag periods or not at all within the time of
study.  The biodegradation of organic chemicals in die-away systems with low
microbial densities has commonly been assessed at initial concentrations of
the test chemicals in the ppm range, but these data demonstrate that many
chemicals that appear to be persistent in surface waters at ppm concentrations
may be rapidly destroyed at lower levels.  This effect of concentration must
thus be taken into account in designs of test systems for evaluating biode-


     These findings also suggest that diethanolamine and 2,4-D may be more per-
sistent at very low than at higher concentrations.  Thus, low substrate con-
centration may also limit biodegradation rates in natural waters.  A possible
explanation for the linear increase with time in the rate of degradation of
diethanolamine at the lowest initial concentration is that the population ac-
tive in its degradation does not increase significantly at the expense of di-
ethanolamine at this level of substrate.  This interpretation suggests that
organic chemicals may be present at such low concentrations in waters that the
rate of microbial decomposition is limited by the initial size of the active

     Although the conversion of part of the  ^C-labeled material to volatile
metabolites other than 14C02 cannot be excluded, the data show that extensive
degradation occurred.  The approximate yields of COa from the various organic
compounds were generally consistent with expected values, and experiments in
which  4C-glucose was added to stream water at several initial concentrations
gave values of 50-60% recovery of 14C as ll*CO2.

     Our findings point to the possibility that many organic pollutants may
persist in aquatic ecosystems owing in part to their low prevailing concentra-
tion.  If persistence is sometimes a consequence of the inability of microor-
ganisms to destroy biodegradable molecules at low concentrations, this failing
may account for the presence of trace levels of certain organic chemicals in
natural waters.  Such trace substances could lead to environmental problems if
they are susceptible to biomagnification and are subsequently toxic to species
at higher trophic levels in food chains.  Furthermore, recent studies have
suggested that some organic chemicals may be toxic to aquatic microorganisms
at ppb levels and lower (5, 10).   Hence, further inquiry is necessary to de-
fine more precisely the influence of concentration on the rate of microbial
destruction of organic molecules in aquatic ecosystems.


     This research was supported in part by Public Health Service Training
grant ES00098 from the Division of Environmental Health Sciences and by Na-
tional Science Foundation grant No. ENV75-19797.  Any opinions, findings, and
conclusions or recommendations expressed in this publication are those of the
authors and do not reflect the views of the sponsoring agencies.

                              LITERATURE CITED

1.  Alexander, M.  1965.   Biodegradation:  problems of molecular recalcitrance
     and microbial fallibility.  Adv. Appl. Microbiol. 7:35-80.

2.  Alexander, M., and B.  K. Lustigman.  1966.  Effect of chemical structure
     on microbial degradation of substituted benzenes.  J. Agric. Food Chem.

3.  American Public Health Association.  1976.  Standard methods for the ex-
     amination of water and waste water, 14th ed.  American Public Health As-
     sociation, Washington, D.C.


 4.   Arnow,  L.  E.   1937.   Colorimetric  determinations of the compounds of 3,4-
      dihydroxyphenylaminotyrosine  mixtures.   J.  Biol.  Chem. 118:531-537.

 5.   Batterton, J.,  K.  Winters,  and C.  Van Baalen.   1978.   Anilines:   selective
      toxicity  to  blue-green  algae.   Science  199:1068-1070.

 6.   Bray,  G. A.   1960.   A simple efficient liquid  scintillator for counting
      aqueous solutions in a  liquid scintillation counter.   Anal.  Biochem.  1:

 7.   DiGeronimo, M.  J. ,  M. Nikaido,  and M.  Alexander.   1978.  Most-probable-
      number technique  for the enumeration of aromatic  degraders in natural en-
      vironments.   Microbial  Ecol.  4:263-266.

 8.   Jannasch,  H.  W.  1967.   Growth of  marine bacteria  at limiting concentra-
      tions of  organic  carbon in seawater.  Limnol.  Oceanogr.  12:264-271.

 9.   MacRae, I. C.,  and M. Alexander.   1965.   Microbial degradation of selected
      herbicides in soil.  J. Agric.  Food Chem.  13:72-76.

10.   Powers, C. D.,  R.  S. Rowland,  H. B.  O'Connors,  Jr.,  and C.  F.  Wurster.
      1977.   Response to polychlorinated biphenyls  of marine phytoplankton  iso-
      lates cultured under natural  conditions.  Appl. Environ.  Microbiol. 34:

11.   Ryckman, D. W., V.  S. Prabhakara Rao,  and J. C.  Buzzell,  Jr.   1966.  Be-
      havior of organic chemicals in the aquatic  environment.   Manufacturing
      Chemists  Association, Washington,  D.C.

12.   Shehata, F. E., and A. G. Marr.  1971.   Effect  of  nutrient concentration
      on the growth of  Escherichia  coli.   J.  Bacteriol.   107:210-216.

13.   U.S.  Environmental Protection  Agency.  1971.   Algal  assay procedure-bottle
      test.   U.S.  Environmental  Protection Agency,  Corvallis,  Oregon.

14.   Walker, N., and D.  Harris.   1970.   Metabolism  of 3-chlorobenzoic  acid  by
      Azotobacter  species. Soil Biol.  Biochem. 2:27-32.



CLESCERI:  Dr. Floodgate, is it possible that the organisms can use some of
the organic nitrogen in the oil itself?  There is a certain amount of hetero-
cyclic-combined nitrogen in oil.  Is this unavailable to the organism?

FLOODGATE:  Yes, more or less.  One or two organisms have been shown,  in batch
culture, to be able to use the nitrogen in this cyclic material, but only in
warm temperatures.  In other words, in nature, there are other substrates
available for them to metabolize.

CLESCERI:  Do you think that means, then, that these compounds are permanently
nondegradable ?

FLOODGATE:  Most of them are nondegradable, or at least with a very long turn-
over time, perhaps decades.

JANNASCH:  Dr. Wright, you presented a slide showing the decrease in specific
heterotrophic activity starting up in the marshes of an estuary and moving to-
ward offshore waters.  I was wondering if you thought that these results might
be due to some effect of the increase in salinity which decreased specific ac-
tivity as opposed to just a decrease in nutrient availability.

WRIGHT:  The range in salinity in the estuary is not great; it's a very low
fresh water input estuary.  The range of the salinity might be a total of 5%;
I don't think that could be a very serious problem.

JANNASCH:  In the acridine orange stain, does that stain both viable and dead
cells, or primarily viable cells?

WRIGHT:  It stains any cell that has nucleic acid in it.  Dr. J. W. Costerton
thinks that there are hardly any dead cells in natural waters.  They die,
autolyze, and are gone pretty rapidly.  But, then, how do you define via-
bility?  Do you define it as whether you can count them on agar plate?  If so,
then a very small percentage are viable.  If you count them on the basis of
what they are doing, in terms of activity, then a large percentage are viable.

P. ROGERS:  You mentioned that Beta bacteria were capable of utilizing excreted
soluble organics.  I was wondering, do you think bacteria are capable of ac-
tually lysing algae or at least attacking algal detritus material directly
through extracellular enzymes?

WRIGHT:  I'm not too familiar with this work but I think that people who have
examined healthy algae under the microscope from sea water and fresh water
rarely find bacteria on their surfaces, unless they are blue-greens that have
large slime sheaths.  I haven't come across research that shows bacteria at-
tacking algae and trying to release the organic compound from them.

ROGERS:  I feel that this is a viable alternative, and I have photographs of
kinetic data showing that this is a possibility.  I believe that plate counts,
for many applications, is definitely a viable alternative to represent hetero-
trophic activity.  I think some of your graphs may have indicated that where
you have no increase initially in your total counts, but plate counts and
other methods of indicating heterotrophic activity showed a linear response
immediately to confinement in the bottle approach.  Am I mistaken?

WRIGHT:  I don't understand what you are saying.

ROGERS:  I think that you are counting a lot of bacteria that are not viable
at that time or never will be viable, at least in fresh water systems.  Recent
work by Jones shows this to be true.

WRIGHT:  How do you account for the large number that eventually show tetra-
zolium salt reduction, up to 70-80%?  Other people have shown with autoradiog-
raphy close to 100% of the bacteria developing on the film, indicating ac-

ROGERS:  Sea water system or fresh water?

WRIGHT:  Sea water.

PASSMAN:  George, you mentioned that under low nutrient conditions, you had
ATP pools that were too low to provide for initiation of enzyme synthesis, or
growth, necessary to begin incorporation of hydrocarbons.   Have you looked at
the cyclic AMP pools?

FLOODGATE:  I was told to be a little controversial or stir things up; that
was material I threw in without the information, as yet.  What happens when
you get a job to do like this, you find yourself writing down your research
program for the next ten years.  That's one of the items on it.  I want to go
back now to see what actually happened to that nitrogen.  Remembering the
graph, it does disappear into the cell somewhere; we have not looked at it
closely; it just disappears.  We've assumed it goes into the cells because
Charlie Brown has shown that it does go into their Pseudomonas.  We have yet
to show that it goes into our oil-degrading "bug."  Whether it actually goes
into the glutamate pool, whether it stays in there or comes out again, or what
happens, we just don't really know yet.

PASSMAN:  Dick [Wright], it was interesting to see that your specific activity
increased under conditions where nutrient levels were higher as we were dis-
cussing yesterday.  Looking at LPS to direct count ratios in the Georges Bank,
there were different patterns.  During the winter the LPS to acridine orange
direct count ratios were low but when we got into the growing season, the LPS
per cell apparently increases.  We also saw this type of phenomenon in dump


sites where hours after drug residues were pumped into Puerto Rican Harbor
there was a rapid decrease in the LPS per cell suggesting that we may be look-
ing at a parallel type of phenomenon here that would be worth further consid-

RAYMOND:  Dr. Floodgate, I missed one of your rate uptake numbers.  You said
30 grams of carbon or oil, is that per cubic meter of seawater per year?

FLOODGATE:  One cubic meter of seawater provides sufficient nitrogen to de-
grade 30 grams of oil in the summer, per year, or 11 grams of oil at winter
temperatures, per year.

RAYMOND:  I would like to throw this out for the audience since we have a lot
of marine microbiologists and chemists here.  Let's talk about mass balances.
Don't we have a much larger organic turnover, mass balance, which is not re-
flected in terms of pollutant turnover?  What amount of material do we have
turned over in marine waters?

FLOODGATE:  The figures are much higher for natural material.  I think it is
10   grams of carbon annually in the ocean.  The amounts of pollutants are
many orders of magnitude less.  So for total heterotrophy in the ocean, man's
effect is insignificant.  But most of our rubbish is not chucked into the en-
tire ocean.  We do not take garbage from London to the Pacific Ocean; we chuck
it into the English Channel.  This is also an area where we have a lot of fish-
ing and oysters.  This is highly productive area.  I don't think we have begun
to approach one of the real problems:  these areas are used for two opposing
purposes—as a source for food and recreation and as places to dump our rub-
bish.  These two ideas are really in opposition.  The North Sea has been used
as a dump for rubbish from the British northeast industrial complex around
Newcastle.  All around the North Sea there is heavy industrialization and even
before we were aware of the environment, we had been dumping stuff in there.
The North Sea is still one of the world's richest fishing grounds.  Does it
really matter?  Do we need all these kinds of regulations either from EPA or
from our side of the European pond, the Commission in Brussels?

RAYMOND:  I think an important point to consider is the type of system we are
using and the methodologies to evaluate whether a compound is degraded.  We
have to select a system and we should not select the very low carbon seawater.

LITCHFIELD:  Dr. DiGeronimo, do you have any idea of the in-situ or natural
concentrations one might expect to find of the p-chlorobenzoate or dimethyl-

DiGERONIMO:  Well, in our sewage systems some of the effluent compounds are
probably found in high concentration.  We felt the concentration in the sewage
may even have been below realistic values seen coming out of a factory.  In
natural waters there is a constant dilution of the compound and an immediate
dilution as it enters from a point source.

LITCHFIELD:  This is providing a carbon source or nutrient for the environ-
ment, particularly in the river system, so you might expect to find these
rates of degradation.  In terms of bacterial decomposition rates, are you


overloading the river system or is the concentration below what the organisms
might detect, or what?

DiGERONIMO:  I don't think I want to say anything about rate.  I'll talk in
terms of persistence.  I think we showed a trend and naturally more and better
rate data are necessary to confirm our systems.

JOHNSON:  Could you elaborate on the most probable number approach you're
using in terms of the method, a three-tube or five-tube system, and some of
the statistics?  I think many of our problems involve an effort to quantitate

DiGERONIMO:  You are absolutely right, but I think anybody who tries to jus-
tify a population study without substrate disappearance would be in trouble.
We developed a system for the aromatic degraders to try to quantitate a par-
ticular population within a larger population.  As the data showed, our popu-
lations of benzoate degraders in sewage were always found over the year at
106/ml.  To quantitate a population that is found at 102/ml, and only reach
106/ml would be virtually impossible to observe and the types of population
shifts  that were taking place.  To answer your specific question, we did use a
five-tube MPN method.

CHAPMAN:  Do you have any information as to the bacterial genera which turn up
in your experiments, with a capacity for growth with chlorobenzoate?

DiGERONIMO:  Yes, but only that they are gram-positive or gram-negative.

CHAPMAN:  In other words, you've not made generic assignments?


CHAPMAN:  This is of interest, of course, because the question arises, if, for
example, your organisms were from the genus Arth.roba.cter, would you logically
expect  those organisms to be selectively enhanced in their population by the
addition of benzoate or would benzoate simply increase the population of many
other organisms, lacking the ability to grow at the expense of chlorobenzoate?

DiGERONIMO:  You are talking about what would be the species of the population?

CHAPMAN:  I think it is of importance to identify certain of these genera, to
know something about their division times in these mixed cultures, and to an-
ticipate whether benzoate might logically be expected to increase their num-
bers .

DiGERONIMO:  I agree.

TIEDJE:  Last night we had a good discussion on cometabolism, but one aspect
we haven't really considered, related to cometabolism, is uptake.  Normally we
think of cometabolism in terms of degradative enzymes, not of transport en-
zymes.  Presumably with low concentrations, we could see a transport system
which would be non-specific.  I'm wondering if one could obtain useful infor-
mation  about cometabolism with low concentrations by simply studying the


kinetics of uptake of the particular compound and its co-substrates.  Have you
given this any thought?  For example, what is the K-t of uptake of the chloro-
benzoate compared to benzoate?

DiGERONIMO:  We have not addressed transport to confirm that.  However, there
can be an extracellular or surface enzyme for transformation of the compound.

TIEDJE:  Most of those studies have been done at higher concentrations where,
presumably, you could have a non-active uptake system.  If you are talking
about low concentrations, as would be expected in effluent systems, you are
going to have to concentrate against a gradient.  Presumably there is a ben-
zoate transport system that is also taking in chlorobenzoate.  If you knew the
K-t for chlorobenzoate transport, then you would be able to predict at what
concentrations you would get uptake or not.

BOLLAG:  When you pre-culture the organisms on the substrate benzoate, is
there a change in the transformation ability of the microbes at different con-

DiGERONIMO:  For low substrate concentrations, we haven't worked with pre-
adapted systems.  I don't know whether pre-adapted cells attack lower concen-
trations differently.

PASSMANN:  Dr. Blackburn, your data on ammonia production versus utilization,
in the first season's graph, showed tremendous utilization but you still have
substantial ammonia in the sediment.  Also you mentioned that in the deeper
sediments you see some curvature rather than a straight increase in the amount
of ammonia.  How precise is that data and, at one particular spot, is that am-
monia deficit, real, or is it simply an anomaly of one observation?

BLACKBURN:  No, there is a certain amount of scatter in this sort of data, but
over the five stations we've studied this is a definite trend.  When we see a
dip in the ammonia profile, this is associated with net uptake of ammonia.
There is a depletion in the ammonia pool rather than an increase in the ammo-
nia pool.

PASSMAN:  At one station there seemed to be a fairly substantial ammonia pool
but a much more pronounced utilization than any of the other stations in that

BLACKBURN:  One measurement is in static pools and the others are rates, so
they are in different units.  I'm not quite sure I understand your question.

PASSMAN:  The question I was asking was, at that particular station was that a
real phenomenon, is there enough replication there to say that there was a
real utilization of ammonia vs production?

BLACKBURN:  This is right.  We see the same trend in the other five stations
we examined, but it is not absolute.  Sometimes the gradient that one sees is
the result of what has gone on in the month before.  The rate is the result of
what is going on now.  So there is a certain time lag before the gradient
changes to the rates that are operating at that point in time.  So if you want
to try to analyze it you can say, the rate before was such and such, the rate


after was such and such, and the rate now is this.  But you don't know quite
how the rates have operated in between those times.  It may have been steady
up to the day before and something changed.  It is somewhat difficult to take
the kinetics and explain absolutely the static data.  In general, there is a
correlation, but the whole thing is confused because the gradients themselves
are mixed up by perturbation, in effect, the whole gradient may be moved up-

PASSMAN:  Thank you.

COLWELL:  Dr. Floodgate is looking too relaxed; I should like to elicit a com-
ment from him.  In your discussion of oil degradation,  you pointed out the re-
quirement for nitrogen in the degradation of certain amounts of oil,  but in a
previous publication of yours, you made a very important statement that oil
spilled during World War II could still be detected in  the Atlantic Ocean.
What were you measuring?  Isn't it a fraction, a portion of the oil that is a
residual portion, namely the resins, asphaltinic, and aromatic compounds which
are really not being attacked in the short time frames  of most studies?  Could
you make some sort of statement about this part of the  oil that probably poses
more of a problem than the short-chain alkanes which everybody, by now, knows
are readily degraded?

FLOODGATE:  Of course, you are absolutely right.  The large ring compounds
certainly have a very long turnover time.  I imagine that these can be in the
order of decades or even longer.  One tends to think of oil as one substrate
when, of course, it is many thousands of different substances.  In fact, if I
remember right, there are something like 200 to 300 n-alkanes alone.   I don't
think we can say very much about the rate of degradation of these very resis-
tant materials.  One can say they seem to stay on the bottom, because I think
most of them seem to end up in the sediment and one gets the impression they
are going to remain there for a very long time.  It could be decades, it could
be centuries in certain situations.  Whether one should be bothered about this
is another matter; after all if they are not going to be attacked by  bacteria
one gets the general impression they are more or less biologically harmless
because they are always covered by water, so they are not going to cause any
immunity problem.  They are just sitting there on the bottom; they would even-
tually presumably be covered up.  One can just hope that they stay there long
enough until some future scientist in the year 2500 or  something comes along,
digs them up and has another look at them.  But I don't think they need bother
us.  They are just there, and they are going to stay there.

COLWELL:  I can accept part of your statement, but not  all of it.  What about
the polynuclear aromatics being taken up by the food chain, as some studies
are beginning to show?

FLOODGATE:  Yes, these are the photo-intermediate  ones, aren't they?  They
possibly are the ones that we should be worrying about  a bit more.

COLWELL:  They are possibly carcinogenic.

FLOODGATE:  Some of them are supposed to be carcinogens.  But there is no evi-
dence, as far as I know, that any of them are causing any trouble.  The amount


of carcinogen in oil is very low, a few micrograms/100 tons.  Some of them get
into the food chain, some of them probably get into us, but probably we pick
up fewer carcinogens from oil than from walking down Main Street or Fifth Ave-
nue.  So should we be bothered about it?

COLWELL:  I am asking you.

FLOODGATE:  I'm not worried about it.

ALEXANDER:  Relative to some of the data that Dr. DiGeronimo presented, it is
important from the viewpoint of the workshop and regulatory agencies to point
out the marked effect of concentration on the rate of decomposition.  Although
the data presented were in percent of the substrate destroyed and converted to
COa, since the concentration declined by orders of magnitude, it meant the
rate also declined by orders of magnitude.  In going from a test system using
parts per million to a real environment with parts per billion or parts per
trillion levels, then we have to assume a decrease in rate by 103 or 101* (or
more) fold.  Conversely, at very low substrate concentrations, as indicated in
the case of 2,4-D, essentially nothing happens.  This is apart from the con-
ventional Michelis-Menten kinetics.  H. Jannasch, A. G. Marr, and others have
demonstrated this in pure culture, but our studies dealt with an effect in a
natural ecosystem.  The question I have relates to some of the comments of
George Floodgate.  George, looking at the issue from the viewpoint of the
regulatory agencies and extrapolating from a model to a natural environment,
what do you feel the effect of particulates would be on the biodegradation of
the non-oil pollutant in the marine environment?  Is a water-soluble substrate
sorbed to a particle—more readily decomposed, for example, because of the
higher concentration?  Or would it be more slowly decomposed because of the
problems of desorption or the need for breaking down some type of bonding?

FLOODGATE:  I think that might be explored.  We've been studying urea, and
certainly it can be absorbed onto clays.  The molecules being small can get
into the interstitial part of clay where the bug cannot get at it.  So to that
extent, clay protects urea and it will stay there forever.  On the other hand,
it can get onto the outside by ionic absorption or Van de Wouls absorption and
it might build up to a level where a bacterium has got something to go for.
But I think the answer is both, it depends on the material it is absorbing and
also the absorbent material.

ALEXANDER:  Have you done any work with any other compounds than the low mo-
lecular weight material which might get into the lattice structure of the clay
or otherwise absorbed?

FLOODGATE:  No, we haven't, but I believe some people have and have seen the
same kind of thing.

ALEXANDER:  In the marine environment?

FLOODGATE:  Yes, I think so.

SOMERVILLE:  First of all, I would like to thank Dr. Floodgate for acting as
an apologist of the oil companies.  I would like to reassure people that

certainly Shell is concerned that potentially harmful compounds have formed
during the natural degradative process,  when oil is released into the environ-
ment.  Also,  I" would like to point out that there is a great deal of natural
seepage of oil.  I have two questions, first of all for Dr. Floodgate, concern-
ing oil spills.  Is there any evidence when you get a massive contamination with
primarily a carbonous energy source,  that this acts as a reservoir for nitro-
gen fixation which then acts in an autocatalytic way to promote degradation?

FLOODGATE:  I don't think so.  More nitrogen fixation in the ocean is done by
an algae, a green algae.  I don't think there is any evidence that the fixation
which is undoubtedly carried out by marine bacteria—Clostridia, Desulfovib-
rios—is sufficient in quantity to make a really big difference.  There are re-
ports of hydrocarbons being the carbon source but I don't think that a big oil
spill is going to encourage much of this fixation.   We have been looking re-
cently at the effect of oil on algae.  We find that small amounts of oil tend
to stimulate algal growth and large amounts tend to decrease algal growth.
Now whether this will be the same kind of thing for the blue-greens which are
the nitrogen fixers, I don't really know.  I think I would like to apply to
Shell for a grant.

SOMERVILLE:  There is a lot of circumstantial evidence for this.  Large spills
on land, for example, do promote nitrogen fixation.

FLOODGATE:  True enough, that work has been done by the people at Westbury; it
happens on dunes and possibly salt marshes but I don't think it happens in the
open ocean.  The nitrogen fixation mechanisms are quite different, and we
haven't anything in the sea that corresponds to Rhizobium or Azotobacter in
the ocean.  These are available in the marshes in the plant life.  It does go
on in the land but not in the sea.

SOMERVILLE:  Thank you.  The other question I have is for Dr. Kadota.  It re-
fers to his limitation of growth rate as a function of carbon concentrations.
I think that both of your cultures reach a max at about 20 mg/liter carbon.  A
quick calculation suggests that this is approximately equivalent to the source
of nitrogen concentrations we have been talking about and suggests that it may
not be just nitrogen limitation which is limiting growth.  Do you know if
growth is limited by anything else or do you have any similar data with

KADOTA:  I have no data on the nitrogen limitation on the program.

PRITCHARD:  Dr. Wright, I am concerned about effects on the properties of the
systems that you see due to bottling or "containerization."  My question, basi-
cally, is, do you think we can settle on a set of conditions or time to study
the degradation processes in a system?  For example, do we study degradation
based on the 0-12 hours or 0-2 days incubation; or do we wait 12 days and then
examine the system for degradation?

WRIGHT:  What we are looking at is a continuous process.  During the first few
hours, I think the response is probably very close to the natural in-situ re-
sponse.  During the first day, there is adaptation and some growth occurring.
After the first day, I think the population is being selected,  and selective
effects are going to probably take precedence over any natural effects.  So if


you want to set hard rules or procedures, you must limit it to 12-24 hours for
natural rates, depending on the temperature—the colder the temperature the
longer the incubation time.  For a fairly natural population response you
might incubate as long as two days.  Beyond this, you obtain a highly selected
response of a highly selected population.

PRITCHARD:  Can I go to the extreme and say if my compound won't break down to
any great extent in two days, I could extrapolate to a persistence in a marine

WRIGHT:  I think one would have to be very careful about that extrapolation.
I would suggest there are probably other processes involved such as absorption
to particles, which will provide for breakdown of a type we don't see typi-
cally in planktonic bacteria.  I don't know the answer to that.

LITCHFIELD:  A question related to this, to what extent to you re-oxygenate
the water, or is this not a consideration when incubating a closed system for
24-48 hours or longer?

WRIGHT:  This is not a problem.  I've been dealing with really quite oligo-
tropic systems.

LITCHFIELD:  You have no oxygen deficit?


SLATER:  Dr. Kadota, have you done any experiments in your continuous enrich-
ments at two concentrations of substrate, where you incubated them for a longer
period of time than the relatively short time on the data you presented, I won-
der if the composition of the two communities changes in these open growth

KADOTA:  I have not incubated for long periods, so I have no data on such

PASSMAN:  Dr. Wright, over the past couple of years there have been a number
of studies in which dialysis was used to incubative systems in situ.  Would
you predict or do you have any information to suggest that wall effects would
be of the same magnitude in these systems as in your systems?

WRIGHT:  I've never used them; but my guess would be that the same things
would happen, perhaps greater because you would probably have leaching of sub-
stances from the dialysis tubing itself.

IVANOVICI:  Dr. Blackburn, I think that developing a method for estimating the
health of a system is very important.  You said that you had done most of your
work in a non-polluted system.  I would like to know if you are going to test
your hypothesis in a polluted system, and if so, what sort of predictions you
might make about the ammonia levels and the other parameters you are measuring.
I would expect you would be able to make predictions for such a method.

BLACKBURN:  Well,  it isn't absolutely a clean system,  it's a brackish fjord
that cuts across northern Denmark.   There probably are inputs from cities,
etc.  The problem is,  you don't know exactly what the  inputs are, and this is
part of the reason we adopted this  approach of determining what is happening
in the sediment and what's getting  out of the sediment.   We can, in fact, de-
termine what's getting into the sediment.  It isn't easy to know exactly what
goes in.  The prediction I would make,  if there is an  increase in input into
the sediment then all the rates of  turnover should increase and one should be
able to determine the change.  If hydrocarbons, cellulose, or something with a
high carbon:nitrogen ratio gets to  the sediment,  then  one should immediately
see this effect.  If it is breaking down then one should get an increased up-
take of ammonia; if it isn't breaking down then one should see an increased
total organic content in the sediment.   It all depends on quantity; and this
again comes into what is your definition of pollution  and if you base it on
quantitative or qualitative considerations.  These are the sorts of things one
might see depending upon what you are putting into the system:  increased
rate, increased organic compounds,  or changes in  the calculated nitrogen:
carbon ratio.

IVANOVICI:  Do you think that you would have some range  of values for rates
that you might associate with an unhealthy system?

BLACKBURN:  No, not really.  Because we have a range of  sediment types, some
have a higher organic input than others,  and they are  turning over.  It's
really a matter of defining what parameters can describe the system,  then, if
it is perturbed in some way one should see a change of those rates or parame-
ters.  I wouldn't like to say that  a certain rate indicated health.  It indi-
cates normality, and some change indicates abnormality.

IVANOVICI:  I would like to. direct  a question to  the panel as a whole.  We've
had presented here a discussion of  the effects of a number of different pollu-
tants.  I think it's really important in such a workshop that we define cer-
tain terms.  Can you put forth a definition of pollution?  What criteria are
you using to define pollution, perhaps concentrations  of the various  chemicals,
or in terms of biological effects,  or in terms of nutrient levels and nutrient

FLOODGATE:  All I can do is recommend that you come to my lectures.  Lecture
No. 2 is on definitions of pollution, and there are five or six completely
different definitions.  I try to use this approach to  show the student that
pollution is what you want it to be.  Now we are  here  with reference  to the
EPA and regulatory agencies in the  United Kingdom,  and the definition accepted
is one agreed on by the FAO in Rome in December 1970.   I can't quote  it ver-
batim, but it is roughly that marine pollution is anything that fouls up the
environment from the human point of view.  That is,  if it destroys amenities,
interferes with fishing, interfers  with quality of water from any human angle.
This is a view of a natural environment which many biologists, the environmen-
talists, find very unpleasant; they don't like it at all.   They feel  that what
promotes human well being is good,  but that isn't the  same thing as what pro-
motes the well being of the natural environment,  the natural creatures.  So
there are other definitions of pollution which take into account the  creatures
that live in that environment.  There are others  which are based on information
theory; there are others aimed toward one particular group,  one is more


microbiological in point of fact.  So the trouble is you are trying to define
something which literally cannot be defined.  It doesn't fit into a logical
category of definable objects, because you can't make limits on these things.
Something like the FAO proposal is the best you can do as a practical working
guide for regulatory organizations like the EPA.

IVANOVICI:  Is that the definition that is being used here, then?

FLOODGATE:  I don't know what definition the EPA uses, but this is the one
that GESAMP (Group of Experts on Scientific Aspects of Marine Pollution) uses.
In fact, as I remember, the FAO definition has several categories.  One is im-
pairment of fishing, another is quality of water, and the third is impairment
of human activities which means things like swimming, etc.  They list each of
the problems under these headings, so a particular substance going into the
sea will be given a high rating under one and a low rating under another head-
ing.  They are probably good working definitions, but they don't suit the en-
vironmental group, they don't like it at all.

IVANOVICI:  I was just curious.

STEWART:  Dr. Wright, isn't it possible that in the acridine orange direct
count method, you are getting or could get population shifts, without mate-
rially altering the total number.  What assurances do you have that this does
not take place?

WRIGHT:  It's very possible.  We just don't know; it's possible they are still
dying and .new cells dividing, adding to the population without a total change.
We just don't know at this point.  There hasn't been enough work with that
method.  Is that what you were referring to?


FLOODGATE:  This fluorescence method is a very interesting one.  Did you say
you were using formaldehyde, I think it is a preservative?  We've been adding
glutaraldehyde, because it does less damage to the more delicate bacteria giv-
ing slightly high numbers.  So I suggest you try glutaraldehyde.

WRIGHT:  I've tried it and encountered real problems with flocculation.  It
seems to create a lot of heterogeneity on the filters.  That's just in our

YOUNG:  Dr. DiGeronimo, in our tertiary-treated waste water, we find mono-,
di-, and trichlorobenzene, not the benzoate.  I'm wondering whether you have
any data on decomposition on these mono-, di-, and trichlorobenzenes.

DiGERONIMO:  No, but in working with engineers at Cornell, we were asked a
rather similar question.  We showed degradation and adaptation of sewage, but
it took three days to get rid of the substrate.  One should consider that some
compounds from industrial effluents may have a residence time in a sewage
treatment plant of only 24 hours.  In 24 hours, only 50% of the compound is
actually being acted upon, and what one sees reaching the environment is the
other 50%.  So, you may have to consider a longer residence time to get

complete destruction of some xenobiotics.

YOUNG:  Regarding chlorobenzenes,  would you anticipate any particular problem
in their decomposition?

DiGERONIMO:  This is speculation,  but I would say no.   When I first started, I
thought I was going to get little  degradation of several compounds, and I
started finding degradation.


          Chairperson, Gordon Chesters


                                 N. J. Poole
                         Department of Microbiology
                           University of Aberdeen
                              Marischal College
                           Aberdeen, Scotland, U.K.

                Anoxic conditions are rare in seawater but where
           anoxia does occur it is usually because of restricted
           water circulation and hence oxygen diffusion, as for ex-
           ample, in certain fjords, deep trenches, estuaries and
           landlocked bays.  The physical restriction of oxygen
           supply is often combined with an increased microbial oxy-
           gen demand resulting from the presence of large quanti-
           ties of utilizable carbon, e.g., a senescent algal bloom,
           domestic sewage or industrial effluent.  In the sediments
           of the continental shelf, however, the situation is dif-
           ferent for with the exception of the so-called "oxygen"
           or "high-energy windows" the thin surface oxic zone over-
           lies an extensive anoxic zone.  Considerable mineraliza-
           tion can occur in the anoxic zone; for example, Jorgen-
           sen (10) calculated that 53% of all the organic carbon
           mineralized in the sediment of a Danish fjord was cata-
           lyzed by the anaerobic sulphate-reducing bacteria.  In
           shallow estuaries receiving industrial effluent the mi-
           crobial activities in anoxic sediments have been shown
           to be a major factor in the development of a "pollution
           syndrome" (1, 19) .   This paper is concerned with the
           anoxic compartment of the marine environment, in particu-
           lar its development, and biological, chemical and physi-
           cal characteristics.  The maintenance of the anoxic con-
           dition and the relationship between the anoxic and oxic
           systems is examined.

     There are three processes by which an organism can generate energy in the
absence of oxygen, i.e., fermentation, bacterial photosynthesis or anaerobic
respiration (see 8, 13, 23).

     In fermentation,  organic compounds serve both as primary electron donors
and terminal electron  acceptors.   Some organisms have to be supplied with
separate sources of electron donor and electron acceptor, for example, Clos-
tridium kluyveri requires acetate plus ethanol; however, other organisms can
obtain both electron donor and electron acceptor from a single compound.  When
compared with aerobic  respiration fermentation is marked by:  a)  a compara-
tively low yield of energy and hence of growth per mole of substrate consumed
(the cell yield is generally below 10%)  and b)  the formation and excretion
into the surrounding medium of "energy-rich" fermentation products, often in
amounts equimolar with the substrate being fermented.

     Except for the cyanobacteria, bacterial photosynthesis is a totally an-
aerobic process and oxygen is not produced by the photolysis of water.  The
photosynthetic bacteria are therefore found in anoxic environments, where
there is sufficient light energy and suitable electron donors, e.g., sulphide.

     In anaerobic respiration an electron acceptor other than oxygen has to be
used.  The principal acceptors are nitrate, nitrite,  fumarate, sulphate and
CO2.  In nitrate respiration the nitrate is reduced to nitrite.  This reaction
is widespread among facultative anaerobic bacteria belonging to a number of
genera, including Pseudomonas, Bacillus, and various  members of the Enterobac-
teriaceae.  In denitrification the nitrate is reduced through nitrite, nitric
oxide and nitrous oxide to nitrogen and even ammonia.  This process is con-
fined to a more limited group of facultative anaerobic bacteria (17) .  Fumar-
ate reduction is effected by a broad range of bacteria  (11)  as well as in some
protozoa and helminths.

     Bacteria which use sulphate or COz as terminal electron acceptors show a
number of significant differences from those able to grow anaerobically with
nitrate or fumarate.  They are strict anaerobes, their respiratory mechanisms
can only be coupled to sulphate or COz.   They also have a restricted nutrition
since only a limited range of compounds can serve as  electron donors.  Bacte-
ria capable of coupling substrate oxidation to the reduction of sulphate (or
in some instances, sulphite, thiosulphate, tetrathionate or sulphur) to sul-
phide, are members of the genera Desulfovibrio, Desulfotomaculum,  Desulfomonas
and Desulfuromonas.  The nutrition of these organisms is limited to the oxida-
tion of hydrogen, pyruvate, lactate, ethanol, formate or malate which are con-
verted to fatty acids  (acetate) and COz.  Desulfotomaculum acetoxidans, how-
ever, has recently been reported as capable of utilizing acetate (25).  Bac-
terial sulphate-reduction in anoxic marine systems is of considerable impor-
tance because of the high sulphate concentration in sea water; in marine sedi-
ments sulphate can be  detected at depths of one meter or more.  In the absence
of sulphate these bacteria are capable of fermenting  ethanol or lactate to
acetate and hydrogen.   This process will only take place, however,  if the par-
tial pressure of hydrogen remains low, as would occur if hydrogen-utilizing
bacteria are also present in the system.

     The methanogenic  group of bacteria are strict anaerobes which obtain
their energy by coupling the oxidation of a very limited range of nutrients to
the reduction of CO2 to methane.   The principal electron donors are hydrogen
and acetate, although  rnethanol and formate can also be used; the latter is
however, generally first cleaved to COz  and hydrogen.

     It is often thought that the anoxic marine environment is the sole pre-
serve of obligate and facultative anaerobic bacteria.  For many anoxic sys-
tems, this is an erroneous view as was demonstrated by Fenchel and Riedl  (6)
who found both ciliates and other invertebrates active in anoxic systems.
                              OF ANOXIC SYSTEMS

     The methanogenic bacteria require more reduced conditions than the sul-
phate-reducers for growth, which in turn require more reduced conditions than
the denitrifiers.  Therefore, as a system goes from the oxic to the anoxic
condition a succession of organisms utilizing anaerobic respiration occurs.
By measuring the redox potential of a system it is therefore possible to de-
termine which electron acceptor is being used in anaerobic respiration.  Such
a succession in both the microbial population and the redox potential is shown
as a function of sediment depth in Figure 1.  This is an idealized situation
since in many sediments all the zones are not distinguishable because of fac-
tors such as a low nitrate concentration and the activities of burrowing ani-
mals.  Also, Figure 1 is a human or "macro-organisms" view of the sediment,
and it does not recognize that microbial activity occurs principally within
microenvironments.  It is therefore to be expected that, for example, denitri-
fication or sulphate-reduction  (9) will occur in what on a gross scale would
be called an oxic environment.  There is little information on the role of
such anoxic microenvironments in mineralization, since the majority of our
techniques  (and philosophy?) are designed to determine the average chemical
and microbial characteristics of a large number of microenvironments.  This
criticism can, of course, also be applied to techniques for determining the
redox potential of a sediment.  Since the platinum electrode behaves reversi-
bly under anoxic conditions it is relatively simple to determine the redox
potential of a sediment.  However, there are some major interpretational and
methodological problems with these measurements  (16, 24), for example, which
redox couple is actually dominating the system, the problem of inserting the
probe without mixing oxic and anoxic zones and the need to prevent interfer-
ence caused by contamination of the platinum proble or reference electrode.
Nevertheless the redox potential, especially when coupled with other biologi-
cal and chemical measurements, provides a useful characterization of the

     Bacterial sulphate-reduction, and hence sulphide production, is an impor-
tant reaction in marine anoxic systems.  Sulphide is also produced by the deg-
radation of sulphur-containing amino acids by either aerobic or anaerobic mi-
croorganisms, but it is not generally considered as a major source of sulphide
in the anoxic system.  Jorgensen  (10), for example, claimed that only 3% of
the total sulphide produced in the sediments of a fjord was derived from or-
ganic sources of sulphur.  Nedwell and Floodgate  (14), however, suggested from
studies of an intertidal mud flat that the sulphate-reducing bacteria were in-
hibited to a larger extent than other heterotrophic bacteria at temperatures
below 10°C and hence at these temperatures organic sulphur became a major
source of sulphide.  Jorgensen  (10) could, however, find no difference between
the Qio of the sulphate-reducers and other bacteria in this temperature range.

                        ?OBM OF RESPIRATION 4          ^IcCEPTORS
                        APPROXIMATE RE00X POTENTIALS

                          AEROBIC RESPIRATION
                                      +400 mV
                          NITRATE RESPIRATION

                                     -10O mV<
                                -160 to -200 mV
                                      -300 mV >
Figure 1.  The relationship between anaerobic respiration, redox potential and
           the depth of  sediment.

     At the pH of most marine  sediments the sulphide is present as  H2S or HS~.
The sulphide will react  with any available heavy metal in forming  ferrous sul-
phides  (7).  Jorgensen  (10)  found that 10% of the sulphide produced in a sedi-
ment was precipitated by metal ions, the rest being oxidized.  Sulphide reacts
rapidly with oxygen, the half  life of sulphide in natural sea water being
usually in the order of  10 min to one hour (1).   The release of sulphide from
sediments  into the water column can therefore cause considerable deoxygenation
(18) which in turn can cause the death of aquatic plants and animals.   The re-
lease of sulphide will be enhanced by changes in factors such as pH, tempera-
ture, pressure and animal activity.  Sulphide is a highly toxic compound and
its presence even at very low  concentrations can have lethal or marked sub-
lethal effects on an animal  or plant community  (20).  Sulphide is  also used by
many of the anaerobic photosynthetic bacteria and as a source of  energy by the

aerobic chemolithotrophic sulphur-oxidizing bacteria.

     Diffusion of oxygen from the water column into the sediment is a rela-
tively slow process, however,- the activities of the burrowing animals in the
sediment are believed to be an important factor in supplying oxygen to the
sediment.  These animals are not evenly distributed through the sediment,
since this depends on a number of biological and physical factors  (22, 26).
The addition of industrial or domestic waste to sediment can cause consider-
able changes in the composition and activity of the burrowing animal community
and hence, indirectly affect the activity of anaerobic bacteria.  Such changes
can be caused by the formation of a smothering blanket of particulate matter,
the presence of toxic compounds or simply by increased aerobic microbial ac-
tivity which causes deoxygenation.

     Most of the anaerobic bacteria are inhibited by even low concentrations
of oxygen.  The activity of the facultative anaerobes will assist in the main-
tenance of an anoxic environment; in addition, the sulphide produced by the
anaerobic sulphate-reducing bacteria will purge oxygen from the system.  The
other products of anaerobic respiration, i.e., nitrogen and methane, do not
react with oxygen, hence oxygen uptake measurements of a sediment give a poor
indication of anaerobic microbial activity.


     One of the major features of microbial activity in anoxic systems, whether
in the marine or other environments, is the development of complex "food webs."
These webs arise because, as stated earlier, fermentation is an inefficient
process, producing incompletely oxidized products.  The sulphate-reducing and
methanogenic bacteria can utilize only a very limited range of compounds as a
source of carbon and energy; these compounds are the products of fermentative
microorganisms.  Degradation of organic compounds in the anoxic environment
therefore occurs in a number of metabolic steps involving distinct physiologi-
cal and biochemical groups of microorganisms which result in denitrification,
sulphate-reduction or methanogenesis (Fig. 2).  There is little information on
the function and activity of the "food webs" in marine systems.  More is known
about the activities of anaerobic bacteria in other anoxic environments, par-
ticularly the rumen; care must be exercised in transposing results from one
environment to another but we can learn from the experiences of workers in
these fields, for example, the need to use rigorous anaerobic techniques to
isolate and grow obligate anaerobes.  A major problem in studying the anaero-
bic "food web" is that some of the necessary experiments require pure cultures;
the difficulty in isolating and maintaining pure cultures of, for example,
marine, obligate anaerobic, cellulolytic bacteria can be seen by the paucity
of these bacteria in culture collections.  While a large number of organic
compounds can be fermented by anaerobic bacteria, certain groups of organic
compounds do appear to be recalcitrant, e.g., the alkanes and many of the aro-
matics.  Aerobic degradation of aromatic compounds often involves an oxygen-
dependent step, though there are a few alternative mechanisms which can be
used by anaerobic microorganisms (4, 13).   The fermentation products are then


used by bacteria relying on anaerobic respiration which involves some inter-
esting, and as yet largely unexplored, microbial interactions.  Evidence from
other anoxic environments would suggest that anaerobic mixed cultures exhibit
some characteristics which are important from the ecological viewpoint.  Mixed
cultures can ferment a wider range of compounds than the component pure cul-
tures, for example, propionate and benzoate can only be degraded anaerobically
by mixed methanogenic cultures.  In mixed cultures more growth and a greater
degree of substrate utilization is usually possible than in pure cultures,
also the fermentation products will differ both qualitatively and quantita-
tively.  For example, in pure culture Rumenococcus albus ferments glucose to
acetate, ethanol, C02 and hydrogen, but in a mixed culture with Vibrio suc-
cinogens, ethanol is no longer produced while the amount of acetate produced
increases.  Mixed cultures, in addition, permit energetically unfavorable re-
actions, e.g., the fermentation of ehtanol to acetate by the "S" organism of
Methanobacterium omelianskii, to be "pulled through" by the removal of hydro-
gen during the metabolism of the second organism.
                                                                       ox io



j}( FERMENT ERS ) 	 ••

C°2, H2

^% ( DENITRIFIER0 ]— 	 *~





	 fc/ SULPHUR - \

Figure 2.  The basis for the formation of anaerobic  food webs.

     The relationship between the sulphate-reducing  and methanogenic  bacteria
is still not certain.  For example,  the sulphate-reducers  can produce acetate
and sulphide, the former can be used as an electron  donor  by the methanogens,
but the latter compound can also be  inhibitory,  hence  a spatial relationship
develops (3).  Bryant et al. (2)  from studies with known mixtures  of  bacteria
suggested that hydrogenase-forming,  sulphate-reducing  bacteria could  be  active
in some methanogenic systems that are low in sulphate.  In the marine sediment
it is still unclear whether methanogenesis occurs only below the sulphate-

reducing zone or if both processes can occur in the sulphate-reducing zone,
the methane produced being used by the sulphate-reducing and other bacteria
(12).  -It is'also probable that the acetate produced by some of the sulphate-
reducing bacteria is used by acetate-utilizing sulphate-reducers; the relation-
ship between cellulolysis and sulphate-reducers is a major interest of my
group.  It is, for example, possible to postualte the following mechanisms for
cellulose degradation, although this is a simplified diagram because the role
of hydrogen and possible inhibition caused by the production of sulphide and
hydrogen is omitted.

                                        cellulolytic bacteria
                                        cellulolytic and non-cellulolytic
                                        fermentative bacteria
                                        Desulfovibrio spp.
     [Methanogenesis]       acetate
     The actual construction of the food web will presumably depend on a wide
range of environmental factors.  The complexity of the "food web" is necessary
for efficient degradation to occur, but it also means that the degradation
process is particularly susceptible to disruption, for example, many microor-
ganisms are inhibited by quite low concentrations of hydrogen and sulphide,
compounds which are often end-products of the degradation process.  Very re-
cently Parkes (15) has suggested some of the seasonal changes in microbial ac-
tivity observed within the sediments of the polluted estuary of the Aberdeen-
shire Don were caused by sulphide inhibition of the bacteria.  Poole et al.
(18) found up to a 24% decrease in cellulose degradation in anoxic sediments
with a high sulphide concentration.

     Many of the products of the anaerobic degradation processes will diffuse
into the oxic zones.  The transport of material between the oxic and anoxic
zones is considerably assisted by the activities of the sediment-ingesting
animals, if they are present.  Rhoads (21) calculated that, in some sediments,
these animals could cause the turnover of the whole sediment annually to a
depth of 5-15 cm.  Many of the aerobic bacteria are able to use the products
of anaerobic metabolism as a source of carbon or energy, e.g., the sulphur-
and methane-oxidizing bacteria.  The "food web" responsible for the anaerobic
degradation of organic ompounds therefore has ramifications in the oxic envi-
ronment.  There is one other aspect of the interchange between oxic and anoxic

systems which has yet to be fully explained,  i.e.,  the mechanisms by which
non-speculating,  obligate anaerobic bacteria,  most  of which are very sensitive
to oxygen, survive in and are dispersed through oxic environments; presumably
the answer is that some of the vegetative cells in  the population are more re-
sistant to oxygen, or that the anaerobes survive in anoxic microenvironments,
possibly in association with facultative anaerobes.

     It can be concluded that the anoxic environment is extensive in the sedi-
ments of the marine system and hence there is a strong possibility that the
fate of any organic compound in the marine sediment will be degradation by an-
aerobic microorganisms.   Deliberate or accidental perturbations of the oxygen
balance in the sediment, e.g., by adding compounds of a toxic nature to the
sediment-ingesting macro-fauna, can have a profound effect on the development
of the anoxic zones.  The anaerobic degradation of organic compounds is
achieved in stages by means of a "food web" comprising distinct biochemical
and physiological groups of organisms.  Little is known about the function of
these "food webs'1 except that mixed cultures of anaerobic bacteria behave dif-
ferently from that which would be expected from a study of the component pure
cultures.  This fact raises considerable problems in the design and method-
ology of experiments aimed at determining the rate of degradation of a com-
pound in anoxic environments.  The problem can perhaps be best summarized in
the words of Francis Bacon, who said in 1620:  "We cannot govern nature except
by obeying her."


     I would like to thank Michele Bryder, Simon Joles, Bob Madden and Dave
Wildish for their contributions to this paper.
                              LITERATURE CITED

1.  Bella, D. A., A. E.  Ramm,  and P.  E.  Peterson.   1972.   Effects  of tidal
     flats on estuarine  water  quality.   J.  Water  Poll.  Contr.  Fed.  44:541-556.

2.  Bryant, M. P., L. L. Campbell,  C. A. Reddy, and M.  R.  Crabill.   1977.
     Growth of Desulfovibrio in lactate  or  ethanol  media low in  sulfate  in
     association with Ha-utilizing methanogenic bacteria.   Appl. Environ.
     Microbiol. 33:1162-1169.

3.  Cappenberg, T. E. 1974.   Interrelationships  between sulfate-reducing  and
     methane-producing bacteria in bottom deposits  of a fresh-water lake.
     I.  Field observations.   Antonie van Leeuwenhoek J. Microbiol. Serol

4.  Dagley, S.  1971. Catabolism of  aromatic  compound  by  microorganisms.
     Adv. Microbial Physiol. 6:1-46.

 5.  Penchel, T. M., and B. B. Jorgensen.  1977.  Detritus food chains of
      aquatic ecosystems:  the role of bacteria, p. 1-57.  In M. Alexander
      (ed.), Advances in microbial ecology. Vol. 1.  Plenum Press, New York.

 6.  Fenchel, T. M., and R. J. Riedl.  1970.  The sulfide system:  a new biotic
      community underneath the oxidized layer of marine sand bottoms.  Mar.
      Biol. 7:255-268.

 7.  Goldhaber, M. B., and I. R. Kaplan.  1974.  The sulfur cycle.  In E. D.
      Goldberg  (ed.), The sea, Vol. 5.  John Wiley and Sons, New York.

 8.  Hamilton, W. A.  1978.  Microbial energetics.  In J. Lynch and N. J. Poole
      (eds.)/ Microbial ecology:  a conceptual approach.  Blackwell Scientific,
      Oxford  (in press).

 9.  Jorgensen, B. B.  1977.  Bacterial sulfate reduction within reduced micro-
      niches of oxidized marine sediments.  Mar. Biol. 41:7-17.

10.  Jorgensen, B. B.  1977.  The sulfur cycle of a coastal marine sediment
      (Limfjorden, Denmark).  Limnol. Oceangr. 22:814-832.

11.  Kroger, A.  1976.  Phosphorylative electron transport with fumarate and
      nitrate as terminal hydrogen acceptors.  Symp. Soc. Gen. Microbiol.

12.  Martens, C. S., and R. A. Berner.  1977.  Interstitial water chemistry  of
      anoxic Long Island Sound sediments.  I. Dissolved gases.  Limnol.
      Oceangr. 22:10-25.

13.  Morris, J. G.  1975.  The physiology of obligate anaerobiosis.  Adv.
      Microbial Physiol.  12:169-246.

14.  Nedwell, D. B., and G. D. Floodgate.  1972.  Temperature-induced changes
      in the formation of sulphide in a marine sediment.  Mar. Biol. 14:18-24.

15.  Parkes, R. J.  1978.  The seasonal variation of bacteria within the sedi-
      ments of a polluted estuary.  Ph.D. Thesis, Aberdeen University.

16.  Parkes, R. J., M. J. Bryder, R. M. Madden, and N. J. Poole.  1978.  Tech-
      niques for investigating the role of anaerobic bacteria in estuarine
      sediments.  In G. Hamilton, Methodology for biomass determinations and
      microbial activities in sediments.  ASTM, Philadelphia  (in press).

17.  Payne, W. J.  1973.   Reduction of nitrogenous oxides by microorganisms.
      Bacteriol. Rev. 37:409-452.

18.  Poole, N. J., D. J.  Wildish, and N. Lister.  1976.  The use of micro-
      ecosystem models to investigate pollution of the estuarine ecosystem by
      pulp mill effluent.  DJanuscr. Rep. Ser. Fish Res. Bd. Can. No. 1403:1-18.

19.  Poole, N. J., R. J.  Parkes, and D. J. Wildish.  1977.  Reaction of estu-
      arine ecosystems to effluent from pulp and paper industry.  Helgolander
      wiss. Meeresunters. 30:47-61.


20.  Poole,  N.  J.,  D.  J.  Wildish,  and D.  D.  Kristmanson.   1978.  The effects
      of the pulp and paper industry on the  aquatic environment.  CRC Crit.
      Rev.  in Environ.  Control 8:153-195.

21.  Rhoads, D. C.   1967.   Biogenic reworking of intertidal and subtidal sedi-
      ments  in Barnstable and Buzzards Bay,  Massachusetts.   J.  Geol. 75:461-476.

22.  Rhoads, D. C.,  and D.  K.  Young.   1970.   The influence  of deposit feeding
      organisms on sediment stability and community trophic structure.   J. Mar.
      Res.  28:150-178.

23.  Thauer, R. K.,  K.  Jungermann,  and K.  Decker.   1977.  Energy conservation
      in chemotrophic anaerobic bacteria.  Bact.  Rev.  41:100-180.

24.  Whitfield, M.   1971.   Ion selective  electrodes for the analysis of natu-
      ral waters.  Australian Marine Sciences Association,  Sydney.

25.  Widdel, F., and N.  Pfennig.   1977.   A new anaerobic, sporing,  acetate-
      oxidizing, sulfate-reducing  bacterium,  Desulfotomaculum (emend.)  ace-
      toxidans.  Arch.  Microbiol.  112:119-122.

26.  Wildish, D. J.   1977.   Factors controlling marine and  estuarine sublit-
      toral  macrofauna.   Helgolander wiss. Meeresunters.  30:445-454.


                            E.  Paul Liechtenstein
                          Department of Entomology
                           University of Wisconsin
                              Madison, WI 53706

                A microcosm apparatus is described for studying the
           fate, metabolism, and movement of synthetic chemicals in
           the environment.  This microcosm is compartmentalized and
           consists of terrestrial and aquatic components  which can
           be held separately under a variety of environmental  con-
           ditions.  Simulated rain delivered occasionally to the
           terrestrial portion results in soil runoff which is  chan-
           neled into the aquatic component with its layer of lake
           bottom mud and its animal and plant inhabitants. The
           microcosm can be used to study the effects of rainfall
           and other environmental conditions on the fate, movement,
           and potential bioaccumulation, and interaction  of one or
           several test compounds after their application  to soils
           and/or crops.  Data are presented relative to the beha-
           vior of insecticides in the terrestrial and aquatic  com-
           ponents of the microcosm.  Problems of unextractable or
           bound   C-residues in soils and in crops, the mechanism
           of binding and the uptake of bound residues by  worms and
           oats are discussed.  Data relative to the fate  of insec-
           ticides in water after soil runoff and the effects of
           lake mud deposits on the metabolism of insecticides  are

     One of the factors contributing to water contamination by pesticides  is
runoff of agricultural soils previously treated with pesticides for  control
purposes.  Thus, soil particles contaminated with insecticide residues  can be
transported with runoff water into lakes and rivers where they may end  up  as
deposits on river or lake bottom mud.  Experiments conducted recently in our
laboratory have dealt with:  problems of the metabolism and movement of 1'*C-
phorate in a flooded soil-water-plant microcosm; the effects of lake bottom
mud on the above processes; and the importance of microorganisms on  the fate


of ll*C-phorate in a soil-lake mud-water microcosm.

     The first series of experiments dealt with investigation of soil flooding
on the fate a~nd metabolism of l "*C-phorate in an agricultural loam soil, on the
movement and metabolism of the insecticide in a soil-water-plant system and
factors affecting these phenomena.   llfC-phorate residues were readily released
from submerged soils into water, amounting to 45% of applied radiocarbon dur-
ing the first three days following flooding.  After a two-week incubation pe-
riod, as much as 50% of the radiocarbon applied to the soil was recovered from
the water.  Phorate was much more persistent under flooded than under non-
flooded conditions.  It was the major compound from submerged soils where it
accounted for approximately 70% of the total residues recovered.  Phorate sulf-
oxide was the major metabolite present in the water.  In nonflooded soils,
phorate sulfone was the principal metabolite, while only traces of it were de-
tected in the flooded system.  However, when Elodea plants were introduced
into the system, after 14 days phorate sulfone amounted to 30% of all benzene-
extractable lV-residues recovered, phorate sulfoxide to 44% and phorate to
27%.  At that time, soil, water and plants contained 32%, 39% and 17%, respec-
tively, of the applied radiocarbon.  While more lipid-soluble volatile metabo-
lites were recovered from nonflooded soils, more 14C02 was evolved from the
flooded soil.  The production of lkCOz was a function of microbiological ac-
tivity.  When 1'*C-phorate treated soil was flooded with increasing amounts of
water, the amounts of radiocarbon residues in the water increased.  However,
  ^C-residues in the water decreased with increasing amounts of soil (10).

     The effects of lake bottom mud on movement and metabolism of 1 "*C-phorate
in a flooded soil-plant system were studied with l "*C-phorate-contaminated loam
soils deposited on lake mud sediments.  A layer of lake mud underlying insec-
ticide-treated loam soil significantly increased the persistence of phorate
and reduced the amounts of   C-phorate residues released from the sediments
into the water.  After a two-week incubation, 62% of the applied phorate was
recovered undegraded, 13% as phorate sulfoxide; however, no phorate sulfone
could be detected in the presence of lake mud.  When insecticide-treated loam
soil was mixed with lake mud sediments, the metabolism of phorate was further
reduced and even less radiocarbon moved into the water.  Varying the amount of
lake mud or of insecticide-treated soil did not affect insecticide distribu-
tion in the system, nor its metabolism.  Both evolution of lkCO2 and the re-
lease of volatile lipid-soluble metabolites were depressed by underlying lake
mud.  Studies pertaining to the rate of metabolism and movement of 14C-phorate
in a soil-lake mud-water-plant system over a two-week incubation period indi-
cated a decline in the radiocarbon content of the soil-lake mud mixture during
the first three days of incubation, with a concomitant increase of lkC in the
water.  Subsequently, the amounts of llfc-compounds remained constant in the
soil-lake mud, decreased in the water, and increased steadily in the Elodea
plants, which picked up these compounds from the water.  Most of the plant-
associated radiocarbon could not be extracted and was bound to the plant tis-
sue, amounting finally to 81% of all the radiocarbon recovered fromEiodea (11).

     Since under agricultural field conditions phorate is oxidized rather rap-
idly in soil to its sulfone and sulfoxide, the experiments were conducted to
investigate the fate of these compounds in soil deposited on flooded lake mud.
In particular, the importance of microorganisms in reduction of phorate sulf-
oxide in a soil-lake mud-water microcosm was studied.  By utilizing a loam-


lake mud-water microcosm, the reduction of phorate sulfoxide, an oxidation
product of the insecticide phorate, was demonstrated.  With flooded phorate
sulfoxide-treated loam soil, only small amounts of phorate were produced.  The
addition of lake mud, however, dramatically increased reduction of phorate
sulfoxide to phorate, which after two weeks of incubation accounted for 44% of
the totally recovered residues.  Exposure of Drosophila melanogaster Meigen to
dry residues of extracts of phorate sulfoxide-treated loam soil, deposited
upon lake mud under flooded conditions, indicated an increase in toxicity due
to production of phorate.  This reduction of phorate sulfoxide to phorate was
the result of the activity of microorganisms, which in turn was dependent on
the supply of organic nutrients.  Thus, addition of glucose to this system
further enhanced the reduction of phorate sulfoxide to phorate.  Phorate sulf-
one, another oxidative product of phorate in aerobic agricultural soils, was
not reduced to phorate sulfoxide or to phorate (9).
                         OF BOUND PESTICIDE RESIDUES

     Results obtained recently  (1,2,3,6,7) in our laboratory pertaining to
soil-bound insecticide residues and their potential release are summarized in
the following discussion.

     For many years depletion curves indicating the persistence or rate of
disappearance of insecticides applied to soils have been published.  Typical
depletion curves  (Fig. 1) were obtained in 1964 by our group after applying
insecticides at the same rate, the same time and by the same methods to loam
soil field plots near Madison, Wisconsin  (5).  These depletion curves indica-
ted that compounds like aldrin or dieldrin persist in soil considerably longer
than malathion, methylparathion or parathion.  Frequently the decline in de-
tectable residues has been associated with such terms as "disappearance" or
"loss" or "volatilization."  However, the apparent disappearance of a pesti-
cide from soil can be due to an inability to detect its residues by conven-
tional procedures.  One reason a chemical cannot be detected is that the com-
pound, or its degradation products, cannot be extracted from soil, thus they
are "invisible."  Those residues which can be extracted long after their ap-
plication are the "visible," persistent ones.

     The use of radiolabeled pesticides in laboratory studies has made it pos-
sible to detect unextractable soil residues.  Combustion or strong hydrolysis
of extracted soils can release these unextracted or bound ll*C-residues.  The
problem is complicated, since present methods for release or liberation of
these bound residues also result in the destruction of their identity.  In-
sight into the mechanism of binding of pesticide residues to soils might shed
some light on the nature of the residues and their potential release.

     Utilizing 14C-ring labeled parathion, the amount of unextractable or
bound *4C-residues in a sandy and a loam soil were determined by combustion to
14CO2, after the soils had been extracted three times with benzene-acetone-
methanol  (1:1:1)  (10).  Depletion curves of extractable parathion residues es-
tablished during a one-month incubation period  (Fig. 2) were similar to those
established under field conditions and resulted finally in recoveries of 30 to


           APPLIED: 5 LBS/5" ACRE C3.I PPM.)

< 20
cc  „

         '.      \              \
         ITHION  \
         i.5  I

                                 I  \
                    2     3
Figure 1.   Depletion curves depicting
   extractable residues of various
   insecticides applied at 6 kg/ha
   (concentration,  3.1 ppm) to loam
   soils in  field plots.
< z
u. o
O i
                                                        S? < 60
                                                        ^ c?
£ ^40
CC ^
                                                               !•• CRANBERRY SOIL
                                                                 BOUND (B)
                                                                                        \  EXTRACTED(E)
                              - LOAM SOIL
                                    BOUND (B)
                                                                     14    21
                                                                     DAYS   OF
                         28   0    7    14
                         SOIL  INCUBATION
 Figure  2.   Binding and extractability of l"*C-ring-
    parathion in two soils during  a 28-day incuba-
    tion period at 27°C (applied dose, 1 ppm).  The
    amounts  of parathion as determined by GLC in
    the  loam soil extract were  nearly identical to
    its  radiocarbon content.   (I)  Sandy soil  from
    cranberry bog;  (II) loam  soil.  Curve B, bound
    parathion; curve E, extracted  parathion; and
    curve E  + B, the total.

36% of the insecticide dose applied to the loam soil.  With a steady decrease
of extractable residues over a one-month incubation period, an increase of un-
extractable, bound 1^C-residues occurred.  This resulted finally in total re-
coveries of extracted plus bound residues, which amounted to 80% of the ap-
plied radiocarbon.  Attempts to exhaustively extract these bound lV-residues
with a variety of solvents, ranging in polarity from benzene to water, failed
to further release a significant amount.  The rate of binding of *V-residues
was highest in the loam soil and was related to the activity of soil micro-
organisms.  In soils sterilized by gamma irradiation or by autoclaving, the
binding of 14C-parathion was reduced by 58 to 84%.  Under anaerobic conditions,
created by flooding soils with water, the rate of binding of *^C-compounds
doubled.  The amount of bound residues decreased from 67 to 16% when soils
were sterilized prior to insecticide treatment and flooding.  Reinoculation of
this soil with microorganisms fully reinstated the soil binding capacity.  In-
cubation of the moist-treated soil under nitrogen also increased the formation
of bound residues, while incubation at 6°C rather than 27°C inhibited binding.
When these experiments were repeated using l **C-ethyl rather than 1!*C-ring la-
beled parathion, identical results were obtained which indicates that the
bound residues apparently contain both the aryl and alkyl portions of the pa-
ration molecule.  This information led us to suspect that amino-parathion,
which contains both the aryl and alkyl portions of the molecule and is formed
under anaerobic conditions, might be the bound residue.  In addition, earlier
data from our laboratory (4) indicated that while parathion residues could be
extracted and detected by thin-layer chromatography in loam soil extracts af-
ter several weeks of soil incubation, amino-parathion and p-aminophenol could
not be recovered after a one-day soil incubation.  When this experiment was
repeated in 1976 using radiolabeled compounds, we found that 49% of applied
^C-aminoparathion could not be extracted from the soil 2 h after its applica-
tion, while only 1.6% of applied parathion was bound in 2 h (3).  Comparison
of binding of all of the nitro and amino analogs of parathion (Fig.  3) during
a brief 2-h incubation with loam soil indicated that in all instances the
amino compounds were bound to a much greater extent than the nitro compounds.

     The role of microorganisms in producing l^C-parathion-derived bound resi-
dues and the mechanism of production of soil-bound residues was also investi-
gated by incubating :^C-ring-parathion in soil-free culture media inoculated
with soil microorganisms (3).  The amounts of 1^C-compounds in culture super-
natants, that upon addition to soil became unextractable, increased up to 12 h
of microbial culture incubation, when 43% of the applied radiocarbon was bound
after a 2-h soil incubation period (Fig. 4).  The increase in soil-bound resi-
dues correlated with a decrease in the amount of parathion in the microbial
culture and a concomitant increase in appearance of the major degradation pro-
duct, aminoparathion (Fig.  5).

     These data indicate that the production of soil-bound residues of para-
thion occurs in two steps.   First, microorganisms convert parathion to amino-
parathion, which becomes rapidly bound to soil.  These soil-bound residues are
unextractable and therefore undetected in routine residue analyses.

     Experiments conducted with 1 I*C-phorate (7) indicated that 26.4% of the
applied residues had become soil-bound within one week of incubation.  Con-
trary to results with   C-parathion, binding of l^C-phorate residues did not
increase after one week.  Further investigations  (6) pertaining to the


x O  50
      =! Q! 30
      O n

      P £20
         S  10
                PA A-Rft     PO  A-PO    PH  A-PH
    Figure 3.  Amounts of 14C-bound residues  in  soils,
       2 h after soil treatment at 1 ppm with ^C-
       parathion (PA), *^C-aminoparathion (A-PA),
         C-paraoxon (PO),  l4C-aminoparaoxon (A-PO),
       p-^C-nitrophenol (PH) ,  or p-^C-aminophenol
                                                                                    BOUND (I)
                                                                              CELL DENSITY (I)
                                                                        12    24   36    48    60
                                                               HOURS OF CULTURE  INCUBATION

                                                     Figure 4.   Binding of    C-compounds  to  soil
                                                        within  2 h after the addition  of  supernatants
                                                        obtained from 0- to 96-h old microbial  cul-
                                                        tures ,  treated with   C-parathion and inocu-
                                                        lated  (I) with soil microorganisms;  NI  = non-
                                                        inoculated controls.

                          Q ---- . --- o ---- _
              0     12    24    36    48    60      96
    Figure 5.  Amounts of parathion and amino-
       parathion determined by gas liquid chromatog-
       raphy  in benzene extracts of supernatants of
       microbial cultures after treatment with  * 4C-
       ring-parathion at 10 ppm and incubation  for
       0  to 96 h.

 - -i
I T(E^B)
	 -«-— ' 'I1- ' 7"" '

4 	 : 	 : ' _j
. ii L
[*--*-?-* 	 J
D. '"C-DDT




           14   21    28   I   r    14    21


Figure 6.  Binding and extractability of
     C-labeled insecticides in a silt loam
   soil during a 28-d incubation period,
   after soil treatment at 1 ppm.  With
   the exception of :V-methylparathion
   the amounts of extractable 1'*C-Dyfonate
   (fonofos) , ^C-Dieldrin, and ltfC-DDT,
   as determined by gas-liquid chromatog-
   raphy, were similar to the amounts of
   extractable radiocarbon.  For compari-
   son purposes,  data are inserted in A
   for the bound residues of :^C-parathion
   in soil.

extractability and formation of bound  ^C-residues in an agricultural loam
soil were conducted with the "non-persistent" insecticides l4C-methylparathion
and llfC-fonofos (Dyfonate®)  and with the "persistent" insecticides ll*C-diel-
drin and 14C-p,p'-DDT (Fig.  6).  With 1 ^C-methylparathion, only 7% of the ap-
plied radiocarbon was extractable 28 d after soil treatment,  while ll+C-bound
residues amounted to 43% of the applied dose.

     Field studies conducted in 1968-69  indicated that fonofos has a half-life
in Piano silt loam soil under Wisconsin summer conditions of about 28 d  (8) .
In the laboratory, this loam soil was treated with 1!*C-ring or ^C-ethyl-
labeled fonofos and incubated for various periods.  After 28 d, about 47% of
the radiocarbon was extractable, most of which was fonofos.  However, unex-
tractable ^C-residues increased with incubation time, resulting after 4 weeks
in 35% of the applied residues being soil bound.  These residues were of
course not detected in the field study.   Results using ^C-ring or ll*C-ethyl
labeled fonofos were very similar indicating that the bound residues probably
do not involve a cleavage product.  Contrary to results obtained with para-
thion, binding of fonofos does not appear to be dependent on microbial ac-
tivity.  While irradiation or autoclaving soil prior to insecticide treatment
and incubation significantly reduced the binding of parathion and flooding en-
hanced it, fonofos binding was not reduced by irradiation.  Autoclaving re-
duced binding somewhat, possibly due to an alteration of soil structure, and
flooding slightly reduced fonofos binding.  Smaller amounts of soil-bound
residues had been formed with the "persistent" insecticides amounting after 28
d to only 6.5% of the applied l V-dieldrin and to 25% of the applied ll+C-p,p'-
DDT, while 95% and 72%, respectively, were still recovered by organic solvent
extraction.  They differed from the organophosphorus compounds in their rela-
tively low binding properties and high extractability from soils.

     The question of the potential biological availability of bound insecti-
cide residues was investigated  (6) by testing the insecticidal activity of
bound residues from l^C-fonofos and l^C-methylparathion-treated soils with
fruit flies  (Drosophila).  With soils containing unextractable radiocarbon at
the insecticide equivalent of 3 ppm, no mortalities were observed during a 24-
h exposure period to the soil, and only slight mortalities occurred during an
additional 48-h exposure.  However, with soils to which the insects were ex-
posed immediately following the insecticide application, at the same concen-
tration as the unextractable radiocarbon  (3 ppm) , 50% of the flies died within
2-3 h after fonofos application, and within 18-20 h after soil treatment with
methylparathion.  It appears, therefore, that bound insecticide residues are
not only unextractable, but they also are less active biologically.

     Experiments also were conducted to study the release and availability of
unextractable, soil-bound residues of l^C-ring-methylparathion and the poten-
tial uptake of these ^C-residues by earthworms and oat plants (1).  Data from
this investigation indicate that unextractable soil-bound insecticide residues
are not entirely excluded from environmental interaction.  After incubation of
soil treated with ^C-methylparathion for 14 d, and exhaustive solvent extrac-
tions, bound residues remaining in this  soil amounted to 32.5% of the applied
insecticide.  However, after worms had lived for 2-6 weeks in this previously
extracted soil containing only bound residues or after several crops of oats
had grown in it, sizable amounts of l ''C-residues were found in the animals.
Earthworms which lived in the soil for 6 weeks contained a total of 2.7% of


the *4C-residues which could not be extracted from these soils, while three
crops of oat plants each grown for two weeks contained a total of 5.1%.  The
majority of previously soil-bound ll*C-residues taken up by earthworms  (58-66%)
again became bound within the animals, while most  (82-95%) of the *4C-residues
in oat plants were extractable.  Greens of oat plants contained 46-62% of the
  C-residues recovered from plants.  Most of the l4C-residues in oat greens
were benzene-soluble while most of the 1^C-residues in the seeds and roots
were water-soluble.

     Because soil-bound insecticide residues can be released from soil by
these organisms, any loss in toxicity due to binding should not be regarded as
permanent.  Even if release of non-toxic compounds occurs, interaction with
other chemicals in the environment cannot be disregarded.  The release and po-
tential biological activity of these bound residues certainly warrants further
study.  In view of the above finding, the expression "disappearance" and "per-
sistence" of pesticides, so widely used during the last two decades, should be
reassessed to consider the bound products.
                              LITERATURE CITED

1.  Fuhremann, T. W., and E. P. Lichtenstein.  1978.  Release of unextractable
     soil bound  1HC-methylparathion residues and their uptake by earthworms
     and oat plants.  J. Agric. Food Chem. 26:605-610.

2.  Katan, Y., T. W. Fuhremann, and E. P. Lichtenstein.  1976.  Binding of
     14C-parathion in soil:  a reassessment of pesticide persistence.  Science

3.  Katan, Y., and E. P. Lichtenstein.  1977.  Mechanism of production of soil
     bound residues  of  l "*C-parathion by microorganisms.  J. Agric. Food Chem.

4.  Lichtenstein, E. P., and K. R. Schulz.  1964.  The effects of moisture and
     microorganisms  on  the persistence and metabolism of some organophosphorus
     insecticides in soils, with special emphasis on parathion.  J. Econ.
     Entomol. 57:618-627.

5.  Lichtenstein, E. P.  1975.  Transport mechanism in soil:  metabolism and
     movement of insecticides  from soil into water and crop plants.  IUPAC
     Congress of Pesticide Chemistry, Helsinki, 1974.  Pure Appl. Chem.

6.  Lichtenstein, E. P., Y. Katan, and B. N. Anderegg.  1977.  Binding of
     "persistent" and "nonpersistent" :^C insecticides in agricultural soil.
     J. Agric. Food  Chem. 25:43-47.

7.  Lichtenstein, E. P., T. T. Liang, and T. W. Fuhremann.  1978.  A compart-
     mentalized microcosm for  studying the fate of chemicals  in the environ-
     ment.  J. Agric. Food Chem. 26:948-953.

8.  Schulz, K. R., and  E. P. Lichtenstein.  1971.  Persistence and movement  of
     Dyfonate in field  soils.  J. Econ. Entomol. 64:283-287.


 9.   Walter-Echols,  G.,  and E.  P.  Lichtenstein.   1977.   Microbial reduction of
      phorate sulfoxide  to  phorate in a soil-lake mud-water microcosm.   J.
      Econ.  Entomol.  70:505-509.

10.   Walter-Echols,  G.,  and E.  P.  Lichtenstein.   1978.   Movement and metabolism
      of 14C-phorate in  a flooded  soil system.   J.  Agric.  Food Chem.  26:(May-
      June) .

11.   Walter-Echols,  G. ,  and E.  P.  Lichtenstein.   1978.   Effects of lake bottom
      mud on the movement and metabolism of  1 "*C-phorate  in a flooded soil-plant
      system.  J. Agric.  Food Chem.  26:599-604.

                       ROLE OF SURFACE  MICROLAYERS

                              Birgitta Norkrans
                      Department of Marine Microbiology
                           University of Goteborg
                          S-413 19 Goteborg, Sweden

                A definition of the different surface microlayers and
           the distribution of "dry" and "wet" surfactants in the air/
           water film has been presented.  Based on results from va-
           rious hydrophobia samplers a thickness of 1 ym for the bac-
           terioneuston microlayer has been established.  Interest
           has focused on the surface lipid film not only as it may
           be per se of basic importance for functioning of the sur-
           face microlayer but also due to the analogous, important
           interface accumulation occurring in the system, namely that
           at the surface of gas bubbles.  Some bacteria have a greater
           tendency to accumulate in the lipid film than others (model
           system studies).  The number of accumulated bacteria differ
           with the lipid content of the film, shown in model systems
           and in the natural environment.  Surface balance studies
           confirm the concept of bacterial interaction at "film-
           bacteria-sites'' where hydrophobic interaction, surface
           charge of bacteria as well as of the film, and enzymic
           activity may be involved.  Biodynamics of the surface mi-
           crolayer have been dealt with considering the reported
           low bacterial biochemical activity, in spite of the high
           concentration of dissolved organic carbon in the layer.

     The surface microlayers in aquatic environments are an infinitesimal part
of the water body, but nevertheless are of great importance.  These micro-
layers are dramatically different from the bulk-subsurface water, for they
comprise an accumulation layer for dissolved substances, particles and micro-
organisms diffusing or brought by rising bubbles, convection and upwelling,
from the subsurface water to the surface.  Microlayers influence the exchange
of gas and transport mechanisms from the water body to the atmosphere and vice
versa.  Furthermore, they are a sink for atmospheric fallout, and concentrate
heavy metals (e.g., 13) and pollutants.  The occurrence of organochlorine re-
sidues is of considerable interest.  Compared to the levels of such compounds
in seawater, Duce et al.  (13) and Larsson et al. (22) report enrichment fac-
tors of 103-10i* for PCB in samples taken from Narragansett Bay, Rhode Island,


and Jorefjorden,  Sweden,  respectively.   The overall picture for the surface
microlayers is well known.   However,  while our knowledge of their chemistry is
gradually growing, real understanding of their biological structure and func-
tion is very far from complete.


     First of all we have to define the term "surface microlayers."  Most
workers  (e.g., 15, 17, 18,  22)  have considered lipids to be the major consti-
tuents of the uppermost surface microlayer.  The amphiphilic molecules have a
preferred orientation with respect to the water surface, with the long-chained
hydrophobic parts extending into the air which enable them to form a strictly
ordered lipid film.  This film consists mainly of free fatty acids in the Ciz-
Caa range, alcohols, and glycerides,  of which triglycerides dominate.  There
are characteristic deviations in the fatty acid patterns of the film compared
to that of the subsurface water.  The relative amount of saturated acids is
higher in the surface film compared to the subsurface water, probably because
of oxidation at the air/lipid interface.  The amount of glycerides relative to
fatty acids is always higher in the surface multifilms than in the subsurface
water.  The lipid film of so-called "dry surfactants" is a real surface micro-
layer, based on the length of fatty acids 10-20 A thick; if multilayered, ap-
proximately 100 A thick.

     Baier et al.  (4), however,  consider all waters, except heavily polluted
ones, to be coated with films of polysaccharide-protein complexes ranging in
thickness from 100 to 300 A.  Their statements are based on germanium-slide
collected^material, analyzed by a "nondestructive, direct analytical tech-
nique," multiple attenuated infrared reflectance, MAIR  (3, 4).  These "wet
surfactants" can form a more irregular, submerged stratum, and are essentially
hydrophilic but stick to the surface water by virtue of their few hydrophobic
chains.  As pointed out by Wangersky (34), it remains to be seen whether the
infrared absorption technique overestimates the contribution of proteinaceous

     These two strata taken together have been called the surface microlayer
by Sieburth et al.  (32), who quote a thickness of 0.1 ym.  As seen in Table 1,
the various sampling devices used collected surface microlayers from 250 to 4
ym thick due to the amount of water withdrawn.  Most authors use the term sur-
face microlayer in a way that makes its thickness dependent on the particular
collection method employed.  Sieburth et al.  (32), however, designate their
screen-collected layer (150 ym)  simply as "screen."

     In view of our special interest in the sampling of the lipid film, the
efficiencies of pronounced hydrophobic samplers with limited water withdrawal
were compared  (Table 2).  These were tested in model systems with various
lipids applied at the air/water interface in different amounts, calculated to
give gaseous or condensed monolayers, or a decalayer.  Good yields were ob-
tained with monolayers, but the yields of gaseous films, less lipid material
than one monolayer,- are significantly lower.  When using the different samp-
lers in natural environment or in a model system, the hydrophobic Nuclepore
membrane gave the highest enrichment factor (E) for bacteria in all comparable
experiments, varying over a range of 102-10l*.   The factor E was defined as the


  number of bacteria/ml  of the collected layer  relative to that of the subsur-
  face  water.  Parker and Hatcher (29) suggested  a  nonlinear change in algal
  population densities with surface microlayer  depth,  some sort of microstrata.
  Here  the operationally dependent dimension, 1 Jim,  seems to coincide fairly
  well  with a biological stratum, the bacterioneuston  layer, yet another surface
Type of
thickness Source

   Bubble bursting    10


  Seawater model system
 (14, 31, 32)


Teflon plate
Seawater estuarine
Seawater model system
(1, 10)
(20, 26, 28)
             AND TEFLON  SAMPLERS USED IN MODEL  SYSTEMS.   Data obtained from  S.
             Kjelleberg,  Th.  A.  Stenstrom, and  G.  Odham (to be published).
                                            Percent recovery of lipid films
Amount of  Calculated
  water      sample
withdrawna t hickness
(pl-ditT2)     (urn)
                        Olive oil
Monolayer  Decalayer  Monolayer Trilayer
                                     Oleic acid
(gaseous film)
Teflon sheet
Teflon plate














aBased on 20 weighings,  "n.d.  = not determined.

     Thus, the term surface microlayer has been involved in the assignment of
a series of strata.  To obtain uniformity, it is suggested that the term sur-
face microlayer be used only with the 0.1 ym layer of dry and wet surfactants,
and otherwise with additions of explanatory pre- or postfixes.  In this man-
ner, the strata sequence from the uppermost layer, very similar to^the Top-
millimeter-range of Maclntyre (24),  would be:  lipid film (10-100 A), surface
microlayer (0.1 ym), bacterioneuston microlayer (1 ym) .  As a further example,
screen microlayer  (e.g., 150 ym, see above) would be used as the term is opera-
tionally defined.  Surface microlayers would represent the layers altogether.


     Our interest has been focused on the lipid film not only as it may be per
se of fundamental importance for the function of the surface microlayers, but
also due to the analogous, important interface accumulation occurring in the
system, namely, that at the surface  of gas bubbles.  The latter are produced
in different ways  (e.g., by living organisms, sea turbulence, etc.) in amounts
sufficient to cover 3 or 4% of the sea surface at any moment  (24) .  During
their slow rise, lipids, other organic molecules,  particulate matter, and mi-
croorganisms adsorb at the surface of the bubbles, finally causing the surface
microlayers to be enriched in organic matter, dead or alive.  At the water sur-
face, the bubbles eventually burst at an assumed rate of 108 sec"1.  When the
surface-free energy of the bubble is transformed to the water beam as kinetic
energy, the aerosol droplets formed are ejected into the air in millisecond
reactions.  This formation of droplets has been described as a surface micro-
tome  (23) , as the ejected droplets carry away material only from the bubble
surface and the top micrometers of the water surface.  The rising bubbles act
to link the subsurface, the surface microlayers and the atmosphere.  They are
of a special importance as conveyors into the atmosphere of persistent sub-
stances like PCB, which withstand most degradation processes except photodeg-


     The microflora of the surface microlayers have been reported to be dif-
ferent from, and in greater concentrations than in the subsurface water (5, 6,
29).  Based on these observations, a marine model system study was undertaken
to find out whether some bacteria have a greater tendency to accumulate in the
surface film than others, and whether the number of accumulated bacteria dif-
fer with the chemical composition or lipid content of the lipid film (26) .

     The model systems comprised an assay tray with a 2.5% NaCl solution as
subsurface water and were inoculated with a defined number of bacteria and
covered with an applied lipid surface film, mono- or decalayered.  The bac-
teria were retrieved by the Teflon plate sampling procedure and enumerated by
plate count methodology.  The enrichment factor (E) for the bacteria was de-
fined, as above, as the number of bacteria/ml in the Teflon-plate collected
layer relative to that in the subsurface water.  E varied over a range from
7-100  (Table 3) in a system with monolayered lipid film of oleic acid  (Ci8-i)
and four different marine bacteria used as test organisms (Aeromonas dourgesi,
Pseudomonas fluorescens, P. halocrenaea, and Serratia marinorubra).  This


terminology does not conform to that of Bergey's 8th Edition.  S. marinorubra
showed an E of 100 or approximately 100, about 10 times higher than values  for
the three other microorganisms.  Still lower values were obtained for two
gram-positive bacteria tested.  E was of the same magnitude whether the film
was formed solely of oleic acid or of an oleic acid or of an oleic-palmitic
acids-triolein mixture, both monolayered.  For a decalyer of oleic acid, E was
approximately twice that obtained in a monolayer.  Field studies also showed a
positive correlation between the number of bacteria in the surface layer and
the amount of lipids simultaneously collected  (Table 4).
Number of bacteria x 106
Test organisms

Aeromonas dourgesi
Pseudomonas fluorescens

Pseudomonas halocrenaea

Serratia marinorubra

per ml in
samples from
Number of bacteria* per ml
      in  samples  from
Total amount (yg)
of fatty acids in
   samples  from







(1 ml)
(1000 ml)

*Average value from generally six samplings.


     The E values of S.  marinorubra decreased with an increase  in  the  number
of cells in the subsurface water,  consisting solely of Serratia cells  (Fig. 1)
This might be due to a decrease in accessible "film-bacteria-sites"  until  a
point of saturation, after which further change in the absolute number of  ac-
cumulated bacteria in the film occurred only within experimental error.  Even
when using an almost constant number of Serratia cells in  the bulk,  in addi-
tion to increasing numbers of Aeromonas cells,  the E values  for Serratia de-
creased (Table 5).  Thus, in spite of the low accumulation tendency  for Aero-
monas cells, they seem to compete  for "film-sites," available binding  groups
in the film, or a minimum of free  space between bacteria,  may be necessary to
eliminate the effect of electrostatic repulsion.
                      NUMBER OF BACTERIA x 106 ml"1
                           SUBSURFACE WATER
Figure 1.  Enrichment factor  (E)  of Serratia marinorubra in model systems with
           monolayered lipid  film of oleic acid at different cell densities in
           the subsurface  water  (26).
     In bubble bursting experiments  (6), where relatively high numbers of bac-
teria were used,  the number  of bacteria transferred into the air were also
relatively constant irrespective of  the number of bacteria in the solution
from which they were ejected.  Even  here a competition for film-sites, namely
around the bubble,  can  be  assumed.   The low E values obtained in Oresund, a
polluted area with  high numbers of subsurface bacteria, 105-108 ml"1  (B
Stehn, personal communication) may be explained in a similar way.

TABLE 5.  ENRICHMENT FACTOR  (E) FOR Serratia marinorubra AND Aeromonas dour-
Number of bacteria x 10
per ml in samples from




Surface film
species Total
205 207
56 96
Subsurface water
species Total
2.3 . 0






     The idea of bacterial  interaction  at  sites  in  or  just beneath  the  surface
film is supported  in surface  balance  studies  with various monolayered films
 (oleic acid, a mixture  of oleic  acid, palmitic acid and  triolein, triolein
alone, or  dipalmitoyl-phosphatidylcholine)  when  their  behavior  with and with-
out bacteria present is compared (20).   During.compression,  the film changes
from a gaseous state, with  a  large  area per molecule,  to a condensed monolayer
where the  molecules are tightly  packed. The  adhesion  of bacteria  (or mole-
cules) is  indicated by  an increase  in the  surface pressure at a given area per
molecule.   Dissolution  of the film,  e.g.,  as  a result  of enzymatic  activity,
decreases  the film pressure.   As could  be  expected  with  regard  to the E values
obtained in similar model systems,  S. marinorubra  (Fig.  2) again showed the
strongest  interaction and greatest  strengthening of the  film, except for  the
cell wall-free Acholeplasma laidlawii  (Fig. 3) which has its lipid  bilayered
plasma membrane accessible.  From enlarged and more sophisticated surface bal-
ance sutdies  (Kjelleberg  and  Stenstrom, to be published) it  can be  assumed
that various factors besides  hydrophobic interaction are involved in the  in-
teraction,  e.g., surface  charge, both of the  monolayered surface film and of
the bacteria, and  enzymatic activity.


     The above forces,  evidently different for different bacteria,  may  con-
tribute to the maintenance  of the bacteria in the microlayers.   But how do
they get there and do they  reproduce in the microlayers?  Aerotaxis and pos-
sibly chemotaxis may play a role in the accumulation.   These factors certainly


                  JO- •
                  M- •-
                  10- •
                     (dynes cm"1)
Figure 2.  Surface pressure, dynes  per centimeter, as a function  of molecular
           area for a monolayer  of  oleic acid on saline solution  with (•—••»)
           and without  (	)  cells of Serratia marinorubra.  Broken lines in-
           dicate viscous  film.   The calculated molecular area  (A2)  is given
           at arrowed horizontal lines (21).
                   30- •
                      (dynes cm"1)
Figure 3.  Surface  pressures,  dynes per centimeter, as  a  function of molecular
           area  for a monolayer of dipalmitoyl-phosphatidylcholine on saline
           solution with and without cells of Acholeplasma  laidlawii A.  Same
           symbols  as in Figure 2 (21).

are more important in model systems where forces like bubble rising, water
movements, and particle transport, which are decisive in nature, do not occur.
In this respect and in many others, the model systems, even in more complica-
ted form than our simple ones, lose the multidimension of a natural ecosystem.
They do, however, offer great advantages in that environmental factors may be
kept constant and individual variables can be studied under controlled condi-
tions.  The results obtained can only increase our understanding of structure
and function of the surface microlayers.

     Natural seawater is an extremely dilute nutrient solution, too dilute for
growth of most marine bacteria.  Since accumulation of organic matter occurs
at all interfaces, accumulation at the surface microlayers sufficient to sus-
tain bacterial reproduction has been assumed.  Aggregates in the surface micro-
layers, as well as subsurface aggregates, can function in the interchange of
metabolites and reduction of exoenzyme escape (19).  Dissolved organic carbon,
in concentrations of .£1.3 mg carbon liter"1 in the subsurface water has been
estimated compared to concentration of 1427 mg liter"1 (2.9 g or organic mat-
ter liter"1) as a mean in the 0.1 ym surface microlayer of the North Atlantic
(32), which corresponds to the amount of nutrients present in media generally
used to grow marine bacteria.  Thus, lack of nutrients may not limit reproduc-
tion.  However, potentially detrimental factors such as intense solar radia-
tion, temperature and salinity flux, high redox potential, and the presence of
toxic organic substances  (32) and heavy metals may present special demands for
survival and reproduction of the microorganisms.  Data now being accumulated
on their biochemical activity point to a low bacterial activity.  Dietz et al.
(12) in their comprehensive data on material from the Vancouver area note that
ATP levels, heterotrophic potential, and heterotrophic activity were lower for
bacterioneuston  (from 70-80 ym depth) than for bacterioplankton, and per CPU,
heterotrophic potential and heterotrophic activity represented only 33 and 10%
respectively, of that for bacterioplankton.  Kjelleberg and Hakansson (20)
found a very clear trend indicating a higher number of biochemically active
bacteria  (lipolytic, proteolytic, and amylolytic activity being tested)  in the
bulk than in the surface microlayers  (20 ym depth) at the Swedish West Coast.
Marumo et al.  (25) , finally, consider the bacteria to accumulate in the sur-
face microlayers by physical forces rather than by reproduction since an ex-
tremely small viable count relative to the total was found; most cells in "the
film were dead or nearly dead."  For easy testing of metabolic state in bac-
teria, it is necessary to find a fluorogenic, real true vital strain, compar-
able to the fluorescein diacetate successfully used for recognizing metaboli-
cally active fungal cells  (33).

     From our work on sediment bacteria, mainly lipolytic, in the deep Norwe-
gian Sea  (27), it occurred to us that conventional substrates do not offer
bacteria, dependent on interfacial enzymes, suitable conditions to display
their biochemical activity.  The same may be valid also for bacterioneuston
bacteria.  There is a special interest concomitant with findings in the sur-
face microlayers, but not in the subsurface water, of bacteria with degrading
potential for compounds which, under prevailing conditions, may be assumed to
accumulate in the surface microlayers  (e.g., NTA-degrading bacteria  [7]).

     The bacterioneuston is poorly characterized, for most often we are in-
formed that marine bacterioneuston is formed by "small, gram negative rods."


This is of minimum value since about 95% of marine bacteria are small, gram
negative rods.  The information "some pigmented" is just as enlightening.
Carty and Colwell (9),  however, classified bacterial isolates from the air/sea
interface obtained during a Pacific Ocean expedition.  They found Pseudomonas,
Vibrio, Aeromonas, and Spirillum species predominent.  The same marine species
were also found in the air above the sea surface otherwise dominated by Bacil-
lus spp.  Even if rare and undescribed species can be expected in this hostile
environment, it seems unimportant to be able to separate the bacterioneuston
into genera and species.  Of more importance is knowledge of its biochemical
activity and means to determine in-situ activity and reproduction rate.

     When collecting microalgae in surface microlayers by drum sampler (60 ym
thick) in freshwater,  Parker and Hatcher  (29)  found the seasonal distribution
of algal genera to be random, with no genus consistently inhabiting either the
surface microlayers or the subsurface water from 10 cm depth.  However, for
individual sampling dates, each genus generally was primarily concentrated
either in the surface or in the subsurface water.  Differences in the organic
and inorganic chemistry of surface microlayers, relative to subsurface water,
may contribute to differences in growth rates and resulting population size.
Knowing the ease with which bacterial enrichment cultures can be obtained,
such differences surely appear also in the bacterioneuston when growth and re-
production on the whole can occur.

     The 1 ym surface microlayer is a heterotrophic microbial community.   Be-
sides bacteria, yeasts and "moulds" have been found (10; Kjelleberg and Sten-
strom, to be published), the latter most probably deriving from an atmospheric
spore fallout.  There are also some heterotrophic microflagellates and all
probably ^serve as food for grazing phagotrophic protists and feed the special
blue-pigmented copepods floating on the surface (11).  Generally we consider
bacteria as decomposers and may overlook their role as food for various con-
sumers.  Sorption to,  or uptake by, bacteria of various persistent compounds
may serve as a means of introducing these toxic compounds into aquatic food
chains  (16, 30), probably of importance in accumulation layers like the sur-
face microlayers.

     Already the size of phytoplanktons surely exclude them from the bacterio-
neuston layer.  Several investigations, however, show the presence of phyto-
plankton in screen microlayer samples  (150-250 ym) .  These investigations have
been based on microscopical work  (e.g., 8) combined with analysis for chloro-
phyll a or phaeopigments  (32).  Estimations of in-situ activity, for example,
expressed as C assimilated per mg photopigment and hour, however, belong to
the future which we hope will add to our still marginal knowledge of the bio-
logical significance of the surface microlayers.
                              LITERATURE CITED

1.  Ahearn, D. G.,  S. A. Crow, and W.  L. Cook.  1977.   Microbial interactions
     with pesticides in estuarine surface slicks.   U.S. Environmental Protec-
     tion Agency Rep. Ecological Res.  Series 600-3.77.050.

2.  Baier, R. E.  1972.  Organic films on natural  waters:  their retrieval,
     identification, and modes of elimination.  J. Geophys. Res. 77:5062-5075


 3.   Baier,  R.  E.   1975.   Applied chemistry at protein interfaces.   Adv.  in
      Chemistry Series 145,  387 pp.

 4.   Baier,  R.  E.,  D.  W.  Goupil, S.  Perlmutter, and R. King.   1974.   Dominant
      chemical  composition of sea-surface films, natural slicks,  and foams.
      J.  Recherches Atmos. 8:571-600.

 5.   Bezdek,  H. F., and A. F. Carlucci.   1972.  Surface concentration of  marine
      bacteria.  Limnol,,  Oceanogr. 17:566-569.

 6.   Blanchard, D.  C,, and L. Syzdek.   1970.  Mechanism for the water-to-air
      transfer  and concentration of bacteria.   Science 170:626-628.

 7.   Bourquin,  A.  W.,  and V. A. Przybyszewski.  1977.  Distribution  of bacteria
      with nitrilotriacetate-degrading potential in an estuarine environment.
      Appl.  Envir.  Microbiol. 34:411-418.

 8.   Brockmann, U.  H., G. Kattner, G.  Hentzschel, K. Wandschneider,  H. D. Junge,
      and H.  Huhnerfuss.   1976.  Natiirliche Oberflachenfilme im Seegebiet vor
      Sylt.   Mar.  Biol. 36:135-146.

 9.   Carty,  C., and R. R. Colwell.  1975.  A microbiological study of air and
      surface water microlayers in the open ocean.  J. Wash.  Acad. Sci.

10.   Crow, S. A.,  D. G. Ahearn, W. L.  Cook, and A. W. Bourquin.  1975.  Densi-
      ties of bacteria and fungi in coastal surface films as determined by
      membrane-absorption procedure.  Limnol.  Oceanogr. 10:602-605.

11.   David,  P.  M.   1965.   The surface fauna of the ocean.  Endeavour 24:95-100.

12.   Dietz,  A.  S.,  L.  J.  Albright, and T. Tuorninen.  1976.  Heterotrophic ac-
      tivities  of bacterioneuston and bacterioplankton.  Can. J.  Microbiol.

13.   Duce, R. A.,  J. G. Quinn, C. E. Olney, S. R. Piotrowicz, B. J.  Ray,  and
      T.  L.  Wade.   1972.   Enrichment of heavy metals and organic compounds in
      the surface microlayer of Narragansett Bay, Rhode Island.  Science

14.   Garrett, W. D.  1965.  Collection of slickforming material from the sea
      surface.   Limnol. Oceanogr. 10:602-605.

15.   Garrett, W. D.  1967.  The organic chemical composition of the  ocean sur-
      face.   Deep Sea Res. 14:221-227.

16.   Grimes, D. J., and S. M. Morrison.   1975.  Bacterial bioconcentration of
      chlorinated insecticides from aqueous systems.  Microb. Ecol.  2:43-59.

17.   Harvey, G. W.   1966.  Microlayer collection from the sea surface:  a new
      method and initial results.  Limnol. Oceanogr. 11:608-613.

18.  Jarvis,  N.  J.   1967.   Absorption of surface-active material at the sea-
      air interface.  Limnol.  Oceanogr.  12:213-221.

19.  Jones, G.  E.,  and H.  W.  Jannasch.   1959.   Aggregates of bacteria in sea
      water as  determined by treatment with surface  active agents.   Limnol.
      Oceanogr.  4:269-276.

20.  Kjelleberg, S., and N. Hakansson.   1977.   Distribution of lipolytic, pro-
      teolytic,  and amylolytic marine bacteria between the lipid film and the
      subsurface water.  Mar.  Biol.  39:103-109.

21.  Kjelleberg, S., B. Norkrans,  H.  Lofgren,  and K.  Larsson.   1976.  Surface
      balance study of the interaction between microorganisms  and lipid mono-
      layer at  the air/water interface.   Appl. Environ. Microbiol.  31:609-611.

22.  Larsson, K.,  G. Odham, and A.  Sodergren.   1974.   On lipid films on the
      sea.  I.  A simple method for sampling and studies of composition.  Mar.
      Chem. 2:49-57.

23.  Maclntyre,  F.   1968.   Bubbles:   a boundary-layer "microtome" for micron-
      thick samples of a liquid surface.  J.  Phys. Chem. 72:589-592.

24.  Maclntyre,  F.   1974.   The top millimeter of the  ocean.  Sci. Amer.

25.  Marumo,  R., N. Tuga,  and T. Nakai.   1971.  Neustonic bacteria  and phyto-
      plankton  in surface microlayers of the equatorial waters.   Bull. Plank-
      ton Soc.  Japan 18:36-41.

26.  Norkrans,  B.,  and F.  Sorensson.   1977.  On the marine lipid surface micro-
      layer—bacterial accumulation in model systems.  Botanica Marina

27.  Norkrans,  B.,  and B.  Stehn.  1978.   Sediment bacteria in  the deep Norwe-
      gian Sea.   Mar. Biol. 47:201-209.

28.  Odham, G.,  B.  Noren,  B.  Norkrans,  A. Sodergren,  and H. LSfgren.  1978.
      Biological and chemical aspects of the aquatic  lipid surface  microlayer.
      Progr.  Chem.  Fats Lipids 16:31-44.

29.  Parker,  B.  C., and R. F.  Hatcher.   1974.   Enrichment of surface fresh-
      water microlayers with algae.   J.  Phycol. 10:185-189.

30.  Paris, D.  F.,  and D.  L.  Lewis.   1976.  Accumulation of methoxychlor by
      microorganisms isolated from aqueous systems.   Bull. Environ. Contam.
      Toxicol.  15:24-32.

31.  Sieburth,  J.  McN.  1965.   Bacteriological samplers for air-water and
      water-sediment interfaces, p.  1064-1068.  In Ocean science and ocean
      engineering,  Trans.  Joint Conf. MTS and ASLO,  Washington,  D.C.

32.  Sieburth, J. McN., P.-J. Willis, K. M. Johnson, C. M. Burney, D.  M.  La-
      voie, K. R. Hinga, D. A. Caron, F. W. French III, P. W. Johnson, and
      P. C. Davis.  1976.  Dissolved organic matter and heterotrophic micro-
      neuston in the surface microlayers of the North Atlantic.   Science

33.  Soderstrom, B. E.  1977.  Vital staining of fungi in pure cultures and in
      soil with fluorescein diacetate.  Soil Biol. Biochem. 9:59-63.

34.  Wangersky, P. J.  1976.  The surface film as a physical environment.
      Ann. Rev. Ecol. Syst. 7:161-176.


              F.  J.  Passman,  T. J. Novitsky  and S. W. Watson

                The  surface microlayer and subsurface water were
           sampled at  7  stations in the Georges Bank region of the
           North Atlantic during August 1977.  Comparison of surface
           microlayer  with subsurface data suggested that bacterial
           and ultraplankton direct counts, phytoplankton biomass,
           heterotrophic and hydrocarbonoclastic viable titers, and
           aliphatic hydrocarbon concentrations were enriched in the
           surface microlayer at several stations.  This was demon-
           strated using the paired-t test (95% confidence level) .
           Glutamate mineralization in the surface microlayers was
           significantly lower than in the subsurface water.  Do-
           decane mineralization rates at 5°C and hydrocarbonoclas-
           tic viable  counts covaried significantly (95% confidence
           level)  with aliphatic hydrocarbon concentrations in sur-
           face microlayers.  In surface microlayer and subsurface
           water,  acetate and glutamate mineralization at 5°C Co-
           varied with acetate and glutamate mineralization at 20°C,
           respectively.  Evidence obtained during one season's samp-
           ling of North Atlantic surface microlayer and subsurface
           waters indicated the enrichment of several microbial pa-
           rameters  in the surface microlayer.  Hydrocarbon-degrading
           activity  was  not enriched.  However, the hydrocarbono-
           clastic viable count and aliphatic hydrocarbon concentra-
           tion were enriched in the surface microlayer.
      Energy Resources Company, Inc., 185 Alewife Brook Parkway, Cambridge, MA
      Woods  Hole  Oceanographic Institution, Woods Hole, MA 02543



     Garrett (5) and Harvey  (6) developed procedures for sampling the surface
microlayers of the sea over a decade ago.  Since that time only a few investi-
gations have addressed the microbial communities of surface microlayers  (2,4,
15) .  In contrast, considerable attention has been drawn to the chemistry and
physics of the sea-air interface  (1, 10, 11, 12, 16, 17).

     Despite the evidence that surface microlayers represent a chemically
unique marine environment, there are few data to suggest that surface micro-
layers support unique microbial populations.  In one study, lipolytic and pro-
teolytic bacteria predominated over amylolytic isolates  (15), while in another
investigation, proteolytic and amylolytic activities were overwhelmingly pre-
ponderant over lipolytic and hydrocarbonolytic activities  (4).

     In the present study, surface microlayer and subsurface samples were col-
lected from the North Atlantic.  Microbiological and hydrocarbon data from
surface microlayer and subsurface waters were compared and interrelationships
among the parameters monitored are discussed.
                            MATERIALS AND METHODS


     Surface microlayer and subsurface samples were obtained at 7 stations in
the Nantucket Shoals, Georges Bank region of the North Atlantic Outer Conti-
nental Shelf  (NAOCS) during the last week of August 1977  (Fig. 1, Table 1).
Sampling was conducted from a rubber boat positioned several hundred yards up-
wind of the mother ship to avoid contamination.  Surface microlayers were col-
lected with a 16-mesh stainless steel sampler  (5).  Subsurface microbiological
samples were collected with a sterile, 2-liter bag sampler  (General Oceanics).
Subsurface samples for hydrocarbon analysis and microbial activities were col-
lected with a 5-liter, Teflon-lined Go-Flo bottle  (General Oceanics).


     The ATP technique of Holm-Hansen and Booth  (8) was used to determine mi-
crobial biomass on a 100 ml aliquant from each sample.  Phytoplankton biomass
was calculated from chlorophyll concentrations which were determined fluoro-
metrically  (9) on 100 ml of each sample.  The lipopolysaccharide  (LPS) assay
of Watson et al.  (19) was used to quantify bacterial biomass in a 10 ml ali-
quant of sample.

     Bacterial and ultraplanktonic direct counts were determined by the fluor-
escence direct count  (FDC) technique  (19).  A 10- and 100-ml aliquant from
each sample was preserved in a final concentration of 0.1%  (w/v) glutaralde-
hyde for bacterial and phytoplankton direct counts, respectively.

Figure 1.  North Atlantic Outer Continental Shelf surface microlayer and sub-
           surface sampling stations,  late August 1977.
Depth (m)
     The membrane filter technique was used to obtain heterotrophic viable
counts.  Dilutions (or volumes)  of seawater estimated to yield 30 to 100 colo-
nies per filter were filtered through 47 mm 0.2 ym porosity Nuclepore mem-
branes.  The membranes were placed on marine agar (Difco) plates and incubated

at 20°C for 2 weeks.  After the original plates were counted, colonies were
replicated onto hydrocarbon agar  (13).  Those appearing on the replica plates
were scored as presumptive hydrocarbon utilizers.

     Mineralization experiments were based upon the procedure of Wright and
Hobbie  (20) as modified by Hobbie and Crawford  (7).  Ten 50-ml aliquots from
each samples were placed in sterile 100 ml serum bottles.  The bottles were
capped with septum caps fitted with Kontes reaction cups.  Whatman No. 1 fil-
ter paper accordions  (1x2 cm) saturated with 0.2 ml i-phenethylamine were in
the Kontes cups.  Four of the bottles were poisoned with 0.1 ml 0.005 M HgCla
solution.  There was one trapping efficiency bottle control and a control for
each nutrient.  The trapping efficiency bottle was incubated with 0.5 ml lkC-
HaHCO3  (specific activity  [S.A.], 0.25 yCi/ml).  Each nutrient was injected
into the control and two test bottles.  Glutamate  and acetate were injected
in 0.5 ml quantities  (S.A. = 0.25 yCi/ml) .  Bottles received 0.1 ml n-d-^C)
dodecane  (S.A. = 0.25 yCi/ml).  The control and one test bottle were incubated
in the dark at 5°C, and the third bottle was incubated at 20°C.  Populations
in glutamate and acetate bottles were killed with  0.5 ml 50% trichloracetic
acid after 4 h incubation.  Populations in dodecane bottles were killed after
2 weeks.  After 1'*CO2 had been trapped, filter papers were transferred to
scintillation vials.  Aquasol scintillation cocktail  (15 ml) was added to each
vial, and  ll*CO2 was quantified by liquid scintillation counting.  Mineraliza-
tion was computed from the fraction of the injected radioactivity that was
trapped in the Jb-phenethylamine-saturated filter paper.


     Hydrocarbon fractions were extracted and separated from seawater samples,
then analyzed by gas chromatography.


     Paired-T tests and correlation coefficients  (r) were computed using pre-
pared statistical programs on a Monroe desk calculator.  Critical values for
T and r were obtained from Rohlf and  Sokal  (14).

     The microbiology data  obtained from the  7  stations  are  summarized in
Tables 2-5.   It  is apparent from Table  2 that bacteria were  approximately
three orders  of  magnitude more abundant than  ultraplankton in the surface mi-
crolayers.  Bacterial and ultraplankton direct  counts and viable counts were
uniform over  the geographic region.  Biomass  data  demonstrated considerably
more variation over Georges Bank and Nantucket  Shoals.   Microbial biomass in
four of the seven surface microlayer samples  was roughly equal to the sum of
the bacterial and phytoplankton biomasses.  This relationship was not consis-
tent, and  in  the subsurface waters  (Table 3), applied to Station 42 alone.

     The hydrocarbonoclastic viable counts  generally represented about 10% of
the heterotrophic viable counts in  both.surface microlayer and subsurface


               AUGUST  1977

Direct count
Bacterial Ultraplankton
(x 1011) (x 109)

(yg C/m3)
Viable count
(CFU/m3, 20°C)

Direct count
Bacterial Ultraplankton
(x 1011) (x 108)

(yg C/m3)

(x 107)

(CFU/m3 ,
(x 107)
Hydro c arbono-
lytic (x 106)

Hydroc arbono-
lytic (x 10 6)

          SHELF SURFACE MICROLAYERS,  IN  LATE AUGUST  1977  (yg C mineralized/m3 per day)a
1.15 x 10"
4.14 x 105



35 x
60 x
35 x
64 x
23 x

10 3

10 3

1.87 x 101
5.50 x 10 3
1.01 x 10"
5.95 x 10"

.78 x
.95 x
.55 x
.51 x
.02 x
.54 x
.65 x




.13 x 102
.01 x 102
.81 x 102
.06 x 102
1.78 x 102
2.62 x 102
1.07 x 102
aND = no data; LTC = less than control.
 Incubation temperature.
          SHELF SUBSURFACE WATERS,  IN LATE AUGUST  1977   (yg  C mineralized/m3  per  day)a
2.03 x 10"
2.45 x 10 3
5.95 x 103
2.34 x 10"
4.48 x 10s

.76 x
.73 x
.58 x
.05 x
.49 x
.45 x

10 3


47 x 10"
57 x 10"
05 x 10"
95 x 103
60 x 10"
25 x 10"

.20 x
.06 x
.66 x
.73 x
.83 x
.80 x
.30 x

3.26 x 102
2.19 x 102
7.96 x 102
3.50 x 102
1.21 x 102
1.23 x 103
aND = no data; LTC = less than control.
^Incubation temperature.

samples.  These estimates are probably an order of magnitude high since the
figures used were from the hydrocarbon replica plates.  Only 10% of the colo-
nies originally isolated from the hydrocarbon replica plates grew in hydrocar-
bon broth or mineralized dodecane (data not shown).

     Surface water temperatures during the late August 1977 cruise ranged from
14°C to 23°C.  During early spring and winter months the temperature ranged
from 4°C to 20°C and 0.2°C to 7°C, respectively.  To determine whether Georges
Bank microbial communities were adapted to colder temperatures, mineralization
rates were determined at 5°C and 20°C.  Overall, mineralization rates for ace-
tate, glutamate, and dodecane in the surface microlayers (Table 4) and subsur-
face waters  (Table 5) were low.  At a number of stations, mineralization rates
in poisoned controls exceeded rates in test bottles.  There were no signifi-
cant differences between 5°C and 20°C mineralization rates for any of the sub-
strates tested.

     The surface microlayer and near-surface data were analyzed to determine
whether any geographic trends could be identified.   Bacterial and ultraplank-
ton direct counts and bacterial biomass were generally above average on the
northwest edge of Georges Bank.  Station 1, south of Nantucket Shoals, had
above average bacterial direct counts, microbial biomass, heterotrophic and
hydrocarbonolytic viable counts, and glutamate and dodecane mineralization
rates in the subsurface sample.  These parameters were not above average in
the surface microlayer sample at that station.  Most of the microbial parame-
ters showed no concentration or activities trends over Georges Bank.

     The primary reason for collecting surface microlayer and subsurface sam-
ples at each station was to determine whether the microbial population in the
microlayer differed significantly from that in the underlying water.  Surface
microlayer and subsurface data from the 7 stations were compared by computing
the ratios for each parameter.  The numbers in Table 6 resulted from dividing
surface microlayer data by subsurface data.  Numbers in boldface type were for
stations where confidence limits for surface microlayer and subsurface data
did not overlap, and where the surface microlayer was enriched.  Only three
parameters, i.e., ultraplankton direct count, phytoplankton biomass, and het-
erotrophic viable count, were enriched significantly in the surface microlayer
at half of the stations or more.  Microbial biomass  and glutamate mineraliza-
tion at 20°C appeared to be enriched in the subsurface waters.

     To substantiate the conclusions inferred from the data in Table 6, the
paired-T statistic was computed for each parameter.   The results appear in
Table 7.  As predicted, ultraplankton direct counts, phytoplankton biomass,
and heterotrophic viable counts were significantly  (p = 0.05)  higher in
Georges Bank surface microlayers than in subsurface  waters.  Glutamate miner-
alization at 20°C was depressed in the surface microlayers.  Differences be-
tween the surface microlayers and subsurface waters  were not significant (p =
0.05) for the other parameters measured.

     Having demonstrated that the surface microlayers were not significantly
enriched over the subsurface waters with respect to most of the parameters
monitored, the next step was to determine whether the geographic distribution
of each parameter differed between the two sample classes.  This was tested by


computing correlation coefficients for each parameter.  As shown in Table 8,
surface microlayer data covaried significantly with subsurface data for all
parameters except bacterial direct counts and microbial biomass.  This
strongly suggests that factors affecting microbial populations in subsurface
waters are the same as those at the surface microlayer.
Bacterial direct count
Ultraplankton direct count
Bacterial biomass
Phytoplankton biomass
Microbial biomass
Heterotrophic viable count
Hydrocarbonolytic viable count
Acetate mineralization:
Glutamate mineralization:
Dodecane mineralization:







aFigures represent ratio of surface microlayer value/subsurface value; >, < is
used where one of the values was less than control, i.e., > if subsurface
value was less than control (LTC); LTC = both values were less than control;
ND = no data.
^Underlined numerals indicate significant  (0.95 confidence level) enrichment
in surface microlayer.

     The 13 parameters used in this study to describe the microbial popula-
tions of Georges Bank surface microlayers and subsurface waters could be di-
vided into four groups of parameters.  These were direct counts, biomasses,
viable counts, and mineralization activities.  Although each group described
a different aspect of the microbial population, it was considered likely that
there were significant relationships among the individual parameters.  This
hypothesis was tested by computing correlation matrices for the surface micro-
layer and subsurface data sets.

     Complete data sets were obtained for 8 of the 13 parameters.  Parameters
for which any station data were missing were omitted from the correlation ma-
trices .  The matrices for the surface microlayers and subsurface waters are
presented in Tables 9 and 10, respectively.


                                     	Mean values5	
            Parameter                  Surface                    Tcrit 0.95
                                     microlayer     Subsurface        1.94

Bacterial direct count (cells/m3)    9.51 x 1011   8.46 x 1011        0.49
Ultraplankton direct count
   (cells/m3)                         1.29 x 109    8.59 x 108         3.11
Bacterial biomass  (LPS-C/m3)            10.0          11.7           -1.13
Phytoplankton biomass (chla-C/m3)       28.4          20.0            2.72
Microbial biomass  (ATP-C/m3)            36.6          72.4           -1.34
Heterotrophic viable count  (CFU/m3)   2.7xl07     1.4xl07         5.23
Hydrocarbonoclastic viable count
Glutamate mineralization at 20°C
(Ug C mineralized/m3 per day)
8.51 x 106

2.96 x 101*
6.82 x 106

5.61 x 104

 *Mean values presented are means for seven stations over Georges Bank.


Bacterial direct count
Ultraplankton direct count
Bacterial biomass
Phytoplankton biomass
Microbial biomass
Heterotrophic viable count
Hydrocarbonoclastic viable count
Glutamate mineralization at 20°C


at p = 0.05
     There were significant  (p = 0.05) correlations between bacterial direct
counts and both types of viable counts in the surface microlayers  (Table 9).
Ultraplankton direct counts also correlated well with viable counts.  The
strong correlation between phytoplankton biomass and hydrocarbonoclastic vi-
able counts may have been due to increased concentrations of hydrocarbons
such as pristine, produced by the phytoplankton.  Much more data would be
needed to demonstrate that this was not coincidental.  Glutamate mineraliza-
tion rates of 20°C were correlated to bacterial biomass, but not to either
bacterial direct counts or viable counts.  It was noteworthy that microbial
biomass was not correlated to any other parameter.



        Parameter                (1)    (2)    (3)    (4)    (5)    (6)    (7)    (8)

Bacterial direct count          1.00
Ultraplankton direct count       —    1.00
Bacterial biomass                —    —   1.00
Phytoplankton biomass            —    —    —    1.00
Microbial biomass                —    —    —    —   1.00
Heterotrophic viable count      0.68   0.67   —    —    —   1.00
Hydrocarbonolytic viable count  0.76   0.56   —    0.88   —   0.89  1.00
Glutamate mineralization at 20°C  —    —   0.60   —    —    —    —   1.00

aOnly correlation coefficients significant at P =  0.05 are shown.
Bacterial direct count
Ultraplankton direct count
Bacterial biomass
Phytoplankton biomass
Microbial biomass
Heterotrophic viable count
Hydrocarbonolytic viable count
Glutamate mineralization at 20°C


(4) (5) (6)

(7) (8)

0.92 1.00
aOnly correlation  coefficients  significant  at  P  =  0.05  are  shown.

     There were more  significant  (p  =  0.05)  relationships in  subsurface waters
than in the surface microlayers (Table 10).  Bacterial,  phytoplankton, and mi-
crobial biomasses; hydrocarbonoclastic viable  counts  and glutamate mineraliza-
tion all  correlated significantly  with bacterial direct counts.  The  correla-
tion between bacterial  direct count  and heterotrophic viable  count was just
below the 95% confidence  level  cutoff.   However, the  0.97 coefficient between
hydrocarbonoclastic and heterotrophic  viable counts suggested that the latter
might be  included  in  the  group  of  parameters covarying  with bacterial direct
counts.   In contrast  to Table 9, subsurface glutamate mineralization  at 20°C
covaried  with bacterial direct  counts  and viable counts, but  not with bacte-
rial biomass.  Bacteria apparently comprised a larger fraction of the subsur-
face microbial population as compared  with  the surface  microlayer since micro-
bial biomass covaried significantly  with bacterial biomass  and bacterial  di-
rect count in the  subsurface waters.


     Bacteria and uItraplankton concentrations in Georges Bank surface micro-
layers and subsurface waters were comparable to concentrations found in other
regions (4, 18).   Viable heterotrophs were significantly more abundant in sur-
face microlayers than in subsurface waters; however, the concentration differ-
ences reported here were not as great as those reported for estuarine samples
(3) .  This may have been due to the differences in sampling methods.

     Crow et al. (3)  collected surface microlayer samples by adsorption onto
47-mm diameter Nuclepore membrane filters.  Each filter adsorbed 5.9 yl of
sample.  The sample thickness calculated from this would have been approxi-
mately 3.5 ym.  The screen sampler collected a layer between 150 and 200 ym
thick, or almost 50 times the thickness of the layer sampled using the mem-
brane technique.  If our surface microlayer data were multiplied by the dilu-
tion factor,  as recommended by Sieburth et al. (16), then the differences be-
tween surface microlayer and subsurface populations over Georges Bank would
be of the same magnitude as those reported for Gulf of Mexico estuaries (3).

     Although there were more viable heterotrophs in the surface microlayers
than in the subsurface water, mineralization rates for acetate, glutamate,
and dodecane were not elevated in surface microlayer samples.  The surface
community was possibly incorporating the substrates at a higher rate than was
the subsurface community.  Unfortunately, only mineralization activity was

     Similar to that of Taguchi and Nakajima (18), our data to some degree
supports the concept that the surface microlayer is a unique habitat in the
marine environment, and deserves more thorough investigation.  In contract to
Sieburth et al. (16)  we did not find the surface microlayer to be enriched
over the subsurface water with respect to every parameter monitored.  The two
regimes were distinguishable for five of the nine parameters, and four of
those five parameters were enriched in the surface microlayers.

     Recently P. Boehm (personal communication) observed that gas chromato-
graphic profiles of aliphatic and aromatic hydrocarbon fractions in surface
microlayers were distinguishable from those in subsurface waters.  We demon-
strated correlations between dodecane mineralization, hydrocarbonoclastic vi-
able counts,  and total aliphatic hydrocarbons in both classes of samples.
However, more comparative bacteriological data are needed to determine whether
the compositional profiles of the complex hydrocarbon mixtures in the surface
microlayers and subsurface waters.

     There were a number of interesting relationships between the two habitats
and among parameters monitored.  Taguchi and Nakajima (18)  suggested that in
estuarine systems the relationships showed seasonality.  Sieburth (15) demon-
strated that functionally, surface microlayer bacterial communities could be
distinguished from subsurface communities.  The surface communities had higher
ratios of lipolytic and peptidolytic isolates than did the subsurface commu-
nity, reflecting the distinctive chemical composition of the surface micro-
layer  (1, 11).

      More extensive, collaborative field investigations will be needed to
 verify the existence and describe the dynamics of microbial communities in
 surface microlayers.


      The authors wish to thank C. Chandler, M. E. Dawson, K. K. McKee, H.
 Quinby, and F. Valois for their technical assistance, and Dr. P. Boehm for
 performing the hydrocarbon analyses.  This research was supported by Contract
 No. AA550-CT6-51, U.S. Department of the Interior, Bureau of Land Management.
                               LITERATURE CITED

 1.  Baier, R. E., D. W. Goupil, S. Perlmutter, and R. King.   1974.   Dominant
      chemical composition of sea-surface films, natural slicks and  foams.
      J. Rech. Atmos. 8:571-600.

 2.  Carlucci, A. F., and P. M. Williams.  1965.  Concentration of bacteria
      from sea water by bubble scavenging.  J. Cons. Perm. Int. Explor.  Mer

 3.  Crow, S. A., D. G. Ahearn, W. L. Cook, and A. W. Bourquin.  1975.   Densi-
      ties of bacteria and fungi in coastal surface films as  determined  by  a
      membrane-adsorption procedure.  Limnol. Oceanogr. 20:644-645.

 4.  Crow^ S. A., D. G. Ahearn, W. L. Cook, and A. W. Bourquin.  1976.   Micro-
      bial populations in coastal surface slicks, p. 93-98.   In J. M.  Sharpley
      and A. M. Kaplan  (eds.) , Proceedings of the Third International Biodegra-
      dation Symposium.  Applied Sciences, London.

 5.  Garrett, W. D.  1965.  Collection of slick-forming materials from the  sea
      surface.  Limnol. Oceanogr. 10:602-605.

 6.  Harvey, G. W.  1966.  Microlayer collection from the sea surface:   a new
      method and initial results.  Limnol. Oceanogr. 11:608-613.

 7.  Hobbie, J. C., and C. C.  Crawford.  1969.  Respiration corrections  for
      bacterial uptake of dissolved organic compounds in natural waters.
      Limnol. Oceanogr. 14:528-532.

 8.  Holm-Hansen, 0., and C. R. Booth.  1966.  The measurement of adenosine
      triphosphate in the ocean and its ecological significance.  Limnol.
      Oceanogr. 11:510-519.

 9.  Holm-Hansen, O., C. J. Lorenzen, R. W. Holmes, and J. D. H. Strickland.
      1965.  Fluorometric determination of chlorophyll.  J.  Cons. Perm.  Int.
      Explor. Mer 30:3-15.

10.  Hunter, K. A., and P. S.  Liss.  1977.  The input of organic material to
      the oceans:  air-sea interactions and the organic chemical composition
      of the sea surface.  Mar. Chem. 5:361-379.


11.  Larsson,  K.,  G.  Odham,  and A.  Sodergren.   1974.  On lipid surface films
      on the sea.   I. A simple method for sampling and studies of composition.
      Mar. Chem.  2:49-57.

12.  Maclntyre,  F.  1974.   The top millimeter  of the coean.  Sci. Amer.

13.  Makula, R.,  and W. R.  Finnerty.   1969.  Microbial assimilation of hydro-
      carbons.  I. Fatty acids derived from normal alkanes.  J. Bacteriol.

14.  Rohlf, F. J., and R.  R. Sokal.  1969.   Statistical tables.  W. H. Free-
      man, San Francisco.

15.  Sieburth, J.  McN.  1971.   Distribution and activity of oceanic bacteria.
      Deep-Sea Res. 18:1111-1121.

16.  Sieburth, J.  McN., P.  J.  Willis,  K.  M.  Johnson, C. M.  Burney, D.  M.  La-
      voie, K. R.  Hinga, D.  A. Caron,  F.  W.  French III, P-  W. Johnson, and
      P. G. Davis.  1976.   Dissolved  organic matter and heterotrophic  micro-
      neuston in the surface microlayers  of the North Atlantic.  Science

17.  Sutcliffe,  W. H., Jr.,  E. R.  Baylor,  and  D.  W. Menzel.  1963.  Sea sur-
      face chemistry and Langmuir  circulation.   Deep-Sea Res. 10:233-243.

18.  Taguchi,  S.,  and K. Nakajima.  1971.   Plankton and seston in the  sea sur-
      face of three inlets  of Japan.   Bull.  Plank.  Soc. Japan 18:20-36.

19.  Watson, S.  W., T. J. Novitsky, H.  Quinby,  and F.  Valois.  1977.   Tech-
      niques for measuring  bacterial  number and biomass in  the marine  environ-
      ment.  Appl. Environ.  Microbiol.  33:940-946.

20.  Wright, R.  T., and J.  E.  Hobbie.   1966.  The use of glucose and acetate
      by bacteria and algae  in aquatic  ecosystems.   Ecology 47:447-464.


                 D. G. Cooper, J. E. Zajic, and D. F. Gerson
                           Biochemical Engineering
                           Faculty of Engineering
                      The University of Western Ontario
                           London, Ontario, Canada

                Interfacial phenomena are extremely important to
           microorganisms metabolizing substrates which are im-
           miscible with water.  This paper reviews examples of
           biosurfactants produced by microorganisms to facilitate
           growth under these conditions.  The most commonly ob-
           served biosurfactants are glycolipids containing reha-
           lose, rhamnose, sophorose or other sugars.  There are
           also examples of lipopeptides, phospholipids, carboxylic
           acids and neutral lipids.  In many cases, microorganisms
           produce mixtures of surface active lipids.
     There are many compartments in the aquatic environment, and the concept
of compartmentalization includes the interfaces which exist in all habitats of
the environment.  Interfaces are of a scale comparable to that of microbes and
possess a defined and active microbial population.  These interfaces are be-
tween either gas-liquid, liquid-liquid, solid-liquid or gas-liquid-solid
phases, and are a focal point for microbial degradation of a wide variety of
insoluble pollutants.

     Interfacial phenomena are complex and the topic is too broad to review
here.  Recent monographs by Marshall (26) , Van Oss et al. (36) and Zviaginsev
(40)  provide considerable detail and background for those being introduced to
the topic.

     Microbes attach to bubbles, liquid droplets and solid surfaces, wherein
problems are encountered relating to substrate solubilization (e.g., oil, sul-
fur,  minerals, synthetics).  Solution of these difficulties often involves
biosynthesis of surface-active molecules which aid in solubilization, emulsi-
fication or wetting of the substrate, or in the attachment of the cell to an
insoluble substrate.

     Since considerable quantities of oil find their way into the aquatic


environment, this paper focuses on the water-oil interface and the biosurfac-
tants produced.by microbes which utilize and degrade oil.

                            WATER-OIL INTERFACES

     Liquid-liquid systems indicate that physically the system is biphasic and
that one of the liquids,  probably containing substrate or nutrients, is insolu-
ble in the dominant phase.  Most of the microbial studies with biphasic liquid
systems center around water and oil.   The mechanisms by which cells take up
hydrocarbon is still not clear; however, some interfacial events that take
place have been elucidated.

     Aiba et al. (1), using Candida guilliermondii grown on n-alkanes, showed
adsorption of oil droplet to cells.  Addition of a surface-active agent gave
improved emulsification but prevented contact, and low growth was observed.
Mimura et al. (29)  firmly established that cells utilizing hydrocarbons pos-
sessed an affinity for them while those not using hydrocarbons did not.  Wang
and Ochoa (37) demonstrated the importance of the interfacial area of the oil
to the specific growth of Candida intermedia on n-hexadecane.  Others have
shown the importance of oil droplet surface area to growth.  Improved surface
area can be obtained by improved agitation, by addition of the proper surfac-
tant, or by natural surfactants synthesized by the microbe.  To demonstrate
the role of the interface in such systems, Yoshida et al. (38) observed the
association of Candida tropicalis cells with hexadecane under various fermen-
tation conditions.   Figure 1 shows Corynebacterium lepus cells imbedded in a
matrix of bioemulsifier (see Gerson et al. [12]).  Candida tropicalis cells
surrounded by droplets of hexadecane were observed by Einsele et al. (9) .   The
size limitation problem for metabolism of hydrocarbons still remains open to
study.  Microbes that adhere to oil droplets grow almost equally as well as
those that accommodate small droplets on their cell surfaces.  Two interfacial
mechanisms appear to be important in hydrocarbon metabolism:  growth on the
hydrocarbon drop (drop-form) and accommodation of hydrocarbon droplets on the
cell surface  (accommodation-form) where the cell adsorbs micro-droplets of oil.
A review of these mechanisms and the interfacial phenomena governing them is
given by Gerson and Zajic (11).

     Bacteria may not utilize exactly the same mechanisms as yeasts in adsorb-
ing hydrocarbons.  Marshall and Cruikshank (27)  have grown Flexibacter auran-
ticus and Hyphomicrobium vulgare in an oil-water system and both cultures tend
to arrange themselves at right angles to the interface.  Examination of the
bulk water phase also shows floe formation.  The floes were of a defined struc-
ture which have been conveniently labelled "rosettes."  The latter are aston-
ishingly similar in appearance, but not scale, to the micelles described for
surface-active agents.

     The microbial affinity for the aqueous or oil phase is similar to another
characteristic of synthetic surface-active agents.  These agents are often
characterized by their hydrophilic-lipophilic balance (HLB), i.e., their ten-
dency to stay in the water phase or bind to the oil phase.  Hydrocarbon-
utilizing microbes may ultimately be analyzed in terms of their HLB character-
istics, which can be determined by measurements of oil-water partition coef-
ficients, contact angles and interfacial free energies.


Figure 1.  Corynebacterium lepus cells imbedded in a matrix of bioemulsifier.
                            SURFACE-ACTIVE AGENTS

     In considering the aquatic environment, both synthetic and biosurfactants
should be discussed.  However, since biosurfactants are conceptually a more
recent topic and give an explanation of how microbes resolve the handling of
certain types of insoluble substrates and liquid-liquid interfaces, attention
is given to microbial surfactants in this section.  Microbial surfactants have
been reviewed by Zajic and Panchal  (39) and Gerson et al.  (12).

     There are a wide variety of extracellular surface-active agents produced
by microorganisms.  The most thoroughly characterized have been the glycolipids
which fall into three major classes, all containing a hydroxy fatty acid co-
valently bonded to a carbohydrate.  These are:  lipids of trehalose and other
carbohydrates with a fatty acid acyl bond; lipids of rhamnose bonded to the
fatty acid by the hydroxyl group; and lipids of sophorose also bonded to the
fatty acid by the hydroxyl group.  Another class of surface-active agents are
lipopeptides, which have a peptide component  (usually less than 10 amino acids)
and a fatty acid component.  Other types of surface-active agents are phospho-
lipids and neutral lipids, the latter including fatty acids, glycerides, alco-
hols and cerides.


     When Arthrobacter paraffineus KY4303 was grown on n-paraffin as the sole
source of carbon, it produced a glyco-lipid with strong emulsifying proper-
ties which was isolated and incompletely characterized.  Each molecule of gly-
colipid contained a disaccharide (trehalose)  and two ft-hydroxy-a-branched
fatty acids esterified at undetermined sites on the sugar.  The fatty acids
had a general formula Rj-CH(OH)-CHR2-COOH, where RI = Ci8 to C23 and R2 = C7
to Ci2.  Glycolipids containing trehalose have been isolated from several
other species of Arthrobacter as well as from Mycobacterium, Brevibacterium,
Corynebacterium and Nocardia when grown on n-paraffins (33,31,16,2).

     It was suggested that the aforementioned ubiquitous glycolipid was pro-
duced by bacteria and used to emulsify the hydrocarbon phase into the aqueous
phase.  This conclusion is supported by several observations.  The glycolipids
isolated from at least two of the bacteria have been shown to have good emul-
sifying properties (34,31).  The addition of penicillin to A. paraffinus
blocked the formation of the glycolipid and depressed the rate of n-paraffin
consumption indicating that the glycolipid is important to hydrocarbon assimi-
lation (33) .  When a species of Nocardia was grown on glycerol, there was only
a trace of a trehalose lipid (31) .  However,  it was produced in good yield
when n-paraffins were the substrate.  Furthermore, additions of glycolipid to
the paraffin media caused a marked enhancement in initial cell growth
(Table 1).

                    % Lipid added   Dry wt (g/1)  at 23 h
aFrom Rapp and Wagner (31).

     An interesting effect on the glycolipids is observed when the appropriate
bacteria were grown on sucrose or fructose instead of n-paraffins (32,16).
The carbohydrate portion, trehalose, of the glycolipid was replaced by sucrose
or fructose.  Thus, when A.  paraffineus KY4303 was grown on sucrose, two gly-
colipids were isolated.   One was identified as 6-0-glucosyl-3-fructoside cory-
nomycolate  (3-hydroxy-a-branched fatty acid), while the other was an ester of
sucrose and two molecules of corynomycolic acid.  When fructose was the sole
carbon source, two different glycolipids were isolated, fructose-6-corynomyco-
late and fructose-l,6-dicorynomycolate.  Similar results were obtained when
species of Corynebacterium,  Nocardia and Mycobacterium were grown on sucrose
or fructose instead of n-paraffins.


     Edwards and Hayashi (8)  completed the identification of  a glycolipid
which had been first isolated by Jarvis and Johnson (17)  from the culture
broth of Pseudomonas aeruginosa grown on glycerol.   This  glycolipid was com-
posed of two molecules of rhamnose and two molecules of g-hydroxy-decanoic
acid and has been given the trivial name of R-l rhamnolipid.  The full struc-
ture is 2-0-a-L-rhamnopyranosyl-a-L-rhamnopyranosyl-B-hydroxydecanoyl-3-
hydroxydecanoate (Fig. 2A).  It was later reported  that when  P. aeruginosa was
grown on hydrocarbon as the sole carbon source, two rhamnolipids were isolated.
One of these was identical to R-l (Fig. 2A).  The other was labelled R-2 and
differed from R-l in that it contained only one molecule  of rhamnose  (i.e.,
a-L-rhamnopyranosyl-B-hydroxydecanoyl-3-hydroxydecanoate  (Fig. 2B).
         HO A   °X0-CH-CH2-C-0-C-CH2-COOH
          OH  OH
       R-I rhamnolipid
  HO  OH   CH3
R-2 rhamnolipid
Figure 2.  Two rhamnolipids isolated from growth of Pseudomonas aeruginosa on
           hydrocarbon as the sole carbon source.

     Histasuka et al.  (14) proposed that the rhamnolipids were produced by
Pseudomonas aeruginosa as emulsifiers to disperse potential hydrocarbon sub-
strated into the aqueous phase.  Rhamnolipid R-l was added in varying concen-
trations to P. aeruginosa S7B1 growing on 1% hexadecane.   As concentration of
the  lipid was increased up to 0.005%, a dramatic increase occurred in the rate
of growth of the bacterium.  As the concentration of lipid was increased above
0.005% there was no further enhancement of the growth rate.  This stimulation
of synthetic surfactants  (e.g., Noigen EA141 or Tween 20).  Furthermore, nei-
ther the rhamnolipid nor the synthetic surfactants enhanced the rate of growth
of P- aeruginosa on glucose.  Thus, it was postulated that the rhamnolipid
acted as an emulsifier, making the hydrocarbon substrate more readily avail-
able for utilization by the bacterium.

     The surface activity and emulsifying power of the rhamnolipid were con-
sistent with the above proposal.   It caused a marked decrease in the surface
tension of water (minimum 34 dynes/cm).   Furthermore, the critical micelle
concentration of about 0.005% corresponded to the concentration which resulted
in the maximum growth stimulation of P.  aeruginosa on hexadecane.  The emul-
sion stability of hydrocarbon-water mixtures containing rhamnolipid was com-
parable to the synthetic surfactants.

     The effect of rhamnolipid on the growth rate of other hydrocarbon-
utilizing bacteria was also tested.  All of the strains of P. aeruginosa tested
showed a marked enhancement of growth.   However, addition of rhamnolipid did
not affect the growth rate of several species of Corynebacterium, Achromobac-
ter, Micrococcus or Bacillus.  Species  specificity was also observed by Itoth
and Suzuki (16), who isolated a mutant of P. aeruginosa (strain PU-1) which
could not metabolize hydrocarbons and did not produce either the R-l or R-2
rhamnolipids.  The addition of either R-l or R-2 to the culture resulted in
growth of PU-1 on hydrocarbon (Table 2).  The addition of trehalose lipids,
which enhances the growth of Nocardia (31) and Corynebacterium (34) on hydro-
carbons, resulted in only relatively poor growth of P. aeruginosa on hydro-

          PU-1 ON HYDROCARBONS (from Itoth and Suzuki [16])
                     Biosurfactant added    O.D.  at 660 my
                          (yg/ml)             (after 50 h)
R-la (10)
R-l (50)
R-2b (10)
R-2 (50)
Trehalose lipid (100)
Trehalose lipid (100)
 Isolated from Nocardium sp.   ^Isolated from Corynebacterium sp.

     When Torulopsis magnoliae is grown on glucose (2.0% w/v)  media, it pro-
duces substantial quantities of a water-insoluble oil (5% v/v) (13) .  Two
major glycolipids were isolated from this oil, both of which contain the di-
saccharide sophorose and a hydroxy fatty acid (Fig. 3A,B).   However, unlike
the 3-hydroxy fatty acids usually observed in glycolipids,  the hydroxyl group
is remote from the carboxylic function.  The two types of fatty acids found,
17-L-hydroxyoctadecanoic acid and 17-L-hydroxy-octadec-9-enoic acid, were both
covalently bonded to sophorose through the hydroxyl function.

                                           a Sophorose Lipid
              lactone of a Sophorose Lipid
Figure 3.  Two major glycolipids  isolated from a water-insoluble oil  produced
           from growth of  Torulopsis  magnoliae on glucose  media.

     The yield of sophorose  lipids  from T. magnoliae was increased  several
fold by reducing the glucose concentration to  10% and  supplementing the media
with various esters of hydrocarbons  (35).  Furthermore, the  nature  of the sup-
plement influenced the distribution of  hydroxy fatty acids in  the sophorose
lipids (Table 3).  When esters were added to the media, usually  the major
lipid found contained the  hydrolyzed  fatty acid with a hydroxyl  function on
the penultimate or terminal  carbon.   However,  there was a  limit  to  this ef-
fect, and most of the fatty  acids found were limited to 16-19  carbon  atoms.

     A study of the sophorose lipids  produced  by T. gropengiesseri  with alkane
and 1-alkene supplements suggested a  metabolic pathway for hydrocarbon incor-
poration (20).  Alkanes are  initially oxidized to terminal alcohols followed
by dehydration to 1-alkenes.  Terminal  alkenes are converted to  fatty acids
which then undergo oxidation to a hydroxy acid.   Rapid incorporation  of the
hydroxy acid into a sophorose lipid protects it from further oxidation.  T.
magnoliae also produces small amounts of crystalline diacetyl  lactones similar
to the oily sophorose lipids but with the fatty acid esterified  to  the sophor-
ose (Fig. 3B) (35).  A study of T. gropengiesseri, which produces larger
.amounts of these cyclic glycolipids,  showed that their production was induced
by hydrocarbon supplements (18).



C17 C17 C18
16-OH 17-OH 17-OH
87 13
24 9 6
54 17 5
9C = C

9C = C



C19 C20 C21
18-OH 19-OH 20-OH Others



8 3
14 4


     Lipopeptides have been isolated  from a wide variety of bacteria and
yeasts,  but only a few have been thoroughly characterized.  The best known ex-
ample is a crystalline compound produced by Bacillus subtilis and which has
been given two trivial names, surfactin or subtilysin.  This contains a pep-
tide chain of seven amino acids in a  cyclic compound with a $-hydroxy fatty
acid (Fig. 4) "(21).  The compound shows strong surface activity (3,6),. and a
0.005% aqueous solution has a surface tension of only 28 dynes/cm,  compared
with surface tension of pure water of 72 dynes/cm at 20°C.  This lipopeptide
has several properties which have been attributed to this surface activity.
It causes lysis of the membranes of mammalian erythrocytes, protoplasts and
spheroplasts of several bacteria, and also inhibits blood clotting.
                      Surfactin  or  Subtilysin
Figure  4.  Crystalline compound produced by Bacillus subtilis.

     An emulsifying factor has been isolated from Candida petrophilum on hydro-
carbon (15).  This compound has not been completely characterized but contains
glutamic acid, aspartic acid, alanine and leucine, .and an unidentified fatty
acid.  When C. petrophilum was grown on glucose the lipopeptide was not found.
If this emulsifying factor were added to C. petrophilum growing on hydrocar-
bons, the rate of growth was enhanced in direct proportion to the amount
added.  The properties of an extracellular surfactant produced by C. lipolytica
were studied by Gerson et al. (10).

     Several other lipopeptides have been isolated from microbia.l broths.
These include lipids containing 3-hydroxy fatty acids isolated from Serratia
marsescens, Pseudomonas viscoas, Bacillus mesentericus and Nocardia asteroides
(4,7).  Less common are lipopeptides containing a simple fatty acid, such as
that isolated from Mycobacterium paratuberculosis (23).


     There are relatively few examples of appreciable amounts of phospholipids
being isolated from cell-free microbial broths.  During growth of Thiobacillus
thiooxidans and concomitant elemental sulphur oxidation, Jones and Benson (19)
noticed a drop in the surface tension of the broth which they attributed to
the phospholipids which could be isolated from the broth.  Several different
phospholipids were found  (Table 4), but the major components were phosphatidyl
glycerol and phosphatidic acid.  It was later shown that each of the isolated
phospholipids had the ability to wet elemental sulphur  (5).

       TABLE 4.  PHOSPHOLIPIDS ISOLATED FROM Thiobacillus thiooxidansa
                                      % of Total    %  of  Total
                                       cellular   extracellular
                Phosphatidyl glycerol      75           49
                Phosphatidyl choline       25           17
                Phosphatidyl ethanolamine   3
                Phosphatidyl inositol      --            1
                Diphosphatidyl glycerol    —            5
                Phosphatidic acid          --           29
      Jones and Benson  (19) .
     When Corynebacterium alkanolyticum was grown on hexadecane, it produced a
small amount of extracellular phospholipids  (1.9 mg/dl).  However, the addi-
tion of 50 yg/ml penicillin to the culture greatly increased the amount of
extracellular phospholipid to 49.4 mg/dl.  The phospholipids produced by C.
alkanolyticum were also found to be sensitive to the hydrocarbon substrate
(22).  When grown on hexadecane, the constituent fatty acids of the phospho-
lipids were mainly C14, C16, CIS and C20.  However, when pentadecane was the
substrate, the fatty acids were mainly C13, CIS, C17 and C19.  Extracellular
emulsifiers were also isolated from Corynebacterium sp., strain PPS-II, by
Panchall and Zajic  (30).


                               NEUTRAL LIPIDS

     Originally the surface-active agents produced by Thiobacillus thiooxidans
were thought to be only phospholipids (19).   A later study showed the presence
of neutral lipids in the extracellular broth (5).   These included hydrocarbons
(22%), triglycerides (4%), fatty acids (4%), hydroxy fatty acids (30%), mono-
glycerides (3%) and others (31%).  Like the  phospholipids, these all had the
ability to wet sulphur and may aid in bacterial attachment to sulphur surfaces,

     Extracellular neutral lipids are often  induced by using hydrocarbons as
a substrate.  A species of Acinetobacter produced very little extracellular
lipid when grown on nutrient broth/yeast extract but yielded a substantial
amount of neutral lipid on hexadecane (Table 5 [25]).  Similarly, the produc-
tion of fatty acids by Micrococcus cerificans is enhanced by using hydrocarbon
media (Table 6 [24]).

TABLE 5.  EXTRACELLULAR LIPIDS PRODUCED BY Acinetobacter SP. (ymol/1)  (from
          Makula et al. [25])
                  .   .,  ,            Nutrient broth/   „   ,
                 Lipid class              .             Hexadecane
                                      yeast extract
Mono- and diglycerides
Fatty acids
Fatty alcohols
Wax esters
TABLE 6.  FATTY ACIDS PRODUCED BY Micrococcus cerificans (from Makula and
          Finnerty  [24])
                                  Total Fatty Acid Produced
                   Medium used    Cellular    Extracellular
                                  (ymol/g)       (ymol/1)

                   Acetate           9.3            5.0
                   Hexadecane        8.2           60.5

                   Heptadecane       8.4           49.0
                   1-Hexadecane      8.6           25.8


     Larger microbial cells such as yeast accommodate small oil droplets which
attach to the cell surface.  In bacteria this process is often reversed and
the bacteria may adsorb to the surface of oil droplets and grow in a fashion
referred to as "drop form."  Attachment of cells to oil, or oil to cells, de-
pends upon the hydrophilic-lipophilic characteristics of the cell surface and
whether either biosurfactants are synthesized or surfactants are added to give
oil emulsification.

     Some cells align themselves at interfaces.  Rosette formation of cells is
observed in the aqueous medium which resembles micelle formation.  Cultures of
Flexibacter and Hyphomicrobium readily show rosette formation.

     A wide variety of surfactants are synthesized by microbes.  The glyco-
lipids fall into three classes, i.e., lipids of trehalose, lipids of rhamnose,
and lipids of sophorose.  Another class of biosurfactants are the lipopeptides
in which the peptide component usually contains 10 amino acids or less.  Other
types of surface-active agents are phospholipids and neutral lipids, the lat-
ter including fatty acids, glycerides, alcohols and cerides.  Microbial sur-
factants offer a new mechanism and a new horizon in our understanding of mi-
crobial interfaces.

                              LITERATURE CITED

1.  Aiba, S., V. Moritz, J. Someya and K. L. Haung.  1969.  Cultivation of
     yeast cells by using n-alkanes as the sole carbon source.  J. Perm.
     Technol. 47:203-210.

2.  Anderson, R. J., and M. S. Newman.  1933.  The chemistry of the lipids of
     the Tubercle bacillus.  J. Biol. Chem. 101:499-504.

3.  Arima, K., A. Kakinuma, and G. Tamura.  1968.  Surfactin, a crystalline
     peptide-lipid surfactant produced by Bacillus subtilis:  isolation, char-
     acterization and its inhbition of fibrin clot formation.  Biochem.
     Biophys. Res. Commun. 31:488-494.

4.  Asselineau, J.  1966.  The bacterial lipids.  Hermann, Paris.  372 pp.

5.  Beebe, J. L., and W. W. Umbreit.  1971.  Extracellular lipid of Thiobacil-
     lus thiooxidans.  J. Bacteriol. 108:612-614.

6.  Bernheimer, A. W., and L. S. Avigad.  1970.  Nature and properties of a
     cytolytic agent produced by Bacillus subtilis.  J. Gen. Microbiol.

7.  Cason, J., and T. S. Prout.  1948.  Concerning the structure of phthioic
     acid.  J. Am. Chem. Soc. 70:879-880.

8.  Edwards, J. R., and J. A. Hagashi.  1965.  Structure of a rhamnolipid from
     Pseudomonas aeruginosa.  Arch. Biochem. Biophys. 3:415-421.

 9.   Einsele,  A.,  H.  Schneider,  and A.  Fiechter.   1975.   Characterization of
      micro-emulsions in a hydrocarbon  fermentation by electronmicroscopy.
      J.  Perm. Technol.  53:241-243.

10.   Gerson,  D. F., J.  E.  Zajic,  and A.  Margaritis.  1976.   Microbial separa-
      tion of bitumen from Athabasca tar sand.  Univ.  Western Ont.  Press.
      241 pp.

11.   Gerson,  D. F., and J. E.  Zajic. 1978.   Surfactant production from hydro-
      carbons by Corynebacterium lepus  sp.  nov. and Pseudomonas asphaltenicus
      sp. nov.  Dev.  Ind.  Microbiol. 19:577-599.

12.   Gerson,  D. F., J.  E.  Zajic,  D.  G.  Cooper,  and M.  D.  Ouchi.  1978.   Micro-
      bial separation of bitumen from Athabasca tar sand,  Part II.   Univ.
      Western Ont. Press.   316 pp.

13.   Gorin, P. A.  J. , F. T. Spencer, and A.  P.  Tullock.   1961.   Hydroxy fatty
      acid glycosides of sophorose  from Torulopsis magnoliae.   Can.  J.  Chem.

14.   Hisatsuka, K., T.  Nakahara,  N.  Sano,  and K.  Yamada.   1971.  Formation  of
      rhamnolipid by  Pseudomonas aeruginosa  and its function in hydrocarbon
      fermentation.   Agr.  Biol.  Chem. 35:686-692.

15.   Iguchi,  T., I. Takeda, and  Ohsawa.   1969.  Emulsifying factor  of hydro-
      carbon produced by a hydrocarbon-assimilating yeast.   Agr.  Biol.  Chem.

16.   Itoth, S., and T.  Suzuki.  1972.   Effect of  rhamnolipid on growth  of
      Pseudomonas aeruginosa mutant deficient in  n-paraffin-utilizing ability.
      Agric.  Biol. Chem. 36:2233-2235.

17.   Jarvis,  T. G., and J. J.  Johnson.   1949.   A  glyco-lipid produced by
      Pseudomonas aeruginosa.  J. Am. Chem.  Soc.  71:4124-4126.

18.   Jones, D. F.   1967.  Novel  macrocyclic  glycolipids  from Torulopsis gro-
      pengaseri.  J.  Chem. Soc.,  p.  479-484.

19.   Jones, G. G., and A.  A. Benson. 1965.   Phosphatidyl  glycerol  in Thio-
      bacillus thiooxidans. J.  Bacteriol.  89:260-261.

20.   Jones, D. F., and R.  Howe.   1968.   Microbiological  oxidation of long-
      chain aliphatic compounds.  Part  I:   Alkanes and alk-1-enes.   J.  Chem.
      Soc., p. 2801-2808.

21.   Kakinuma, A., M. Hori, M. Isono, D.  Tanura,  and K. Arima.   1969.  Deter-
      mination of amino acid sequence in surfactin, a  crystalline peptide-
      lipid surfactant produced  by  Bacillus  subtilis.  Agr.  Biol. Chem

22.  Kikuchi, M., T. Kanamaru, and G. Nakao.  1973.  Relation between the
      extracellular accumulation of L-glutamic acid and the excretion of phos-
      pholipids by penicillin-treated Corynebacterium alkanolyticum.  Agr.
      Biol.' Chem. 37:2405-2408.

23.  Laneelle, G., and J. Asselineau.  1962.  Isolement de peptido-lipides a
      partir de Mycobacterium paratuberculosis.  Biochim. Biophys.  Acta

24.  Makula, R. A., and W. R. Finnerty.  1972.  Microbial assimilation of hy-
      drocarbon:  cellular distribution of fatty acids.  J. Bacteriol.

25.  Makula, R. A., P. J. Lockwood, and W. R. Finnerty.  1975.  Comparative
      analysis of the lipids of Acinetobacter species grown on hexadecane.
      J. Bacteriol. 121:250-258.

26.  Marshall, K. C.  1976.  Interfaces in microbial ecology.  Harvard Univ.
      Press.  165 pp.

27.  Marshall, K. C., and R. H. Cruickshank.  1973.  Cell surface hydropho-
      bicity and the orientation of certain bacteria at interfaces.  Arch.
      Mikrobiol. 91:29-40.

28.  Marshall, K. C., R. Stout, and R. Mitchell.  1971.  Mechanisms of initial
      events in the sorption of marine bacteria to surfaces.  J. Gen.
      Microbiol. 68:337-348.

29.  Mimura, A., S. Watanabe, and I. Takeda.  1971.  Biochemical engineering
      analysis of hydrocarbon fermentation.  J. Ferm. Technol. 49:255-271.

30.  Panchal, C., and J. E. Zajic.  1978.  Isolation of emulsifying agents
      from a species of Corynebacterium.  Dev. Ind. Microbiol. 19:569-576.

31.  Rapp, P., and F- Wagner.  1976.  Formation of a trehalose lipid by Nocar-
      dia sp. grown on n-alkane.  6th Internat. Ferm. Sym.  (Berlin), p. 63.

32.  Suzuki, T., K. Tanaka, and S. Itoth.  1974.  Sucrose lipids of Arthro-
      bacteria, Corynebacteria and Nocardia grown on sucrose.  Agr. Biol.
      Chem.  38:557-563.

33.  Suzuki, T., K. Tanaka, and S. Kinoshita.  1969.  The extracellular accu-
      mulation of trehalose and glucose by bacteria grown on n-alkanes.  Agr.
      Biol.  Chem. 33:190-195.

34.  Suzuki, T., K. Tanaka, I. Matsubara, and S. Kinoshita.  1969.  Trehalose
      lipid  and a-branched-g-hydroxy fatty acid formed by a bacteria grown on
      n-alkanes.  Agr. Biol. Chem. 33:1619-1627.

35.  Tulloch, A. P., J. F. T. Spencer, and P. A. J. Gorin.   1962.   The fermen-
      tation of long-chain compounds by Torulopsis magnoliae.  Can. J. Chem.

36.  van Oss, C. J. ,  C.  F.  Gillman,  and A.  W.  Neumann.   1975.  Phagocytic en-
      gulfment and cell adhesiveness.   Dekker,  New York.  160 pp.

37.  Wang, D. I. C.,  and A.  Ochoa.   1972.   Measurements on the interfacial
      areas of hydrocarbon  in yeast  fermentations and relationship to specific
      growth rates.  Biotech. Bioeng.  14:345-360.

38.  Yoshida, F., T.  Yamane, and H.  Yagi.   1971.   Mechanism of uptake of liquid
      hydrocarbons by microorganisms.   Biotech. Bioeng. 8:215-228.

39.  Zajic, J. E., and C. Panchal.   1976.   Bioemulsifiers.   CRC Grit. Rev.
      Microbiol. 5:39-66.

40.  Zviaginsev, D. Z.   1973.  The interaction  of microorganisms with solid
      surfaces.  University  Press, Moscow.   175 pp.



PASSMAN:  Dr. Poole, Jorgensen has been doing some work using radiolabelled
sulfate, sulfide, etc., in polychromide tubes which she is sticking into sedi-
ment.  Have you used this method, and can this data from a fairly inexpensive
radiolabelled study be extrapolated to biodegradation, in general?

POOLE:  Oh, yes, I'm a great believer in using labelled sulfate.  She devel-
oped a rather cheap method for looking at sulfate reduction.  At the Univer-
sity, we had no money, no fume hoods, we just couldn't do the distillation to
separate the labelled sulfate and sulfide, but you can make it acid, then
standardize the pressures and remove a gas sample.  If you inject labelled sul-
fide into toluene at -7°C, you can therefore count it in a toluene cocktail
cheaply.  We can do 35 samples in half a day with the method.  Using the stan-
dard method, we do three.

PRITCHARD:  Are aromatic rings likely to be degraded in the anaerobic environ-
ment or more likely to be funneled back out into the aerobic water layer?

POOLE:  I don't know.  I think maybe a biochemist could better answer your
question.  But aromatics can be degraded in anaerobic environments.

RAYMOND:  You said that sediments in cold areas may not be very active.  Do
you have some evidence on the effect of temperature on anaerobic biodegrada-

POOLE:  Yes, we have some evidence that says during our winters with tempera-
tures of 0-4°C in the sediments, there is very little biodegradation occurring.

RAYMOND:  Do you suppose this is due to H2S because normally H2S is not more
toxic at colder temperatures?

POOLE:  No, actually it is more toxic at warmer temperatures.

DAGLEY:  I would like to comment on the degradation of aromatic compounds in
anaerobic environments.  The subject has been reviewed extensively by W. C.
Evans in Nature (London), Vol. 270, pp. 17-22.

TIEDJE:  Dr. Lichtenstein, I'm concerned with what you are following, maybe a
lot of these products are standard metabolic intermediates which are no longer
of concern in the toxicology sense.

LICHTENSTEIN:  This could be possible.   With a * **C label you can follow the
fate of the compound.  I stated that we should be concerned about the parent


compound and the biologically active metabolites.   You first have to determine
what is present, then which of the compounds is biologically active.  If you
obtain soluble   C compounds in water,  it is probably no longer insecticidal
to mosquito larvae.  Toxic to mosquito  larvae does not mean that the compounds
are also toxic to humans.   The biological activity of the benzene-soluble com-
pounds toward insects is relatively easy to determine.  I agree, we have to
determine what the materials are and if they are biologically active.  We are
restricted, of course, because we can neither conduct tests with human beings
nor with each of the 1,400,000 species  of insects.

MATSUMURA:  I wasn't too sure what you  meant when you said the microorganisms
were responsible for this increased binding.  Do you mean that aminoparathion
was formed by microorganisms and thereby increasing binding?

LIECHTENSTEIN:  Aminoparathion is not found in sterilized soil, nor do we en-
counter binding phenomena.  Based on data with sterilized soil, reductive pro-
cesses in soil are dependent on microorganisms.  This, however, is not true
for all compounds but appears to be so  with compounds containing nitro-groups.
We repeated these tests with Dyfonate,  another organophosphorous compound,
utilizing irradiation and re-inoculation.  The binding of Dyfonate is not re-
lated to the activity of microorganisms.

MATSUMURA:  Have you ever changed the pH and attempted to re-extract the amino-

LICHTENSTEIN:  Not in these experiments, where we purposely kept the soil
within its natural state.   We have tried to extract bound residues at differ-
ent pH but we were unsuccessful.  In this particular situation, we tried to
keep the soil as natural as possible.  We made six different extractions with
various combinations of solvents.  Usually we were able to extract 90% with
the first three extractions.

LIU:  How do you know that the bound material in the soil is one compound or
two compounds or what?  We have done some experiments with   C pesticides in
soils.  We fractionated the soil or organic matter, following the standard
procedure, into six different fractions—a humus,  6 humus, etc.  After a few
weeks incubation, we extracted the major portion of the radioactivity from one
of these six fractions.  The bound material was a single compound which appar-
ently chemically or physically attached to that soil organic matter.  However,
after 6 months incubation, the radioactivity was evenly distributed in the six
different fractions.  This is evidence, that gradually these radioactive com-
pounds are being incorporated into the  soil organic matter.  It is no longer
one metabolite.  I'm suggesting the soil organic matter takes up the radio-
activity and then goes to the oats.  If you analyze the oats you will find
this radioactivity is in amino acids, sugars, and other biological compounds.

LICHTENSTEIN:  I am not sure if you asked me a question or if you wanted to
add something.  I think, though, that it would be a big contribution if one
could repeat this experiment and concentrate on the oats.  In a routine ana-
lytical laboratory one usually does not extract soil more than three times.
All we have shown is that you don't extract everything and that a sizable por-
tion of the  ^C was unextractable.  Even in the case of parathion, we don't
have direct proof of what is bound.  We have indirect proof that only amino

compounds are being bound and this process involves microorganisms.  With
other organophpsphorous materials, residues are also bound but microorganisms
do not appear to be involved.

BOLLAG:  The problem of bound residues was discussed intensively in a sympo-
sium in Colorado.  Unquestionably, it is one of the major problems in pesti-
cide research today and, in general, when dealing with xenobiotic compounds.
They may be bound in soil, but little is known as to how this occurs.  I be-
lieve that microbial activities are very important in the binding of parathion.
It is well known that amino compounds are quickly bound in soil.  Some time
ago we did an experiment where you added amino compounds to soil for 1/2 hour;
when we tried to extract them, we found they were bound.  I do not think, in
this case, it is a biological binding but perhaps a chemical or physical

LIECHTENSTEIN:  Once you have the amino compound.

BOLLAG:  Yes.  This is true if you have an amino compound and also if you have
a phenol which may be formed.  In my opinion, it is the activity of the micro-
organism which forms the intermediates, analinic compounds and phenolic com-
pounds, and these may then be bound by physical or chemical factors within
soils.  It may be the clays or organic matter, etc., but that is usually diffi-
cult to determine.  It is also possible to imagine binding through enzymes in
soil which are active in formations of analines and phenolics.

GARNAS:  Did you ever run a sterile control and put one of the reduction pro-
ducts into it?

LICHTENSTEIN:  Yes, we added parathion to sterile soil, but no binding was ob-
served under these conditions.

PRITCHARD:  In relation to Dick Garnas' question, did you try sterilizing soil
by gamma radiation in which you had bound pesticides and did you grow the oats
on that sterilized soil?

LICHTENSTEIN:  No, we didn't do that.

PRITCHARD:  What you seem to infer is that the oats are responsible for bring-
ing the aminoparathion out of the bound state, where, in fact, it might be the
microbial populations which are responsible.

LICHTENSTEIN:  This could be.  We conducted studies with crop plants contain-
ing residues which had grown in soil treated with insecticide.  Bound residues
were found in some of the crop plants.  In recent studies conducted at the
Biotron under different climatic conditions, we found that binding is to some
extent dependent on environmental conditions.

CHESTERS:  I've got a general question that I would like to address to the
panel.  We have been talking about several unqiue microenvironments.  It is
pretty clear these unique zones in the aquatic environment appear to effect
biodegradation.  We talked about adsorption and sediment surfaces, we talked
about surface microlayers, and we talked about anoxic environments.  In  light
of the purpose of this conference, which is to suggest methodologies or


strategies with a development of a protocol for examining toxic organic chemi-
cals in the environment,  would you, in reality, suggest that these unique
zones be taken into account when devising the strategy?  Should any of these
unique microenvironments  be included in that testing protocol?   And, if so,
how would you do it?

POOLE:  Are we saying that the word unique suggests something special?  We are
talking about the marine  environment, one part of the environment is surface
layer, another part the anoxic, etc.  I fail to see how you cannot take these
areas into account.  If you don't take them into account you are going to fail.

CHESTERS:  How are you going to account for them?

POOLE:  That's where I duck.  Taking anoxic situations, I think you have to
rely on microcosms.  There are people here much more qualified in microcosms
than I am.  That is the easiest way to start studying the problem, then per-
haps you want to go back into the environment for further study.  You've got
to relate the two.

CLESCERI:  Dr. Poole, regarding your comment on the microcosm, I think some-
thing should be added.  The microcosms originate from the so-called unique en-
vironment.  Right?  I mean, this is not something like your pet microcosm
that's been in your laboratory for the past few years.  I think this is an ex-
tremely important feature.

POOLE:  Yes.  The whole idea is to make it as realistic as possible.  You are
completely right.  You've got to be as realistic as possible.  You must allow
for seasonal effects.

LIECHTENSTEIN:  I have a feeling that the word microcosm could get five differ-
ent definitions from five people.  We talked about microcosms for a whole week
last year in Corvallis.  Basically, people are trying to simulate what goes on
in nature.  Whatever we do, however, one has to realize that only restrictive
answers to certain problems will be obtained.  I cannot see that we will ever
come up with a microcosm which would answer the whole multitude of questions
we are facing.  We will be rather limited, even if we should be able to come
up with a good microcosm.

CHESTERS:  It seems to me that we are talking about a lot of compounds.  If we
are going to make assessments of hazardous compounds, and they are going to be
realistic, in the sense of those that are truly hazardous being declared haz-
ardous, then our testing methodologies must not be so sensitive that every-
thing becomes hazardous.   We can't be talking about things that are going to
cost more than the defense budget.  We can't be talking about doing things
that are going to require more personnel than are available in the history of
the world.  And how are we going to boil some of these unique facets that
we've talked about here into some consensus opinion of how to proceed.  Lich-
tenstein says we need more in the EPA budget.

CLESCERI:  Dr. Passman, with respect to your last comment, if one of the pol-
lutants has a very low density and accumulates at the surface, clearly then
the biology of that surface is going to reflect that.  Right?

PASSMAN:  For example?

CLESCERI:  I was thinking of oil.

PASSMAN:  Well, oil has a gaseous fraction that is going to volatilize fairly
quickly.  A considerable portion of the non-volatile fractions are photo-
oxidized in a fairly short order, and much of the remaining materials sink
fairly quickly.  Some of the work our colleagues have been doing on the Argo-
Merchant spill and on several of the other New England spills show that oil
disappears pretty quickly from the surface and either winds up in the sedi-
ments or is diluted throughout the water column.  Two months after the Argo-
Merchant spill, we could not identify Argo-Merchant oil in the sediment within
200 meters of the wreck.

CLESCERI:  These time frames are long for that period.  So clearly within
those weeks or months that you are speaking of there is going to be a differ-
ence in the biological composition of that layer.

PASSMAN:  Oh, true, from an esoteric perspective, I'd say the surface micro-
layers offer some fascinating areas of investigation.  What we are addressing
here is the question of whether EPA should be concerned with the microbial
degradation of pollutants in the surface zone or other unique microenviron-
ments.  Again, we are talking about cost effectiveness of a certain type of
study.  When you address this question, as Dr. Chesters also mentioned, you've
got to decide what needs to be studied and set priorities.  From both our own
data and that in the literature, I don't think there is sufficient rationali-
zation for putting a high priority on studying the role of degradation in the
surface microlayer, per se.  I think the residence time in sediment is going
to be much greater, and you should get more fruitful returns from investigat-
ing that area.

CLESCERI:  Some of this is dependent upon what the ultimate test procedures
are going to be.

PASSMAN:  That's true.  Again, we are talking about dollars and public health/
public safety.

COLWELL:  There really isn't any mystery in the association of hydrocarbono-
clastic bacteria with surface film and finding a correlation with occurrence
of phytoplankton.  Plankton can contain fatty acids and various kinds of lipid-
like materials, including aliphatics, etc.  Such correlations have been re-
corded, particularly in marshes at the end of the fall season, where release
of material occurs with hydrocarbons of biogenic origin increasingly entering
the systems.  I find it interesting that you make the statement that there is
no need to worry about surface films because this is, in fact, an example of a
microcosm, where mixtures of pollutants can have serious effects.  For example,
heavy metal and pesticide will concentrate in the surface film, with 3 or 4-
hundred-thousand-fold concentration in the film, creating a significant effect
on the bacteria in the film, as well as with their capability of dealing with
whatever the pollutant may be.

PASSMAN:  While I agree with you, the question becomes, what is the residence
time of these materials in the film?  How long will they continue to persist
in the film before the film is broken, and they are either diluted into the
water column or trapped in other micro habitats?

COLWELL:  Well, that was one of the reasons we used a very large "microlayer,"
in fact a "surface layer," as was pointed out by the earlier speaker.  We had
to determine whether the bacteria in that mixed layer were the same in our
study as in the air above the water.  It is important to determine whether, in
the turbulent surface layer, you are dealing with the same microorganisms.

FLOODGATE:  There are two comments I would like to add, if I may.  First of
all, on the stability of that surface layer, it is much more stable than
people imagine.  It isn't something present only during flat calm seas.  When
you have a lot of breaking waves, the surface layer is reformed almost instan-
taneously.  It is a very stable layer, and doesn't mix up nearly as much as
you might imagine.  Mixing occurs only in a few microns, perhaps tens of mi-
crons, on the surface.  The next comment I would like to make is on this word
"unique," which has been flying around with careless abandon.  I think one
must recall that Winogradsky, who was a better microbiologist than any of us
here, said that he could isolate any bug he wanted from the sole of his boots.
So you can find any bug you want from anywhere in the sea.   So, it is true
that you can find certain microorganisms in the surface layers and the same
ones immediately below it.  But that doesn't mean to say you shouldn't look in
the surface layer to find out what is going on there.

DiGERONIMO:  I have no working experience in the marine environment, but in
the surface layer that you said should be discounted, would these .microorga-
nisms found there originate from the surface layer or do they originate from
sediment?  Do you think the populations found in the surface contribute to the
subsurface layer?

PASSMAN:  Your question is, are the microorganisms in surface films from aero-
sols, aerosols coming from special sources, or are they from resuspended sedi-

DiGERONIMO:  You explained there was no difference between the subsurface
layer and the surface layer.  I was wondering if the lack of difference was
because either the surface layer is being formed by the subsurface layer or
the surface layer is contributing to the subsurface layer.   I was wondering
which is important.  Can you comment on that?

PASSMAN:  I can comment.  We haven't looked at our taxonomic data,  so I won't
address whether the speciation differences are due to organisms coming from
one source or another.  Some of the work done by Carlucci in bubbling type ex-
periments indicates that organisms originally suspended in the water will be
differentially layered onto the surface due to size and affinity for organic
constituents of the surface microlayer.  So that says the original source was
the water column, and they were selected out onto the surface layer because of
bubbling or because of transport mechanisms.  In the microlayer itself, a 0.1
micron layer, few or none of the organisms there are of the dimension where
they can be totally contained by that monolayer.  We are sampling the screen


layer, 150-300 microns, and we know what their distribution is in that layer.
This question could be addressed by using different types of sampling methods.
For instance, I believe Al Bourquin's Nuclepore membrane method picked up a
layer of 3.7 microns?

BOURQUIN:  That's correct, depending upon the slick or film, of course.

NORKRANS:  We have gotten one-micron thick samples using the Nuclepore mem-
brane .

PASSMAN:  What types of counts do you get on them?  Are they of the same mag-
nitude that we see on the screen?  We get approximately 106-107 bacteria/ml.
Using the Nuclepore membrane, we got approximately the same magnitude of cells.
So chances are, then, it's safe to say the cells are existing right near that
monolayer, probably protruding through the monolayer or existing directly below
the monolayer, but you get approximately the same number of cells in the one-
micron sample as you do in the 120-125 micron sample.  Where do they come
from?  There are several sources.

DiGERONIMO:  Have you investigated the areas for total activity?  Is there a
difference at this microlayer vs the sublayer?  Is there more activity at the
microlayer, or below it?

PASSMAN:  We found the glutamate activity was depressed in the microlayer as
compared with the subsurface water from approximately one meter.  This is from
one season of study, and I wouldn't want to make any pervasive statements sug-
gesting this is how it is over the Georges Banks.  I would expect to find some
seasonality; I need a broader data base before saying specifically what's hap-
pening.  During the late summer, however, we found glutamate mineralization
depressed.  We didn't look at incorporation.  This may have been enriched in
the surface microfilm.  There were a number of limitations with the study.

NORKRANS:  Did you make specific determinations of biomass?

PASSMAN:  Not in that study.  We did isolate several yeasts and fungi, and
greater than 70% of the hydrocarbonolytic bacteria isolated grew on a complex
mixture of approximately 20 aliphatic and aromatic hydrocarbons.  The minerali-
zation was strictly with l-lltC n-dodecane.  So we are studying just one n-
alkane as opposed to a variety of alkane and aromatic hydrocarbons in miner-
alization.  We were looking only at mineralization, not incorporation.  I
think there has been some work that indicates that cometabolism may be an im-
portant parameter.  Studying the secondary metabolites might be the route for
looking at hydrocarbon metabolism in natural systems.

BOLLAG:  Are fungi included in total biomass?



    Chairperson, F. Matsumura

                      FATE STUDIES IN MICROCOSMS+

             P-  H.  Pritchard,  A.  W.  Bourquin,  H. L. Frederickson
                               and T. Maziarz
                    U.S.  Environmental  Protection Agency
                Environmental  Research  Laboratory/Gulf Breeze
                            Gulf  Breeze,  FL 32561

                Two microcosms used in environmental  fate  studies are
           described.   One is a static system which utilizes a sedi-
           ment/water  core and the other is a continuous-flow system
           using a structured sediment/water growth vessel with con-
           tinuous addition of seawater.  The effects of design char-
           acteristics of both systems on the fate of methyl para-
           thion (MPS) was studied.  Sediment/water cores  taken
           directly from the environment were generally slower to
           degrade MPS than cores "structured" with sediment and
           water in the laboratory.  Degradation rates were slower
           when sediment to water ratios were increased  (water de-
           creased) in either type core.  Laboratory-aged  cores were
           less reactive than "fresh" cores when  COa and degrada-
           tion products of C^-MPS was measured.
                Continuous-flow microcosms which were acclimated to
           methyl parathion over a 50-day period were more active in
           removal from the water column and metabolism of MPS than
           aged systems not exposed to MPS for the preceding 25 days.
                Total  radioactivity measurements in the water of
           continuous-flow systems showed a greater removal from
           active systems than sterile ones.  This is attributed pri-
           marily to sorption to sediments.  Product  analysis showed
           removal of  MPS with production of aminomethyl parathion
           and   COz,  whereas the only product in the control system
           was p-nitro phenol.

                This paper supports the contention that design fea-
           tures will  partly determine the outcome of a fate experi-
           ment.  Intact sediments, aged systems, acclimated
     tcontribution no.      from the Gulf Breeze Environmental  Research  Labora-
tory .

           systems and size of the microcoms were shown here to
           affect the data from two different microcosms.

     The setting of water quality criteria for pesticides and toxic substances
requires the establishment of guidelines for the registration of these chemi-
cals.  Guidelines for aquatic environments require development of protocol in
two major areas:  toxicology, and fate and transport.  Interaction between
these two areas has become increasingly complex due to decreased use of very
persistent polyhalogenated organic chemicals and an increased use of a variety
of new chemicals which have a finite lifetime in the environment.  Laboratory
toxicological studies must now be continually scrutinized and evaluated ac-
cording to the information generated on fate and transport.  Since microbial
degradation is one of the most important aspects of fate and transport, it is
critical that good, representative methods be available to determine the bio-
degradation potential of the environment in question.

     We recognize three key issues that influence an aquatic degradation study:
A) Due to the specialized nature of aquatic environments, terrestrial fate
studies are not directly applicable.  B) Methodology for studying the fate of
chemicals in aquatic environments is seriously lacking.  C) Field studies are
more difficult to perform in aquatic environments than in terrestrial systems.

     Consequently, our research has centered on the development of microcosm
or model ecosystem technology to generate information on degradation processes
in estuarine salt marsh environments.  For simplicity,  a microcosm is defined
as a containerized portion of an aquatic environment maintained in the labora-
tory under controlled conditions.  It is designed to simulate,  but not neces-
sarily duplicate, natural environmental situations.  In the case of an estuar-
ine system, characteristics such as sediment-water interfaces,  high dilution
capacity, and salinity changes should be modeled into the microcosm study.

     It is also important to design systems for easy adaptation or assimilation
into other research programs, particularly those that provide a data base for
development of registration guidelines.  The systems should be simple, repro-
ducible, and inexpensive.  Before establishing a system a priori, system de-
sign characteristics should be varied and tested to determine their effect
relative to an established degradation process and to reveal possible defi-
ciencies of the microcosm (9,4,22).  The purpose of this paper is to describe
two of our microcosm systems, one previously described by Bourquin et al. (5),
and to indicate features that should be studied in detail before predictive
capability can be realized.

                            MATERIALS AND METHODS


     Two different microcosms have been developed:  one, referred to as "eco-
core," uses cores of water and sediment; the second, referred to as a con-
tinuous-flow microcosm, uses a structured water/sediment system into which


seawater and pesticide are fed continuously.

     Eco-core microcosms  (Fig. 1) were  set up by obtaining  an undisturbed  (in-
tact) core of sediment  (50 g), water  (200 ml),  and  a  sterile glass  tube  (3.5 x
30 cm).  Cores were taken from a  local  pristine salt  marsh  (10-17 ppt  salinity,
24-28°C); the sediment cores  were approximately 95% sand  (by weight) with  a
3-5 mm layer of fluffy, high  organic  content  (15-20%)  detritus on top.   Cores
were maintained at constant temperature (24°C±2).   Aeration gave maximum mix-
ing in the water column without resuspending the sediments.

     For each eco-core experiment,  the  system was  "spiked"  with both radio-
labelled and cold pesticides  to give  a  final water  concentration of 0.5  mg/
liter.  The distribution of parent  compound and degradation products in  the
water and sediment was then followed  with  time. Water overlying the sediments
was sampled  (5 ml) and analyzed on  a  periodic basis;  after  a designated  incu-
bation period, the microcosm  was  dismantled and the sediments analyzed for
parent compound, degradation  products,  and bound residues.  Sterile controls
 (2% formalin) were run simultaneously.   Air exiting each  system was passed
through polystyrene bead  resin  (Amberlite  XAD-4) to monitor for volatile pro-
ducts, and radiolabelled CC>2  was  trapped in aqueous sodium  hydroxide  (2N) .
Figure  1.   Schematic diagram of eco-core microcosm.
     The  continuous-flow microcosm (Fig. 2)  consisted of a reactor vessel into
which  water (350 ml) and sediment (100 g)  collected from the same.salt marsh
were placed.   After the sediment had settled (^2 h), a 3-4 cm layer of detri-
tus settled over an equally thick layer of sand.  Both fresh filtered seawater

and sterile artificial seawater containing  the radio-labelled pesticide were
metered into the reactor vessel at  equal  rates.   Turnover time for the con-
tents of the reactor vessel was 25  h.  Concentration of pesticide entering the
vessel was 20 ug/liter; salinity was between 15  and 17 ppt.  The system was
continuously aerated for mixing and oxygenation  (dissolved oxygen at sediment-
water interface was 20% of saturation); an  undisturbed sediment-water inter-
face was maintained.  Air exiting the  system was monitored as described above.
At periodic intervals, water samples  (20  ml)  and sediment cores (3 ml) were
analyzed for parent compound and degradation products.  Sterile controls were
obtained by adding formalin  (2% v/v, final  concentration)  to the original
water-sediment mixture followed by  the continuous addition of formalized (0.1%)
artificial seawater.  This was sufficient to eliminate viable bacteria in the
water in the growth vessel even with the  continuous addition of non-sterile


                                                           XAD- RESIN
         / TUBING"^
             TUBING  SAMPLE
                                   t t
                                 vf-f .V'r^i-rV?:
                                                                      C02 TRAPS
                                      • DETRITUS (5.0 cm)

                                      •SAND (3.5cm)
Figure 2.  Schematic diagram of continuous-flow microcosm.


     C-14 methyl parathion  (2, 6 ring-labelled, specific  activity = 30 Ci/mole,
Amersham-serale Corp.) was the pesticide for these  studies.   It was the sole
source of carbon for each microcosm except for the  carrier  (acetone),  which
was added at a final concentration of 1 mg/liter  for  eco-core microcosms and
40 yg/liter for the continuous flow microcosm.  All other chemicals were nano-
grade purity.


     Figure  3 gives  a  schematic of the analytical methodology employed.  All
water samples were extracted with methylene chloride, twice at ambient pH
 (7.6) and twice under  acidified (pH 2.0 with concentrated HC1) conditions.
All sediment samples were washed twice by centrifuging and resuspending with
water and then extracted four times with acetonitrile.  Acetonitrile was back-
extracted with methylene chloride and the latter sample checked for products.
Radioactivity in  all samples was measured with PCS™ scintillation cocktail
 (Amersham Corp.,  Arlington Heights, 111.).
                           ANALYTICAL SCHEME, MICROCOSMS
                                                      	••[ TOTAL 14C
                                                         TOTAL "C



                                                            TOTAL '«C
                                                            TOTAL '«C
Figure  3.   Analytical scheme for analysis of ll*C-pesticide fate in microcosms.

      Methylene chloride extracts were chromatographed on silica gel thin  layer
plates  by  an etherrhexane (3:1, v:v) solvent system, and spots visualized by
autoradiography in a spark chamber  (Birchover Inst., Ltd., Letahworth,  Herts, .
England).   Degradation products were verified by co-chromatographing  standards
on a gas chromatograph equipped with a flame photometric detector by  UV spec-
trometry and by mass spectrometry.

     Radiolabelled carbon dioxide in alkaline  traps  was determined by scintil-
lation counting of a 0.5 ml sample in Harvey's scintillation cocktail (Harvey
Instrument Co.) .  All combustions were done with a Harvey Instrument Combus-
tion Chamber at 900°C.



     Chemical analyses of water overlying the  sediment in an eco-core micro-
cosm experiment are shown in Figure 4.   Total  radioactivity in the water  (un-
extracted water) of both the experimental and  the sterile controls dropped
rapidly during the first 2-3 d and then  leveled off.  This initial drop was
probably due to diffusion of the methyl  parathion (MPS) into the sediments as
there was no evidence of adsorption to glass.   Water in the sterile control
always reached a constant concentration  at a value lower than the experimental.
Total 14C levels in the sterile control  remained constant with continued  incu-
bation, while those in the experimental  systems decreased at a slow rate.
This steady decrease appeared to be caused by  increasing adsorption to sedi-
ments (mediated by some biological component)  since  significant amounts of
carbon were not lost from the system through degradation (see below).  Dupli-
cate experimental cores  (dotted lines) gave excellent replication.


                                 IUNEXTRACTED WATER]
••'-'-i* •*;^~	••	• Sterile control
                                                  -gj Experimental
                                                    Sterile control
                        8    12   16    20   24    28   32    36   40
                                    TIME (days)
Figure  4.   Radioactivity in the water column above sediment of an  eco-core
            microcosm spiked with methyl parathion.  Unextracted water  repre-
            sents  total radioactivity in water sample  (dotted lines are results
            from duplicate cores).   Extracted water represents radioactivity
            remaining in water sample after solvent extraction  (see Materials
            and Methods).   Sterile control contained 2% formalin.

     The radioactivity remaining in the water after solvent extraction (ex-
tracted water or polar products) increased steadily, above sterile control
values, from day zero.  This was the result of slow metabolism of the hydroly-
sis products of methyl parathion (MPS) to more polar unextractable materials.
The presence of unextractable radioactivity in water from sterile controls was
due to the incomplete extraction of paranitrophenol (see below) by methylene
chloride.  Unextractable products in experimental systems increased at a rate
approximately equal to the loss of total radioactivity in the water column.
All eco-core microcosms without sediment  (water only)  to date were incapable
of generating polar products above sterile control levels or of inducing loss
of total radioactivity from the water column.  This indicates that components
of the sediments may mediate the production of the polar unextractable pro-
ducts through the initial attack on MPS.

     The fate of parent compound and the production of breakdown products in
sterile and nonsterile microcosms are shown in Figures 5 and 6.  MPS disap-
peared rapidly in the nonsterile systems; by day 12 only 0.1% of the original
MPS remained.  During the first 10 d of incubation, 80-90% of the loss was ac-
counted for by the products produced in the water column, i.e., paraaminophe-
nol  (PAP), aminomethyl parathion (AMPS), polar unextractable products, and
C02.  Thereafter, the products accounted for only 50-70% of the loss of methyl
parathion.  The remaining 30-50% was bound to sediment (see below).
                                     16    20    24
                                    TIME (days)
Figure  5.  Products  identified by thin-layer chromatography  in methylene  chlor-
           ide  extractable material  from water above  the  sediment  in a  non-
           sterile eco-core microcosm  spiked with methyl  parathion (0.5 mg/
           liter).   MPS = methyl parathion, AMPS =  aminomethyl parathion,
           PAP  =  p-aminophenol.  Products  are expressed as dpm/ml  of water
           column.     CO2 is  total produced per collection period  averaged per
           number of days in  that period.



i \j



\ STEf
^••^•. 	 •
^**^- __

0 4 8 12 16 20 24 28



32 36
                                   TIME (days)
Figure 6.  Products identified by thin-layer chromatography in methylene chlor-
           extractable material from water above the sediment in a sterile eco-
           core microcosm spiked with methyl parathion (0.5 mg/liter).
           MPS = methyl parathion, AMPS = aminomethyl parathion, PAP = p-
           aminophenol.  Products are expressed as dpm/ml of water column.
           11*C02 is total produced averaged per days in collection period.
     Only low levels of paranitrophenol (PNP)  were produced in nonsterile
cores; its absence may have reflected rapid metabolism or incorporation into
microbial biomass.  The production of PAP, in fact, could be from PNP rather
than from AMPS.  In other tests (see below), cores were shown to be capable of
PNP metabolism after extended incubation periods.

     PNP was the only product seen in sterile controls (the dimethoxy-thiophos-
phate hydrolysis product was not monitored in these studies), indicating sig-
nificant hydrolysis of MPS by nonbiological processes.  In all instances, the
decrease of MPS in the sterile controls was accounted for by production of PNP.
We have detected no enhanced hydrolysis due to the presence of formalin.

     The production of ll*C-C02 (see Fig. 5) always commenced between days 4
and 10, peaking shortly thereafter.  The initial lag may have resulted from a
preferential metabolism of the acetone carrier.  With continued incubation,
production of COa leveled off and eventually  (day 28-32)  began to decrease
slowly.  In most of the cores where methyl parathion was the carbon source,
10-15% of the parent compound was converted to CO2.

     Table 1 gives the material balance for an eco-core that incubated for 32
days.  Most of the radioactivity in the sediments was associated with the de-
tritus fraction and not with the sand fraction  (low organic  [1-2%], non-


suspendable particulate).  In the nonsterile systems, 90% of the radioactivity
associated with the detritus was tightly bound  (not extractable with aceto-
nitrile) , and the remainder was loosely bound  (extractable with acetonitrile).
In sterile eco-cores the opposite was true:  only a small amount of radio-
activity was associated with the detritus fraction.  Sand had little affinity
for the radioactivity in either system.  Overall, about the same total amount
of radioactivity was found in the sediment, including interstitial water, re-
gardless of whether it was sterile or nonsterile.  However, in the sterile
systems, 75% was found in the interstitial water; whereas, less than 40% was
found in the interstitial water in nonsterile eco-cores.  Degradation products
found in interstitial water and associated with sediment  (solvent extractable
fraction) were the same as those found in the water overlying the sediment.
             Type of sample
Percentage of total 1'*C
  Sterile     Nonsterile
Carbon dioxide
Overlying water
Extracted water
Extractable products*3
Interstitial water
Core wall washes
Resin trap
Percent recovery








aSpiked with  llfC-MPS  (0.5 mg/liter) , analyzed after  32 d of incubation.

Percentage figures are percent of  total  11+C in overlying water.

     Several other experiments were carried out to determine the effects of
microcosm design factors with the previous information as background for the
typical fate of methyl parathion in these eco-cores.  The necessity of using
an intact sediment-water core was determined by assembling microcosms in two
different ways:  by the normal coring procedure  (referred to as "intact" micro-
cosms) or with a sediment-water mixture in the glass  columns in which

structuring occurred and allowed the mixture to settle out (referred to as
''mixed").  Each set was spiked with 14C MPS and the degradation processes moni-
tored with time as described above.

     Table 2 shows a partial summary of the results from the two types of eco-
core microcosms.  At the end of each incubation period (8 and 15 d) an entire
microcosm was dismantled and analyzed.   Activity was generally higher in the
mixed systems.  Total radioactivity in the water overlying the sediment was
consistently lower in the mixed systems; this was complemented by increased
levels of radioactivity bound to the sediments.  Of the other compartments
analyzed, only slight differences were noted; C02 production was also slightly
higher in the mixed systems.  Table 3 shows the results from product analyses
of the mixed and intact eco-core experiments.  No differences were observed in
the types of products produced.  Since CO2 levels were lower in the intact
core, a slower degradative process may be occurring resulting in the accumula-
tion of intermediates.
                                     Percentage of total

Extracted water
Day 8

Day 15

Day 8 Day 15


 Total activity:  10 yCi (^C)  in 0.5 mg/liter MPS per core.

     The differences of intact versus mixed eco-cores were determined further
by monitoring CO2 production from MPS over a 16-day incubation period when the
water to sediment ratios (v/v)  were varied.  Results are shown in Figure 7.
Mixed cores showed the greatest C02 production regardless of the water to sedi-
ment ratio.  However, if water above the sediment (sediment volume maintained
at approximately 50 ml)  was reduced from 200 ml (5:1 water to sediment ratio)
to 50 ml (1:1), C02 production decreased.  Thus,  water to sediment ratios af-
fect C02 production and merit consideration in the design of microcosms.

     During one of these experiments (see Fig. 7), lifc ring-labelled p-nitro-
phenol (0.1 mg/liter) was added to the cores after 16 d of incubation with
  C-MPS.   An immediate increase in C02 production occurred, indicating that
the microcosm system was acclimated to PNP metabolism while metabolizing MPS.
Product analysis showed PNP disappearance was coupled with increases in polar
degradation products (not shown).   The mixed core systems again showed higher
metabolic activity (CO2  production) after PNP supplementation.


 Radioactivity  (dpm x  10  /ml)
MPS   PNP   AMPS   PAP   Base










152 68

437 80
19 91
1 4


 ^Missing  values,  radioactivity was not significantly above background,
 Figure 7.   Cumulative 1HCO2 production from eco-core microcosms spiked with
            0.5 -mg/liter methyl parathion  (MPS) as a function of sediment  in-
            tact versus mixed, and as a function of sediment to water  ratios
            (v/v) ,  1:1 and 5:1.  p-Nitrophenol  (llfC) was  added at  day  17 and
            ll*CO2 from both PNP and MPS was observed thereafter.


     Aging or acclimation to laboratory conditions  is  an  important considera-
tion in microcosm experimentation.  Accordingly, two types  of intact eco-core
systems were set up:  one set was spiked with   V-MPS  in  previous  experiments
within 2 h after obtaining the fresh cores, the other  set was spiked after the
cores had aged  (with continuous aeration) for 7 d in the  laboratory at 25°C.
Removal of MPS from the water above the sediment  (Fig.  8) was slower in the
aged microcosm, particularly over the first 4 days.  Similar  responses were
apparent in the production of polar unextractable degradation products (not
shown).  Aminomethyl parathion and carbon dioxide production  were  also greater
in the unaged or fresh system.  In both instances  (aged and unaged), degrada-
tion products, including COa, accounted for 85-95%  of  the loss of  methyl para-
thion in the water, indicating that very little bound  to  sediments within the
16-d incubation period.  The small amount of binding that did occur (5-10%)
was the same regardless of aging.  Overall, however, an unaged or  fresh eco-
core showed higher degradative activity.


•	• Unacclimated
o	o Acclimated
                                                     /  cumulative
                                                    /.carbon dioxide
                                   8    10     12    14    16
                                     TIME (days)
 Figure  8.   Changes  in  levels  of methyl  parathion,  and aminomethyl parathion in
            overlying water and carbon dioxide  in eco-core microcosms with and
            without  laboratory aging  (see  text).

      With these microcosms,  the  pesticide  enters  continuously.  Since it takes
 time  for the  pesticide  to diffuse  into  the sediment,  a gradient will be estab-
 lished  in the sediment.   Consequently,  the concentration of pesticide in the

water overlying  the  sediment will increase with time,  eventually equalling the
influent concentration when equilibrium conditions result.   In a sterile sys-
tem where biologically mediated sorption to sediment or biodegradation will
not occur,  faster  equilibrium should be attained.   Changes  in total radioac-
tivity in water  overlying sediment for sterile and nonsterile continuous flow
microcosms  are seen  in Figure 9.   Radioactivity in the water increased
steadily in the  sterile microcosm, eventually leveling off  around 11.8 x 103
dpm/3 ml.   Influent  radioactivity was 12.1 x 103 cpm/3 ml,  indicating the
sterile control  had  reached equilibrium.  In the nonsterile microcosm, the
radioactivity in the water leveled off at values considerably lower than those
observed in the  sterile control.   As demonstrated below,  degradation contribu-
ted very little  to loss of radioactivity from the water column.   Thus, the
leveling off of  the  radioactivity could be due to biologically mediated bind-
ing to sediment.
                      0    100   200   300   400   500   600
                                    TIME (hours)
Figure 9.
Total radioactivity in water above sediment in sterile and nonster-
ile continuous flow microcosm receiving continuous addition of  If*C-
methyl parathion  (50 yg/liter).  Turnover time was 25 h.

     Analysis of the degradation products generated in the water overlying the
 sediments is shown in Figure 10.  In the sterile microcosm, the only degrada-
 tion product was PNP.  The decrease and eventual steady state in MPS concen-
 tration was compensated entirely by production of PNP-  In nonsterile micro-
 cosms, the decrease in MPS concentration was greater and reached a much lower
 steady concentration level.  The production of AMPS accounted for approxi-
 mately 75% of the decrease in MPS; the remaining 25% was lost to polar degra-
 dation products and carbon dioxide.  Only trace amounts of PNP were detected.


— • Sterile
-o Nonsterile
60    °
40   x


                   200    300    400    500   600
                       TIME (hours)
Figure 10.  Products identified by  thin-layer  chromatography  (% of total  radio-
            activity)  in methylene  chloride  extractable material  from water
            above sediment in continuous  flow  microcosm receiving lV methyl
            parathion (50 yg/liter).  MPS =  methyl parathion, AMPS = amino-
            methyl parathion,  and PNP = p-nitrophenol.

     Table 4 shows the distribution of radioactivity in water and sediment of
both sterile and nonsterile microcosms sampled at 425 h.  Radioactivity, unex-
tractable from water  (polar degradation products), was always higher in non-
sterile systerts.  Very little difference in total radioactivity was seen be-
tween water overlying the sediment and water within the sediment  (intersti-
tial) .  As seen with the eco-core microcosms, PNP was the only degradation
product in the sterile controls.  In nonsterile  systems, no PNP was typically
seen in interstitial water, perhaps signifying its metabolism to polar pro-
ducts  (base), the latter also being higher in interstitial water.  Higher
levels of AMPS were also observed in nonsterile  systems.


         Substrate                     Sterile               Nonsterile

  Overlying  (dpm/ml)                 4,058  (19%)b            3,089  (31%)b
  Interstitial  (dpm/ml)              2,020  (4%)              3,725  (34%)

Extractable (dpm/g solids) 72,244
Bound (dpm/g solids) 14,700

distribution in water
Overlying Interstitial
aWater and sediment  sampled at  425  h  from continuous  flow microcosm.

^Percent of  1'IC unextractable from  overlying  water.

     Large quantities of  radioactivity  were bound  (not  solvent extractable) to
sediment material, particularly in  nonsterile systems.  The binding phenomenon
was compound-dependent, since,  in other studies  with  the same microcosms
 (Bourquin and Pritchard,  unpublished),  compounds like Kepone and diflubenzuron
became bound to a much  lesser extent.


     To determine the degree to which the systems  acclimate to the degradation
or transformation or a  particular pollutant,  two types  of continuous  flow  mi-
crocosms were studied:  one had a continuous  exposure to unlabelled MPS  (20
yg/liter) for 50 d,  the other had an  interval exposure  to MPS of 25 d followed
by a 25-d period of  no  exposure.  On  the 50th day,  each system received  14C

MPS continuously from a reservoir and the levels of total radioactivity in the
water overlying the sediments and in the alkaline CO2 traps were monitored.
The results, as depicted in Figure 11,  showed that the microcosms actually
lost their original degradative capacity for MPS when not continuously exposed
to the pesticide.  Release of radiolabelled carbon dioxide in the acclimated
system reached the same levels as observed in earlier experiments (Fig. 10).
In the unacclimated system, however, the lack of exposure to methyl parathion
resulted in lower levels of C02 production.  The fact that C02 production re-
mained in steady state indicated degradation capacity did not increase with
time as might be expected.
                                   6      8       10
Figure 11.  Total radioactivity in water (per 3 ml)  above sediment and J'
            production (per day) from acclimated (continuous exposure to
            methyl parathion for 50 h) and unacclimated (continuous exposure
            to methyl parathion for 25 h followed by no exposure for 25 h) and
            continuous flow microcosm receiving lf*C methyl parathion (20 yg/


     The application of microcosm techniques to studies of degradative pro-
cesses in aquatic environments is relatively untested but holds promise as a
potential laboratory tool to predict the rate and extents of fate processes in
the real world  (4,12,15,1).  Use of radiolabelled materials and sterile con-
trols further enhanced our evaluation of biological and nonbiological degrada-
tion processes; since aquatic environments are open systems with large dilution
capacities, controlled field experiments, as executed in terrestrial environ-
ments (19), are prohibitive.  Microcosms may be the next best mechanism for
extracting information about fate and transport in aquatic habitats.

     Microcosms also are an excellent method for assessing the interactive
processes involved in fate and transport.  Aquatic ecosystems are complex en-
tities with many individual reactions dependent on output and input processes
occurring in the surrounding milieu  (8).  We feel it is not prudent to base a
study on fate of a chemical strictly on the results obtained from experiments
on individual reactions examined apart  from their more complex origin and
chosen for study because of their amenability to defined laboratory experimen-
tation.  Such is often true with models developed to predict exposure concen-
trations in aquatic environments  (6,2); individual processes like photolysis,
hydrolysis, biodegradation are studied  individually and then summed together
to predict an overall effect.  Laboratory isolation of individual components
of a complex system often necessitates  structuring the experiments so that the
interpretation of results becomes meaningless.  For example, descriptive analy-
ses of biodegradation rates invariably  fall back on Monod kinetics (3); this
choice necessitates:  a) the isolation  of a pure culture that will grow on a
pollutant as its sole source of carbon  and energy; b) the determination of
maximum growth rate  (y max) and substrate saturation constants  (Ks),  and c)
the assumption that biodegradation is primarily a function of microbial growth
under steady-state conditions.  Each of these conditions is artificial to many
"real-life" aquatic situations.  For example, methyl parathion does not readily
degrade in seawater in the absence of sediment  (Pritchard and Bourquin, unpub-
lished) .  When sediment is present, we  have not been able to isolate a single
organisms that was responsible for the  degradation observed.  Likewise, at
very low substrate concentrations, degradation may be only indirectly related
to microbial growth, making y max and Ks determinations meaningless.  Our
studies, in fact, suggest that only small changes in microbial population den-
sity occurred in the continuous flow microcosms because acclimation did not
occur in these systems  (Fig. 11).  Therefore, problems of interpretation arise
when the experimental isolation of individual rate processes is required.

     We envisaged a microcosm study that would monitor inputs and outputs
rather than individual components; thus, interactive forces could be observed
as a total effect.  Our studies clearly showed that the rate and extent of
formation of amino methyl parathion  (AMPS), although documented in the litera-
ture  (21,13,20,18,10), can probably be modeled in a microcosm study if reduced
conditions in sediments can be regulated according to in-situ measurements.
Naturally occurring sediments  (range point) used in our studies typically have
mild reducing conditions  (Eh = +50 to -150) at the sediment-water interface
before collections.  Since these conditions are approximately duplicated in
the microcosms and since we readily detect the production of AMPS in the


microcosm, we would predict the production of AMPS under in-situ conditions.
The formation of AMPS also represents a detoxification mechanism of MPS  (16)
and thus the effect on predicting exposure concentration is considerable.

     Our microcosm studies also emphasize the need to more completely study
the binding of MPS to aquatic sediments, particularly if sorption processes
are to be modeled correctly for exposure concentration evaluation.  It is ap-
parent from our studies that the formation of bound residues (solvent unex-
tractable) is greatly enhanced by biological activity, as exemplified by our
use of sterilized microcosms.  Katan et al. (13)  and others have described
similar binding phenomena for soils.  The extent of this binding, its substrate
specificity, and the environmental factors that control it, should be further
studied.  Data from our eco-core experiments (Fig. 5 and 6, Table 1) indicated
that biologically mediated binding increased with continued incubation, i.e.,
loss of MPS from the water column initially was compensated for by the produc-
tion of degradation products, not by binding to sediments.   However, with con-
tinued incubation, more of the loss of MPS was accounted for by sediment-bound
residues.  Since an apparent acclimation was occurring, threshold concentration
limits could exist in which minimal binding occurs if the MPS concentration in
the sediment is below a certain level.

     Katan and Lichtenstein (14) reported that the binding moiety of MPS was
probably the reduced product, AMPS.  Approximately half of the radiolabelled
AMPS added to their loam soil became irreversibly bound within 24 h.  Our re-
sults do not explicitly implicate AMPS as the bound component in aquatic sedi-
ments, but they do indicate that if AMPS is bound, it occurs at a much slower
rate because it accumulates in the water phase (Fig. 5, 10).  Overall, these
results suggest that soil studies  (sorption in particular)  cannot be directly
extrapolated to aquatic sediments.

     Microcosms can be constructed in basically two ways:  first by removing a
portion of the environment with its complex components and studying them under
controlled laboratory conditions or, second, by amassing together a limited
number of individual organisms in some container in an attempt to model the
complexity of a large ecosystem.  We are of the impression that these latter
types of microcosm studies are difficult to interpret, and susceptible to ex-
perimental bias (11,17) when applied to fate and transport studies.  Microcosms
which start with more complex components such as portions of a natural envir-
onment in a laboratory container are ultimately more amenable to studying fate
processes and making predictions.  These systems should have more semblance to
real world situations because they potentially contain many of the interactive
processes occurring in microenvironments.  However, containerization processes
may greatly affect validity of the data generated from any microcosm,  espe-
cially when attempting to extract a simplistic answer from a complex system.
Design features of a microcosm will also impact interpretation of the data
generated therein.  The size of the microcosm,  water to sediment ratio, sedi-
ment surface area, sediment mixing, aeration,  and nutrient regeneration are
all critical factors which must be addressed before a microcosm can be suc-
cessfully used and standardized.

     Our data support the contention that design features will partly deter-
mine the final outcome of a fate study.  With the eco-core microcosm,  an


intact sediment/water core had slower overall biological -activity but larger
amounts of degradation products than a mixed sediment/water sample.  Actual
intact cores probably give a more accurate assessment of the fate of MPS.
These microcosm studies also reveal that degradation processes may be quite
sensitive to perturbations in the sediments, such as natural dispersion and
scouring brought on by different climatic regimes.  A reduction of the water
to sediment volume ratio also reduced degradation activity  (as measured by CO2
production) (Fig. 7).  Aromatic ring cleavage mechanisms of p-nitrophenol, a
precursor to CO2 production, are probably restricted to aerobic habitats and
unlikely to proceed in sediments  (10) .  A reduction in water volume decreases
the oxidizing zone and slows the microbial attack and eventual CO2 release.
Since natural estuarine environments are tidal, fluctuations in water level
are typical, and degradation activities may vary accordingly.

     Acclimation is another factor significantly affecting degradation activi-
ties.  Laboratory acclimation of the eco-core microcosm, a closed system, had
a depressive effect  (significant changes in 7 d) on the processes contained
therein.  Within a closed system, only a finite concentration of microbial
growth factors are available before regenerative processes supply these fac-
tors internally.  We believe biodegradation studied during the regenerative
phase  (long periods of laboratory acclimation) will reflect potential charac-
teristics of the laboratory system and not of the natural environment.  With a
closed system like the eco-core, our work indicates the most informative re-
sults are generated with as fresh a system as possible.

     The continuous-flow microcosms demonstrated that acclimation to degrada-
tion of the pesticide did not increase after an initial incubation period  (see
Fig. 11) , but in fact decreased if the pesticide was not continuously supplied
at low levels.  The results are not mutually exclusive regarding acclimation
in the eco-core, but it is difficult to determine what actually happens in the
natural environment.  Since fresh nutrient, bacterial cells, and invertebrate
animals continuously enter the microcosm (inflowing seawater), some unknown
factor(s) is present in the original sediment that is not maintained unless
methyl parathion is continuously supplied.  We suspect this is related to a
sediment-binding phenomenon because of the difference in equilibrium kinetics
between an acclimated and unacclimated microcosm  (see Fig. 11, radioactivity
in water over sediment).

     Additionally, we feel that carbon dioxide production considered alone in
marine microcosms is a poor indicator of the extent of degradation  (2% of
total carbon added).  These findings contrast sharply with many soil studies
in which COa may represent much more of the total test carbon added  (19).  In
our studies with methyl parathion and other pesticides  (unpublished), it was
apparent that hydrolysis of the parent compound was far more indicative of the
overall degradation process and that product analysis, other than C02 , should
be carried out in conjunction with a microcosm study.

     The merits of an eco-core type microcosm relative to a continuous-flow
microcosm are being defined and investigated.  We developed a core-type micro-
cosm as an aquatic equivalent to similar core-type microcosms developed  for
fate studies in terrestrial environments (7).  These microcosms are simple,
inexpensive, relatively rapid, and amenable to standardization for  test


protocols in registration programs.   They are basically tier-two testing (tier-
one being sewage sludge BOD test and/or batch enrichment studies),  but they are
limited to assessments of degradation potential only.

     Continuous-flow microcosms are significantly more complicated but lend
themselves to assessments of the degradation capacity of aquatic environments,
i.e., the extent a particular aquatic environment will degrade or transform a
pollutant and how fast these processes will occur.  The virtue of assuming
some sort of steady-state kinetics coupled with rate processes that are not
limited by depletion of internal factors make continuous-flow microcosms poten-
tially amenable to assessments of degradation potentials that truly reflect
real life situations.

                              LITERATURE CITED

1.  Ambrosi, D., A. R. Isensee, and J. A. Macchia.  1978.  Distribution of
     oxadiazon and phosalone in an aquatic ecosystem.  J. Agric. Food Chem.

2.  Baughman, G. L., and R. R. Lassiter.  1978.  Prediction of environmental
     pollutant concentration, p. 35-54.  In J. Cairns, K. L. Dickson, and
     A. W. Maki (eds.), Estimating the hazard of chemical substances to aquatic
     life.  Amer. Soc. for Testing and Materials, STP 657, Philadelphia, Pa.
     278 pp.

3.  Bazin, M. J., P. T. Saunders, and J. I, Prosser.  1976.  Models of micro-
     bial interactions in the soil, p. 463-498.  In CRC Critical Reviews in

4.  Bondietti, E. A., S. Draggan, J. R. Trabalka, and J. P. Witherspoon.
     1974.  State-of-the-art and recommended testing for environmental trans-
     port of toxic substances.  U.S. Environmental Protection Agency, Environ-
     mental Sciences Division, Pub. No. 893.  ORNL/EPA-1.  152 pp.

5.  Bourquin, A. W., M. A. Hood, and R. L. Garnas.  1977.  An artificial micro-
     bial ecosystem for determining effects and fate of toxicants in a salt-
     marsh environment.  Dev. Ind. Microbiol. 18:185-191.

6.  Branson, D. R.  1978.  Predicting the fate of chemicals in the aquatic en-
     vironment from laboratory data, p. 55-70.  In J. Cairns, K. L. Dickson,
     and A. W. Maki (eds.), Estimating the hazard of chemical substances to
     aquatic life.  Amer. Soc. for Testing and Materials, STP 657.   Philadel-
     phia, Pa.  278 pp.

7.  Draggan, S.  1976.  The microprobe analysis of 60CO uptake in sand micro-
     cosms.  Xenobiotica 6:557-563.

8.  Floodgate, G. D.  1976.  Decomposition processes in the sea with special
     reference to man-made wastes, p. 223-245.  In 17th Symposium,  The role of
     terrestrial and aquatic organisms in decomposition processes.   Blackwell
     Scientific Publications, London.

 9.  Giddings, J. M., and G. K. Eddlemon.  1977.  The effects of microcosm
      size and substrate type on aquatic microcosm behavior and arsenic trans-
      port.  Arch. Environ. Contam. Toxicol. 6:491-505.

10.  Graetz, D. A., G. Chesters, T. C. Daniel, L. W. Newland, and G.  B. Lee.
      1970.  Parathion degradation in lake sediments.  J.  Water Pollut. Control
      Fed. 2:R76-R94.

11.  Isensee, A. R., E. R. Holden, E. A. Woolson, and G. E. Jones.   1976.  Soil
      persistence and aquatic bioaccumulation potential of hexachlorobenzene
      (HCB).  J. Agric. Food Chem. 24:1210-1214.

12.  Isensee, A. R., P. C. Kearney, E. A. Woolson, G. E. Jones, and V.  P.  Wil-
      liams.  1973.  Distribution of alkyl arsenicals in model ecosystem.
      Environ. Sci. Technol. 7:841-845.

13.  Katan, J., T. W. Fuhremann, and E. P. Lichtenstein.  1976.  Binding of
        C parathion in soil:  a reassessment of pesticide persistence.
      Science 193:891-894.

14.  Katan, J., and E. P- Lichtenstein.  1977.  Mechanism of production of
      soil-bound residues
      Chem. 25:1404-1408.
soil-bound residues of   C parathion by microorganisms.   J.  Agric.  Food
15.  Kearney, P. C., J. E. Oliver, C. S. Helling, A. R. Isensee,  and A.  Kont-
      son.  1977.  Distribution, movement, persistence, and metabolism of N-
      nitrosoatrazine in soils and a model aquatic ecosystem.   J. Agric. Food
      Chem. 25:1177-1181.

16.  Melnikov, N. N.  1971.  Chemistry of pesticides.  Springer-Verlag,  New

17.  Metcalf, R. L., G. K. Sangha, and I. P. Kapoor.  1971.  Model ecosystem
      for the evaluation of pesticide biodegradability and ecological magni-
      fication.  Environ. Sci. Technol. 5:709-713.

18.  Munnecke, D. M., and D. Hsich.  1974.  Microbial decontamination of para-
      thion and p-nitrophenol in aqueous media.  Appl. Microbiol. 28:212-217.

19.  Sanborn, J. R., B. M. Francis, and R. L. Metcalf.  1977.   Degradation of
      selected pesticides in soil:  a review of the literature.  U.S. Environ-
      mental Protection Agency Ecological Research Series, EPA-600/9-77-022,
      Cincinnati, Ohio.  615 pp.

20.  Schulz, K. R., and E. P. Lichtenstein.  1971.  Field studies on the per-
      sistence and movement of dyfonate in soil.  J. Econ. Entomol.

21.  Sethunathan, N.  1973.  Microbial degradation of insecticides in flooded
      soil and in anaerobic cultures.  Residue Rev. 47:143-165.

22.   Taub,  F.  B.   1969.   A  continuous gnotobiotic  ecosystem, p.  101-120.   In
      The structure  and  function of  fresh-water microbial  communities.  Res.
      Monogr.  3,  Virginia Polytechnic Inst.,  Blacksburg, Va.


                             Lenore S. Clesceri
                            Department of Biology
                      Rensselaer Polytechnic Institute
                               Troy, NY 12181

                The use of mathematical modeling as a tool to study
           biodegradation in natural environments is discussed.
           Both suitability and limitations of the approach are  con-
           sidered in the light of other biological systems.   The
           usefulness of simulation in hypothesis testing is  illus-
           trated by manipulating the microbial component of a lake
           ecosystem model.  The effects of the removal of microbial
           mineral mobilization, phytoplankton productivity and
           allochthonous organic loadings on organic matter and
           nutrient pools are demonstrated.
     Perhaps the nature of ecology is the best argument for the role  of  sys-
tems theory in ecological inquiry.  Biology has been defined by Weaver (5) as
"organized complexity" among the natural sciences.  This is never seen so
clearly as in ecosystem studies.  Some general statements can be made about
ecosystems (and microcosms) that indicate their suitability for the method-
ology of mathematical modeling:  a)  they are complex and multivariable;
b) many functional components are imperfectly measured or even unidentified;
and c) multilevel activities are strongly interconnected functioning  to  main-
tain a reasonably steady state overall.  Thus, these systems have properties
similar to biochemical and physiological systems and as such are amenable to
application of organized analytical and modeling techniques.

     An example of the modeling of a biochemical system is Hess1  model  (4) of
glycolysis.   More than 89 reactions among 65 chemicals are described  in  this
model, and good agreement with experimental data, i.e., flux, concentration of
intermediates, etc., is observed; however, it took half a century of  biochemi-
cal investigation to gather the data for this feat.

     In ecosystem studies, there is an additional constraint to applying mathe-
matical models and in particular computer program models.  This is the  innate
uncertainty that the data are reproducible, and that what has been measured is


representative of the system.   Yet ecosystems are not the only systems that
suffer from variability.   Toxicity studies are done with hundreds of mice for
statistical evaluation of the data, since great differences in susceptibility
are seen in laboratory mice.

     In our example of glycolysis, we assume that the homeostatic apparatus of
the organism will maintain the temperature, the hydrogen ion concentration,
ion balance, glucose pool, etc., such that the cellular environment is reason-
ably constant.  Within reason, the processes of natural selection and adapta-
tion produce the same kind of stability in the environmental system.  However,
this means that our environmental system is changing, perhaps not as much in
function as in structure.  In the first instance, the environment is kept
"constant" by the activities of cells and in the second, the environment forces
adaptation or selection to the system.

     This, of course, is the basis for concern, since much of the experimental
data used in model construction have been gathered in laboratory investiga-
tions of environmental samples.  One can assume that community structure of
water or sediment is determined by certain major environmental variables, and
that given those variables, a certain community structure will result (subject
to change by the laboratory environment).  This laboratory effect is probably
no worse than comparing mice to men in toxicity studies.  Laboratory data,
however, are influenced by seasonality of the sample and population shifts
from changing growth conditions upon bringing the sample into the laboratory.
While laboratory data are valuable, it may be different from that obtained in
the field.  This is no doubt similar to differences in biochemical activities
of intact and excised tissue, yet most biochemical studies are done with ex-
cised tissue.

     Testing certain hypotheses derived from the laboratory, or performing
certain experiments in ecosystem studies may be virtually impossible.  One
cannot sacrifice an ecosystem very readily.  There are examples of such large-
scale experimentation (e.g., Hubbard Brook), but for most needs these systems
are not available.  This use, simulation, is a unique application of computer
programmed models.

     We have simulated several conditions that could not be experimentally
tested to probe the sensitivity of a general lake ecosystem model to various
perturbations (1).  The microbiological detail of the lake ecosystem model
CLEANER, illustrated in Figure 1, shows the various compartments of the micro-
bial growth and decomposition components of CLEANER.  Figure 2 shows the rela-
tionship of the microbial component to the rest of the compartments in the
lake ecosystem model.

     Using a computer programmed mathematical model developed for Lake George
in upstate New York as part of the International Biological Program, we deter-
mined the differential rates of uptake by the microflora during the course of
a year as modified by environmental variations.  As shown in Figure 3, the
model predicts dynamic changes in uptake rates as a result of inputs from
spring runoff and spring primary productivity.  The evidence that spring run-
off is an important input to the lake is given in Figure 4.  One of the major
tributaries to Lake George, Northwest Bay Brook, drains a large percentage of


Figure 1.  Compartmental diagram of microbial growth and decomposition model.
          WRDOM, Refractory dissolved organic matter in water
          WLDOM, Labile dissolved organic matter in water
          WDEC, Decomposing microorganisms freely suspended in water
          DDEC, Decomposing microorganisms attached to particulate organic
          DDOM, Dissolved organic matter sorbed to particulate organic
          POM, Particular organic matter
          DIM, Dissolved inorganic matter

                                          JUV OMN
                             ADULT OHNIV
                                                DEM FISH

<- 	 -^

                                              CHIR    SED DO    SED C02
Figure 2.  Compartmental diagram of lake
  MAC 1,2,3, Three macrophytes of dif-
    fering growth habits
  NAN, Nannophytoplankton
  NET, Net phytoplankton
  BLGR, Non-N2 fixing blue-green algae
  N-FIXBG, Na-fixing blue-green algae
  COPE, Copepods
  CLAD, Cladocerans
  JUVOMN, Juvenile omnivorous
  ROT, Rotifers
  ADULTOMN, Adult omnivorous
  PISC, Piscivorous fish
  N-PISC, Non-piscivorous fish
  DEMFISH, Demersal fish
  LOOM, Labile dissolved organic matter
  RDOM, Refractory dissolved organic
  WDEC, Decomposing microorganisms
    freely suspended in water
  DDEC, Decomposing microorganisms
    attached to particulate organic
  LDDOM,  Labile dissolved organic
    matter sorbed to particulate
    organic matter
  LPOM, Labile particulate organic
ecosystem model  (3).
RPOM, Refractory particulate  organic
RDDOM, Refractory dissolved organic
  matter sorbed to  particulate
  organic matter
IRDOM, Interstitial water refractory
  dissolved organic matter
CHIR, Chironimids
SLDOM, Sediment labile dissolved
  organic matter
SRDOM, Sediment refractory dissolved
  organic matter
SLPOM, Sediment labile particulate
  organic matter
SRPOM, Sediment refractory particu-
  late organic matter
SEDDEC, Sediment decomposing  micro-
DO, Dissolved oxygen
CO2, Carbon dioxide
N, Nitrogen compounds in water
P, Phosphorus compounds in water
S, Sulphur compounds in water
SEDDO, Sediment dissolved oxygen
SEDN, Sediment nitrogen
SEDCOa, Sediment carbon dioxide
SEDP, Sediment phosphorus

                                            s s s J JjrrtfJ
Figure  3.  Decomposer uptake of dissolved organic matter.
          DDOM,  Dissolved organic matter sorbed to particulate organic
          WRDOM, Refractory dissolved organic matter in water
          WLDOM, Labile dissolved organic matter in water

                                                            NORTHWEST BAY  BROOK

                                                            POC  CONCENTRATION AND LOADING
                                                                Apr  '  May  '  June  '  July
Feb  '  Mar
Sept   '  Oct  '  Nov   '  Dec
     Figure  4.  Northwest bay brook particulate organic carbon  (POC) concentration and loading.

the basin, and as such is an important source of allochthonous material to the
lake.  We have been able to validate these uptake rates for DOM using growth
rate data from continuous culture and isotope uptake experiments  (2).

     The sensitivity of the model to changes to conditions that could not be
studied experimentally are illustrated in Figures 5 and 6.  Figure 5 shows the
response of the decomposer and organic matter compartments to normal and per-
turbed conditions in Lake George, New York, over a period of 827 d.  In the
normal simulation  (Fig. 5A), the annual cycles are evident, and the model is
seen to be stable with a large, but realistic, fluctuation in the freely sus-
pended decomposers.

     When the decomposers are removed from the system, the response is sig-
nificantly different  (Fig. 5B).  Labile DOM increases rapidly, while refrac-
tory DOM increases more slowly.  POM exhibits a stable seasonal fluctuation
but, as will be shown later, does not reach the peak concentrations that are
attained with normal seasonal contributions from algal production.  In the ab-
sence of heterotrophic uptake, sorbed DOM exhibits slight annual peaks that do
not occur under normal conditions.

     A simulation without influence of phytoplankton is shown in Figure 5C.
In this detritus-based system, the decomposer and organic matter compartments
maintain stable, annual patterns of oscillations.  However, the patterns re-
flect the seasonal loadings of organic matter.  Without autotrophic production
in the late spring and summer, the freely suspended decomposers, and to a
lesser extent the POM and attached decomposers, exhibit pronounced troughs.
The abnormally low microbial activity, in turn, results in a slight seasonal
rise in refractory DOM that is not seen in a normal simulation.

     The complete absence of allochthonous organic loadings (Fig. 5D) leads to
an entirely different series of patterns of declining concentration.  This
suggests that with the present calibration the decomposition submodel is  overly
sensitive to external loadings.  This sensitivity probably reflects our inade-
quate linkage between littoral and pelagic zones.  Interestingly, the DOM com-
partments show little change; they are receiving some input from autochthonous
sources but are not subject to the solubilization and sedimentation that af-
fect the POM pool.

     Figure 6 demonstrates the control that remineralization and autotrophic
uptake exercise over the nutrient pools.  In Lake George the dissolved oxygen
and COz levels are rather uninteresting.  However, in a normal simulation
(Fig. 6A), phosphate exhibits the effects of remineralization followed by se-
vere depletion as a result of phytoplankton growth.  Nitrogen has a less
marked seasonal fluctuation.

     Figure 6B indicates that there is no remineralization by decomposers, and
phosphate disappears as soon as the growing season begins.  Soluble inorganic
nitrogen is remineralized by higher organisms and also is brought in through
precipitation and runoff so that it gradually increases in concentration.

     In the absence of uptake by phytoplankton  (Fig. 6C), nitrogen, phosphorus,
and carbon increase in concentration.  Without organic loadings  (Fig. 6D),


1 .00 44
-, 316.69 *•

636.38 •*

827.00 44
0 + 	
fsj 1.00 44
o 2
- B'
318.69 +4 B
> - B
~* - BB

636.38 4*
827.00 4 +
1.00-02 1 .00400 1 ,00<
1.00-01 1.00+01
6 4FF6— B^ 99, 4 A
66 f ^BB 99, A*
«,F BB; 9 A A,
X6 BBB 9,9 A
/ f *—* """" * *>
B-B66F* 99' A
B-6B 666 FF 999 A
BB 66 F 99 A
8 6C F 9 'A
^B-6 6 F4 9 + A
66 — B-F'F-B-^^ 9, A
S6 — eJ>-F^"~^B^^ X 	 9 XAA
GF ?^9 N?A
6F BB 9A
61 F BBB 99'AA
A'j,V 	 B 99 A
/BE6X 99 A
B-6,66 F' 99 A
BB-B 666 FF 99 A
BB 6 FF 99' 4 |
B-BS 6' f ? AA
B-B6-B f 9 A
GC TTt 0 ?T t\
"^e-G^r ^"^-B 9-9s A-A'
*6F B-BX 9 A
+6F e' 9A
1.00-02 1.00 + 00 1 .00
1.00-01 1.00+01


E 4

6 4FFB-BS 99 4 AA E 4
* F B 9 AA E
6 F p-B 9 A (
1 XfBr t E
BX6.B'F 99
BB' t f ? E
B-B 66 F _99 E
B-B~B' 666' F' 99' A ^
B' 6«' F' 99 A E
GG fF 99 A E
66 FF 4 9,9 A4 E
6' FF 99 A E
6 FF 99 A ^
e' F 9 A £
66 F' 9 A E
666 FF 99 A E
' f 9 A E
FF ?9 A E
? ? * f
F.F 99 A E
FF + 9.9 ** F.
F 9 A E
f' ,«' * t
F.' ? A E
/ « * f
F 9 A E

F 9 A4 E 4
1.00 44
318.69 44

636. 3B «*

827.00 ++

1 .00-03
1.00 +4
318.69 44

636.36 4+
827.00 4+
0 + 	
1.00-02 1 .004QO 1 -00
1 .00-01 1 .00401

Ffe— B. 99^ + A
F ^~8B 99, AA
6F /B-B ?AA
6^F BB 9,AA,
>66,BB^ -99JJ
i"" 9?"
Gi^6 ?99' A
9 6*V '" *
B Gi F ? *
6BJ «F ?9 * A,
6>-BF-B, 9 *


p 64

F ^B-B 99 /A
6.F ^B 99 /
GF BB' ?A,
F \X-B >?
^X-X6 ,?9
k^ 99A
X'F 99' ^
{ 99 4 A
F ,9 ?
t ? M
F>B— B^ 99, A
F ^B-B 99, AA
6F B-B 9AA
6F BB' 9AA
1 .00400 1 .00
1 .00-01 1 .00401






i 4
FF 9, + A EE +
\ \ ^ \ -.
F 99 'AA
f ?9 A
fF ?9 A
FF 99 A
FF 99 A
F ?
99 *
F 9
( 9 '»
F-Ff 9-99 A
! '
F' 9'
f' ?9
FF 99
FF 99 +


F 9
4F 9 +

E 4

^ ~

E +

' I -
'A E
A E 4
Figure 5.  Example of a simulation.  The response of decomposer and organic matter compartments are
           shown under  (A) control conditions, (B) in the absence of decomposers,  (C) in the absence of
           phytoplankton,  (D) in the absence of allochthonous organic loading.  Symbols:  A, WLDOM;
           B, WDEC; E, WRDOM; F, DOOM; G, DDEC; 9, POM.

. 1.00 +*

D 318.69 ++
~* _
636.38 ++
827.00 ++
{£ 1.00 ..
[ 1 ~

- c.c— c 	
D 318.69 ++
636.38 +«


827.00 ++
Figure 6. Example of a
1.00+00 1.00+02 1.00-02 1.00+00 1
1.00-01 1.00+01 1.00+03 1.00-03 1.00-01 1.00*01
C + CX DO + B L « 1.00 +« C»C DO, »
C-C DO BB L - - NC-C B
'ccv BO * L - - 'cc \
C BB BL^ - - 	 .CC ^BO
CCf DD BL - - C 	 C~- 	 " tl
C, D' BL - C— C DO'
? B, BL - -, C^^^ B
C B L - - 00
< 0 L - - B
f if L B
» C • B L' » 318.69 **'U * X «
t B L' - C 	 C 	 C BBV
cc 'D B- - c~cx DD
C, B LLB - CC 00
C 'D L B - - f f'f fO
i B LLH — ~ f 	 ^ 	 C D D— D
C B LBB - - C-C OB
^ Lll B - -''C 	 — C BO
C L i B.
+ c + L h + 636.38 «< « 0 «
C L' B 0
C L (1 - - C-C 	 C^ 00
C D L B - - ^-C- C% 00
(i; bo u B - c — c, DO,
C 0 L' B - - CC 0
+ t + 0 L > + 827.00 ++ + C 0 +
1.00+00 1.00+OZ 1.00-02 1.00+00 1.
1.00-01 1.00+01 1.00+03 1.00-03 1.00-01 1.00*01
C + C Dp * B L * 1.00 ** C* D «
C DO B L - - ji 0
___ 	 C-C DB B L - - 	 	 C 	 C b
DD B LI - - 00
? ">>;-- P
B B L 00
'P "(•'-- P
B BL p
B BL b
« DB* BL + 318.69 •* • OB
B B\ b
DD 'BL b
b B( 'B
B BL 0
DO ft - - 0
0 X - - D
P1 ! - - P
P ? P
DD X - - 0
+ * b X + 636.38 ++ + + o'
b L'B D
0 L.BB - - B
B LB 0
b ^B - - b
0 LB 0
, < b LB + 827.00 •* * • b
simulation. The response of nutrient pools are shown under
*H L
"»' L
" L

P (

B Ik
1 "-L
B L'

B L'

? *-'
" <
1 L








tions, (B) in the absence of decomposers, (C) in the absence of phytoplankton, (D) in the ab-
sence of allochthonous organic loading.  Symbols:  C, soluble reactive phosphate; D, soluble
inorganic nitrogen; L, dissolved oxygen; M,  carbon dioxide.

there is an initial mobilization of phosphate by decomposers; otherwise, the
effects of decomposition are limited, and the result is similar to that of no
decomposers (Fig. 6B).

     How does this exercise help us understand actual situations in aquatic
ecosystems?  It should be possible to classify aquatic systems according to a
group of key functional variables, and within a class certain types of respon-
ses should be expected, such that responses determined in one system should be
applicable to others in the class.  The problem, of course,  is that we don't
know enough about how aquatic ecosystems are regulated to have this confidence.
But the problem of environmental degradation is prevalent and the means  for
treating  it are limited.  Therefore we have to implement our available tools
with the  recognition of their shortcomings.  Mathematical modeling is one  of
those  tools.  Careful development, constant testing, continual refining, and
cautious  use of computer program models can lead to information  on biodegrada-
tion in aquatic environments.  However, the information cannot be more than
hypothetical in perturbation studies and predictions.  Nevertheless, a good
hypothesis is a valuable step in scientific investigation and the basis  for
many a decision.

     The  tight coupling of mathematical models with laboratory and field in-
vestigations is  the  best approach when applying models for the purpose of  de-
riving information related to biodegradation in the environment.  The enormous
complexity of the  system suggests the use of a systematic methodology.   How-
ever,  modeling should not be considered a substitute for field and laboratory
investigation.   Rather, models are used most beneficially in facilitating  the
organization of  diverse data, the detection of conceptual weaknesses, and  the
testing of hypotheses that may be virtually impossible to test in other  ways.
With the  same kind of reservation used in applying cellular  studies to a whole
organism, models  can be used to predict system responses.
                               LITERATURE CITED

 1.   Clesceri,  L.  S.,  R.  A.  Park,  and J. A. Bloomfield.   1977.  General  model
      of microbial growth and  decomposition in aquatic ecosystems.  Appl.
      Environ.  Microbiol.  33:1047-1058.

 2.   Clesceri,  L.  S.   1972.  Role  of the heterotrophic microflora  in  the cycling
      of materials.   Oak  Ridge National Laboratory, Eastern Deciduous Forest
      Biome—IBP Memo Report No. 72-64.  Oak Ridge, Tenn.

 3.   deCaprariis,  P.,  R.  A.  Park,  R. Haimes, J. Albanese, C. Collins,  C.  Desor-
      meau, T.  Groden,  D.  Leung, and B. Youngberg.  1977.  Utility of the com-
      plex ecosystem  model MS.CLEANER, p. 87-89.  In Proceedings of the  Inter-
      national  Conference on Cybernetics and Society.

 4.   Garfinkel, D., and B. Hess.   1964.  Metabolic  control mechanisms.   VII. A
      detailed  computer model  of the glycolytic pathway  in Ascites cells.
      J. Biol.  Chem.  239:971-983.

 5.   Weaver,  W.  1948.  Science and complexity.  Am. Scientist  36:536-544.



                     J. H. Slater
         Department of Environmental Sciences
      University of Warwick, Coventry CV4 7AL, UK

     It is generally recognized that most natural environ-
ments contain a heterogeneous assemblage of microbes.  In
some instances this has resulted in evolution of mutually
beneficial associations of different microorganisms, al-
though the extent of their occurrence, distribution and
general significance remains to be fully assessed.  Micro-
bial communities show varying levels of structural com-
plexity and degrees of interdependence between the com-
ponent populations.  They may be based on non-specific
interactions which result in loosely structured communi-
ties with variable component populations.  In contrast,
microbial communities can be tight associations based
upon a specific, perhaps obligatory, interaction.
     Mixed cultures may have a crucial role in biodegrada-
tion of complex natural products or xenobiotics.  In some
instances it is difficult to isolate a single species with
the complete capability to metabolize a given complex sub-
strate.  Nevertheless, such degradation does occur in
nature at a reasonable rate and this may be the result of
the combined activities of several organisms with separate,
but complementary, degradative capabilities.  Similarly,
the breakdown of a xenobiotic could involve several steps
catalyzed by enzymes produced by different microbes or the
evolution of novel or improved catalytic activites in an
existing enzyme system.  In either situation, multispecies
communities provide a stable, "permissive" environment for
development of the requisite capability.  One important
aspect may be cometabolism where products which cannot be
further metabolized in a pure culture can be used by other
organisms in a mixed culture.  Furthermore, the signifi-
cance of the role of degradative plasmids may be enhanced
within a microbial community since there is a greater
genetic potential.

     Traditional enrichment and isolation techniques favor
pure cultures, whereas continuous-flow culture techniques


           can be used to isolate interacting associations of micro-
           organisms.  The advantages and disadvantages of this tech-
           nique in examining adaptive processes in mixed cultures is

     Although it is generally accepted that most natural environments contain
heterogeneous assemblages of microorganisms (4), little is known about the
consequences of growth and coexistence of different microbial populations in
the same niche.  It may be that much of the activity of different populations
is comparatively independent, in which instance the traditional methods of
isolating and examining pure cultures should provide the necessary information
on the role of that population in nature.  However, it seems improbable that,
during the course of evolution, interactions and relationships between differ-
ent populations have not developed.   Indeed, it is now becoming clear that mi-
crobial communities, i.e., mutually beneficial units composed of different
species of microorganism, do occur in nature; these may be isolated if appro-
priate enrichment and selection techniques are employed.  Perhaps because of
the emphasis upon the isolation of single species of microbe, relatively few
examples of structured microbial communities have been described.  Certainly,
the extent of their occurrence, distribution and overall significance in na-
tural habitats remains to be fully assessed.  Thus, scant attention has been
paid to the physiological, biochemical and genetic consequences of interac-
tions between different populations.

     This communication describes some of the types of microbial communities
that have been isolated and characterized and discusses the use of continuous-
flow culture techniques in isolation of stable mixed cultures.  In addition,
the relevance of the growth of mixed cultures with respect to biodegradation
of environmentally undesirable compounds and the evolutionary potential of mi-
croorganisms within a community structure are discussed.

     Although a number of basic types of interaction between different micro-
bial populations have been recognized (28, 35, 36), it is at present difficult
to assert that all the potential categories of microbial community have been
observed.  A basic feature, however, of microbial communities is that they
constitute a group of microorganisms which grow together in the same environ-
ment.  In many instances, it is also possible to recognize a mutual benefit
between the component populations.   Thus, microbial communities show a wide
range of structural complexity and degrees of interdependence between the com-
ponent populations.  On the one hand they may be based on non-specific inter-
actions which result in formation of loosely structured communiities.  Indeed,
it may be argued that some of these loose associations ought not to be consid-
ered as authentic communities.  For example, there is clearly some interaction
between the various populations involved in successions or particular sequen-
ces in biogeochemical cycles  (i.e. , nitrification) where the growth of one
population sufficiently alters the environmental conditions permitting the

growth of a second, succeeding population.  In most situations, there is a
characteristic sequence of population changes, temporally separated, but ob-
viously comprising an interacting group of organisms.  Furthermore, under ap-
propriate culture conditions it is possible to maintain a stable mixed culture
of part or the whole of the succession or cycle.  It can be argued that from
such a loose association of different populations potentially important speci-
fic communities may be evolved under the appropriate environmental conditions.

     Alternatively, microbial communities can exist that are based on speci-
fic, obligatory interactions such that none of the component populations are
able to grow alone.  It is these categories of microbial communities which are
likely to be overlooked since conventional enrichment and selection procedures
favor the isolation of single species able to grow in a particular environ-
ment.  A number of different categories of tight microbial communities have
been recognized, at least for the simpler, smaller communities (35).

     Microbial communities may be based on reciprocal, specific nutritional
interactions between the component populations.  In the simplest example,
namely a two-membered community- two auxotrophic populations can have their
requirements met in part by the other population.  For example, Nurmikko (31)
isolated a community composed of a phenylalanine-requiring strain of Lactoba-
cillus plantarum and a folic acid-requiring strain of Streptococcus faecalis.
The two populations could be grown together, but not separately,  in a medium
lacking both growth requirements since the Lactobacillus excreted folic acid
and the Streptococcus provided phenylalanine.  Among a number of similar com-
munities (3, 10, 23, 26, 39, 42) ,  one of the more recent has been the characteri-
zation of a simple two-membered community capable of utilizing cyclohexane as
the sole carbon and energy source (37).  Considerable difficulties have been
experience in the past in attempting to isolate a pure culture able to use
cyclohexane, although the compound has been known to be readily degraded in
the natural environment.  This particular community was isolated from an es-
tuarine mudflat and the primary organism capable of metabolizing cyclohexane
was identified as a species of Nocardia.  However, the primary organism grew
only in the presence of a pseudomonad which was subsequently shown to be sup-
plying the actinomycete with biotin.  In turn, the pseudomonad was supplied
with metabolizable carbon and energy sources derived from either the immedi-
ate degradation of cyclohexane or cell lysis products.

     More complex interactions may occur leading to the formation of closely
coupled communities.  In 1943, Schopfer (33) described an association of a
species of Mucor and Rhodotorula.  Both organisms required the vitamin thia-
mine but separately each organisms was unable to synthesize the compound.
Thiamine has two major precursors, namely pyrimidine and thiazole, and it was
shown that the Mucor species could only synthesize the former component while
the Rhodotorula species produced the latter.  Thus, as a community these two
organisms were able to complete the synthesis of the common requirement for
growth.  More recently, Bates and Liu (2)  found that a mixed culture of two
pseudomonads resulted in production of lecithinase, whereas separately neither
culture contained this activity.  The implication is that the two strains pro-
duced different enzyme subunits which together resulted in the formation of a
fully functional enzyme.

     The ubiquity or otherwise of these specific kinds or association in natu-
ral habitats has not been examined in any detail, although it is our view that
they are likely to be much more common than has so far been appreciated.
First, there is considerable circumstantial evidence that specific nutrition-
ally based associations do occur.  For example, many marine bacteria have been
observed to grow in close proximity to algae and the heterotrophs are known to
excrete vitamin Bi2 which may enhance the growth of the algae (4).  Moreover,
it is probable that the bacterial populations derive some benefit from the as-
sociation through the excreted reduced carbon compounds and oxygen produced as
a result of algal photosynthesis.  It would be extremely interesting to know
how significant these relationships are in promoting the growth of the mixed
cultures compared with growth as individual populations.

     Second, specific associations are likely to be overlooked because of the
selection and isolation procedures used to examine the microbial flora of a
particular habitat.  Conventional techniques result in the disruption of popu-
lation interactions leading to lack of growth of the component populations
during the selection procedures.

     Stable associations can be based upon the removal of a growth inhibitory
compound which is either produced by one of the members of the community or
may be present in the growth environment initially.  For example, an organism
originally known as Chloropseudomonas ethylica, a photosynthetic bacterium
with the unusual capacity to utilize ethanol as a photosynthetic electron
donor, has now been shown to be a tightly coupled microbial community composed
of a photosynthetic green sulphur bacterium and a sulphate-reducing bacterium
(13).  The most important interaction, although there may be others concerned
with the carbon nutrition of the two organisms, regarding the utilization of
oxidized and reduced sulphur compounds.  The photosynthetic organism required
hydrogen sulphide as an electron donor, resulting in its oxidation to sulphate,
while the sulphate-reducing bacterium required sulphate as a terminal electron
acceptor, regenerating hydrogen sulphide.  Thus, by growing as a community the
two organisms were able to recycle limited quantities of sulphur, thereby sus-
taining growth of both populations.  The importance of this relationship to
both organisms is that neither has to grow in a high sulphide-containing en-
vironment which is toxic to both populations.  A number of other less obliga-
tory relationships fulfilling the same metabolic function have been identi-
fied  (40) .

     Wilkinson et al. (38) have described another microbial community contain-
ing at least four different species and capable of growing on methane as the
sole carbon and energy source.  Only one member of the community, an unidenti-
fied pseudomonad, was able to oxidize methane which resulted in some excretion
of methanol when the organism was grown in pure culture.  Methanol accumula-
tion resulted in inhibition of the pseudomonad.  However, within the community,
a species of Hyphomicrobium oxidized the available methanol as its primary
carbon and energy source, thereby preventing the accumulation of methanol to
toxic levels and so maintaining the growth of the methane-oxidizing pseudo-
monad.  Together these two populations comprised greater than 95% of the total
community under all the growth conditions examined.  However, it is uncertain
what function the two minor populations, a Flavobacterium sp. and an Acineto-
bacter sp., had in the structure of the community.  It was postulated that
these species grew on cell lysis products and their presence ensured the


maximum utilization of all the available carbon and energy sources.  If this
is true, then the presence of these two populations may confer little direct
advantage to the major groups and simply be retained within the growth envi-
ronment because they occupy an available growth niche  (16).

     A number of different mixed cultures have been described, particularly
for rumen microorganisms, which depend on the transfer of hydrogen ions (or
electrons)  between the component populations.  For example, the organism known
as Methanobacillus omelianskii was found to be a tight association of two bac-
teria with the overall capability of utilizing ethanol as a carbon and energy
source, resulting in the production of methane (6).  One component of the com-
munity was termed the S organism and oxidized ethanol to acetate with the con-
comitant production of hydrogen.  In pure culture the presence of hydrogen
caused the cessation of S organism growth, whereas in the mixed culture, growth
continued because the hydrogen ions were used as an energy source by the sec-
ond component, namely Methanobacterium strain MOH.  The methanogenic bacterium
used carbon dioxide as an electron acceptor producing methane as a final pro-
duct.  The specific association of these two organisms maximized the amount of
new biomass produced from the oxidation of unit amount of ethanol and, as with
"Chloropseudomonas ethylica," ensured the immediate removal of an inhibitory
waste product in order to sustain growth of the hydrogen-producing organism.
The association of organisms in Methanobacillus omelianskii was not obligatory
since other methanogenic bacteria, for example Methanobacterium ruminantium,
could grow in association with the S organism.  Subsequently, a number of
other communities, in which excess reducing power produced by one population is
immediately used by a second, have been isolated  (5,9, 20) .

     Insufficient attention has been given to the possibility that associa-
tions, particularly those based on loosely grouped organisms, may occur on the
basis of overall basic growth characteristics determined by the two fundamen-
tal growth parameters, namely maximum specific growth rate and the growth-
limiting substrate saturation constant.  The maximum specific growth rate of a
group of organisms may be greater than the individual growth rates of the sepa-
rate populations grown under the same conditions.  Osman et al. (32)  isolated
a stable, three-membered mixed culture using a chemostat continuous-flow cul-
ture system with orcinol as the sole carbon source.  A single primary popula-
tion, Pseudomonas stutzeri, metabolized orcinol and grew in pure culture on
this substrate and in the mixed culture was the dominant population.   The two
secondary populations, unable to grow on orcinol, comprised up to 20% of the
total biota depending on the growth conditions.  Acetate was identified as a
major metabolite excreted by the primary population which, together with cell
lysis products, supported the growth of the secondary populations.  A signifi-
cant point was that P. stutzeri was capable of good growth on orcinol in pure
culture indicating that there was little direct or obligatory interaction with
the two secondary populations.  Two possible reasons may explain the continu-
ing presence of the secondary populations and the extreme stability of the
three-membered mixed culture which was continuously cultured for several thou-
sand hours during which period it was subjected to a number of environmental
stresses, including temperature changes and alterations in the concentration
of the growth-limiting substrate.  First, the association could have been the
result of a fortuitous presence of the secondary populations.  If the concen-
trations of growth substrates required to support the growth of the secondary

populations were high enough to sustain growth rates which were greater than
the specific growth rate of the primary population, then the secondary popu-
lations could be retained without any need for a direct, mutually beneficial
interaction with the primary population.  Second, it was possible that the
community as a whole had the most competitive growth kinetic constants.  In
continuous-flow culture systems, the most important parameter is the satura-
tion constant which measures the organisms' or the community's affinity for
the limiting substrate (35) .  In monoculture the primary population had a
saturation constant, Ks,  of 100 mg orcinol liter"1, whereas the complete mixed
culture had a Ks of 71 mg orcinol liter"1, indicating a significantly greater
affinity for orcinol by the mixed culture than the primary population alone.
This appeared to explain the selection of the mixed culture as compared with a
pure culture of P. stutzeri.  However, further experiments showed that a two-
membered community of P.  stutzeri and Brevibacterium linens exhibited an even
greater orcinol affinity with a Ks of 56 mg orcinol liter"1.  This simpler
community was not selected possibly because no growth conditions could be es-
tablished leading to the competitive exclusion of the second secondary popu-
lation, a species of Curtobacterium.  The mechanism of interaction between the
pseudomonad and the B. linens population, which seemed to be the significant
relationship resulting in the increase or orcinol affinity, has not been elu-

                              POPULATION GROWTH

     The examples cited previously have illustrated the value of specific in-
teractions between different populations which have led to the formation of
stable microbial communities.  There are, however, considerable potential ad-
vantages to be gained from the growth of mixed populations in the same habi-
tat, particularly when organisms have to respond to unusual or novel environ-
mental conditions.


     Mixed cultures may have a crucial role to fulfill in the degradation of
complex natural products or xenobiotics.  For example, several different mi-
croorganisms between them may have the complete capability to degrade a par-
ticular compound although none of the individual species have the full gene-
tic information for the synthesis of all the necessary enzymes.  Thus, under
appropriate growth conditions a microbial community would be formed.  One par-
ticular variation which may be important in nature, particularly for the deg-
radation of unusual compounds, occurs if one member of the community is able
to cometabolize a particular compound producing a metabolite which is then
further metabolized by a second population.

     Johanides and Hrsak (24) have isolated a bacterial community growing on a
mixture of at least twenty isomers and homologues of linear alkylbenzensulpho-
nates which together comprised the growth-limiting substrate in a continuous
enrichment.  The resulting mixed culture predominantly contained Pseudomonas
and Alcaligenes species in porportions which varied depending on the precise

growth conditions.  None of the isolated organisms were capable of growth on
the mixed substrates in monoculture.  It was suggested that the degradation of
the surfactant depended on a combined metabolic attack by several organisms.
It cannot, however, be eliminated at this stage that possibly some of the com-
munity interactions did not involve cometabolism of part of or the whole of
the molecule  (19) but specific nutritional interactions between the component

     Some years ago, Gunner and Zuckerman (14) described a synergistic rela-
tionship between a species of Arthrobacter and a Streptomyces species to ac-
count for the degradation of the insecticide Diazinon  (0,0-diethyl 0-2-
isopropyl-4-methyl-6-pyrimidyl thiophosphate) which strongly indicated that
some form of combined metabolic attack was involved.  Treatment of soil with
the insecticide enriched these two organisms although individually, neither
organism was able to utilize the compound as the sole carbon and energy source.
The Arthrobacter species alone was able to metabolize the ethyl ester moiety
but   C-labelled studies revealed that neither organism alone was able to me-
tabolize the pyrimidine ring.  As a community, however, rapid degradation oc-
curred producing two unidentified metabolites which were not further degraded
by these two organisms.  Neither the mechanism of ring cleavage nor the pre-
cise enzymatic relationship between the two organisms was elucidated but these
results have indicated the potential importance of such interactions in the
degradation of complex compounds.

     Another example of possible cooperative metabolism by two microorganisms
comes from the leaching of iron and copper from sulphur-containing ores such
as pyrite and chalcopyrite (1, 27).   It has been shown that Leptospirillum
ferrooxidans can generate energy from oxidation of ferrous iron but not from
reduced sulphur compounds.  Conversely, Thiobacillus organoparus cannot uti-
lize ferrous iron but can obtain energy from the oxidation of reduced sulphur
compounds.  As pure cultures neither organism utilized pyrite or chalcopyrite
while a mixed culture was able to rapidly oxidize the appropriate components
of the ores.  The nature of the interaction has not been explained.

     Cometabolism occurs widely  (18) and some recent evidence suggests that
interactions based on cometabolism may be important in natural populations.
Munneke et al. (29, 30) have adapted mixed populations to grow on the toxic
insecticide parathion (0,0-diethyl 0-p-nitrophenyl phosphorothioate) using
chemostat systems and isolated apparently stable multispecies communities.
Most recently, Daughton and Hsieh (12)  have characterized the metabolic core
of one such community which had been selected after two years continuous
growth on parathion.  An initial examination suggested that the community ap-
parently contained three organisms,  none of which could hydrolyze parathion.
A more reigorous examination of some of the slow-growing isolates revealed the
presence of a fourth organism, Pseudomonas stutzeri, which had an extremely
active parathion-hydrolyzing capacity.   It was, however, unable to grow on
either of the two hydrolysis products,  namely diethyl thiophosphate and p-
nitrophenol.  One of the original isolates,  P. aeruginosa, could grow on p-
nitrophenol.  Thus in the complete community, P. aeruginosa depended on the
hydrolytic activity of the primary P.  stutzeri which in turn depended on the
provision of utilizable metabolites  from the P. aeruginosa population.  No
role was ascribed to the two other non-parathion utilizing populations, a

coryneform and an unidentified pseudomonad.  None of the organisms were able
to make use of the second breakdown product, diethyl thiophosphate.

     Recently, using continuous-flow culture systems, we have isolated an un-
identified mixed culture growing on a mixture of glycerol and dichloroacetic
acid (DCA) as the combined growth-limiting substrates (Lovatt and Slater, un-
published data).  At a growth rate of 0.1 h"1, approximately 80% of the DCA
was dechlorinated but DCA alone could not support the growth of the mixed cul-
ture in the chemostat and none of the organisms was able to grow individually
on DCA.  It seems probable that DCA was cometabolized although we have not
completely excluded the possibility that the growth of a DCA utilizer depended
on a nutritional interaction with another population in the mixed culture.


     An interesting problem at the present time concerns the capacity of mi-
croorganisms to adapt to utilize novel carbon source for growth.  For natur-
ally occurring materials, microorganisms have had several billion years to
evolve the necessary catabolic pathways.  However, in the contemporary envi-
ronment, microbial populations have been exposed to an increasing diverse
range of unusual or environmentally foreign compounds over an exceedingly
short period of time on an evolutionary scale.  This situation raises the in-
triguing questions of how rapidly and by what mechanisms do microorganisms
evolve the necessary, new catabolic potentials and whether or not there are
any limitations to microbial degradative capabilities.

     There are basically two classes of adaptation phenomena which have to be
considered.  First, the comparatively simple adaptations through mutation and
appropriate selection, of existing catabolic mechanisms.  Second, the more
complex problems involved in the evolution of complete pathways for the meta-
bolism of a new growth substrate.  The former may involve changes in enzyme
specificities, control and regulation mechanisms or transport systems for the
assimilation of the new substrate  (11).  Pure culture studies with the ali-
phatic amidases of Pseudomonas aeruginosa  (11) and the pentitol dehydrogenases
of Klebsiella aerogenes  (17, 41), have illustrated the importance of altera-
tions at these levels in the acquisition of a new catabolic capacity.  However,
little is known in detail about the frequency and nature of adaptations at the
molecular level in naturally occurring populations.

     Senior et al.  (34) isolated a multispecies community growing on the her-
bicide Dalapon, 2,2'-dichloropropionic acid  (22DCPA).  Although the original
community contained three organisms growing on 22DCPA, it was observed that
after a period of time one of the original secondary organisms, Pseudomonas
putida S3, unable to grow on 22DCPA, acquired the potential to grow on the
compound.  Growth on chlorinated aliphatic acids required the presence of the
enzyme dehalogenase and the new 22DCPA-utilizing strain, P. putida P3, was
shown to contain 22DCPA specific dehalogenase activity with a maximum activity
of 7.8 ymol 22DCPA converted/mg protein"1/!}"1 .  Subsequently it was found that
the parent strain could grow slowly on 2-monochloropropionic acid and under
certain conditions very low levels of 22DCPA specific dehalogenase activity,


approximately 0.5 ymol 22DCPA converted/mg protein l/ti l.  This adaptation in-
volved either a change in the regulatory mechanisms or perhaps in the trans-
port of 22DCPA.  The important point is that the complete community provided
a suitable environment, a "permissive" environment, for the retention of the
parent organism in a situation where alone it would have been unable to grow
and evolve.

     More recently, we have examined the degradation of the herbicide lontrel,
3,6-dichloropicolinic acid  (36DCPA)  (Lovatt, Gilchrist and Slater, unpublished
observations).  It proved impossible to isolate organisms from soil, using
either continuous-flow culture techniques or conventional dilution plates,
able to grow on 36DCPA.  We have, therefore, attempted to adapt a microbial
community growing on picolinic acid to utilize 36DCPA.  The original pico-
linic acid community was composed of three primary utilizers, Pseudomonas
aeruginosa, Alcaligenes faecalis and a second Alcaligenes species, and three
secondary organisms, namely Bacillus licheniformis, a species of Rhodococcus,
and a Corynebacterium aguaticum-like organism.  The community could not be
grown on 36DCPA as the only carbon source and the initial attempts to adapt
the culture to a mixture of picolinic acid and 36DCPA (in the ratio 5:1)  also
resulted in culture washout.  However, after several cycles of growth on pico-
linic acid alone followed by a period of growth on the mixed substrates,  the
community adapted to grow in the presence of 36DCPA.  From this culture,  none
of the organisms was able to grow on 36DCPA in pure culture although it was
found that 36DCPA stimulated oxygen uptake in the mixed culture.  There was no
increase in the culture biomass suggesting that none of the 36DCPA carbon was
utilized for biomass production.  The composition of the 36DCPA-tolerant com-
munity, in terms of the primary organisms, was markedly different from that of
the original community.  Initially, at a growth rate of 0.05 h"1, the commu-
nity was dominated by A. faecalis at about 90% of the community, with P.  aeru-
ginosa the next most abundant primary picolinic acid-utilizer.  In the adapted
community considerable fluctuations in the primary populations were observed
together with the second Alcaligenes species achieving an equivalent abundance
with A. faecalis.  We are examining the precise nature of this adaptation with
one possibility being the selection of an organism with the potential to co-
metabolize 36DCPA.


     Continuous-flow culture techniques provide the most suitable experimental
systems for isolation and analysis of interacting mixed cultures.  An impor-
tant feature of continuous cultures is that the method ensures that interact-
ing associations are selected provided that as a group the association was
more competitive than individual organisms growing under the same conditions.
Furthermore, one of the properties of open growth systems is that other orga-
nisms not involved in the required mixed culture are competitively selected
against and rapidly eliminated from the growth vessel (34, 36).   In addition,
continuous-flow culture enrichments have an important advantage in great
flexibility of the selection conditions which can be chosen.  For example, by
controlling the flow rate of fresh medium into the growth vessel, different
growth rates may be predetermined.  It is now well established from pure cul-
ture studies that this parameter is particularly significant in determining


the outcome of continuous enrichments (15, 21, 22, 25).   It is also feasible to
alter other environmental conditions such as the nature of the growth-limiting
substrate and its concentration and physical parameters, although so far there
have not been any systematic studies on the influence of these factors on the
types of mixed culture selected for.  Finally, open growth systems can be
carefully controlled with the possibility that stable, steady-state conditions
may be established  (7).

     The possibility of applying continuous-flow culture systems to problems
in microbial ecology have been recognized for some time (e.g. 15, 21) and the
development of these techniques should provide valuable tools in attempting to
understand the growth and behavior of microorganisms in nature.  It can be
argued that these culture systems can closely reflect the conditions occurring
in nature and future developments, for example,  developing systems able to
contain particulate materials  (8), should increase their relevance to such


     I wish to acknowledge the work of my colleagues and many informative dis-
cussions with them.  I thank the Science Research Council, the Dow Chemical
Co., Ltd., the Engineering and the Royal Society for financial support of work
in my laboratory.

                              LITERATURE CITED

1.  Balashova, V. V., I. Y. Vedenina, G. E. Markosyan, and G. A.  Zavarzin.
     1973.  The auxotrophic growth of Leptospirillum ferrooxidans.  Mikro-
     biologiya 43:581-585.

2.  Bates, J. L., and P. V. Liu.  1963.   Complementation of lecithinase ac-
     tivities by closely related pseudomonads:  its taxonomic implication.
     J. Bacteriol. 86:585-592.

3.  Bergter, F., and D.  Noack.  1966.  Ein mikrobiologisches beispiel fur
     oszillationen in einer zwel-arten-gesellschaft.  Studia Biophys.

4.  Brock, T. D.  1966.   Principles of microbial ecology.   Prentice-Hall,
     Inc., Englewood Cliffs, N.J.

5.  Bryant, M. P., L. L. Campbell, C. A. Reddy,  and M. R.  Crabill.  1977.
     Growth of Desulfovibrio in lactate  or ethanol media low in sulphate in
     association with hydrogen utilizing methanogenic bacteria.  Appl.
     Environ. Microbiol. 33:1162-1169.

6.  Bryant, M. P., E. A. Wolin. M. J. Wolin, and R. S. Wolfe.  1967.   Methano-
     bacillus omelianskii, a symbiotic association of two  species of bacteria.
     Arch.  Mikrobiol. 59:20-31.

 7.  Bull, A. T.  1975.  Microbial growth, p. 415-442.  In A. T. Bull, J. R.
      Lagnado,. J. O. Thomas and K. F. Tipton (eds.), Companion to biochemistry.

 8.  Bull, A. T., and R. E. Ashby.  1977.  A column fermenter for the continu-
      ous cultivation of microorganisms attached to surfaces.  Lab. Prac.

 9.  Cappenberg, Th. E.  1975.   A study of mixed continuous cultures of sul-
      phate-reducing and methane-producing bacteria.  Microbial Ecol. 2:60-72.

10.  Chao, C. C., and P. J. Reilly.  1972.  Symbiotic growth of Acetobacter
      suboxydans and Saccharomyces carlbergansis in a chemostat.  Biotech.
      Bioeng. 14:75-92.

11.  Clarke, P.  H.  1974.  The volution of enzymes for the utilization of novel
      substrates, p. 183-217.  In M. J. Carlile and J. J. Skehel (eds.), Evolu-
      tion in the microbial world.  24th Symposium of the Society for General

12.  Daughton, C. G., and D. P. H. Hsieh.  1977.  Parathion utilization by bac-
      terial symbionts in a chemostat.  Appl. Environ. Microbiol. 34:175-184.

13.  Gray, B. H., C. F. Fowler, N. A. Nugent, N. Rigopoulos, and R. C. Fuller.
      1973.  Reevaluation of Chloropseudomonas ethylica.   Int. J. Syst.
      Bacteriol. 23:256-264.

14.  Gunner, H.  B.,  and B. M. Zuckerman.  1968.  Degradation of 'Diazinon' by
      synergistic microbial action.  Nature (London) 217:1183-1184.

15.  Harder, W., and H. Veldkamp.  1971.  Competition of marine psychophilic
      bacteria at low temperatures.  Antonie van Leeuwenhoek 37:51-63.

16.  Harrison, D. E. F., and S. Wren.  1976.  Mixed microbial cultures as a
      basis for future fermentation processes.   Process Biochem. 11:30-32.

17.  Hartley, B. S.   1974.  Enzyme families, p. 151-182.   In M. J.  Carlile and
      J. J. Skehel  (eds.), Evolution in the microbial world.  24th  Symposium of
      the Society for General Microbiology.

18.  Horvath, R. S.   1972.  Microbial co-metabolism and the degradation of or-
      ganic compounds in nature.  Bacteriol. Rev. 36:146-155.

19.  Horvath, R. S., and B. W.  Koft.  1972.  Degradation of alkyl benzene sul-
      fonate by Pseudowonas species.  Appl. Microbiol. 23:407-414.

20.  lanotti, E. L., D. Kafkewitch, M. J. Wolin, and M. P. Bryant.   1973.  Glu-
      cose fermentation products of Ruminococcus albus grown in continuous cul-
      ture with Vibrio succinogenes:  changes caused by interspecies transfer
      of hydrogen.  J. Bacteriol. 114:1231-1239.

21.  Jannasch, H. W.  1967.  Enrichment of aquatic bacteria in continuous cul-
      ture.  Arch. Mikrobiol. 59:165-173.

22.   Jannasch,  H.  W.  1968.  Competitive elimination of Enterobacteriaceae
      from seawater.  Appl. Microbiol. 16:1616-1618.

23.   Jensen, H. L.  1957.   Decomposition of chloro-substituted aliphatic acids
      by soil bacteria.   Can. J.  Microbiol. 3:151-164.

24.   Johanides, V., and D. Hrsak.  1976.  Changes in mixed bacterial cultures
      during linear alkylbenzensulphonate  (LAS)  biodegradation, p. 426.  In
      H. Dellweg  (ed.),  5th International Symposium on Fermentation.

25.   Kuenen, J. G., J. Boonstra,  H.  G. J. Schroder, and H. Veldkamp.  1977.
      Competition for inorganic substrates among chemoorganotrophic and chemo-
      lithotrophic bacteria.  Microbial Ecol.  3:119-130.

26.   Magee, R.  D., J. F. Drake, A.  G. Fredrickson, and H.  M. Tsuchiya.  1972.
      Studies in intermicrobial symbiosis:  S. cerevisiae  and L.  casei.
      Can. J. Microbiol. 18:1733-1742.

27.   Markosyan, G. E.  1973.  A new mixotrophic sulfur bacterium developing in
      acid media,  Thiobacillus organoparus sp. N.  Doklady Akademii Nauk SSSR

28.   Meers, J.  L.   1973.  Growth of bacteria in mixed culture.  CRC Crit. Rev.
      Microbiol. 2:139-184.

29.   Munneke, D. M., and D. P. H. Hsieh.  1974.   Microbial decontamination of
      parathion and p-nitrophenol in aqueous media.  Appl.  Microbiol.

30.   Munneke, D. M., and D. P. H. Hsieh.  1975.   Microbial metabolism of a
      parathion-xylene pesticide.  Appl. Microbiol. 30:575-580.

31.   Nurmikko,  V.   1954.  Symbiosis experiments concerning the production and
      biosynthesis of certain amino acids and vitamins in  associations of lac-
      tic acid bacteria.  Ann. Acad. Sci. Fenniciae 54:7-58.

32.   Osman, A., A. T. Bull, and J.  H. Slater.   1976.  Growth of mixed micro-
      bial populations on orcinol in continuous culture, p. 124.   In H. Dell-
      weg  (ed.), 5th International Symposium on Fermentation.

33.   Schopfer,  W.  H.  1943.  In Plants and vitamins.  Chronica Botanica,
      Waltham,  Mass.

34.   Senior, E., A. T. Bull, and J.  H. Slater.  1976.  Enzyme evolution in a
      microbial community growing on the herbicide Dalapon.  Nature (London)

35.   Slater, J. H., and A. T. Bull.   1978.  Biochemical basis of microbial
      interactions.  Ann.  Appl. Biol. (in press).

36.   Slater, J. H., and A. T. Bull.   1978.  Interactions between microbial
      populations, p. 181-206.  In A. T. Bull and P. M. Meadow (eds.), Com-
      panion to Microbiology.


37.   Stirling, L. A., R.  J. Watkinson,  and I.  J.  Higgins.   1976.   The micro-
      bial utilization of cyclohexane.   Soc.  Gen.  Microbiol.  4:28.

38.   Wilkinson, T. G.,  H. H. Topiwala,  and C.  Hamer.  1974.   Interactions in a
      mixed bacterial population growing on methane in continuous culture.
      Biotech. Bioeng.  16:41-47.

39.   Wilson, T. E.,  and P. W. Goaz.   1959.  The oral yeast-Lactobacillus rela-
      tionship.  J.  Dent. Res. 38:1044-1051.

40.   Wolfe, R. S., and N. Pfennig.  1977.  Reduction of sulphur by Spirillum
      5175 and syntrophism with Chlorobium.  Appl. Environ.  Microbiol.

41.   Wu, T. T., E. C. C.  Lin, and S. Tanaka.   1968.  Mutants  of Aerobacter
      aerogenes capable of utilizing xylitol as a novel carbon source.
      J. Bacteriol.  96:447-456.

42.   Yeoh, H. T., H. R. Bungay, and N.  R. Kreig.   1968.  A microbial inter-
      action involving a combined mutualism and inhibition.   Can. J.
      Microbiol. 14:491-492.


                       P.  W.  Rodgers  and  J. V.  DePinto
              Department of Civil  and Environmental  Engineering
                       Clarkson College of Technology
                              Potsdam, NY 13676

                A procedure for systematically  developing a predic-
           tive model for synthetic  organics  is demonstrated  and
           discussed.  Specifically  a  conceptual model  illustrating
           the pathways of an organic  in  an aquatic  ecosystem is
           presented.  This conceptual model  is further translated
           into a time-varying mathematical framework.   This  poten-
           tial predictive model requires extensive  field and labo-
           ratory experimentation.  Two experimental approaches, a
           Dual Culture Diffusion Apparatus and a  Two-Stage Con-
           tinuous Flow System, are  suggested for  providing certain
           descriptive and kinetic data necessary  to the predictive

     There exists a massive number  of industrially generated  synthetic or-
ganics, possessing widely different chemical  natures, which are released  into
our ecosystem.  Eight hundred million pounds  of  pesticides are applied an-
nually in the United States and their use  has risen  rapidly throughout the
world.  Presently, the external costs of using these pesticides may be esti-
mated as high as two to three billion dollars annually  and identifiable human
pesticide poisonings are nearly 60,000 per year  (18).   Over 700 synthetic or-
ganics have so far been identified  in drinking waters and many of  these chemi-
cals have been found to be carcinogenic in animals or to exhibit toxic proper-
ties  (1).  Many of these organics  (PCBs and DDT,  for example) have been de-
tected in many remote areas of the  world (8, 16,  17, 22,  24) , indicating that
these organics may have universally contaminated our biosphere.

     Thus far, the mechanisms responsible  for the transport of synthetic  or-
ganics are not accurately described and quantified.  Even  the analytical  mea-
surement and biological consequences of these organics  are not fully  eluci-
dated.  The present situation requires effective methods for  evaluating the
potential hazards of the many organics now being produced.  The development


and application of a deterministic model, simulating the natural phenomena af-
fecting the fate of organics within a mathematical framework, would facilitate
the necessary evaluations.

     There currently exists a number of conceptual and developing mathematical
models intended to suggest the fate of organics  (7,9, 16, 24).   The conceptual
models have succeeded in identifying and assimilating many potential pathways
of synthetic organics.  For the most part, these general pathways are the same,
regardless of whether a given organic is of natural or synthetic origin.

     In the process of translating these conceptual models to descriptive, de-
terministic models it becomes evident that although the pathways may be simi-
lar the experimental quantification of controlling processes and the subse-
quent calibration of the model are the two remaining tasks in developing pre-
dictive capabilities.  The kinetics which govern these processes remain, for
the most part, unknown quantities requiring comprehensive investigation.

     The scope of this paper is to demonstrate a methodology for constructing
a model describing the fate of organics.  This model emphasizes the need for
experimental work and the necessity for kinetic data.  The paper offers two
experimental apparatus which may be utilized for quantifying some of the pro-
cesses .

                              CONCEPTUAL MODEL

     A conceptual model recognizes known concepts of an ecosystem and proceeds
to translate them into an interacting functional diagram (9).  Conceptual
models are the first step in organizing scientific knowledge and hypothesis
for the purpose of predicting the fate and effects of an agent released into
the environment.  These models may provide answers to questions concerning the
variables requiring measurement, the accuracy required of the measurement, and
the frequency and location of sampling  (9).   Conceptual models also serve as a
guideline for the development of mathematical models.

     The construction of a conceptual model describing the fate of organics
requires the identification of pertinent phenomena.  In Table 1 we see a syn-
opsis of the relevant processes, the parameters influencing these processes,
and the compartments in which they are active.  The specific nature of an or-
ganic will determine which processes and compartments will be utilized in a
particular model.

     For a functional diagram to be representative of a given ecosystem's dy-
namics and compatible with existing mathematical frameworks, it is often de-
sirable to further divide compartments into smaller box models  (3, 6, 14, 23).
A conceptual representation of a two-box model describing the fate of an or-
ganic within a lake ecosystem is illustrated in schematic form in Figure 1.
This schematic model represents a summer stratified lake, which includes all
the processes presented in Table 1.  The upper box represents the epilimnion
and the lower box the hypolimnion.  Within each box a given organic may be
either bound in a solid phase or in free solution.  Both boxes also contain
their own set of relevant sources and sinks for each organic form.  Mass
transport between boxes takes place by diffusion and/or sedimentation across
the thermocline.


                  Controlling parameters




    Chemical Decay

    Cultural Inputs
Chemical Structure, Ultraviolet Intensity

Activity Coefficients, Partial Vapor Pressure,
Diffusion Rates, Desorption Rates, Temperature

Hydrodynamic Diffusion and Dispersion,
Adsorption-Desorption, Degradation Rates, Evaporation

Heat of Adsorption, Functional Groups, Surface Area,
Water Solubility of Organic, Organic Content of Soil
pH and Temperature

Advection, Diffusion, Aerosol Properties,
Chemical Reactions, Meteorological Factors
Topographical Features

Chemical Nature of Organics, Diffusion, Adsorption,
Temperature, Microbial Dynamics.

Chemical Structure, Catalyzers, pH, Temperature

Agricultural, Industrial, and Political Practices








                      POINT SOURCE INPUTS

                          RAINFALL  FALLOUT  DIFFUSION
                            f-H	r
                                        CHEMICAL DECAY	
                                        BACTERIAL DEGRADATION -


                     LEACHING- -
                                        CHEMICAL DECAY	
                                        BACTERIAL DEGRADATION-

Figure 1.   Schematic outline  of two-box organics model for lakes.
                               DETERMINISTIC MODEL

     After  gaining a conceptual grasp of the  organic pathways  in  an ecosystem,
it is necessary to further  translate this scientific knowledge to a mathemati-
cal representation.  The  time-varying fate of an organic compound is mathe-
matically described by formulating a mass balance, which includes the proces-
ses listed  in Table 1.

     This mass balance approach has been used to translate the conceptual two-
box model in Figure 1 into  a mathematical framework.  Each process within a
box is represented as a mathematical term.  Since there are  two organic forms
(the solid  phase [PORG] and the solution phase [SORG]) and each may reside in
two boxes  (epilimnion and hypolimnion), a complete mass balance results in
four interdependent differential equations  (22).  (Nomenclature at end of
[SORG]  in  the Epilimnion
                                 (eq.  1)

                  Sum of Point Inputs
   Rainfall  or Washout
+  I A  [SORG]

                 Atmospheric Diffusion      Leaching

               + k,([SORG]   - [SORG]  )V   + k  [SORG] .V
                  I       S         G  G     J-J       J G

                 Diffusion                    Adsorption


               + -=2- Ath([SORG]h - tSORG]g) - ka[SORG]e6Ve


                 Evaporation                Chemical Decay

               - EP.M -106/18GP       -     k  [SORG] V
                   lew             c      e e

                 Biological Degradation     Photodecomposition

                                                                  — aO
                 k,[SORG]  V                 V k [SORG]  rte    -e
                  d      e e                 e p      e k H
                 Biological Partitioning    Outflow
                 Z U[Biomass]  [SORG]     -   Q[SORG]
[PORG]  in the Epilimnion                                                (eq.  2)

                 Sum of Point Inputs       Rainfall or Washout


   V  	—	 = Z Q  [PORG]              + I A [PORG]
    e    dt         n       n                 e      r

                 Fallout                   Adsorption

               + Mf-A                    + k [SORG] V
                     e                      a      e e

                 Biological Partitioning   Diffusion


               + Z u[Biomass]  [SORG]     +	A   ([PORG] -[PORG]  )
                             e      e      -    th       h       e


                 Sedimentation             Biological Degradation

               - 9 A^ [PORG]             - Z k, [PORG]  V
                  e tn       e                d       e e

                 Chemical Decay            Outflow

               - Z kc [PORG]gVe          - QJPORG]


[SORG]  in the Hypolimnion

                                                       (eq.  3)
      d[SORG]    k

   V.  	-	 = 	 A   ([SORG]  -  [SORG],)  +k  [SORG] .V


               - k  [SORGLGV,,
                  a      h  n
                 Biological Partitioning

               - E U[Biomassh [SORGL
                             h       h

                 Interfacial Diffusion
               + k.A  , ([SORG] . -  [SORG], )
                  i sd        j          h
[PORG] in the Hypolimnion

                - [PORG]h)
                  — Ath


                + k  [SORGL6V,
                   a      h  h
                  Biological Degradation
                - I kd[PORG]hVh
                  Loss to Sediments
                - Vsd  [P°RG]h
                                               Biological  Degradation
                            - kd[S°RG]h\
                              Chemical Decay
                            - k  [SORGLV
                               c      h  h
                                                       (eq. 4)
                              Biological Degradation
                            + I. U[Biomass]. [SORG].
                                           h       h
                              Chemical Decay
                            - E k  [PORG], V,
                                 c       h  h
Similar equations to the ones above could be derived  for  a  non-stratified win-

ter model, which would be represented by a one-box model.   The summer and win-

ter models would be joined by boundary conditions existing  at fall  overturn

and spring stratification.  At the fall overturn the  [SORG]  would be
                                         [SORG]  V

                                                                        (eq.  5)

and the [PORG]  would be

                          [PORG]   V  + [PORG],  V,
                                e  e         h  h
                          	r	_

The concentrations of the epilimnetic and hypolimnetic crganics,   [SORG]e,
[SORG]n, [PORG]e, [PORG]n, are those calculated at the end of the stratified
period  (or Two-Box Model).  These concentrations at the beqinninq of stratifi-
cation are set equal to the concentration of crganics ([PORG] or  [SOPG])  at
the conclusion of the winter one-box model.

     Any given mathematical representation offered in the differential equa-
tions is for the present a hypothetical representation and is not meant to be
a definitive statement.  However, some of these terms have been shown in the
literature to be a good representation of the process.  For instance, Snod-
grass and O'Melia (23) had success with their representation of diffusion and
sedimentation  (which took into account lake depth) when developing a model tc
predict the fate of phosphorus in Lake Ontario.  The term used for evaporation
was developed by Mackay et al. (15)  and provides a method for calculating the
potential evaporation for compounds whose vapor pressure and water solubility
are known.  The. adsorption term used in the model was formulated by Hamaker
and Thompson (11).  Hamaker (10)  also suggested the leaching term which de-
pends strictly on empirical data.  Photodecomposition rates could vary as a
function of light (or UV) intensity; therefore, a light extinction factor uti
lized in phytoplankton models was employed (3).

     Inspection of the deterministic model presented in equations 1 through 4
reveals that it is necessary to quantify many parameters and that knowledge of
the kinetics of these processes is of utmost importance.  Kinetic parameters
needing experimental evaluation include determination of the rate coefficients
describing bacterial degradation, adsorption, biological partitioning, chemi-
cal decay, photodecomposition and leaching.

     Experimentation dealing with certain processes may reveal that additional
parameters may have to be treated as state variables and modeled separately.
One such process may be the microbial degradation of organics.  The dynamics
of microbial populations may greatly affect the extent and rate of organic: de-
composition.  The specific assemblage of microorganisms will certainly deter-
mine the by-products of decomposition.  Environmental perturbations that would
not directly affect a specific organic pool could certainly effect the micro-
bial dynamics which in turn would influence the fate of the organic.

     The inclusion of bacteria as a state variable in a model describing the
fate of organics requires extensive experimental efforts to quantify processes
such as:  the uptake rates of soluble organics; bacterial specific growth
rates as a function of nutrient utilization;  stoichiometric bacterial biomass
production from soluble organics; the rate of bacteria-mediated by-product
formation; and the ability of bacteria to utilize organics adsorbed on par-
ticulate matter.  Several investigators have recently developed models which
are intended to model microbial growth and decomposition activity within an
aquatic ecosystem (2, 4, 5, 13).   In order to utilize and to improve these
microbial growth models extensive field and laboratory investigation must be


initiated in order to  supply  specific  data for  the  various  groups  of synthe-
tic organics.

                           EXPERIMENTAL APPROACHES

     Two experimental  approaches which lend themselves  to process  research and
kinetic evaluation of  organics  in  natural  water are proposed  and discussed in
this paper.  The  first approach employs a  Dual  Culture  Diffusion Apparatus
 (DCDA)  (Fig. 2).  The  DCDA allows  two  batch cultures to share the  same media
while keeping non-soluble substances separated  by a polycarbonate  membrane.
                                  MEMBRANE FILTER
                                     -AIR VENT-.
                                                      SIDE VIEW
                                                      TOP VIEW
Figure 2.  Diagram of Dual Culture Diffusion Apparatus  (DCDA).
     Thus far the DCDA has been utilized to measure  the  rate  of  inorganic
phosphorus regeneration due to bacteria-mediated and endogenous  decay  of phy-
toplankton.  The general approach has been to maintain one  side  of  the DCDA
in the dark  (the culture vessel) and the adjoining side  in  an appropriate  lit
environment  (the assay vessel).  The culture vessel  supported a  previously
grown algal culture, Scenedesmus sp., with or without a  bacterial community
present.  The assay vessel was then inoculated with  a phosphorus-starved
Scenedesmus culture and maintained in the light.  The intent  of  the experimen-
tal design was to'measure the rate of phosphorus mineralization  that might
occur in the dark culture vessel via bacterial decomposition  and algal respi-
ration.  To accomplish this purpose phosphorus fractions were periodically
sampled in the assay vessel.  Any increase in total  phosphorus was  assumed to
have originated from mineralization in the culture vessel which  had subse-

 quently diffused  through the membrane and had been accumulated by the P-
 starved Scenedesmus.  By calculating a mass balance on phosphorus in both ves-
 sels  of the DCDA  regeneration rates of phosphorus could be obtained.  This ex-
 perimental approach has two primary advantages over previous nutrient regen-
 eration studies.  In batch cultures it has been found that nutrient uptake is
 a  competitive parallel reaction with nutrient regeneration.  Therefore, samp-
 ling  a classical  batch culture for regenerated soluble nutrients allowed only
 the net regeneration rates to be obtained (21).  In comparison, the DCDA main-
 tains a competitive rate with nutrient uptake in the culture vessel, namely, a
 faster rate of diffusion of soluble nutrients.  When these diffused nutrients
 are immobilized in the assay vessel, rates which more closely represent gross
 regeneration rates are obtainable.  This technique also demonstrates directly
 that  by-products  of degradation  (in the culture vessel)  are subsequently ca-
 pable of being reassembled into algal biomass (in the assay vessel).

      The DCDA functionally has three rates which ultimately determine the
 change in particulate phosphorus in the assay culture:  a) the rate of phos-
 phorus mineralization in the culture vessel; b)  the rate of diffusion across
 the membrane which divides the vessels; and c) the rate of uptake of phos-
 phorus by the P-starved Scenedesmus in the assay vessel.  Phosphorus uptake
 was assumed to be very rapid relative to the other two rates and the presence
 of soluble phosphorus in the assay vessel to be minimal at all times (18) .  A
 mass  balance on the dark side (or culture vessel)  appears in equations 7 and 8.

      Accumulation of     _   rate of release   _   rate of diffusion
    soluble phosphorus        from biomass          across membrane

            dCi                                         DACi
          V —          =         W(t)        -        —           (eq. 8)

 For first order regeneration with initial particulate phosphorus = P0.

                                             f   -vl
                              Ci(t) V = P0 V |l-e    J                  (eq. 9)

 therefore the rate of release from the biomass would be

                                        -k t
                        W(t)  = kr PQ V e  r                           (eq. 10)

 Substituting eq.  10 into eq.  8,  the material balance becomes

                        DC                 -k t
                          1                  r
OCi = kr PQ e                           (eq.
where                           a = —

The solution to eq. 11 for Ci = 0, when t = 0 is

                                  k P     -k t
                         c> <« • -
     Since it is the lit side  (or assay vessel) which is sampled during an ex-
periment it is necessary to formulate a mass balance on the lit vessel for the
accumulation of algal particulate phosphorus (02), presented in equations 13
and 14:

 Rate of C2 accumulation = rate of Ci diffusion across membrane       (eq. 13)

                                 V -- = aCi                          (eq. 14)
Solution to eq. 14 for C2 = 0, when t = 0 is

                                        -        U-e-«,
If the value of a » k  then eq. 15 reduces to
                      r       ^
                        C2(t) = P
                                 o "—
(eq.  16)
     Equation 16 simply states that if the diffusion rate is significantly
greater than the release rate then the accumulation of particulate phosphorus
(C2) in the lit side would not be diffusion limited.  In Figure 3 a graphical
comparison of eq. 15 and 16, using laboratory data, is seen.  In this case
the lag in phosphorus accumulation in the assay vessel is the result of some
diffusion limitation and not due to a lag in phosphorus regeneration in the
culture vessel.

     Experimentation has shown that the rate of diffusion, a, from the culture
vessel (volume = 900 ml) across the membrane to the assay vessel of low levels
of phosphate (10-150 ugP/liter) is as high as 0.4 day"1.  By utilizing the
mass balance scheme presented above and assuming that a » k (utilizing eq.
16), we obtained first-order phosphorus regeneration rates that varied from
0.001-0.05 day"1.  The rates appear to be dependent on the physiological state
of the algae and the size fractionation of the decomposer community.

     We believe that the DCDA experimental approach presented above offers a
method for evaluating certain processes that determine the fate of organics
in aquatic ecosystems.  Adsorption or desorption rates could be measured with-
out disturbing the vessel housing the process of interest.  Suppose a certain
particle has been saturated with a particular organic and is placed in the
culture vessel.  Then an adsorbent with no sites previously occupied is placed
in the assay vessel, which is sampled periodically.  By utilizing the previous
mass balance approach the rate of adsorption in the assay vessel can be used


to calculate the rate of deadsorption in the culture vessel.
——a = 0.245 days
 Kr=0.015 days"1
 Kr=0.015 days"1
                           10    15    20    25    30    35
Figure 3.  Phosphorus regeneration from aerobic decompositi    f Scenedesmus
           in a DCDA.  Dashed and solid lines represent the predicted phos-
           phorus accumulation in the assay vessel for a first-order regenera-
           rate of 0.015 day"1 and diffusion rates of °° and 0.245 day"1, re-
     Adsorption rates could also be investigated in less time and utilizing
fewer vessels by placing the adsorbent in the culture vessel and spiking  the
vessels with a known quantity of organic and monitoring the decrease of or-
ganics in the assay vessel compared to a control.  This approach would be
feasible only if the rate of adsorption was not significantly higher than the
rate of diffusion.  Rates of adsorption for 2,4-D have been found to vary from
1.728 x 10~2 day"1 to 3.8 x 10": day"1 depending on the adsorbent clay  (11).

     Since specific by-product formation is highly dependent upon the particu-
lar microbial population and environmental conditions, the DCDA would offer a
method of study that would not require harvesting or disturbing the decompos-
ing culture.  If the decomposing culture, including the original spike of or-
ganic, were housed in the culture vessel then the assay vessel could be peri-
odically sampled and replaced with fresh media.  The withdrawn media could be
analyzed for the by-products while the decomposing culture is never disturbed.

     Other possible organic process-related applications of the DCDA include:
microbial mediated recycle or transformation of organics bound to particles;
biological partitioning and microbial degradation or release of these parti-
tioned organics; and possibly certain species interactions.  Hopefully addi-
tional applications will be recognized where separation of process phases are

desirable or where it would be advantageous not to disturb or harvest a cul-
ture in order to evaluate its dynamics.

     A second experimental apparatus which lends itself to the study of or-
ganic kinetics is the continuous-flow apparatus.  A continuous-flow culture
offers several advantages over a batch culture.  Since steady-state is reached,
a mass-balance describing the process is simplified from a differential equa-
tion to a linear algebraic equation where kinetic data can be more easily ex-
tracted.  To illustrate this point a hypothetical mass balance describing the
decomposition and endogenous respiration of algae results in

                       dA                        yRB
                     VTT=Q(A  - A) - V N A	— V                (eq. 17)
                       dt       o                 Y

By defining Q/V = D  (dilution rate) and considering a steady-state situation
(i.e., dA/dt = 0) the mass balance becomes

                     0 = D(S  - S) - N S	—                       (eq. 18)

Clearly the above equation is easily analyzed and offers a number of opportu-
nities for linearization.  This system allows extraction of kinetic data which
is directly applicable to time-varying models such as the deterministic model
presented previously.

     Our experimental work has expanded the classical flow-through design to
include two stages.  The first stage is fed algal growth media at a constant
rate and overflow.   Steady-state is evaluated with in vivo Chlorophyll "a11,
electronic particle  counts and particle volume concentration determinations.
The spent media from the first-stage flow-through growth culture overflows
into a second-stage  flow-through culture where growth is prohibited by a dark-
ened environment and algal respiration and bacterial metabolic activity in-
fluences a change in total biomass and physiological state of the algae.

     Two-stage continuous-flow cultures could be employed to simultaneously
follow phytoplankton uptake of a given organic compound in the first stage
and bacteria-mediated release and/or degradation of the biologically bound
(partitioned)  organic in the second stage.

     The continuous-flow system also allows the investigator to assay the ef-
fect an organic would have on the growth rate, physiological state and nutri-
ent utilization of a range of microorganisms.  This is accomplished because
the feed characteristics and dilution rate force a culture with a defined
mean physiological state.

     Furthermore, the continuous-flow system offers a system which would fa-
cilitate the modeling of bacteria as a state variable.  The growth rates and
population dynamics could be obtained as a function of the feed composition.
Media characteristics,  which might include presence of by-products and persis-
tence of different forms of organics as a function of residence time and mi-
crobial assemblage, could be measured.  The chernostate has long been used for

evaluation of bacterial growth kinetics and has the potential for assisting
the development of a sub-model describing the dynamics of a decomposer commu-
nity within an aquatic system.


     Because of the ever-increasing production of new synthetic organic chemi-
cals, the prediction of the fate of these materials in our ecosystems has be-
come a health and economic necessity.  In this paper we have shown that a
mathematical ecological model offers an efficient method of tracing the fate
of synthetic organics discharged into aquatic environments.  The usefulness of
the model proposed herein is twofold.  First, it may be used to correlate
field and laboratory experimental data.  In this way the model can be used as
a research tool to establish a feedback loop between process experimentation
and overall model development.  Second, the verified model provides a basis
for making informed decisions regarding agricultural, industrial, and munici-
pal management of synthetic organic chemicals.

     Further refinement and development of the proposed model will not only
include evaluation of rate coefficients, but expansion of the scope of the
model to include specific biological functional groups (such as bacteria, phy-
toplankton, and zooplankton) as state variables.  In this way we can better
predict the fate of an organic in addition to evaluating any perturbation of
the aquatic community due to the compound in question.  The ultimate goal of
such a model is to be able to make management decisions regarding a given syn-
thetic chemical before it is released into the environment.

     Another goal of this paper was to point out the pressing need for kinetic
data for the development of models such as the one proposed herein.  The DCDA
and the continuous-flow system (single or multi-stage), whose operation has
been demonstrated, are experimental tools which may provide a method for char-
acterizing and quantifying some of the processes governing the fate of or-
ganics.  These experimental techniques combined with many others, in the labo-
ratory and in the field, will make the development of a predictive model for
organics a feasible goal.


     We gratefully acknowledge the help of Fred McKosky and James Bonner in
development of experimental approaches.  Research associated with this paper
was, in part, sponsored by the Large Lakes Research Laboratory, Grosse lie,
Michigan.  EPA grant number R-804937.

A    = surface area, L
E    = water evaporated, M/T
e    = epilimnion subscript
g    = sedimentation coefficient,  L/T
G    = grams of water containinq Me,  M

h    = hypolimnion subscript
II    = depth of euphotic zone-, L
I    - intensity of rain, L/T
Ia   = li'jht intensity at surface
Is   = optimum intensity at surface
j    = soil origin subscript
ka   = adsorption rate constant, 1/T
kc   = first order chemical decay coefficient, 1/T
kd   = first order biological degradation coefficient, 1/T
kf   = overall diffusion coefficient, 1/T
kg   = extinction coefficient of light  (or UV),  M/L'
ki   = mass transfer coefficient, L/T
kL   = (cm moved by organic)/(cm of water entering the soil)(time),  1/T
kp   = first order photodecomposition coefficient, 1/T
kth  = roean vertical eddy diffusion coefficient, L2/T
L    = length
M    = mass
Me   = grams of organic compound, M
Mf   = fallout flux, M/L2-T
n    = input from point sources subscript
N    = endogenous respiration rate, 1/T
P    = vapor pressure of organic compound  (mm Hg)
Pw   = vapor pressure of water  (mm Hg)
PORG = organics in particulate or solid phase
Q    = flow, L3/T
r    = rain input subscript
S    = saturation value subscript
sd   = sediment subscript
SORG = organics in free solution phase
T    = time
th   = thermocline subscript
                                        T  O
U    = second order accumulation rate, L -L /M-T
V    = volume, L
z^h  = average depth of thermocline, L
 [ ]  = concentration, M/L3
6    = the fraction of adsorptive area uncovered
aO   = Ia/ls
al   = la/Is exp(-kg-h)

                              LITERATURE CITED

1.  Anonymous.  1978.  EPA proposes far-reaching regulations for reducing syn-
     thetic organics in drinking water.  Series on Water Resources:   No. 5,
     Civil Eng. ASCE, p. 69-71.

2.  Barsdate, R. J., R. T. Prentki, and T. Fenchel.  1974.  Phosphorus cycle
     of model ecosystems:  significance for decomposer flood chains and effect
     of bacterial grazers.  Oikos 25:239-251.

3.  Bierman, V. J., Jr.  1976.  Mathematical model of the selective enhance-
     ment of blue-green algae by nutrient enrichment, p. 1-32.  In R. P.
     Canale  (ed.), Modeling biochemical processes in aquatic ecosystems.  Ann
     Arbor Science Publishers, Ann Arbor, Michigan.

 4.   Clesceri,  L.  S.,  R.  A.  Park,  and J.  A.  Bloomfield.   1977.   General model
      of microbial growth and decomposition  in aquatic ecosystems.   Appl.
      Environ.  Microbiol. 33:1047-1058.

 5.   Cordeiro,  C.  F.   1975.   Mathematical modeling of lake systems  with nutrient
      cycling.   Ph.D.  dissertation,  University of Notre Dame,  Notre Dame,  Ind.

 6.   Crews,  W.  B., J.  W.  Brewer,  and T.  J. Petersen.   1975.   Modeling of atmos-
      pheric behavior:  a submodel of the dynamics of pesticides, p. 79-96.
      In R.  Hague  and V.  H.  Freed (eds.), Environmental dynamics of pesticides.
      Plenum Press, New York.

 7.   Davidson,  J.  M.,  and J.  R. McDougal. 1973.   Experimental  and  predicted
      movement  of  three herbicides in a  water-saturated soil.   J. Environ.
      Quality 2:428-433.

 8.   Durham, W. F., and J.  F. Armstrong.   1961.   Insecticide content of diet
      and body  fat of Alaskan natives.   Science  134:1880-1881.

 9.   Gillett, J. W.,  J. Hill IV,  A.  W. Jarvinen,  and  W.  P.  Schoor.   1974.  A
      conceptual model for the movement  of pesticides through  the environment.
      Ecological Research Series,  U.S. Environmental  Protection Agency

10.   Hamaker, J. W.  1975.   The interpretation of soil leaching experiments,
      p. 115-133.   Jn R.  Haque and V. H.  Freed (eds.), Environmental dynamics
      of pesticides.   Plenum Press,  New  York.

11.   Hamaker, J. W.,  and J.  M. Thompson.   1977.   Adsorption, p. 49-144.  In
      C. A.  I.  Goring and J.  W. Hamaker  (eds.),  Organic chemicals in the soil
      environment, Vol. 1.   Marcel Dekker, Inc.,  New  York.

12.   Haque,  R., F. T.  Lindstrom,  V.  H. Freed,  and R.  Secton.   1968.   Kinetic
      study of  the sorption of 2,4-D on  some clays.   Environ.  Sci.  Technol.

13.   Krogius, F. V.,  E. M.  Krokhin,  and  V. V.  Menshutkin.   1972.  The modeling
      of the ecosystem of Lake Dalnee on an  electronic computer, p.  149-164.
      In Z.  Kajak  and A.  Hillbricht-Dkowska  (eds.), Productivity problems  of
      fresh waters.

14.   MacCracken, M. C., T.  V. Crawford,  K. R.  Peterson,  and  J.  B. Knox.  1971.
      Development  of a multi-box  air pollution model  and initial validation
      for the San  Francisco Bay Area. URCL-73348, Lawrence  Radiation Labora-
      tory,  Livermore, California.

15.   Mackay, D., and W. W.  Wolkoff.   1973.  Rate of evaporation of  low-solu-
      bility contaminants from water bodies  to atmosphere.   Environ. Sci.
      Technol.  7:611-614.

16.   Nisbet, I. C. T., and A. F.  Sarofim. 1972.   Rates and  routes  of transport
      of PCBs in the environment.   Environ.  Health Perspect.  1:21.


17.  Peterle, T. J.  1969.  DDT in Antarctic show.  Nature 224:620.

18.  Pimentel, D.  1978.  Benefits and costs of pesticide use in U.S. food
      production.  Unpublished MS.

19.  Rhee, G-Y.  1973.  A continuous culture study of phosphate uptake,  growth
      rate and polyphosphate in Scenedesmus sp.  J. Phycol. 9:495-506.

20.  Risebrough, R. W., R. J. Huggett, J. J. Griffin, and E. D. Goldberg.
      1968.  Pesticides:  transatlantic movements in the northeast trades.
      Science 159:1233-1236.

21.  Rodgers, P. W.,  and J. V. DePinto.  1976.  Laboratory study of freshwater
      algae-bacteria interactions and accompanying nutrient dynamics.  Mas-
      ter *s thesis, Clarkson College of Technology, Potsdam, N.Y.

22.  Sladen, W. J. L., C. M. Menzie, and W. L. Reichel.  1966.  DDT residues
      in Adelie Penguins and a Crabeater Seal from Antarctica.  Nature 210:

23.  Snodgrass, W. J., and C. R. O'Melia.  1975.  Predictive model for phos-
      phorus in lakes.  Environ. Sci. Technol. 9:937-944.

24.  Woodwell, G. M., P. P. Craig, and H. A. Johnson.  1971.  DDT in the bio-
      sphere:  where does it go?  Science 174:1101-1107.


                   Jeffrey M.  Giddings,  Barbara T.  Walton,
                  Gerald K.  Eddlemon,  and Kathleen  G.  Olson
                       Environmental Sciences' Division
                        Oak  Ridge  National Laboratory
                            Oak Ridge, TN  37830

                Polycyclic aromatic hydrocarbons  (PAH),  many of which
           are known carcinogens,  are expected to  be  important compo-
           nents of wastewater from coal  conversion facilities.   No
           large-scale conversion  facilities  are currently  in opera-
           tion, and information about the environmental behavior of
           PAH is scarce.   Microcosms which mimic  the structure and
           function of pond ecosystems can provide some  of  this in-
           formation.   Two 80-liter pond  microcosms were treated with
           0.5 yg/liter [9-14C]  anthracene, a representative PAH, to
           measure:   the rate of sorption by  sediments,  bioaccumula-
           tion by several aquatic organisms,  and  transformation of
           anthracene by abiotic processes and by  aquatic organisms,
           including bacteria.  Samples of water,  sediment,  and orga-
           nisms were removed periodically for determination of llfC
           activity.  Additional samples  were extracted  with organic
           solvents for separation and quantitation of anthracene and
           transformation products by thin-layer chromatography and
           radioautography.  Anthracene disappeared rapidly from pond
           water but accumulated in sediments. Over  a 12-week period,
           80% of the anthracene was transformed,  mainly by photo-
           lysis and biological activity.   An unidentified  derivative
           of anthracene persisted in all microcosm components exam-
           ined, while anthracene  and another degradation product
           persisted in the upper  layer of sediment.
      Research sponsored by the Division of Biomedical  and Environmental Re-
search, U.S. Department of Energy,  under contract W-7405-eng-26  with Union
Carbide Corporation.   Publication No.  1200, Environmental  Sciences  Division,


     A major aspect of our national energy program is development of technolo-
gies for converting coal to liquid and gaseous fuels.  The aqueous effluents
from coal conversion facilities are expected to include hundreds or thousands
of organic compounds, including phenols, aromatic amines, monoaromatic hydro-
carbons, thiophenes, and polycyclic aromatic hydrocarbons  (PAH).  PAH are of
particular concern for several reasons:  (a) many are carcinogenic; (b) they
may not be removed efficiently by biological waste treatment;  (c) some appear
to be resistant to degradation; (d) their environmental fate is practically
unknown  (4) .

     We have selected anthracene, a 3-ring PAH, for intense study as a repre-
sentative compound of this class.  Compared with other PAH, anthracene is in-
termediate in size (M.W. = 188), solubility in water  (73 yg/liter, [6]), and
octanol:water partitioning coefficient (2 x 10 \ [5]).  The characteristics of
anthracene which determine its fate in aquatic environments have been summar-
ized by Southworth (7).  Anthracene is moderately volatile, with a predicted
volatilization half-life of 300 h in a 1-m deep water column under quiescent
conditions.  It sorbs strongly to suspended particulate organic matter  (K^ =
concentration on particulates/concentration in water = 25,000), an order of
magnitude less to clay particles (K^ = 1600) , and only slightly to silt par-
ticles  (KD = 100) (7) .  Anthracene is rapidly degraded in sunlight.  The photo-
lytic half-life in a 2-cm column of distilled water is approximately 35 min
under midday midsummer sunlight at 35°N.  Because photolysis is a function of
light intensity, absorption of light by the water would probably limit photo-
lysis to the upper 100 cm in most natural waters (7).  Hydrolysis of anthra-
cene has not been found to occur.

     Microbial degradation of anthracene has been measured in short-term radio-
tracer  experiments.  A mixed culture of microbial strains derived from soil
around  an oil drilling site rapidly converted anthracene to polar compounds,
with over 90% conversion within 90 min (3).  In samples of sediment from an
oil-contaminated stream, anthracene was readily broken down  (half-life = 12 d),
but only 10% of the radiotracer appeared in the polar fraction.  In sediment
from an uncontaminated stream, the degradation rate was an order of magnitude
slower  (2).

     These basic studies on the fate of anthracene were all short-term, one
week or less, measurements of single processes.  They leave several questions
unanswered:  What is the fate of anthracene when the various transformation
processes are acting simultaneously?  What is the fate of anthracene in a sys-
tem containing not one, but many, functional components of an aquatic ecosys-
tem?  What is the fate of anthracene and its transformation products over a
longer  period of time?

     We have now examined the fate of anthracene over a 12-week period in
laboratory microcosms modelled after pond ecosystems.  These microcosms pre-
serve the natural complexity of the decomposer and autotroph communities, and
persist in stable condition for six months or longer  (1).  This paper describes
the potential long-term fate of anthracene in an aquatic ecosystem with many of
the biotic and abiotic transformation processes occurring simultaneously.


                            MATERIALS AND METHODS


     The microcosms were initiated in January, 1977, using water, sediment,
and organisms collected from a shallow pond.  Water was collected in buckets.
Sediment and masses of the dominant Elodea community were placed in tubs,
where large debris was removed.  Adiverse animal community was included in
both the sediment and Elodea collections.

     Two 80-liter slate-bottomed glass aquaria were acid-washed and placed in
an environmental chamber.  A 5-cm layer of sediment was added to each aquarium;
a band of opaque tape prevented light from reaching the sediment from the
sides.  Pond water was then slowly siphoned into each aquarium through a home-
made diffusion device to a depth of 30 cm.  One 100-g (drained wet weight)
portion of the Elodea community was placed in each microcosm; a few roots were
pushed into the sediment to anchor the plants.  The microcosms were covered
with glass plates to reduce evaporation.  The chamber temperature was ini-
tially set at 8°C  (the ambient temperature of the pond),  raised to 12°C in
April, and to 16°C in May.  A bank of cool white fluorescent lights provided
16,000 lux on a 12-h-light:12-h-dark cycle.


     In mid-July, the microcosms were treated with [9-ll*C]  anthracene (speci-
fic activity = 32 yCi/iamole, Amersham/Searle).  Approximately 1 uCi of anthra-
cene was dissolved in 20 ml acetone and diluted to 500 ml with distilled water.
The solution was divided into two 250-ml aliquots.  One ml of each aliquot was
withdrawn for radioassay, and the remainder was mixed with about one liter of
microcosm water and slowly poured into the microcosm.  The initial concentra-
tion of anthracene in the water was calculated to be 0.5  yg/liter.


     Two 1-ml water samples were removed from each microcosm two to five times
weekly, placed in a scintillation vial with 10 ml Bray's  solution, and counted
on a Packard Tri-Carb Liquid Scintillation Spectrometer.   Samples of sediment,
Elodea, and filamentous algae were taken approximately once a week.  Only the
upper 1 cm of sediment was collected, except for the final samples in which
entire cores were removed and sectioned.  Samples of sediment and organisms
were dried overnight at 55°C, weighed, oxidized with a Packard Tri-Carb Sample
Oxidizer, and counted.

     The final samples were taken 70-84 days after anthracene addition.  In
addition to the samples described above, periphyton from the aquarium walls,
snails, snail eggs, zooplankton, and water mites were collected.  The snails
were removed from their shells, and all samples were then dried and oxidized
as described above.


     Water samples for extraction and thin-layer chromatography  (TLC) were
taken from the microcosms 11 times during the course of the experiment.  Sedi-
ment samples for extraction were taken at the end of the experiment; only the
upper 2 cm were used.  Algae samples were removed and extracted on day 42.
Daphnia magna were abundant at the time of anthracene addition, and 10-12 in-
dividuals were removed daily until the population was depleted.

     Samples were extracted as follows:  a) Water samples were acidified with
a mineral acid and extracted twice with ethyl acetate in a 1:5 ratio of sol-
vent to water.  For each extraction, solvent-water mixtures were agitated for
45 min on a wrist-action shaker,  b) Wet sediment samples were extracted over-
night with acetone in a Soxhlet apparatus.  Acetone was evaporated from the
extracts, and the remaining liquid was extracted three times with equal vol-
umes of ethyl acetate.  c) Excess water was blotted from Daphnia and algae.
Samples were weighed, homogenized in acetone, and centrifuged.

     The extracts from each sample were pooled, dried over anhydrous calcium
sulfate, centrifuged, and reduced to 2 ml under nitrogen.  One ml of the ex-
tract was counted by liquid scintillation spectrometry to determine total ex-
tractable ll*C, and the remainder of the extract was subjected to thin-layer
chromatography on silica gel 60 TLC plates.  Radioautograms were prepared with
X-ray no-screen film.  Residues remaining after extraction of algae and sedi-
ment were oxidized to determine unextractable radioactivity.
                           RESULTS AND DISCUSSION


     The 14C activity in the water declined rapidly for the first 15 days, but
more slowly for 50 days thereafter (Fig. 1).  The mean first-order kinetic
rate constant was 0.055 day"1 for days 0-15 and 0.007 day"1 for days 20-65.
Rate constants determined for several metals in similar microcosms  (unpub-
lished data) ranged from 0.026 day"1   (selenium) to 0.23 day"1  (mercury).
There was a suggestion of accelerated l **C loss over the last 20 days of the

     Table 1 shows the X'tC activity in all microcosm components 12 weeks after
anthracene was added.  Approximately 30 cpm/ml (15% of the initial  ll*C) re-
mained in the water.  Most of the J4C accumulated in the upper 2 cm of sedi-
ment, with much lower activity below this level.   The total activity in the
sediment was estimated to be 10.2 x 106 and 8.0 x 106 cpm for the two repli-
cate microcosms.  The recovery of label from the microcosm (92-114%) indicated
that volatilization of 11+C compounds was negligible (Table 2).

     All organisms accumulated activity to about 103 times the activity in the
water.  In each microcosm, highest activities were in water mites and snail
tissue.  These animals probably represent the ends of the grazing and detrital
food chains, respectively, implying that anthracene or its degradation pro-
ducts may undergo biomagnification (accumulation up the food chain).


                                                       ORNL-DWG 78-5620
              200 c:
30    40    50
     TIME (days)
                            C Activity in Water.-Anthracene  Microcosms
Figure 1.   14C  activity (cpm/ml) in mocrocosm water.  Solid and dotted lines
            represent averages of duplicate samples from  replicates I and  II,

          TION  OF 10.4 x 106 cpm AS ANTHRACENE
Upper 2 cm
Lower 5 cm
Algae (benthic)
Water mites
Snails (tissue)
Snails (eggs)
Microcosm I



Dry weight
Microcosm II



             Water activity in cpm/ml,

                            TABLE 2.
                     *C BUDGET
                                          Anthracene microcosms
% Recovered






10 6













     Chromatographic analysis of water, sediment, Daphnia, and algal extracts
showed the presence of anthracene, four radiolabelled derivatives, and an un-
resolved polar fraction (Table 3).  Co-chromatography of sample extracts with
internal non-radioactive standards permitted identification of anthracene
(Rf = 0.84); however, Rf's of 9-anthraldehyde  (Rf = 0.40), anthraquinone
(Rf = 0.31), and anthracene-9-carboxylic acid  (Rf = 0.01) did not agree with
Rf values of radioactive unknowns.  The contribution of each compound to total
extractable activity in the water during the experiment is shown in Figure 2.
Only trace amounts of anthracene were recovered from the water after day 20.
The disappearance of compound 5 closely followed the disappearance of anthra-
cene, suggesting that anthracene was converted directly to compound 5, which
was not persistent.  Compound 3 was present in large quantity throughout the
experiment.  This compound constituted as much as 49% of the extractable ra-
dioactivity in the water (day 28), and although its relative abundance de-
creased thereafter, it was still present when the experiment was terminated on
day 84.  Forty-five percent of the total
on day 84.
                         C in the water was not extractable
                              Day 6
                              Day  42
Day 84
                         Daphnia   Water    Algae   Water    Sediment   Water

                                                         ORNL-DWG 78-6482
                                                       60   70
                                   20    30    40    50
                                          TIME (days)

                           Distribution of 14C Activity in Extracts from Water
Figure 2.  Distribution of 14C activity in extracts from microcosm water.
           Compounds are numbered as in Table 3.  Data are averages  from rep-
           licate microcosms.
     Distilled water, to which [9-11+C] anthracene was added, yielded anthra-
cene, a polar fraction, the four derivatives found in microcosm extracts,  and
an additional compound with Rf = 0.34, after 7 days in the environmental cham-
ber.  The latter compound, which constituted 14% of the radioactivity ex-
tracted from the distilled water, was never detected in microcosm extracts.
Figure 3 is a comparison of extractable fractions from microcosm water and
from distilled water.  After 7 days, the   C profile from distilled water  was
very similar to that of microcosm water on day 12.  Anthracene was nearly  ab-
sent from both systems.  The large polar fraction recovered from distilled
water  (45%) as well as from microcosm water  (48%), is particularly striking
since it indicates that anthracene is readily degraded to polar compounds  in
the absence of microcosm biota.  In studies of abiotic degradation of anthra-
cene, Southworth  (7) found that photolysis was rapid but hydrolytic breakdown
was slow; therefore, the anthracene transformation products detected in the
microcosm water and in the distilled water were probably due to photolysis.
                     C accounted for 40% of the activity in the upper  2  cm of
                                                                       At least
sediment, and 30% of the activity in the lower sediment, after 84 d.
part of this  "*C was probably incorporated into bacterial cells.  One-third of
the extractable activity, or 17% of the initial addition, was anthracene,  and
another large fraction was compound 5  (Table 3).  The polar fraction was only
16% of the extractable activity.  These findings are consistent with observa-
tions of Herbes and Schwall (2).  The data imply that microorganisms in the
sediment do not readily degrade anthracene, and that anthracene in the sedi-
ment could provide a source for continued exposure of benthic organisms to the
parent compounts.

     Daphnia were found to contain higher levels of extractable anthracene
than was found in water  (Table  3).  The distribution of activity in extracts

of algae was not markedly different from the distribution in water extracts
taken at the same time, except that more of compound 3 was present (Table 3)
Sixty-five percent of the 1
by organisms interacting with sediments (8).

     Pond microcosm experiments can provide  valuable information about the
long-term environmental fate of an organic pollutant.  In a microcosm, the
pollutant is subjected to the interactive effects of many abiotic and biotic
transformation processes.  Thus, the microcosms represent a level of complexity
between single-process experiments and actual environmental events, and pro-
vide one means (mechanistic models are another) of integrating simultaneous
processes.  Because pond microcosms are stable for many months (1), the per-
sistence of degradation and transformation products as well as the parent com-
pound can be assessed over extended periods  of time.  Microcosm experiments
with PAH and other constituents of coal conversion effluents will provide a
link between single-process laboratory studies and future studies in the

                              LITERATURE CITED

1.  Giddings, J.  M., and G. K. Eddlemon.  1977.  Some ecological and experi-
     mental properties of complex aquatic microcosms.  Presented at the Eco-
     logical Society of America Symposium on the Role of Microcosms in Eco-
     logical Research, East Lansing, Michigan, August 23, 1977.

2.  Herbes, S. E. , and L. R. Schwall.   1978.   Microbial transformation of poly-
     cyclic aromatic hydrocarbons in pristine and petroleum-contaminated sedi-
     ments.  Presented at the ASTM Conference on Sediment Microbial Activity,
     January 30,  1978, Fort Lauderdale, Florida.

3.  Herbes, S. E., L. R. Schwall, and G. A.  Williams.  1977.  Rate of micro-
     bial transformation of polycyclic aromatic hydrocarbons:   a chromato-
     graphic quantification procedure.  Appl. Environ.  Microbiol.  32:244-246.

4.  Herbes, S. E., G. R. Southworth, and C.  W. Gehrs.  1976.  Organic contami-
     nants in aqueous coal conversion effluents:  environmental consequences
     and research priorities, pp. 295-303.  In D. D. Hemphill, ed., Trace sub-
     stances in environmental health,  vol. 10.

5.  Leo, A. J.  1975.  Calculation of partition coefficients useful in the
     evaluation of the relative hazards of various chemicals in the environ-
     ment.  In G. D. Veith and D. E. Konascwich, eds.,  Symposium on Structure-
     Activity Correlations in Studies of Toxicity and Bioconcentration with
     Aquatic Organisms.  International Joint Commission,  Burlington, Ontario.

6.  Mackay, D., and W. Y. Shiu.  1977.  Aqueous solubility of polynuclear aro-
     matic hydrocarbons.  J. Chem. Eng. Data 22:399-402.

7.  Southworth, G. R.  1977.  Transport and  transformations of anthracene in
     natural waters:  process rate studies.   Presented at the ASTM 2nd Annual
     Symposium on Aquatic Toxicology,  Cleveland, Ohio,  October 1977.

8.  Walter, M. T., and H. Johnson.  1977.  A model system to study the desorp-
     tion and biological availability of PCB  in hydrosils, pp. 178-195.  In
     F.  L. Mayer  and J. L. Hamelink, eds., Aquatic toxicology and hazard
     evaluation.   ASTM STP 634.


                        BIODEGRADATION IN MICROCOSMS

MATSUMURA:  Microcosms are like the Internal Revenue Service; whenever someone
hears the word, they wince.  They are difficult to define, but we must try to
agree on something.  In the past, people have confused microcosms with model
ecosystems.  We have conveniently defined a model ecosystem as something that
we are trying to model containing more than one biological species.  We are
talking about microcosms in a much wider approach using any method which simu-
lates natural events.  Of course, there is little difference between bringing
some of the "poison" into the laboratory to study the complex field events and
taking something from nature and trying to study it simplified in a lab.  For
instance, when you attempt to study some compound underneath a fluorescent
light to determine photodegradation, by our definitions, this is a microcosm.
The idea is to simulate sunlight and study the compound's fate in this envir-
onment.  Basically there are four microcosm approaches we can think of:
1) the physical-chemical approach, often times called "benchmark" approach,
whereby scientists determine the chemical properties—lipid solubility, water
solubility, volatility, etc., whatever physical-chemical character—which may
affect the behavior of chemicals.  2) The second approach is a pure culture or
simplistic type approach whereby one strain of microorganism or one tube of a
photocell or whatever is utilized; this is not a complex system; you are
studying only one reaction at a time.  3) The third approach is the model eco-
system type whereby you have more than one species and you are trying to study
some complex parameter like accumulation.  4) The fourth approach is concerned
with mathematical modeling.  In this, scientists try to analyze the reactions
from a systems point of view.  Basically you can say that there is no differ-
ence in any scientific approach and anything you use to study field conditions
in the laboratory is a microcosm.  That is my point of view.  Of course, like
the blind man trying to assess what the elephant looks like, one is looking at
only one side.

P. ROGERS:  Dr. Clesceri, you talked of the validation of a model.  Can you
elaborate on what you feel validation of a model is?  Addressing yourself to
questions such as:  what kind of sampling is reasonable before one can say
that a model is verified?  Would you expect if there were a major perturba-
tion, in Lake George for instance, that a verified model, or your model, would
be capable of predicting the results of this perturbation, because it changes
an input?  And should a model be tested with different lakes before being de-
clared as verified?

CLESCERI:  First of all, of course our model is good, and yes, I think that it
could do a reasonably good job at predictions.  How often should one go out to
validate?  That's really a hard question; as far as I'm concerned, the more


the better.  I think the real problem is the tremendous seasonal variation.
With a seasonal type of sampling,  validating four times a year is nowhere near
enough.  I think the sampling variability is too great and one has to do an in-
tensive kind of sampling regime in order to get a good validation.  That's my
personal opinion, and perhaps I differ from other mathematical modelers.  Be-
cause I have come up with some of the field work.  Consequently, I feel quite
wedded to the field approach.  Your other question was, can the model apply to
different lakes?  I think so, yes, lakes of a similar class.  I would feel
that a model could be applied to oligotrophic,  or nesotrophic or eutrophic
lakes or what have you.  But the point is, it would be to a class of lakes,
and you would verify within that class.

WRIGHT:  I would think you have the goal of subjecting the system to a more
realistic perturbation.  For example, the effects of a drought, or the effects
of a period of heavy winds, a temperature change on the lake which leads to a
change from diatom dominance to blue-green algae dominance.  Is this your goal?

CLESCERI:  Very definitely.

WRIGHT:  Because these other perturbations are very unreal.

CLESCERI:  The reason I brought these out was to show the usefulness with re-
gard to this kind of experiment.  It simply never could be done, and it's al-
most an academic exercise.  The real usefulness of the model is to ask those
realistic questions such as the ones you mentioned.  Yes, clearly, that is our

BAUGHMAN:  I wonder if you have considered using models as an aid in arriving
at decisions if one has results from the kinds of tests (biodegradability, mi-
crobial activity, etc.) that we've been talking about in this meeting.  One of
the things we have not addressed is:  given results from these tests, how does
one draw a conclusion?  It seems fairly obvious that mathematical models can
be used to aid in the interpretation of these data or to arrive at a decision
of what should be done with a compound based on the kinds of effects that may
result therefrom.

CLESCERI:  This is where I think we all get cold feet.

BAUGHMAN:  Well, somebody is going to have awfully hot feet.  These are the
hard decisions that have to be made.  Can we use this approach to answer some
of these questions?

CLESCERI:  I would say, given the proper data,  and building the model on solid
data, yes!  I would go so far as to say they can be, and should be, used for
that kind of prediction.  But what worries me,  there are a lot of models that
are not built on proper data, and under those conditions I think we have to be
very careful.

FERGUSON:  There is a point you brought up I want to reiterate.  It is ex-
tremely important because a lot of management decisions are waiting for the
appropriate model so managers can decide what to do and what not to do in
terms of allowing corruption of the natural environment.  What many of the
more complex models are doing is trying to simulate a natural unimpacted


environment.  The great multitude of potential impacts precludes generating a
model which is able to make predictions on more than one of these potential
impacts.  It's important to list the kinds of impacts that the model you are
talking about is capable of making predictions on.  For example, dumping in a
particular toxic material is not going to be handled very readily in most of
the models we are generating.  I constructed a model of trophic dynamics in an
eel grass bed.  The problem was the possible effects of dredging on this bed.
While we can say if you dredge the eel grass bed, then it's gone; the model
wasn't really needed to make that kind of prediction.  Another possibility is
maybe the dredging caused some turbidity which reduced light penetration into
the eel grass bed.  The model we made didn't predict solar insulation of the
grass bed, so we said, let's pretend the grass was only growing at half the
rate.  That is an example of the kinds of problems we have.  We need to ex-
plain to government agencies that the models we develop are extremely limited
in their perspective.  That is a characteristic of models.  It has to be de-
signed to answer particular questions.  At this point, we don't know which
question is the most important to answer.  There are so many possibilities
that it's important when we do establish some sort of model, we make a list of
the potential capabilities in prediction.  I'm not sure where your model
stands in that regard.  I wonder if you could explain what kinds of perturba-
tion your model could potentially make predictions about?

CLESCERI:  I agree with everything you said.  This model can handle any range
of conditions for any one of those compartments.  As Dr. Wright mentioned,
changes that would happen due to a drought, could be predicted with this model.
What would happen, for example, if there were a large dumping of phosphate or
nitrogen or a sudden deficiency of oxygen?  These could be handled by this
model by simply varying those compartments.  Obviously, there is a rate with
each one of those changes which then would be coupled to the model.  I would
just like to mention, we are presently working on a pesticide model to be
coupled to this model.  Because of the solubility problems with toxic organic
materials, the question of where the toxics are going is extremely important.
So coupling with a pesticide model should produce that kind of information
and then we can predict what the ecosystem response will be.  First of all you
have to know where the toxics are going, and the pesticide model will provide

SIKKA:  Dr. Pritchard, I was wondering if you acidified and extracted the
water phase with methylene chloride.  I would guess the radioactivity in water
is partially due to p-nitrophenol which would have gone into the organic phase
had you acidified before extraction.

PRITCHARD:  Yes, our water samples were extracted under both basic and acidic
conditions, so I think we are getting all the p-nitrophenol out.  I'm unsure
as to the nature of the unextracted materials, but it could be pesticide bound
to biological materials such as bacteria.

STERN:  Incidentally, if I knew you were working on that first system we would
have insisted on greater support for your laboratory.

PRITCHARD:  If Tom Duke is here,  I hope he takes note of that.

STERN:  In the last slide, you indicated that your aged material in the


flowing system was more active than in the fresh,  and the converse is true in
the cored system.  It could be that in your cored system you get adsorption to
biomass or solid materials.  You may get adsorption to a level which might be
inhibitory whereas in your continuous system this might not happen.  So that
what you are seeing is a localized concentration to inhibitory levels in your
biomass and therefore a drop of viability of your cells.

PRITCHAKD:  We have seen very little inhibitory effects by methyl parathion
except at very high concentrations (10 ppm)  and I therefore feel toxic effects
could not account for these differences.  Also, since continuous flow experi-
ments have not yet been performed with an intact core,  it is difficult to an-
swer your question.  What we may be seeing is once you scoop out the sediment
and water and put it into the continuous flow system, it may require an accli-
mation period before normal conditions are obtained.   Thus, the faster rate
might be the more indicative of a natural environmental condition.

STERN:  The other point is that in a continuous flow you may need a certain
adaptation period before you start to get things happening.

PRITCHARD:  No, it seems that that was not the case and that's what surprises
me.  Some of our data indicate degradation occurs  very rapidly once you start
the experiment.  I feel we need to take a close look at those initial rates.
These may be the most reflective rates of what is  actually going on in the
environment.  I would also like to make a comment regarding bacteria in both
the coring and the flow-through systems.  We have  been unsuccessful in our at-
tempts to isolate bacteria which will degrade methyl parathion,  amino methyl
parathion. or p-nitrophenol.  We tried a variety of approaches but still have
not picked up any sole carbon source utilizers.  Likewise,  in following the
concentrations of bacteria in the water column and in the sediment, they show
no trend that can be correlated with degradation processes.

LEE:  I was interested in your comment about mixing vs  a straight core.  In
our area of Georgia we work with a six-foot daily tide range.  In such an area
it seems mixed type sediments are more realistic than just taking a core,  not
allowing any mixing.  So it depends on the area; certainly in our area we get
mixing of several centimeters in a good tidal range.

PRITCHARD:  Exactly, and my point is just that—there are a number of factors
one should look at in terms of designing microcosms.

DRAGGAN:  One of the usual justifications for using a mixed core rather than
an intact entire core is that you overcome,  in some measure,  the complexity of
heterogeneity of the environment which should increase  the replicability within
the experiment and enhance the ability to duplicate subsequent experiments. Do
you find that your eco-cores are any more or less  replicable?  Is the confi-
dence of the data any better?  I think there is some evidence now that intact
cores, like eco-core, can be just as replicable.

PRJTCHARD:  Yes, I agree.  They are definitely replicable,  and we've seen very
good replication with our systems.   If we run triplicate cores,  for example,
they look virtually identical.   This replicability, I feel, is not due to a
reduction in complexity.

CHOU:  Have you looked into the fate of dimethyl thiophosphate?  At UC-Davis,
scientists found that dimethyl thiophosphate was not degraded in their para-
thion systerns.

PRITCHARD:  No, we haven't looked but you're right, it did not break down.

RAWLINS:  Have you any data on the incorporation of C-14 into the cells of the

PRITCHARD:  I think some of the radioactivity we are seeing in polar unex-
tracted material would represent just that type of incorporation.

CHAPMAN:  Let me say. Dr. Slater, what a stimulating talk I thought yours was.
In comment, we have some observations I'd like to share in connection with
2,4-D utilizing organisms.  In the course of isolating cultures of this kind,
one initially establishes stable communities and ultimately determines that
the communities co-exist because of fastidious 2,4-D degraders.  One can then
isolate 2,4-D degrading organisms by supplying them with appropriate nutrients,
either amino acids or co-factors, and work with them in pure culture.   Ribbons
and I worked with pure cultures on the catabolic pathways of orcinol and the
closely related molecule, resorcinol.  We reported different metabolic path-
ways by which these compounds were degraded.  It's interesting that acetate,
which we identified as an orcinol metabolite, is the basis for the community
existence you described.  I would like to suggest that if you observe that
your Pseudomonas, which is capable of utilizing orcinol, will also grow with
resorcinol, then one can make predictions based on the metabolic pathways as
to what might happen to the community structure.  The metabolic pathway by
which resorcinol is degraded does not yield as much acetate, and it is quite
possible that by switching very simply from orcinol to resorcinol, the commu-
nity structure will fall apart.

SLATER:  Yes, in talking to Dr. Ribbons, he made the same good point.   We have
done none of the biochemical work with this system, and it's something we wish
to do.

FLOODGATE:  How far do you think all these pretty relationships you show in
the fermenter really represent what does on in the sea?  I always get a bit
bothered about spatial relationships—how the little molecules manage to get
from one "bug" to the other in a swirling mass of water?  Does it all happen
on the sides of the fermenter as it might happen on a beach?  Or is this some-
thing which we can really project and say it is going on in the oceans right

SLATER:  My fermenters are a swirling mass.  There have been very few attempts
to compare directly the type of community that we observe in these really very
synthetic laboratory systems with what may actually be going on in the natural
environment.  My feeling is that these are probably valid comparisons.  There
is nothing essentially different between the type of enrichment you do in a
continuous culture system and the type of enrichment you do when you put a
compound into a particular natural habitat.  I think the diazinon work, old
work not done in a fermenter, is a good example.  The original observation of
the relationship between the Arthrobacter and Streptomyces was simply careful


microbiology.  Those two populations were enhanced over others when diazinon
was applied to soil.  Now we haven't done a reciprocal of trying a diazinon
enrichment in a fermenter.  You have very good evidence of the type of enrich-
ment happening in soil, and if you can do it there, I am quite sure you could
do it in a fermenter.

PRITCHARD:  At what point are you going to draw a line in terms of accessing
persistence in an enrichment like this?  If you let the system incubate long
enouch, a population may eventually evolve which will break it down.  But is
this evolutionary process environmentally realistic?

SLATER:  Well, this refers to George Floodgate's point a little.  I don't
think we give enough consideration to one of the key points in microbial sys-
tems.  That is, they are very versatile and very adaptive.  In the dipicolinic
experiments, clearly, there is some sort of adaptive process.  We had to run
the fermenter for 180 days, a considerable length of time.  From a natural en-
vironment point of view, I think you've got to consider these points.  These
considerations are difficult in screening and testing programs but they can be
done.  Nevertheless, those sorts of activities are quite likely to occur in
natural environments to some point.  You then get what may appear to be essen-
tially a non-degradable compound becoming a degradable compound.  A much
greater authority than myself is Professor Pat Clark in London.  Her view is
very much that we've been subjecting the environment to all sorts of indigni-
ties over the last 100 years with respect to antibiotics, etc.  The tremendous
potential is there to adapt and to involve the new systems, new mechanisms can
deal with it.  There is obviously the question of limitations at some point.

PRITCHARD:  I've heard the criticism many times that in continuous-flow sys-
tems the useful organisms may wash out due to the selective nature of the sys-
tem.  You will quite often end up selecting for a population that is growing
on a contaminant in your media.  My question is, are you saying we might not
have to worry about that as much?

SLATER:  Well, I think you do have to worry.  You may well be missing orga-
nisms.  We've done a number of different compounds and it's really very im-
pressive how very rapidly a complex soil system in these enrichments can change
to a very few number of organisms.  There are a number of things that you can
do about it.  You can continue to reinoculate for periods of time to see
whether you can put back an organism which may fit into a simple enriched com-
munity which will achieve degradation of a compound which is not being de-
graded.  So I think that is a real problem.  There are other things we ought
to begin thinking about.  How can we develop continuous-flow systems so that
we don't get this simplification?  One of the things that Prof. A. Balkins is
looking at is designing a system which allows the introduction of particulate
matter in a homogeneous sense—not a colloidal but dispersed sand or clay par-
ticles or whatever.  It seems probable that if you have an interface or some
sort of increase in complexity you will sustain a great number of organisms in
the community.  So I think that the thing that you can do technically is to
try to make sure that you get a better starting community.

PRITCHARD:  Can I make just one more comment, relative to what you just said?
I think it is really very critical at this point.  We have found that methyl


parathion is not degraded in a water only system, even when the influent water
is providing a continuous inoculation.  However, as soon as you add the sedi-
ment to the system, degradation commences.

SLATER:  The one thing that worries me is the attitude that one must have a
pure culture which is degrading this compound.  I'm not saying that is your
attitude, but I think a lot of people do have this attitude.  We must consider
corporate activities between organisms where you have genetic potential to a
particular function split between different populations.  Now, if I may be a
little bit speculative, there is very little evidence apart from parathion and
one or two others.  My feeling is that those sorts of relationships are more
common than we are prepared to accept at this moment, because of our condition-
ing to the way in which we do enrichments.

DiGERONIMO:  A lot of times in the natural environment the introduction of
xenobiotics may be episodic.  Could you comment on enriching if the test com-
pounds are going to disappear and appear at different times, not added on a
continuous basis, but on an episodic one?

SLATER:  The basis of this position is that you want a selection pressure of
some sort so that you can get the adaptation processing.  The problem is
whether that kind of system would be continuous enough to force the organisms
in the direction you want to be going.  I don't really know.  I don't think
we've got any experimental evidence which would show whether or not that is a
better or worse system for enriching.  I think, however, from plasmid work
which is perhaps related in a fringe way, that once you have the capability
located in the population, even if you remove the selection, it will stay
there and be in a relatively low level.  So I don't know, it's an open ques-

RAYMOND:  What would happen if you did not "insult" your system with such high
concentrations?  Don't you think that probably leads to your limiting the num-
ber of types of microorganisms that survive in the system?  Why only two or
three?  I noticed your concentrations were 0.5 g/liter or higher.

SLATER:  Those are the concentrations you apply to the system, they are not
necessarily the concentrations at which you conduct the enrichment.  The value
of a continuous-flow system is that you can conduct an enrichment at greatly
reduced concentrations.  If you recall my orcinol experiments, at low dilution
rates, were exceedingly low.  What is, I think, a criticism of what we have
done is that we supplement at high nitrogen and phosphorus concentrations.
That may not be relevant at all in a natural situation.  The ratios could be
variable and so it's just a matter of doing natural experiments.  We chose
higher concentrations since it provides a higher biomass so that we could do
what we wanted to with the cells.  But I think it is a point that should be
taken into account when you are doing these community enrichments.

COLWELL:  May I make a comment in regard to the need for mixed population
studies?  I should like to amplify the point you made about plasmids.  I do
think that plasmids play quite an important role in degradation in the environ-
ment.  For example, we have found that in mercury-transforming microorganisms,
90% of the organisms may carry plasmids.  We've been able to cure the bacteria


of the plasmids and show that loss of mercury resistance or metabolism is as-
sociated with loss of the plasmid.  Furthermore, we have examined drug resis-
tance in E.  coli and been able to transfer this characteristic to autochtho-
nous Vibrio species in Chesapeake Bay.  We're in the process now of working on
the transfer of mercury metabolism amongst Pseudomonads.  With this kind of
genetic exchange there is a need to study mixed populations possible in natu-
ral systems.  When examining degradation studies in a given system I will be
very interested in Professor Balkins' results with his particulates, because I
think that technique may create a population mass, insuring plasmid transfer.
However, I would take issue with the statement that once the process is in the
population it will stay there.  Not quite so, because we have discovered that
you can isolate acid and gas-producing Pseudomonas aeruginosa quite readily
via an MPN test in a polluted area.  Now, we all know that gas-producing Ps.
aeruginosa are quite extraordinary.  It is possible that the Pseudomonad picks
up a lac gene via a plasmid and loses the plasmid quite readily.  Thus the po-
tential gene exchange may be quite significant but won't be quite as stable in
situations as yours, accounting for erratic kinds of results observed for some

SLATER:  I would agree with that and my statement was an over-simplification.
We have someone working in the lab on stability of drug-resistant plasmids,
and we did some early work on the stability of organisms carrying lac operon,
and they were unstable.  Place the organisms into conditions of carbon limita-
tion and they would fall apart.  There was a reason for this, but the interest-
ing thing is that with the natural plasmid, if I may use that word, of E.  coli
carrying certain drug resistances, they are exceedingly stable.  We tried car-
bon and phosphorus limitations, and there are certain modifications but the
plasmid, as such, stayed there.  Ellwood's group also studied thi-s and they
found a similar sort of stability.  Can I go back to the original point about
the plasmids in the first place?  Plasmids beg the question to an extent be-
cause they already contain the necessary genetic information for particular
transformation in a particular pathway.  The question is, where did they come
from in the first place?  My view is that mixed systems are the point at which
various capabilities can be developed perhaps in separate organisms.  What you
may be seeing with plasmid mediated systems is the next step up where there is
an aggregation of those activities into a cooperative state.

COLWELL:  Sort of a community corporation?

SLATER:  Yes, they put together a plasmid which can then be transferred around
a population, and you get coordinate control.  That is really speculative, but
I think the point is they have to get the basic genetic information in the
first place.  It seems to me quite likely that the wider the genetic base in
the community the more chance there is that various components can evolve par-
ticular capabilities to deal with a new compound, and then perhaps later put
them together on a plasmid.

COLWELL:  I quite agree, but I would just add one point.   It's significant
that we frequently find plasmids in Pseudomonas.  Stability may well be a
function of the plasmid moving amongst species of the genus, as well as amongst
related genera.

LASKOWSKI:  Dr. DePinto, I would like to add to your•comment regarding role of
models.  I think a very good thing might happen in modeling.  We'll probably
arrive at a series or battery of models that will describe various physical
situations.  If one knows key physical and chemical properties of materials
you can begin to get some idea of what might happen if those materials were
placed in these different environments.  I agree, I don't think there will be
any one model that will serve this purpose, but it will be extremely useful to
have a battery of models not only in the mathematical sense, but also in the
physical sense, that these materials could be run through.

PRITCHARD:  I have a comment relative to your use of the term chemostat.  I've
gotten now so that I shudder every time anybody mentions the word chemostat in
a context like this.  I think the explanation is very simple.  A chemostat is
a highly defined, very specific type of laboratory technique in which bacteria
are growing as a function of a single limiting substrate under steady-state
conditions.  As soon as we begin to work with mixed cultures of any kind or
get wall growth of any sort, or use multiple undefined substrates, then the
term chemostat should not be used.  The second thing about chemostats is that
the growth of bacteria therein can be easily described with Monod kinetics.
Many other types of continuous culture systems cannot be readily modeled with
Monod kinetics.

DePINTO:  I would like to make two comments on what you just said.  First of
all, if you survey recent literature, such as by Reeves, there are other mathe-
matics describing use of chemostat that do not depend on Monod.  In many cases
these are being applied to uptake rates and decomposition rates in the model-
ing of phytoplankton in the Great Lakes area.  They do not depend on Monod
equations.  Secondly, I recognize that there are limitations, certainly, of a
flow-through system, but I think that a chemostat or flow-through system offers
the ability not to duplicate but to simulate the natural system.

PRITCHARD:  I'm not arguing that the continuous culture is not a great tech-
nique to use.  Certainly, I wish more people would apply it, but I'm saying
the use of the term chemostat has to be reserved for a very specific process.

DePINTO:  Certainly, flow-through system is definitely better.

SLATER:  Can I just back up what Hap said?  Professor Pirt, who I think of all
people is perhaps one of the few that can justifiably get involved in the se-
mantics of the terminologies of continuous systems, suggested that the generic
word to be used ought to be continuous-flow culture system.  From that, you
then derive a whole gang of different systems, chemostat being one, turbidi-
stat, etc., there's a whole range of other sorts of systems.  So the preferred
term ought to be continuous-flow culture system.
Can I just make one point about modelling that worries me greatly?  I have
considerable doubt at this stage of development about the validity of some of
these models.  I'm not going to talk about attempting to model or simulate
complex ecosystems.  Let's just talk solely about continuous-flow culture sys-
tems where we may have a relatively simple community or a few cultures.  From
the attempts people have been making to model and explain in mathematical
terms the various activities that are going on in a system, to describe sub-
strate concentrations, population levels, and so on, it is transparently


obvious that they fall short of reality.   The point I would like to make is
that we are trying to develop models without the fundamental understanding of
the biological processes that are involved.   For example, with the Monod ki-
netics there is no reason to suppose that the Ks value, saturation value, is
growth rate independent.  We always treat it as such, but there is some evi-
dence that there may well be a variation in  Ks as a function of growth.  The
way we apply it, we assume it's a constant.   What I would advocate is, at the
moment, we ought to concentrate far more on  looking at the biological proces-
ses, try to work out how those factors are modified.  The second point I would
like to make relates .to an argument that I have had with Dr. Bazin at the Queen
Elizabeth College in London, who you might know does a lot of work on modelling
of nitrification in soil columns.  And this  is that you can improve your model
by just heaving in whole heaps of constants  and eventually make it look like
what you've got in nature, etc.  Those constants have no biological basis what-
soever; so, if I had a preference,  at this time, it would be to look much more
closely at the biological relationship between population and the relationship
with the physical environment before we go overboard on too many models.

DePINTO:  Let me say first of all I'm not a  modeller; I consider myself to be
a microbiologist.  I started using models in order to get more of a conceptual
grasp of what is going on in the system and  to try to assimilate data from a
wide range of experimentalists.  I think a model is a research tool and not
necessarily an answer, at least for quite a  long time.

BOLLAG:  I would like to add some comments from a biological viewpoint.  For
me, it is difficult to understand how it is  possible to use a model predicting
the fate of an organic molecule if it is not yet known which kind of transfor-
mation can take place.  It is my opinion that almost no xenobiotic exists from
which all transformation possibilities are known.  How is it possible to model
the fate of a xenobiotic whose pathway has not been been determined?

DePINTO:  In many ways I disagree with you,  because I think it's never too pre-
mature to start developing a model to be used as a research tool in coordinat-
ing experimental data.  Secondly, I think in meetings ten years ago, sponsored
by the EPA, they were interested in modelling the fate of phosphorus and phy-
toplankton in lakes, one could hear some of  the same kind of things you just
stated.  Ten years later, they have these models working for the Great Lakes,
and they are being verified and improved upon for Lake Erie.  They've found it
a very desirable tool to be used for communication purposes.

LIU:  We have about ten people who are interested in modelling.  I find one
very important thing:  they should pay more  attention to understanding the
lake system.  The system is very complicated; for example, the bacteria bio-
mass distribution is not uniform, particularly in the lake sediment and under
the interface.  You have 108-109/ml bacteria less than 1/2 centimeter of sur-
face, but in the lake water column only about 103-10l*/ml.  The sediment or the
surface is not stable.  For example, on Lake Ontario, you have currents of ten
knots per hour.  Actually, lakes are constantly changing; the hypolimnion and
epilimnion also change with the seasons.   When you develop a model, you cannot
use such data as total organic carbon because of so-called labile organic mat-
ter.  What are they measuring, not all organic matter can be utilized?

BAUGHMAN:  I find myself in an awkward position coming to the support of the
modellers, because I certainly don't consider myself a mathematical modeller.
It seems to me the chief complaint against models is that they don't always
accurately predict behavior in some environment that we are interested in.  I
find little difference between use of a mathematical model as a tool or guide
to our research and decision making as a basis for complaint than perhaps the
fact that Dr. Bollag did not use some Bacillus instead of his fungi.  The
model has a purpose it is intended to serve and certainly will not serve a dif-
ferent purpose very well.  There is a point that should be made here; models
can serve at least two purposes in the context that we talk about.  One is to
simulate some actual environment, like Lake George or Pensacola Bay or some
aspects of these.  The kind of model that was referred to here is one that is
intended to simulate the kinds of things that can happen rather than to pre-
dict exactly what does happen in some specific site.  It appears to me that
one of the things we are badly missing is some basis for establishing what is
important.  For example, we have seen that films occur on water surfaces and
organisms are concentrated there, that if we have sediments in a system, if
they are mixed or non-mixed, you get different results.  My question is, are
those differences important and what is the basis for telling if they are im-
portant?  If the fact that we can simply see something different is to be the
basis for importance, then we have an infinite problem on our hands.  I would
suggest that our models won't improve and, perhaps, some of our bases for in-
terpretation won't get any better, unless we start to address this question.

BAUGHMAN:  Dr. Giddings, I have a couple of questions.  What kind of lights
were used in your study?  Were the tanks usually covered?  And what was the
depth of the water?

GIDDINGS:  Cool white fluorescent, the depth was 30 cm, and the tanks were
covered.  There was light in the sediment.

BAUGHMAN:  I'm curious why you expect photochemical degradation of anthracene,
because anthracene absorbs fairly low in the ultraviolet region.  Cool white
light is very short for anthracene, particularly if you covered your tank with
plastic.  You would expect photochemical degradation of anthracene to be tri-
vial unless there is significant sensitized degradation, which you should not
get in the distilled water system.

GIDDINGS:  I can't answer that except that is what we found.

BAUGHMAN:  I was wondering, did you attempt to look at the ratio of anthracene
in the sediments to the ratio in the water during the course of these experi-

GIDDINGS:  We weren't successful in taking sediment samples until near the end
of the experiment so we don't have that data.

BAUGHMAN:  One would fully expect anthracene to sorb to the sediment, thereby
drastically reducing the rates of processes like volatilization and photo-
chemical degradation in those systems.

GIDDINGS:  Yes, that's what we thought.

DAVIS:  I happened to hear you talk about the photocatalytic activity.  There
is something working against all of this.  Photocatalytic activity is indeed
occurring out there in these estuaries, especially highly turbid Texas estu-
aries, that is a protective mechanism, when you have a secchi disk reading in
ours, for example, Florida estuaries are beautiful by comparison, you know.
Secchi disks of about four to six inches normally with high silt loads from
the rivers and precipitation or sedimentation later.  Before long these things
settle out and build up, and such things as the oyster accumulate these com-
pounds.  And I think it is important what you pointed out.  What was the sa-
linity of your microcosms?  Were they entirely fresh water?

GIDDINGS:  Yes, this was spring water.

DAVIS:  Do you have any data on conductance?

GIDDINGS:  About 100 microohms initially but it goes down after that.

DAVIS:  Prom what I've found, not on anthracene but on related compounds, as
you go up into 3% estuarine salt levels you can remove a factor of 10 in the
degradation rates.  In other words, with the same number of "bugs" per milli-
liter you don't get degradation of the same amount per day.

GIDDINGS:  Thank you.

JOHNSON:  I have a question.  How do you plan to use this microcosm, what will
you do with your data?

GIDDINGS:  Our initial objectives are to see how it works.  Initially we plan
to run a number of the representative PAH's as I pointed out,  through the mi-
crocosms, see if we detect the same sort of trends that we saw in simpler test

JOHNSON:  Assuming that you are successful, how would you report these data to
a federal agency?

GIDDINGS:  I'm not sure.  The microcosms task group has been beating this
around quite a bit.  Regulations we are talking about from here seem mostly to
require some sort of standardized test or battery of standardized tests.
These microcosms, I've pointed out, aren't standardized and aren't standariz-
able.  The real world is often not standardized or standardizable.

JOHNSON:  Would you feel this would be representative of a community or some-
thing of this sort?


JOHNSON:   I am a microbiolegist, but since I do represent a laboratory that is
concerned with fish I was wondering how could you exclude fish from your pond?

GIDDINGS:  Well,  that was one of the concessions we made to the laboratory
conditions.   We've tried including fish in the microcosms but this was not a
degradation study; rather, what effect did they have on the ecology of the


microcosms?  They have less of an effect than I would have expected consider-
ing the fish density is much greater than you would expect in a real pond.  We
found increased rates of production in the presence of fish.  We found that
bluegill or fathead minnow would eat up snails and that would have additional
secondary and tertiary effects on the ecology of the pond.  But we haven't
done a degradation experiment to see what effects that would have.

JOHNSON:  You made a statement rather in passing that fish or animals were not
an important sink for the compounds.  I was wondering how you could make that

GIDDINGS:  Well, I'm speaking in terms of the distribution of the compound and
its derivatives at the end of the experiment.  Little was incorporated into
biological tissue, unless we count the 35% that was bound to the sediment.
What was in the plants and animals was an insignificant fraction from a point
of view of where the anthracene went.

JOHNSON:  But you also made a statement, that after 84 days you knew something
about this compound in regard to what would happen in a stream or pond.  And
actually what you would know is what would happen in 84 days in this little
microcosm that excluded fish and other animals, and probably other inverte-
brates also.

GIDDINGS:  Perhaps, we haven't sampled the pond.  The way to answer you would
be to compare what happens to anthracene in a microcosm with what happens to
anthracene in a pond.  That latter study isn't practical.  About all we can do
is to compare the ecology of the microcosm to the ecology of the pond, the
water chemistry of the microcosm to the water chemistry of the pond.  We find
great similarity between them.  I think that the microcosm environment is very
similar to the pond environment.  The ultimate test of a model, an LC50 test,
or a microcosm test is whether that really does apply to the world.  That's
one of the big problems we've got to work on.

COONEY:  On the question of light effects, I think there was a comment from a
gentleman over here who implied that the chromophore of anthracene might not
be long enough to account for the photochemistry-  I think that one does not
have to rely upon a UV-absorbing chromophore because when white light is inci-
dent upon an aquatic system, you get sufficient singlet oxygen generated so
that it can be lethal to large numbers of microbes.  There is a recent report
indicating that in estuaries there is sufficient singlet oxygen present to ac-
count for transformation of organic pollutants to a measurable degree.

PRITCHARD:  Could you get the same results in a smaller system?

GIDDINGS:  Well, we've experimented with smaller systems and they have certain
advantages:  they're easier to work with; they're easier to set up; they re-
quire less material; you can put more of them in a smaller space; you don't
need as much anthracene; you don't need as much C-14.  But the difficulties we
found with smaller microcosms were that they are less stable, took longer to
equilibrate, were less replicable, and basically were harder to sample re-
peatedly without causing strange effects in controls and things like that.
They were not amenable to repeated sampling the way that we need to sample.


GARNAS:   In listening to your discussion on your system, I failed to see the
main purpose why you used the system.  Had you been looking at fate of a chemi-
cal due to photochemistry, hydrolysis, or sorption, I would tend to believe
you would study those properties in your laboratory environment while trying
to simulate the real world.  You've taken the substrate from a pond, you've
eliminated a certain trophic level, and put them in a glass box in your labora-
tory.  You really haven't defined to us where the real world is in that system.

GIDDINGS:  I would submit that the microcosm is an ecosystem similar in func-
tion and structure to a pond ecosystem with the exception of fish.  We can de-
bate how important fish are to microbial degradation.  We've also found that
the microcosms are similar to the ponds from which they're constructed.  But,
I don't think that's essential to accepting them as a valid ecosystem which is
much more amenable to experimentation than a real pond is.  George Southworth
and Steve Hervies have generated these data on single processes and have tried
to put it all together into a model that predicts what's going to happen to
anthracene when all these processes are interacting at once like in nature.
Except for some qualitative statements like, "photolysis is likely to be im-
portant," and "microbial degradation is likely to be important," they weren't
able to say what the results of all these things were going to be.  One ap-
proach is to just continue to modify your model, refine your model, stick in
as many mechanisms to try to come up with a prediction of the natural fate of
anthracene.  We've shortcut that by treating the microcosm as a black box.
We're using a microcosm instead of a mathematical model to integrate these
processes.  They are all happening simultaneously, physically, not mathemati-

GARNAS:  But, it would appear that just from the fate of the chemicals, you
don't really have a good grasp as to what is happening.  Is it photochemical;
is there indirect evidence that there is or is not volatility; is it a bio-
degradation process or some other process?  The difficulty that you've run
into now is that the complexity is so great that you can't zero in on any in-
dividual component or individual process.

GIDDINGS:  That's right.  For instance, the experiment we did with the dis-
tilled water made it much easier to understand what was happening in the micro-
cosm water.  I use that as an example to show that we have to interact with
the lower levels of testing to understand what is happening to the microcosm.
The microcosm shows us some things that we could not have predicted, or could
not have been sure of, by putting together the results of single process
studies.  For instance, the competition between sorption and photolysis on the
fate of anthracene in the water, you can try to get rates defined real care-
fully, and then see what happens when you have them running simultaneously.
Does absorption beat out photolysis?  You can try to do it mathematically, but
you can do it just as simply in the actual biological system.


Chairperson, Donald G. Ahearn


       E. M. Davis, J. Bishop, and R.  K. Guthrie
          The University of Texas at Houston
                School of Public Health
                    P.O. Box 20186
                   Houston, TX 77025

     Organic compounds which usually degrade rapidly in
fresh water environments were shown to become almost re-
fractory in saline wastewaters.  Salt concentrations were
maintained at or near that of estuarine environments.
Additionally, saline wastewater BOD data can be ques-
tioned if the test is conducted in accordance with the
method currently being employed.  Microbial populations
were found to be the reason for as much as a 30% differ-
ence in BOD results with tests conducted at different
salinities.  Salt-tolerant bacteria, used as seed orga-
nisms in BOD testing, showed that BOD results were simi-
lar at 0%, 1%, and 3% salt concentrations when using
standard  (nonsaline) dilution water.  Acclimated sewage
seed produced BOD values significantly lower when saline
wastes were tested.  Equal concentrations of bacteria in
the BOD test did not reduce equal amounts of organic ma-
terials when exposed to various salt concentrations.
Sewage seed degraded 44 mg/liter/d organic material at 0%
salt, 38 mg/liter/d at 1% salt, and 22 mg/liter/d at 3%
salt.  Bacterial genera which predominated in 3% salt
were Pseudomonas, Bacillus, Staphylococcus, and Flavo-
bacterium, three of the four genera being potentially

     Hypersaline wastewaters also may not be amenable to
conventional treatment methods.  Activated sludge method-
ology applied to an organic amine wastewater containing
3.5% salt showed only 48% BOD removal and 52% TOG removal
at an F:M loading rate of 0.2-0.5 Ib BOD/d/lb MLSS.  Set-
tleability was seriously affected by increased salt con-
centrations.  The SVI of most units can be expected to
range from 0.4 ml/g despite high bacteria densities of
3xl06/ml.  Extended aeration is more efficient but

           requires  detention times  of up to 80 d.   Application of
           nitrification methodology following activated sludge
           treatment of the hypersaline wastewater  was necessary.
           Ammonia increased from 13 mg/liter to 70 mg/liter in the
           effluent.  Ammonia removal averaged 65%  in nitrification
           units with some sludge recycle.   Anaerobic digestion of
           hypersaline wastewaters is an alternative provided the
           salt content is maintained below 1.3%.   At a detention
           time of 15 d, anaerobic efficiency increased with de-
           creases in salinity.

     Complications occur naturally in saline aquatic environments in the over-
all degradation of organic pollutants.   The microbial population present in
either estuaries or hypersaline industrial wastewaters may not function to
their full capacity in degrading the organics for several reasons.   The per-
centage of bacteria representing the most functional group for degradation
purposes may not be the dominant microbiota in a heterotrophic population (10).
Additionally, even relatively small amounts of heavy metals can reduce organic
degradation rates (11).  Temperature of the ecosystem corresponding to season
of the year plays an extremely important role (2).   Specific research investi-
gations have demonstrated the complexity of functional metabolism of microbial
populations in respiration and concurrent degradation of pollutants.  Bio-
chemical oxygen demand (BOD)  testing has been used to evaluate waste loading
and for design bases for waste treatment units.   Yet, only recently has it ap-
peared that that test (9) may not show the full waste loading or its presence
in a saline environment.

     In one study (5) the soluble waste portion in a salt water carrier was
metabolized at the same rate by the microbial population in domestic sewage.
However, the insoluble portion produced different rates, with that of domestic
sewage in fresh water being three times greater than that of domestic sewage
in a salt water carrier.   It was further shown that only the degradation of
suspended organic material was inhibited by salt and the dissolved organic
fraction of the BOD was unaffected (8).  Even salt-tolerant bacteria function
in BOD removal rates according to the salt concentrations (3).  Increases in
BOD resulted when the salt level was decreased.   Ten-day BOD's on salt water
wastes were 3 to 6 times higher with standard dilution water than with a dilu-
tion water having the same salinity as the waste.   Further,  the rate of bio-
logical oxidation, K, may vary considerably under different test conditions
(4).  Wastewaters containing less than 10,000 mg/liter chloride diluted with
fresh water were shown to have a K value greater than that of fresh water
waste.  The K value increased to the point equivalent to 50% of that of sea
water.  When the 50% level was exceeded, K decreased, until in 100% sea water,
it was lower than in fresh water.  This led to the hypothesis that a low sa-
linity may stimulate microbial activity and result in increased BOD values,
and in some instances the organic pollutants may persist because cellular con-
stituents of lytic activity are preferred as a nutrient source to the organic
substrate which has been introduced as a pollutant (6, 7).

     This study was undertaken to further identify and quantify some of the
reasons for reduced degradability of organic pollutants in marine environments
and to ascertain whether the BOD test truly represents the organic pollution
load in saline ecosystems.  Microbial populations were examined for their
relative efficiency of wastewater treatment in wastewaters having salinities
comparable to those of the marine environment.
                            MATERIALS AND METHODS

     The BOD tests reported herein were conducted in accordance with method-
ology listed in Standard Methods  (9) except for those results obtained by mano-
metric analyses (Each Chemical Co. BOD Apparatus).  Bench-scale activated
sludge units were used for testing of the industrial waste  (3).  Nitrification
and anaerobic digestion units additionally were bench-scale in size.  Micro-
bial genera in the wastewater and in the various testing units were identified
by the methods outlined in Diagnostic Microbiology  (1).  The principal organic
fraction of the industrial hypersaline wastewater tested in this research was
an ethylamine complex.  The wastewater had a salt concentration of 3.7%, a
total bacterial population of 2.2xl02/ml, and a BOD of 30 mg/liter, using the
existing population in the wastewater as seed.  Salt-tolerant bacterial seed
for activated sludge units comprised Bacterium T-52 and Paracoccus halodenitri-
ficans (3).

                           RESULTS AND DISCUSSION

     Quantification of the effect of sodium chloride concentrations on a sew-
age bacteria-seeded glucose standard are shown in Figure 1.  Also shown is the
representative daily BOD for a 600 mg/liter sodium acetate solution.  The data
show dramatically how two salt concentrations, which approach the salinity of
sea water, affect the results of a 5-d BOD.  Those data were determined by the
manometric technique.  Using the salt-tolerant bacterial Paracoccus halodeni-
trificans (Fig. 2), similar effects were observed at the 5-d period, except
for more rapid respiration rates in the early periods.  The implication of the
data shown in Figure 1 is that BOD results of saline wastewater should be
viewed with caution regarding the total waste load, and may not be representa-
tive of the total organic concentration.  Stimulation of salt-tolerant bacte-
rial species using nonsaline BOD dilution water is indicated in Figure 2, es-
pecially for the acetate standard.  Additionally, the survival of all sewage
seed bacterial genera in saline environments is not assured in BOD testing.  A
noteworthy example of that is presented in Table 1.  Declines in populations
even in nonsaline BOD's to 5 d is indicated.  More importantly, genera which
survive at 3% salt concentration, similar to marine situations, are illus-
trated in Series V of that table.  Potentially pathogenic genera, e.g. , Pseudo-
monas and Staphylococcus, appear to have survived best.  Distinct changes oc-
curred in sewage seed bacteria populations during their acclimation to the
salt concentrations and to the organic glucose standard.  The raw sewage seed
had an initial bacterial population of 107/ml.  During the acclimation period
to the glucose standard population levels dropped to 106/ml.  Acclimation to
the glucose standard containing 1% and 3% salt resulted in a bacterial popula-
tion of loVml and 103/ml, respectively.  This behavior resulted in lower

       350 r
                             350 r-
      Figure 1.

        • 0%

       -• 1%

        A 3%

BOD values for glucose-glutamic acid and sodium acetate standards

in three-salt cx>nce«fer^ttens-nis±iTg^sewage~bacteriaT~seed.

BOD values foi ylucost

in three salt concentrations using Paracoccus halodenitrifleans  as



and days
Series I Series II Series III Series IV Series V
E . col i
Enter obacter
Total No. /ml

15% 77% 70% 40% 8%
8% 7% 10% 30% 7%
12% 2% 5% — 7%
10% 30% 7%

6% — 5%
4% 5% — — 1%
10% 5%
8% 2%
1% 2%
S.exlO1* 8.5xl08 6.5xl03 IxlO7 6.6x10
220 — 235
Glucose Standard
Series Salt Cone., %
I 0
II 1
IV 1
V 3
13% 70% 71% 8% 14%
35% 10% 11% 7% 16%
5% — 7% 20%
46% 10% 7% 7%

6% — -- 70% 44%
5% 11%
1% 6%
— — __ — —
— — — — —
— — — — —
— — — — —
2 2.3xl07 6.5xl03 5. OxlO6 6.6xl02 3. OxlO5
246 -- 194 — 170
Dilution Water
Salt Cone. , %

initial populations in the BOD's conducted at 1% and 3% salt.  During BOD
testing, bacterial populations of the sewage seed increased from 102-103 in
the saline diluted organic standard and increased 10 - 105 during the same
time period in the organic standards using standard (nonsaline) dilution water.
The greater increase in the standard dilution water accounted for the increase
in BOD values.  Linear regression analysis was applied to the data for the glu-
cose standard using sewage seed and the two salt concentrations comparing bac-
teria counts versus 5-d BOD values.  Since the initial bacteria counts varied
at the beginning of each BOD test, all initial counts were based at zero and
the log increase in bacterial counts were regressed against BOD values.  The
linear regression of those log increases in bacteria on the resultant BOD
values yielded a significant "r" value of 0.87 (Fig. 3).  The "r" value is the
linear regression analysis correlation factor and for the number of data ana-
lyzed to be significant at the 0.05 level, the "r" value must have been 0.49.
At the 0.01 level it must have been 0.62.  This explains some of the differ-
ences in the BOD values for saline wastewaters diluted with standard and sa-
line dilution waters.  Even with low initial bacteria counts, population in-
creases in saline organic standards using the standard dilution water caused
the BOD values to be higher than those for the same saline organic standard of
the same salinity as that of the waste, which had a smaller population in-

              JE  3.0
               8  2-°
               S  1.0
                          60     120    180   240
                                   BOD, mg/1
Figure 3.  Sewage seed population increases in the 5-d glucose BOD's using
           various salt concentrations.
     Average bacterial increases in the manometric BOD flasks and the corre-
sponding BOD removals were regressed against time to determine if the same
number of sewage bacteria utilized the same aount of glucose at different salt
levels.  The slope of those regression lines yielded the bacterial increase
per day and the BOD removal per day (Fig.  4).   with the same increase in bac-•
terial population, the BOD removal rate was 44.4 mg/liter/d at 0% salt, 38.6
mg/liter/d at 1% salt and only 22.8 mg/liter/d at the 3% salt level.  Sewage

   seed acclimated to the saline glucose standard resulted not only in low ini-
   tial counts and slower growth rates but also in a slower utilization rate of
   organic, material.
   Figure 4.  Slopes of BOD removal and bacterial population increases in mano-
              metric BOD tests using glucose and sewage seed at three salt con-
        Activated sludge treatment, comprising extended aeration, was applied to
   the industrial wastewater for different loading rates and detention times.

        Bench-scale activated sludge units were used for monitoring the sedimen-
   tation, clarification, and organic removal in the industrial wastewater.   Each
   unit had a capacity of 5 liters and an aeration rate of 0.0033 Ib O2/ft /h.
   The seed culture was a mixture of Bacterium T-52 and Paracoccus halodentrifi-
   cans.  The units were fed conventional loading rates (F:M) of 0.2 and 0.5 Ib
   BOD/d/lb MLSS.  After 60 d of treatment, feed rates were changed to yield an
   F:M ratio of 1.5 and 0.06.  Total suspended solids  (TSS) and volatile sus-
   pended solids (VSS) and the sludge volume index (SVI) were monitored to deter-
   mine the amount and settleability of the sludge.  Bacterial populations were
   monitored in the units to determine concentrations present at the different
   feed rates and if they were being flushed from the units.  Total organic car-
   bon  (TOC)  and BOD were measured to determine the efficiency of organic removal
   at the different feed rates.  The BOD of the industrial wastewater was con-
   ducted using standard dilution water since our previous experimentation with
   the standard dilution water and a saline dilution water of equivalent salinity
   yielded equiva-lent values.

        Analysis of the activated sludge unit treatment data through day 60
   showed essentially no difference in BOD or TOC removal for the two conventional
   F:M loadings of 0.5 and 0.2 Ib BOD/d/lb MLSS.  Effluent BOD and TOC values for
   both units averaged 75 and 100 mg/liter, respectively, at those feed rates.

This represented a 48% removal of BOD and a 52% removal of TOC.  Total sus-
pended solids (TSS)  remained high in both effluents, averaging 115 mg/liter.
MLSS in the units ranged from 160 to 200 mg/liter.  Settleability was poor
with the Sludge Volume Index (SVI)  ranging from 0.4 to 2.8 ml/g.  Microscopic
examination of the effluent showed a fine pinpoint type floe which remained
suspended in the wastewater.  The floe was granular in appearance, contained
no bound water and was devoid of filaments.  Bacterial populations averaged
3xl06/ml in both units.  Detention times varied from 8 to 20 d.  Changing to a
higher feed rate of 1.5 Ib BOD/d/lb MLSS resulted in higher effluent BOD, TOC,
and TSS values.  BOD and TOC values averaged 86 and 122 mg/liter, respectively,
after reseeding of the units.  This represented a 36% removal of BOD and a 42%
removal of TOC.  TSS in the effluent was high (121 mg/liter) due to the fact
that the 2-3 d detection time did not allow the settling chamber to remove all
of the slow settling sludge.  Most floe was still of the small pinpoint type.
MLSS were higher and ranged as high as 400 mg/liter.  Settleability of the
solids was poor with the SVI ranging from 2 to 4 mg/g.  An extended aeration
rate of 0.06 Ib BOD/d/lb MLSS produced lower effluent BOD and TOC averages of
45 mg/liter and 85 mg/liter  (68% and 58% removal, respectively).  The pinpoint
floe still remained in the effluent.  TSS values never dropped below 60 mg/
liter, and averaged 70 mg/liter with a SVI average of 1 mg/liter.  MLSS in the
unit were low, ranging from 100 to 120 mg/liter.  A feed ratio of 0.06 Ib BOD/
d/lb MLSS yielded the highest quality effluent of the four feed rates examined.
This feed rate is not practical for an activated sludge unit because of the
long detention times which ranged from 60 to 80 d.  The conventional feed
rates of 0.2 and 0.5 Ib BOD/d/lb MLSS produced similar effluent qualities.  The
higher feed rate  (0.5) would be preferred due to the shorter detention times
of 7 to 10 d.  The rapid feed rate with a 2-3 d detention time yielded a poor
quality effluent that was high in BOD, TOC, and TSS.  Of the four feed rates,
0.5 Ib BOD/d/lb MLSS yielded the highest quality effluent for the least amount
of time..  Due to the characteristics of this particular wastewater, a problem
was encountered which made the activated sludge system ineffective as a single
stage treatment method.  Ammonia values increased from 13 mg/liter in the raw
wastewater to an average of 70 mg/liter in the effluent from the units at con-
ventional feed rates.  Ammonia-nitrogen from the ethylamine complexes is an
end product of carbonaceous degradation.

     Removal of ammonia-nitrogen from the effluent of the activated sludge
units by complete-mix nitrification reactors averaged 65% regardless of deten-
tion time and feed rates (Fig.  5).   Ammonia concentrations were reduced from
an average of 70 mg/liter NHa-N to 25 mg/liter NHa-N.   Denitrification was
started in the units but was lost due to the system setup which had no recycle
of sludge.  Denitrification bacteria were flushed from the units within 20 d.
Nitrate-nitrogen averaged 49 mg/liter in the influent and 75 mg/liter in the
effluent.  BOD values were reduced in the units from an average of 73 mg/liter
to 66 mg/liter and TOC was reduced from an average of 100 to 90 mg/liter.  The
plug-flow unit with sludge recycle removed 60% of the ammonia from the waste-
water at a detention time of 15 d.   The ammonia concentration decreased from
an average of 75 to 30 mg/liter NHa-N.  Denitrification was accomplished but
the removal efficiency was low, averaging only 25%.  BOD values were reduced
from 86 to 78 mg/liter through the unit and TOC was reduced from 122 to 106 mg/
liter.  Floe particles in these units would settle well and the TSS average
was about 55 mg/liter.


                    ,- 200
                                           100 -,10
                            GAS PRODUCTION
                            BACTERIA POPULATION
                                           60 £-
                                                               20  -
                    6  7
                                                   10 11  12
Figure 5.   Nitrification capability for  a  saline wastewater at  four detention
      Observations during the operation of  the anaerobic digesters  showed a de-
crease  in gas production and BOD removal with an increase in salt  concentra-
tion  (Fig.  6).  When a  salt concentration  of  1.3% was reached, gas production
ceased.   At this time,  there was only a small decrease in bacterial  popula-
tions,  therefore a salt concentration of 1.3% was inhibitive against one par-
ticular  group of organisms,  the gas producers.   The bacterial population of
the digester sludge was  4xl02/ml when the salt concentration reached  2%.   This
indicates that anaerobic digestion may not be feasible for hypersaline waste-
waters .

             DETENTION TIME   7 D
              NHj-N REMOVAL   60%
             10 D
                         15 D
20 D
                                                            15 D
Figure 6.   Anaerobic digestion performance witn  increases in salt concentration.


     The BOD values obtained in sewage-seeded organic standards showed a sig-
nificant trend,  i.e.,  as salt concentration increased,  BOD values decreased.
Changes were significant at the 0.05 statistical level.  Dilution of sewage-
seeded saline organic  standards with standard dilution water resulted in BOD
values higher than those of corresponding non-saline organic standards due to
increases in bacteria  populations and increased organic removal (BOD values)
in the presence of low levels of salt.   Bacteria populations of sewage seed
correlated with corresponding BOD values and concentrations of salt in organic
standard solutions showed that the addition of 1% and 3% salt resulted in a
decreased initial bacterial population,  decreased growth rates, and a decreased
ability of an equivalent number of bacteria to degrade an equivalent amount of
organic material.  Each salt-tolerant bacterial species yielded comparable BOD
values for each organic standard conducted at three salt concentrations.  Popu-
lation increases and K values were similar at the three salt concentrations
and when using either  the standard dilution water or the saline dilution water.
Since sewage seed does not compare favorably with salt-tolerant bacteria in
determining the BOD of saline wastewaters, for determination of the BOD of a
wastewater with elevated or varying salt concentrations, a salt-tolerant bac-
terial seed should be  used to insure an  accurate and reproducible BOD value.

     Bench-scale, continuous feed activated sludge units were capable of 48%
BOD removal and 52% TOC removal at conventional feed rates using selected salt-
tolerant bacteria.  The best BOD and TOC removal occurred with the extended
aeration process but the 60 to 80 d detention times are not feasible for acti-
vated sludge units.  The feed rate of 0.5 Ib BOD/d/lb MLSS yielded the highest
quality effluent with  a detention time of 8 and 10 d that would be acceptable
for activated sludge units..  A fine pinpoint floe with poor settling charac-
teristics was noted at all feed rates utilized.   This pinpoint floe may be re-
lated more to the characteristics of the bacteria in the saline media than to
feed rate.  Nitrification and denitrification in a hypersaline wastewater can
be accomplished utilizing salt-tolerant  nitrifying and denitrifying bacteria.
While not as efficient as freshwater systems, those bacteria were capable of
removing 60-70% of the ammonia-nitrogen  and 25% of the nitrate-nitrogen from
the hypersaline wastewater.  Anaerobic digestion was not feasible with the hy-
persaline wastewater.   The gas-producing bacteria in the digester sludge did
not acclimate to the 3.5% salt content of the wastewater.   Gas production
ceased at 1.3% salt.  This was the only  treatment method that utilizes fresh
water organisms and attempted to acclimate them to salt.  If a culture of salt-
tolerant anaerobic bacteria could be found, or developed,  it would seem that
this treatment method  would also be feasible.
                               LITERATURE CITED

1.  Bailey, R. W.,  and E.  G.  Scott.   1974.   Diagnostic microbiology.   C.  V.
     Mosby Co., St. Louis.

2.  Carney, J. R.,  and R.  R.  Colwell.   1976.   Heterotrophic utilization of
     glucose and glutamate in an estuary:  effect of season and nutrient load.
     Appl. Environ. Microbiol.  31:227-233.


 3.   Davis,  E.  M.,  J.  K.  Petros,  and E.  L.  Powers.   1977.   Organic biodegrada-
      tion in hypersaline wastewaters.   Indust.  Wastes 23:22-25.

 4.   Gotaas,  H. B.   1949.  Effects of seawater on the biochemical oxidation of
      sewage.  Sewage  Works J.  21:818.

 5.   Kessick, M. A.,  and K. L.  Manchen.   1976.  Salt water domestic waste
      treatment.  J.  Wat. Pollut. Control Fed. 48:2131-2136.

 6.   Kincannon, D.  F., and A.  F.  Gaudy.   1966.  Sequential substrate removal
      after change in salt concentration.  Biotech.  Eng.  8:371.

 7.   Kincannon, D.  F., and A.  F.  Gaudy.   1966.  Some effects  of high salt con-
      centrations on activated sludge.   J.  Wat.  Pollut.  Control Fed.  38:

 8.   Manchen, L. K.  1974.  Unit operations in saltwater sewage treatment.
      M.S. Thesis,  Rice University.

 9.   American Public Health Association.  1976.   Standard methods for the ex-
      amination of water and wastewater, 14th ed.  American Public Health Asso-
      ciation, Washington, D.C.

10.   Walker,  J. D., and R. R.  Colwell.   1976.  Measuring the  potential activity
      of hydrocarbon-degrading bacteria.  Appl.  Environ.  Microbiol.  31:189-196.

11.   Walker,  J. D., and R. R.  Colwell.   1976.  Oil,  mercury and bacterial in-
      teractions.  Environ. Sci.  Technol. 10:1145-1147.


                      J. B.  Healy, Jr.  and L.  Y.  Young
                    Environmental Engineering  and Science
                       Department of  Civil Engineering
                             Stanford University
                             Stanford,  CA 94305

                Decomposition of  a range  of  ligno-aromatic  compounds
           to methane under strict anaerobic conditions was observed
           over varying periods of time.  Compounds  examined include
           vanillin,  ferulic acid, cinnamic  acid, protocatechuic  acid,
           catechol,  phenol, and  syringealdehyde.  Evidence for ring
           cleavage was provided  by gas analysis and mass balance
           calculations, with conversion  of  substrate  carbon to gas
           ranging from 60% to 98%.   Methanogenic cultures  maintained
           on ferulic acid comprise several  species  forming an an-
           aerobic food chain.  Inhibition of methane  formation by
           bromoethanesulfonic acid (BESA) did not appear to affect
           decomposition of ferulic or propionic acids.  Gas chroma-
           tography demonstrated  the temporary buildup of acetate
           and proprionate as intermediates  in methane production.
           Cultures inhibited with BESA also contained butyrate,  iso-
           butyrate and isovalerate,  suggesting that pathways of  de-
           composition may be different from those of  benzoate.   The
           resulting products of  fermentation agree  with the calcu-
           lated stoichiometry, e.g., ferulic acid:
           Ci0Hi004 + 5.5 H2O -> 4.75 C02  + 5.25
     Methanogenesis plays a key role in the carbon  cycle  by  aiding  mineraliza-
tion of organic matter in highly anaerobic  environments.   Compounds which are
not degraded in the water column may eventually  find  their way  into marine
sediments, most of which are anaerobic  below the first  few centimeters.   The
fate of ligno-aromatic compounds in such environments is  of  concern because
they are pollutants discharged with wood pulping wastes,  and may constitute a
carbon sink if they are refractory under anaerobic  conditions.   This  report
provides evidence that a variety of lignin-associated aromatic  compounds are
biodegradable to methane under anaerobic conditions.

     Lignin is a complex three-dimensional  aromatic polymer  consisting largely


of phenylpropane building blocks  (Fig.  1).   Its biodegradation in  any  environ-
ment is considered to be a slow process (1).   When the complex structure  of
lignin is broken down by physical,  chemical,  or biological action  (15)  a  va-
riety of aromatic compounds is yielded.   For example, the alkaline heat treat-
ment of lignin,  as practiced by the pulp  and paper industry, is expected  to
release a range  of simple aromatics including those shown in Figure  1.
— CH



H2COH 0 0=C
H'c_l H2C
1 i
\ fpn \Cn
r\\ L ^ ^ J I ^"^ Lnr

-0 CH2
•u 0
^Vc-c-c i
-/ i i j j



                                p-HYDROXYBENZOIC ACID
                  PROTOCATECHUIC ACID
                    CINNAMIC ACID
                                    FERULIC ACID


    -OCH 3
                                                  VANILLIC ACID
Figure 1.  Model lignin polymer and  compounds resulting from alkaline heat
     The  fate  of these compounds in anaerobic environments is unclear.   Exten-
sive studies on aerobic metabolism of  aromatic compounds  (7, 14, 5) have  taken
place, while relatively little attention has been given to their  anaerobic
fate.  Early evidence suggested that some simple aromatic compounds were fer-
mentable  to C02 and CH4 (17, 4, 3, 10).    Other investigators  isolated  indi-
vidual species which anaerobically degraded aromatic compounds by photo-

metabolism (16, 11, 6)  and nitrate respiration  (18, 19).   A stabilized consor-
tium of three predominant microorganisms also has been found to degrade benzo-
ate to methane  (9).   Biochemical pathways proposed for the different kinds of
anaerobic benzoate metabolism vary in the types of aliphatic acids formed  (8).
The manner in which these acids are further metabolized depends upon the par-
ticular organism or organisms involved.  Our report focuses on the character-
istics of the methanogenic fermentation of a model lignin derivative, ferulic
acid, by an anaerobic microbial food chain.
                            MATERIALS AND METHODS


     Pre-reduced defined media contained (per liter) :  0.001 g resazurin,
0.04 g  (NHiJaPOit, 0.2 g NHi+Cl, 1.8g MgCl2.6H2O, 1.3 g KCl, 0.02 g MnCl2.4H2O,
0.03 g CoCl2.6H20, 0.0057 g H3B03, 0.0027 g CaCl2.2H20, 0.0025 g Na2MoOit «2H2O,
0.0021 g ZnCl2, 0.368 g FeCl2'4H2O, 2.64 g NaHCO3,  0.5 g Na2S«9H20, and  1%
 (v/v) vitamin solution (20).  The media was buffered at pH 7.0 by a bicarbon-
ate/CO2 system with a gas atmosphere of 30% C02 and 70% N2.  Oxygen was  re-
moved from the media by coiling followed by addition of the sodium sulfide re-
ducing agent.  Resazurin was used as an oxidation-reduction indicator.   A
C:N:P molar ratio of 100:15:1 was provided for aromatic substrate concentra-
tions of 300 mg/1 used in serum bottle enrichments.  Pre-reduced replacement
media, used for maintenance of stock cultures, contained an aromatic substrate
concentration of 10 g/1.   Additional amounts of nitrogen and phosphorus  thus
were required to sustain the same C:N:P ratio as in defined media.


     Defined media were inoculated with 10% (v/v) seed from a laboratory an-
aerobic digester fed primary settled sewage on a 15-d detention time.  Serum
bottles (250 ml) were flushed with oxygen-free gas for 20 min before addition
of the inoculated media.   Cultures were incubated in the dark at 35°C.

     Syringes were used to add aromatic substrate,  remove portions of the cul-
ture for analyses, and to monitor the volume of gas produced.  Substrate con-
centration was determined by diluting a centrifuged sample and assaying  it in
a spectrophotometer at a characteristic UV wavelength for each particular aro-
matic compound.  Gas composition was determined on a Fischer-Hamilton gas par-
titioner.   Gas production and substrate concentration were corrected for back-
ground levels by subtracting values measured in a control culture devoid of
aromatic substrate.

Cross Acclimations

     Aliquots  (7.5 ml) of acclimated stock culture were transferred to Hungate
tubes, previously flushed for 10 min with oxygen-free gas.  These cultures had
not produced gas for several days before the transfer.  Then 2.5 ml of differ-
ent pre-reduced aromatic substrates was added to each tube, returning the 10-
ml culture to original enrichment nutrient levels.   A control tube contained


no aromatic substrate.  The onset, rate, and extent of gas production in the
tube with the particular substrate to which the stock culture was acclimated
was compared with gas production in the other tubes.


     Acclimated stock cultures were maintained in serum bottles by regular re-
placement of 1/5 of the culture volume with fresh pre-reduced media.  This re-
placement was made after the culture finished converting its previous supply
of substrate to gas.  After a 150-ml culture was shaken into a homogeneous
mixture, 30 ml were withdrawn with a syringe and either wasted or transferred
as inoculum to another serum bottle.  An appropriate volume of replacement
media was then added, with the balance of 30 ml being defined media.


     2-Bromoethanesulfonic acid  (BESA) was used to  inhibit methane production
in experimental cultures.  BESA, dissolved in pre-reduced defined media, was
added to cultures at the time of media replacement  in a concentration of 10" M.


     Samples for gas chromatography of volatile fatty acids  (Ci-C?) were pre-
pared by acidifying a 2-ml portion of culture fluid with 2 drops of concen-
trated sulfuric acid and extracting this portion with 1 ml of ethyl ether.
One yl of- this ether extract was injected into a glass column packed with 10%
SP-1000/1% HsPOi, on 100/120 chromosorb W AW  (Supelco, Inc.).  A hydrogen flame
ionization detector was used in a Tracer MT 220 gas chromatograph.


     Initial enrichment studies  indicated that  a  range of aromatic compounds
are degradable under strict anaerobic  conditions  (Table  1).  Decomposition in
enrichments was slow,  first requiring  an acclimation period of about 2 weeks,
followed by gas production over  an  additional 2-  to 4-week period.  Cultures
acclimated relatively  easily to  syringic acid and syringaldehyde in 5 d com-
pared to 27 d for the  more refractory  catechol.   The conversion of substrate
carbon to gas was observed to vary,  ranging  from  59% with catechol to 98% with
syringealdehyde.  In most instances, more than  80% of the carbon was converted
to gas, indicative that ring cleavage  occurred  under these strict anaerobic

     When additional substrate was  delivered to the same enrichment cultures,
degradation to methane and carbon dioxide occurred without a lag.  This is il-
lustrated by the temporal relationship of phenol  degradation  (Fig..2).  Ini-
tial phenol enrichments required a  16-d lag  (Table 1), while subsequent en-
richments showed a relatively short 1-d lag.  The conversion of substrate car-
bon to gas in Figure 2 is very rapid as evidenced by the coincidence of


substrate  disappearance and gas production.  Methane is produced continuously
throughout the entire gas-producing period and constitutes more than half the
total gas  produced.



Vanillic acid
Ferulic acid
Cinnamic acid
Benzoic acid
Protocatechuic acid
p-Hydroxybenzoic acid
Syringic acid
Period of
gas production
% Conversion
substrate carbon
to gas


                               12   15    18   21  24   27
Figure  2.  Phenol mass balance.  Conversion of phenol  carbon  to CO2 cind Cli,, .

     The stoichiometry for the conversion of several  ring compounds  to  gas is
shown in Figure 3.  As illustrated by these equations,  50 to  60% of  the or-
ganic carbon can be converted to methane.  Based upon these stoichiometry, the
experimental conversion of these compounds to methane is  summarized  in  Table 2.
The amount of methane produced closely agrees with theoretical  values  (95-
100%); while total gas values ranged from 77 to 94% of  theoretical.
                  C6H602 +  3.5H20  -> 2.75C02  + 3,25CH[|
                                   +  5 , 5H0 -> 4.75C0   + 5,250

Figure 3.  Mass balance stoichiometry for four example  aromatic  compounds.
                          TABLE 2.   CARBON BALANCES
                  Total gas produced
                                               Methane  produced
Phenol 2.29
Catechol 1.76-2.67
Vanillic acid 1.13
Ferulic acid ' 10.7
Theoretical Actual
2.44 1
2.87 1.21
2.03 1
13.7 7
     Cross acclimation studies were undertaken by feeding  a  culture previously
grown on one aromatic carbon source a new and different  aromatic substrate.
Ut the ramje ot" compounds tested,  qenerally those with  similar  structure were
readily utilized and converted to  gaseous end products.   Figure 4 is an exam-
ple illustrating that a ferulic acid enrichment does  not use phenol, but
readily utilises cinnamic acid resulting in immediate gas  production.  It
should be noted that without this  prior acclimation to  ferulic  acid, growth on
cinnamic acid would require a 14-d lag period (Table  1)    Cultures grown on
Uu-ulLc ,-icid were observed to readily utilize cinnamic  acid, vanillic acid,
vanillin, and acetate in cross acclimation studies (Fig. 5). As illustrated
in Figure '>, Hie structure of a compound such as cinnamic  acid  may be similar
enough that the? organisms can readily utilize it.  Immediate uptake of vanil-
lin diid vanillic acid, however, suggest that these compounds may serve as

intermediates  in decomposition  of  ferulic acid.  This is supported by the ob-
servation that cultures  acclimated  to vanillin or vanillic acid cannot simul-
taneously utilize ferulic  acid.
Fiqure 4.  Gas production in cross  acclimation  studies  for  ferulic acid accli-
           mated studies.
                     o FERULIC ACID
                     D CONTROL
                     A PHENOL
                     o CINNAMIC ACID
Figure 5.  Cross acclimation results  with  ferulic  acid  cultures

     Since acetate is a known precursor for methane formation, and it is ob-
served to be readily utilized by the ferulic acid culture, its presence as an
intermediate in ferulic acid degradation was further examined by gas chroma-
tography.  Figure 6 illustrates the conversion of ferulic acid to gaseous end
products with a temporary buildup of acetate.  Significant acetate levels
first appear after most of the ferulic acid has been degraded and gas produc-
tion has begun.  This is followed by the disappearance of acetate as gas pro-
duction ceases.  These results are similar to the observations on the methano-
genic degradation of benzoic acid  (9).
                  14 r
Figure 6.  Ferulic acid conversion to methane with temporary appearance of
     Detection of other volatile acid intermediates was thought to be hampered
by their presence in very low concentrations.  To enhance the buildup of these
intermediates, active enrichments were inhibited with BESA, an analogue of co-
enzyme M  (personal communication, R. Wolfe).  It acts specifically by blocking
methyl transfer in the conversion of acetate to CO2 and CHi,.  As shown in
Figure 7A, ferulic acid is rapidly degraded; however, total gas production is
less than 20% of that in an uninhibited culture.  In addition, acetate concen-
trations were elevated to concentrations almost 4 times higher than that in
uninhibited systems.  Four other compounds, propionic, isobutyric, butyric,
and isovaleric acid were detectable only in cultures inhibted with BESA  (Fig.
7B), suggesting that when methane formation is not blocked, rapid conversion
of these organic acids takes place preventing their detection.

                ORGANIC CARBON (fi mote)
                   ro    *    o>    oo
              o    o    o    o    o
            o cz-	1	1	1     r

O    *    00     N»    O)
Figure 7.  A—Acetate buildup in BESA inhibited ferulic acid cultures.
           B—Temporary buildup of propionic acid and appearance of isobutyric,
              butyric, and isovaleric acids.

     Methanogenic enrichment cultures are able to degrade a wide range of
ligno-aromatic compounds with conversion of the substrate carbon to C02 and
CHit.  Well acclimated cultures can readily produce stoichiometric amounts of
total gas and CHi* from various aromatic compounds provided as the sole carbon
source.  The fact that these microbial populations are simultaneously accli-
mated to more than one aromatic compound suggests that compounds of similar
structure may be degraded to methane via similar pathways or partial pathways.
In addition, some simultaneously utilized compounds may serve as intermediates
during degradation, such as with vanillin or vanillic acid during ferulic acid

     For methane formation, aliphatic acids and acetate are important precur-
sors and result from cleavage of the ring structure.   The role of acetate as a
methane precursor is common in other methanogenic systems (12) .   The immediate
uptake of acetate by the ferulic acid culture indicates that an active group
of acetate-utilizing methanogens is present.

     Methane bacteria are known to utilize a very limited number of simple

compounds such as acetate, formate  and  the  reduction  of  C02  to  CH^.   The  con-
version of ligno-aromatic compounds such  as ferulic acid,  therefore,  must rely
first on other anaerobic species  for  initial decomposition,  ring  cleavage,  and
formation of intermediate volatile  acids.   Without these reactions  and  the
production of precursors, methane formation could not take place.   Therefore,
fermentation of ligno-aromatic compounds  to COa  and CHit  is dependent  upon a
microbial food chain or microbial consortium as  termed by Ferry and Wolfe (9).

     Propionate, isobutyrate, butyrate, and isovalerate  were detected in  cul-
tures inhibited with 10-l*M BESA,  without  which detectable levels  would  not
have been obtained.  The buildup  of propionate  (Fig.  7B), followed by  its  dis-
appearance when acetate buildup leveled off,  supports previous  reports  that
propionate is readily converted to  acetate  and CC>2  (13)  and  that  it is  an in-
termediate in aromatic degradation  (2).   Since rapid  disappearance  of ferulic
acid was unaffected by the BESA and decomposition of  propionate also  proceeds,
it appears that the BESA has  little or  no effect on the  other organisms in the
microbial food chain.

     The fate of the ferulic  acid molecule  may take one  or several  pathways
which is unclear at this point.   The  presence of isobutyric  and isovaleric
acid suggests that the pathway(s) may be  somewhat different  from  that which
has been elucidated for benzoate  (8).   An example of  the stoichiometry  describ-
ing conversion of ferulic acid through  one  of the volatile acid intermediates,
isovaleric acid, is shown in  Figure 8.  The presence  of  all  of  the  intermedi-
ates, except for methanol, has been detected in  our enrichment  systems.   In
addition, carbon balance data (Fig. 2)  agree closely  with the net reaction.
              FERULIC ACID
                                   2C2H402  +  CH40
                       ISO-VALERATE  ACETATE  METHANOL
                   C5HIO°2 * 4H2°
                       2CH4 + 2C02
                     CH40  + 2H
                        C02 + 2H
                        + 5.5H20 - - 4.75C02 + 5.25CH4
Figure 8.   Stoichiometry of ferulic  conversion to COa and CH4 through iso-

      In conclusion,  decomposition  of ferulic acid to CO2 and CHi* appears to be
 mediated by a - consortium of anaerobic organisms.   Compounds resulting from
 ring cleavage include  propionate,  butyrate,  isobutyrate and isovalerate which
 are likely to be converted to acetate and C02,  which in turn are used by me-
 thanogens for production of carbon dioxide and  methane.

      This work was  supported by a grant from the U.S.  Department of Energy,
 Fuels from Biomass  Program.
                               LITERATURE CITED

 1.  Alexander,  M.   1965.   Biodegradation:   problems of molecular recalcitrance
      and microbial fallibility.   Adv.  Appl.  Microbiol. 7:35.

 2.  Balba,  M.  T.,  and W.  C.  Evans.   1977.   The methanogenic fermentation of
      aromatic  substrates.   Biochem.  Soc.  Trans.  5:302-304.

 3.  Barker, H.  A.   1956.   Biological formation of methane.   In Bacterial fer-
      mentations.   John Wiley and Sons, New York.

 4.  Clark,  F.  M.,  and L.  R.  Fina.   1952.   The anaerobic decomposition of ben-
      zoic acid during methane fermentation.   Arch.  Biochem.  36:26-32.

 5.  Dagley, S.   1967.  The microbial metabolism of phenolics.   In A.  D.  Mc-
      laren and G.  H.  Peterson (eds.),  Soil biochemistry.  Edward Arnold,

 6.  Dutton, P.  L., and W.  C. Evans.   1969.  The metabolism  of  aromatic com-
      pounds by Rhodopseudomonas  palustris.  Biochem.  J. 113:525-536.

 7.  Evans,  W.  C.   1963.  The microbiological degradation of aromatic  com-
      pounds.  J.  Gen. Microbiol. 32:177-184.

 8.  Evans,  W.  C.   1977.  Biochemistry of the bacterial catabolism of  aromatic
      compounds in anaerobic environments.   Nature 270:17-22.

 9.  Ferry,  J.  G.,  and R.  S.  Wolfe.   1976.   Anaerobic  degradation of benzoate
      to methane by a  microbial consortium.  Arch. Microbiol. 107:33-40.

10.  Fina, L. R.,  and  A. M. Fiskin.   1960.   The anaerobic decomposition of ben-
      zoic acid during methane fermentation.   II.  Fate of carbons one  and seven.
      Arch.  Biochem. Biophys. 91:163-165.

11.  Guyer,  M.,  and G. Hegeman.  1969.   Evidence for a reductive pathway for
      the anaerobic metabolism of benzoate.  J. Bacteriol. 99:906-907.

12.  Jeris,  J.  S.,  and P.  L.  McCarty.   1965.   The biochemistry  of methane fer-
      mentation using  C   tracers.   J.  Water Poll. Contr.  Fed.  37:178-192.


13.  McCarty, P. L., J. S. Jeris, and W. Murdoch.  1963.  Individual volatile
      acids in anaerobic treatment.  J. Water Poll. Contr. Fed. 35:1501-1516.

14.  Ornston, L. N., and R. Y. Stanier.  1964.  Mechanism of 3-keto-adipate
      formation by bacteria.  Nature 204:1279-1283.

15.  Pearl, I. A.  1967.  The chemistry of lignin.  Marcel Dekker, Inc., New

16.  Proctor, M. H., and S. Scher.  1960.  Decomposition of benzoate by a
      photosynthetic bacterium.  Biochem. J. 76:33 pp.

17.  Tarvin, D., and A. M. Buswell.  1934.  The methane fermentation of or-
      ganic acids and  carbohydrates.  J. Am. Chem. Soc. 56:1751-1755.

18.  Taylor, B. F., W. L. Campbell, and I. Chinoy.  1970.  Anaerobic degrada-
      tion of the benzene nucleus by a facultatively anaerobic microorganism.
      J. Bacteriol. 102:430-437.

19.  Williams, R. J.,  and W. C. Evans.  1975.  The metabolism of benzoate by
      Moraxella species through anaerobic nitrate respirations:  Evidence for
      a reductive pathway.  Biochem. J- 148:1-10.

20.  Wolin, E. A., M.  J. Wolin, and R. S. Wolfe.  1963.  Formation of methane
      by bacterial extracts.  J. Biol. Chem. 238:2882-2886.


           Carl E.  Cerniglia,  Richard  L.  Hebert,  Robert  H.  Dodge,
                    Paul  J.  Szaniszlo  and David T. Gibson
                         Department  of Microbiology
                      The University of Texas  at  Austin
                              Austin,  TX  78712

                A wide taxonomic  and  phylogenetic spectrum of  fungi
           were shown to transform naphthalene.  The ability to oxi-
           dize naphthalene predominated  in  the Mucorales, but sig-
           nificant hydroxylation also  occurred in species of  Neuro-
           spora, Claviceps and Psilocybe.   The predominant metabo-
           lite formed was 1-naphthol.  Other products identified
           were 4-hydroxy-l-tetralone,  trans-l,2-dihydroxy-l,2-
           dihydronaphthalene,  2-naphthol, 1,2- and 1,4-naphtho-
           quinone.  Cunninghamella elegans  oxidized naphthalene,
           biphenyl and dibenzofuran  by reactions similar to those
           observed with mammalian enzyme systems.
     Aromatic hydrocarbons are ubiquitous  in nature.  They are  formed by me-
dium and high temperature pyrolysis  of  organic material  (2).  In marine eco-
systems, these compounds have been isolated from  ocean water, marine sediments
and certain marine organisms.  Sources  include natural submarine seepage of
petroleum and accidental oil spillage which occurs  in the production, trans-
port, and utilization of petroleum.  In addition, trace  amounts of polycyclic
aromatic hydrocarbons are released into the environment  by a wide variety of
natural and man-made sources, such as forest fires, incomplete  combustion of
organic material,  motor vehicle emissions, coal-liquefaction and gasification
processes, cigarette smoke,  and industrial wastes  (2).

     Studies in our laboratory have  been directed toward elucidating mecha-
nisms used by microorganisms for degradation of aromatic hydrocarbons and re-
lated compounds.   Such investigations have shown  that bacteria  incorporate one
molecule of atmospheric oxygen into  the benzenoid nucleus to form dihydrodiols
with a relative cis-stereochemistry  (16, 18).   Further oxidation of these in-
termediates leads  to the formation of catechols which are substrates for enzy-
matic fission of the aromatic ring (12, 29).   The above  observations are in
direct contrast to the mechanisms used  by mammalian systems for the oxidation
of aromatic hydrocarbons.   Mammalian enzymes incorporate one atom of molecular


oxygen into the aromatic substrate to form reactive arene oxides which can
isomerize to phenols or undergo enzymatic hydration to yield trans-dihydro-
diols (23, 25, 27).  There'is ample evidence in the literature to suggest that
fungi oxidize aromatic hydrocarbons in a manner similar to that observed in
higher organisms  (1, 13, 14, 28, 31).  Consequently, we have initiated a compre-
hensive research program to investigate the biodegradative potential of a wide
range of fungal species.  In addition to providing information on the differ-
ences between mammals and microorganisms with respect to oxidation mechanisms,
the results are also relevant in terms of the fate and possible effects of
aromatic hydrocarbons in the environment.  In this communication we summarize
some of our results on fungal oxidation of naphthalene and show how the tech-
niques used for the study of this substrate can be extended to other aromatic
                            MATERIALS AND METHODS

     The isolation and characterization of Cunninghamella elegans has been de-
scribed previously (8).  The sources of the fungi and the growth conditions
used in the survey for naphthalene oxidation also have been reported (11).
Small-scale fermentations were conducted in 125-ml Erlenmeyer flasks which
contained 30 ml of Sabouraud dextrose broth.  After 48 h, cultures were har-
vested as described previously  (9) and transferred to 30 ml of fresh medium.
Naphthalene, dibenzofuran and biphenyl were added to separate flasks to give a
final concentration of 0.5 mg/ml.  Two control flasks were incubated, one with
sterile Sabouraud dextrose broth, and the other with sterile dextrose broth
and substrate.  All flasks were incubated at 30°C for 48 h on a rotary shaker
at 250 rpm.  At the end of each experiment, cultures were examined microscopi-
cally for bacterial contamination.  No contamination was observed.


     After 48 h the mycelium was filtered, and the culture filtrate extracted
with 3 volumes of ethyl acetate.  The organic layer was dried over anhydrous
sodium sulfate, and the solvent removed in vacuo at 30°C.  The residue was ex-
amined for metabolic products by thin-layer, high-pressure liquid and silica
gel column chromatography and compared with the residues obtained from each
control.  A schematic representation of the separation methods is depicted in
Figure 1.


     All analytical procedures except for those noted below were as previously
described  (11).  High-pressure liquid chromatography  (HPLC) was used to sepa-
rate napththalene, biphenyl and dibenzofuran metabolites.  HPLC was performed
on a Water Associate Model 6000 A solvent delivery system.  A Model U-6K sep-
tumless injector and Model 440 absorbance detector operated at 254 nm.  Sepa-
ration was achieved with a uBondapak C\Q (3.9 mm x 30 cm).


                                               ETHYL  ACETATE
                                            I  REMOVE SOLVENT

                     RECRYSTALIZE    SOLVENT
                      COMPOUNDS  ^
                           HIGH PRESSURE LIQUID
                             //PORASIL  OR
                             •fi BONDAPAK
                                       (U.V., I.R. PMR, MASS SPEC, DERIVATIVES)
Figure 1.  Schematic representation of techniques used  to  separate,  isolate
           and identify naphthalene, biphenyl and dibenzofuran  metabolites.
     Naphthalene metabolites were separated by gradient elution.   The initial
solvent composition was 50% methanol and 50% water.  The  final  solvent concen-
tration was 95% methanol and 5% water.  A convex gradient at  curvature setting
four was employed with a flow rate of 0.4 ml/min.  The biphenyl and dibenzo-
furan metabolites were also separated by gradient elution.  The initial sol-
vent composition was 30% acetonitrile and 70% water.  The final solvent con-
centration was 70% acetonitrile and 30% water.  A gradiant curvature setting
eight was employed at a flow rate of 1.5 ml/min.  As metabolites eluted from
the column, samples were collected and immediately analyzed by  ultraviolet
spectrophotometry and mass spectrometry.  Extinction coefficients at 254 nm
were used to determine the relative amount of each metabolite produced by the
different microorganisms.


     Naphthalene and dibenzofuran were obtained from Aldrich  Chemical Company.
Biphenyl was obtained from Mallinckrodt Chemical Works.   trans-1,2-Dihydroxy-
1,2-dihydronaphthalene was a generous gift from L. A. Kapicak,  Union Carbide.
Both cis-1,2-dihydroxy-l,2-dihydronaphthalene and 4-hydroxy-l-tetralone were
prepared as described previously  (19, 9).  1-Naphthol, 2-naphthol, 1,4-naphtho-
quinone and 1,2-naphthoquinone  (J. T. Baker Company) were purified by vacuum
sublimation before use.  2-Hydroxydibenzofuran, 3-hydroxydibenzofuran and 4-
hydroxydibenzofuran were generous gifts from N. E. Stjernstrom, Astra Pharma-
ceuticals AB, Sweden.  2-Hydroxybiphenyl was obtained from Aldrich Chemical
Company, and 3-hydroxybiphenyl and 2,5-dihydroxybiphenyl  from RFR Corporation.
All solvents used for HPLC were purchased from Burdick and Jackson Labora-
tories, Inc.

                           RESULTS AND DISCUSSION

     Cells of Cunninghamella elegans grown in the presence of naphthalene pro-
duced six metabolites.  The separation of these products by HPLC is shown in
Figure 2A.  The major metabolites were 1-naphthol  (67.9%) and 4-hydroxy-l-
tetralone  (16.7%).  Minor products isolated were 1,4-naphthoquinone  (2.8%),
1,2-naphthoquinone  (0.2%), 2-naphthol  (6.3%), and trans-1,2-dihydroxy-l,2-
dihydronaphthalene  ('5.3%).  The  isolation and identification of the napthalene
metabolites has been described earlier  (9).  It has been well documented that
mammals oxidize naphthalene to naphthalene-l,2-oxide, 1-naphthol and trans-1,
2-dihydroxy-l,2-dihydronaphthalene  (20, 21, 25, 26).   Naphthol arises from the
intermediate naphthalene-1,2-oxide by spontaneous rearrangement, whereas the
trans-naphthalene diol arises by enzymatic hydration.  In contrast, bacteria
metabolize naphthalene to cis-dihydrodiols by the action of a dioxygenase (7,
19, 22) .   Our results suggest that C. elegans oxidizes naphthalene in a manner
analogous to mammalian enzyme systems.  To further demonstrate that the metabo-
lism of naphthalene by C. elegans is analogous to mammalian cytochrome P-450
systems, we prepared subcellular fractions of C. elegans and assayed for naph-
thalene oxygenase activity.  The metabolites formed from the microsomal hy-
droxylation of naphthalene were  trans-1,2-dihydroxy-l,2-dihydronaphthalene,
1-naphthol, and 2-naphthol.  The major metabolite was 1-naphthol (10).  Enzy-
matic activity was dependent on  the presence of reduced nicotinamide-adenine
dinucleotide phosphate and oxygen.  Reduced microsomal preparations when
treated with carbon monoxide showed absorption maxima at 450 and 420 nm.  The
results suggest that the metabolism of naphthalene by fungal microsomes may be
analogous to the cytochrome P-450-dependent monooxygenase activity associated
with mammalian liver microsomes.

     The techniques developed during our studies with C. elegans permitted a
detailed study of naphthalene oxidation in organisms belonging to five major
fungal taxa.  Eighty-six species of fungi in thirteen classes were investi-
gated.  Metabolites were detected by thin layer chromatography and quantita-
tively analyzed by high pressure liquid chromatography.  Approximately 55% of
the organisms investigated oxidized naphthalene  (Table 1).  Naphthalene oxida-
tion predominated in the Zygomycetes.  Species of Cunninghamella, Syncephalas-
trum and Mucor oxidized naphthalene to similar compounds as described for C.
elegans.  Other fungi showing a  similar degradative activity were Neurospora
crassa, Claviceps paspali and Psilocybe strains.  Approximately 35% of the
fungi tested produced trans-naphthalene diol.  The fungi which produced sig-
nificant 1-naphthol levels also  showed the production of 4-hydroxy-l-tetralone.
This is consistent with our previous study with C. elegans  (9) which indicated
that 4-hydroxy-l-tetralone is formed from 1-naphthol.  Previously Bollag et al.
(3) had demonstrated the formation of 4-hydroxy-l-tetralone as a bacterial oxi-
dation product of 1-naphthol.  trans-Naphthalene diol and/or 1-naphthol were
produced from some of the lower  fungi  (Chytridiomycetes and Oomycetes)
screened.  This suggests that the enzymes responsible for naphthalene metabo-
lism may have originated very early during the evolution of eucaryotic cells.
Fungi from a variety of trophic  habitats and environments were able to oxidize
naphthalene to some extent and in all cases it appears that the reactions are
similar to those that occur in mammals.

     The techniques described above were also utilized to study the metabolism
of biphenyl and dibenzofuran by  C. elegrans.


                                   ARENE  OXIDE     trans - DIHYORODIOL
                                  cis - DIHYDRODIOL
Figure 2.  Separation of metabolites formed from naphthalene  (A), biphenyl  (B)
           and dibenzofuran (C) by high pressure liquid chromatography.  Sepa-
           ration methods described under Materials and Methods.
     Biphenyl is used commercially as a fungistat in the shipping of citrus
fruits.  Although the metabolism of biphenyl by mammals  (30) and bacteria  (4,
5, 6, 17, 24)  has been widely studied, relatively little is known about  its
metabolism by fungi  (28, 31).   Biphenyl oxidation by C. elegans  produced  2-,
3- and 4-hydroxybiphenyl.  The major site of hydroxylation is at the 4-position
to form 4-hydroxybiphenyl and secondary oxidation also produced significant
amounts of 4,4'-dihydroxybiphenyl.  Less hydroxylation occurs at the 2- and

 3-positions.  When 4-hydroxybiphenyl or 2-hydroxybiphenyl replaces biphenyl as
 the substrate, C.  elegans  produces  4,4'-dihydroxybiphenyl and 2,5-dihydroxy-
 biphenyl,  respectively.  HPLC analysis of  the metabolites produced by C.  ele-
 gans grown in the presence of biphenyl is  shown in  Figure 2B.   In addition to
 free phenolic metabolites,  experiments with 1^C-biphenyl indicate that  44% of
 the known  metabolites present in  the culture medium are glucuronide conju-
 gates.   These results are  similar to those formed in the transformation of bi-
 phenyl by  mammals.
    Phlyctochytrium reinboldtae   1,2,5
    Rhizophlyctis rosea          5
    R.  harder!                   5
    Saprolegnia parasitica       5
    Phytophthora cinnamomi       5
    Thraustochytrium sp.         5
    Hyphochytrium catenoides     1,5

    Cunninghamella elegans       1,2,3,4,5,6
    C.  echinulata                1,2,3,4,5,6
    C.  japonica                  1,2,3,4,5,6
    Syncephalastrum sp.          1,2,3,4,5,6
    S.  racemosum                 1,2,3,4,5,6
    Afucor sp.                    1,2,3,4,6
    M.  heimalis                  1,2,3,5,6
    Gilbertella persicaria       1,5,6
    Absidia sp.                  1,2,5,6
    A.  glauca                    1,2,5,6
    Zygorhynchus moelleri        1,5,6
    Cokeromyces poitrassi        1,5,6
    Choanephora campincta        1,5,6
    Phycomyces blakesleenaus     1,5,6
    Circinella sp.               1,5,6
    Thamnidium anamolum          1,5,6
    Rhizopus arrhizus            1,5,6
    Basidiobolus ranarum         1,5,6
    Conidiobolus gonimodes       1,5,6
                  Smittium culisetae          5
                  S. simulii                  5
                  S. culicis                  5

                  Saccharomyces cerevisiae    1,5,6
                  Emericellopsis sp.          5
                  Neurospora crassa           1,2,3,4,5,6
                  Sordaria fimicola           5,6
                  Claviceps paspali           1,2,3,4,5,6

                  Psilocybe strictipes        1,2,3,4,5,6
                  P. subaeruginascens         1,2,3,4,5,6
                  P. cubensis                 1,2,3,4,5,6
                  P. stuntzii                 1,2,3,4,5,6
                  Panaeolus subbalteatus      5,6
                  P. cambodginensis           5,6

                  Aspergillus niger           5,6
                  Penicillium notatum         5,6
                  Gliocladium sp.             5,6
                  Epicoccum nigrium           5,6
                  Pestalotia sp.              5
*1 = trans-naphthalene diol;  2  =  4-hydroxy-l-tetralone;  3 = 1,2-naphthoquinone;
 4 = 1,4-naphthoquinone;  5 =  1-naphthol; 6 = 2-naphthol.
 Metabolites were separated and quantitated as described previously (11).

     Dibenzofuran is an oxygen-containing heteroaromatic compound identified
as a constituent of coal tar.  Chlorinated derivatives of dibenzofuran have
been identified as trace contaminants in commercial preparations of polychlori-
nated biphenyl, pentachlorophenol and hexachlorobenzene.  These chlorinated di-
benzofurans have been reported to be more toxic to mammals than the compounds
that they contaminate (15).   Despite their widespread occurrence and toxico-
logical properties, little is known about the microbial degradation of diben-
zofuran.  C. elegans grown in the presence of dibenzofuran produced four me-
tabolites.  The separation of these metabolites by HPLC is shown in Figure 2C.
C. elegans oxidized dibenzofuran to form trans-2, 3-dihydroxy-2,3-dihydrodiben-
zofuran, 2-hydroxydibenzofuran, 3-hydroxydibenzofuran and 2,3-dihydroxydiben-
zofuran.  Each product was isolated and identified by conventional chemical
techniques.  The formation of 2-hydroxydibenzofuran and 3-hydroxydibenzofuran
also suggests the formation of 2,3-dibenzofuran oxide since it is characteris-
tic of arene oxides to spontaneously isomerize to  form phenols.  These phenols
did not arise from the decomposition of trans-2,3-dihydroxy-2,3-dihydrodiben-
zoburan since control experiments indicated that this dihydrodiol was very
stable under our culture conditions and isolation  techniques.
                        ABSORPTION  (254 nm)
Figure 3.   Initial reactions used by mammals,  fungi  and bacteria to  oxidize
           aromatic hydrocarbons.

     This paper provides evidence for the fungal metabolism of aromatic hydro-
carbons.  The fungi surveyed in this report include organisms found in terres-
trial, freshwater and marine ecosystems.  Many of the fungi tested oxidized
naphthalene to some extent.  C. elegans oxidized naphthalene, biphenyl and di-
benzofuran to metabolites which are quite similar to those reported in mamma-
lian enzyme systems.  These mechanisms are compared to the metabolism of aro-
matic hydrocarbons by bacteria in Figure 3.

     The results obtained using pure cultures of fungi, chemically defined
media, and simple aromatic substrates, reveal general features of degradation
and information that may be valuable in predicting the fate of these and more
complex molecules in the environment.
     These investigations were  supported by Grants RA 04525 from the Environ-
mental Protection Agency and CA 19078  awarded by  the National Cancer Institute,
DHEW.  CEC and RD are postdoctoral  and predoctoral trainees respectively sup-
ported by Grant T32 CA  09182 awarded by the National Cancer Institute, DHEW.
We thank Joe Morgan for technical assistance and  Roberta DeAngelis for assis-
tance in preparing the  manuscript.
                               LITERATURE  CITED

 1.  Auret, B. J., D.  R. Boyd,  P. M.  Robinson,  C. G. Watson, J. W. Daly, and
     D. M. Jerina.   1971.   The NIH  Shift  during  the hydroxylation of aromatic
     substrates  by  fungi.   Chem. Commun.,  p. 1585.

 2.  Blumer, M.   1976.  Polycyclic aromatic hydrocarbons  in nature.  Sci. Am.

 3.  Bollag, J.-M.,  E.  J. Czaplicki,  and R.  D. Minard.  1975.  Bacterial metabo-
     lism of  1-naphthol.  Agric. Food  Chem.  23:85-90.

 4.  Catelani, D., C.  Sorlini,  and V. Treccani.   1971.  Experientia 27(10):

 5.  Catelani, D., A.  Colombi,  C. Sorlini,  and V. Treccani.   1973.  Biochem. J.

 6.  Catelani, D., and A. Colombi.   1974.   Metabolism of  biphenyl.  Biochem. J.

 7.  Catterall, F. A.,  K. Murray, and P. A.  Williams.   1971.  The configuration
     of the l,2-dihydroxy-l,2-dihydronaphthalene formed  in the bacterial me-
     tabolism of naphthalene.   Biochem. Biophys. Acta  237:361-364.

 8.  Cerniglia, C. E.,  and J. J. Perry.  1973.  Crude oil degradation by micro-
     organisms isolated from the marine environment.   Z.  Allg. Mikrobiol.  13:


 9.   Cerniglia,  C.  E.,  and D.  T.  Gibson.   1977.   Metabolism of naphthalene by
      Cunninghamella elegans.   Appl.  Envir.  Microbiol.  34:363-370.

10.   Cerniglia,  C.  E.;  and D.  T.  Gibson.   1978.   Metabolism of naphthalene by
      cell extracts of  Cunninghamella elegans.   Arch.  Biochem. Biophys.  186:

11.   Cerniglia,  C.  E.,  R.  L.  Hebert,  P. J.  Szaniszlo,  and D. T.  Gibson.   1978.
      Fungal transformation of naphthalene.   Arch.  Microbiol.  (in press).

12.   Dagley, S.   1971.   Catabolism of aromatic compounds  by microorganisms.
      Adv. Microbial Physiol.  6:1-46.

13.   Ferris, J.  P., M.  J.  Fasco,  F. L.  Stylianopoulou,  D.  M. Jerina,  J.  W.
      Daly, and  A.  M. Jeffrey.   1973.   Monooxygenase activity in Cunning-
      hamella bainieri: Evidence for a fungal system  similar to liver micro-
      cosms.  Arch. Biochem.  Biophys.  156-97-103.

14.   Ferris, J.  P., L.  H.  MacDonald,  M. A.  Patrie,  and  M.  A. Martin.   1976.
      Aryl hydrocarbon  hydroxylase activity in the  fungus  Cunninghamella
      bainieri:   Evidence  for the presence  of cytochrome  P-450.   Arch. Biochem.
      Biophys.  174:443-452.

15.   Fishbein,  L.   1976.   Trace organic contaminants in the environment.   I.
      Chlorinated dioxins  and dibenzofurans.  Int.  J. Ecol.  Environ.  Sci.  2:

16.   Gibson, D.  T.   1972.   The microbial  oxidation  of aromatic hydrocarbons.
      Crit. Rev. Microbiol. 1:199-223.

17.   Gibson, D.  T., R.  L.  Roberts, M.  C.  Wells,  and V.  M.  Kobal.   1973.  Oxida-
      tion of biphenyl  by  a Beijerinckia  species.   Biochem.  Biophys.  Res.
      Comm. 50:211-219.

18.   Gibson, D.  T.   1977.   Biodegradation of aromatic petroleum  hydrocarbons
      p. 36-46.   In D.  A.  Wolfe,  ed.),  Fate and  effects of petroleum  hydrocar-
      bons.  Pergamon Press,  New York.

19.   Jeffrey, A. M., H. J. C.  Yen, D.  M.  Jerina,  T. R.  Patel,  J.  F. Davey, and
      D. T. Gibson.  1975. Initial reactions in the oxidation of naphthalene
      by Pseudomonas putida.   Biochemistry  14:575-584.

20.   Jerina, D.  M., J.  W.  Daly, B. Witkop,  P. Zaltzman-Nirenberg, and S. Uden-
      friend.  1968. The  role of oxide-oxepin systems  in  the  metabolism of
      aromatic substrates. III.  Formation  of 1,2-naphthalene  oxide from naph-
      thalene by liver  microsomes. J.  Am.  Chem.  Soc. 90:6525-6527.

21.   Jerina, D.  M., J.  W.  Daly, B. Witkop,  P. Zaltzman-Nirenberg, and S. Uden-
      friend.  1970. 1,2-naphthalene oxide as an intermediate in the micro-
      somal hydroxylation  of  naphthalene.   Biochemistry 9:147-156.

22.  Jerina, D. M., J. W. Daly, A. M. Jeffrey, and D. T. Gibson.  1971.  cis-
      l,2-Dihydroxy-l,2-dihydronaphthalene:  a bacterial metabolite from naph-
      thalene.  Arch. Biochem. Biophys. 142:394-396.

23.  Jerina, D. M., and J. W. Daly.  1974.  Arene oxides:  a new aspect of
      drug metabolism.  Science 185:573-582.

24.  Lunt, D., and W. C. Evans.  1970.  The microbial metabolism of biphenyl.
      Biochem. J. 118:54p-55p.

25.  Oesch, F., D. M. Jerina, and J. W. Daly.  1971.  Conversion of naphthalene
      to trans-naphthalene dihydrodiol:  Evidence for the presence of a coupled
      aryl monooxygenase-epoxide hydrase system in hepatic microsomes.
      Biochem. Biophys. Res. Comm. 46:1713-1720.

26.  Oesch, F., D. M. Jerina, J. W. Daly, A. Y. H. Lu, R. Kuntzman, and A. H.
      Conney.  1972.  A reconstituted microsomal enzyme system that converts
      naphthalene to trans-l,2-dihydroxy-l,2-dihydronaphthalene via napththa-
      lene-l,2-oxide:  Presence of epoxide hydrase in cytochrome P-450 and
      P-448 fractions.  Arch. Biochem. Biophys. 153:62-67.

27.  Oesch, F.  1973.  Mammalian epoxide hydrases:  Inducible enzymes catalyz-
      ing the  inactivation of carcinogenic metabolites derived from aromatic
      and olefin compounds.  Xenobiotica 3:305-340.

28.  Smith, R. V., and J. P. Rosazza.  1974.  Microbial models of mammalian
      metabolism.  Aromatic hydroxylation.  Arch. Biochem. Biophys. 161:

29.  Stanier,  R. Y., and L. N. Ornston.  1974.  The g-ketoadipate pathway.
      Adv. Microbial Physiol. 9:89-151.

30.  Sundstrom, G., O. Hutzinger, and S. Safe.  1976.  The metabolism of
      chlorobiphenyls—a review.  Chemosphere 5:267-298.

31.  Wiseman,  A., J. A. Gondal, and P. Sims.  1975.  4'-Hydroxylation of bi-
      phenyl by yeast containing cytochrome P-450:  Radiation and thermal sta-
      bility,  comparisons with liver enzyme  (oxidized and reduced forms).
      Biochem. Soc. Trans. 3:278-281.

                       IN BIODEGRADATION STUDIES

                                  D. Liu
                         Toxic Substances Section
                     National Water Research Institute
                      Canada Centre for Inland Waters
                        Burlington, Ontario, Canada

                A selective enrichment technique has been developed
           for rapid  isolation of lipophilic compound-degrading mi-
           croorganisms.  This method is based on the fact that bio-
           degradation  of lipophilic substances takes place mainly at
           the substance-water interface.  Thus, the areal extent of
           this interface will determine the availability of sub-
           strate to  the degrading microorganisms and thereby con-
           trol the primary biodegradation rate as well as the micro-
           bial biomass.  The isolation procedure involves finely
           emulsifying  the test  substance and stabilization of the
           emulsion with sodium  ligninsulfonate.  Microorganisms
           capable of degrading  n-alkanes, aromatic hydrocarbons,
           phenols and  polychlorinated biphenyls were isolated using
           this technique.

     By definition,  an enrichment  technique implies the creation of an artifi-
cial environment which will  either permit the desirable microorganisms to mul-
tiply at a faster rate than  others or will suppress the growth of all but the
desirable microorganisms  (1).  Enrichment techniques have made an important
contribution to development  of modern microbiology in studies of microbial
degradation and pathways.  The literature on enrichment technique is vast and
widely scattered, however, most  procedures deal with isolation of microorga-
nisms which will grow on hydrophilic compounds.  Only limited information is
available on isolation of microorganisms capable of degrading lipophilic com-
pounds .

     As more and more new organic  compounds are being synthesized, the threat
of environmental pollution caused  by these new chemicals increases accord-
ingly.  Most are of  lipophilic nature  (hydrophobic) which are inherently more
resistant to biodegradation  than are hydrophilic ones.  This is probably due


to the requirement of most microorganisms for an aqueous phase to carry out
their metabolic activity; furthermore, the hydrophobic nature of these new
chemicals makes them less susceptible to microbial attack.  Studies in our
laboratories have been concerned with the biodegradation of persistent toxic
organic substances and the isolation of lipophilic-compound-degrading micro-
organisms by an enrichment technique.
                            MATERIALS AND METHODS

     Source of Inoculum.—Activated sludge from the local municipal sewage
plant was used as the major source of inoculum in the biodegrdation study
since the bacterial types  (Alcaligenes, Acinetobacter, Flavobacterium, Pseudo-
monas and Escherichia) and their concentration (7 x 107 to 2 x 108 ml"1) in
the sludge were observed to remain relatively stable throughout the year.  Oc-
casionally, samples of activated sludge or soil from local industrial plants
also were used as the bacterial seedings.

     Growth Medium.—To control the variables normally encountered in a growth
medium containing complex organic nutrients such as yeast extract and peptone,
the following chemically defined mineral broth was chosen as the basal medium
(g liter"1):  KzHPO^, 1.3; KH2POit, 0.82;  (NHit)2SOit, 1.0; MgSOi, = 7H2O, 0.05;
FeSOi+«7H2O, 0.01; CuSOi^SHzO, 0.01; CoCl2«6H2O, 0.001; MnCl2-4H20, 0.001;
NaCl, 0.05.  The final pH was 6.9.  The medium was sterilized at 121°C for 15
min.  Sodium ligninsulfonate and the test substances were sterilized sepa-
rately before being added to the basal medium.

     Preparation of Emulsion.—One hundred milligrams of the test substance
was added to a sonic cup containing 10 ml of distilled water and 5 mg of so-
dium ligninsulfonate.  Solid test substance was dissolved in a minimal amount
of n-hexane.  The mixture was subjected to pulse ultra-sonification for ap-
proximately 1 min until it was emulsified.  The emulsion was freshly prepared
prior to the initiation of each experiment.

     Preparation of Emulsion-Agar Plate.—Fifty milligrams of the test sub-
stance were made into 10 ml of a fine emulsion as described above under asep-
tic conditions, and the emulsion was then mixed with 500 ml of sterile basal
agar (approx. 50°C) for pour plate.  Plates made in this manner were semi-
translucent and colony development was slow  (2-7 d).

     Enrichment Culture.—Various amounts of the test substance in emulsion
form were added to 125-ml Erlenmeyer flasks containing 50 ml of mineral medium
and 5 mg of sodium ligninsulfonate to give final concentrations of the test
substance of 10, 20, 50, 100 and 300 mg liter"1.  Fresh activated sludge  (0.1
ml) was added to each flask as the inoculum, with incubation at room tempera-
ture (21°C) on a gyratory shaker for 1-4 weeks or longer.  The samples were
checked daily under a phase contrast microscope and the test substance was
monitored for biodegradation using gas chromatography.  If evidence of bio-
degradation was noticed, the culture broth was then used for isolation of the
degrading microorganism on emulsion agar plates.


     Biodegradation Measurement.—In all experiments, only the primary biodeg-
radation of the test substance was followed, i.e., the loss of parent com-
pounds.  Five-miHiliter samples of the culture broth and 1 ml of n-hexane
were mixed vigorously in a 15-ml conical glass centrifuge tube on a vortex
mixer for 1 min.  The emulsion was broken by centrifugation at 2,000 x g for
10 min and the clear hexane extract was used for GC analysis, using two gas
chromatographs equipped with FID detectors.  A Beckman GC-65 was fitted with
dual 1.8 m x 6.3 mm o.d. stainless steel columns containing 2% OV-1 on Chrom-
sorb G-AW-DMCH.  A Hewlett-Packard 5730A was fitted with dual 1.8 m x 6.3 mm
o.d. glass columns containing 10% OV-1 on Chromsorb W-HP.  The oven tempera-
ture was programmed from 100-250°C at 5-8°C min"1  and the temperatures at the
injector port and detector were at 250 and 300°C,  respectively.

     Growth Determination.—Cultures were in 2-liter Bellco spinner flasks at
room temperature (21°C) and at various intervals samples were withdrawn for
dry weight determination  (5).

     Manometric Technique.—Oxygen consumption was measured at 20°C in a Gil-
son differential respirometer.  Cells were harvested from the growth culture,
washed and resuspended in cold 0.05 M phosphate buffer (pH 7.0).  A typical
reaction mixture contained 1 ml of cells, 1 ml of  0.05 M phosphate buffer,
0.9 ml of water, 0.1 ml of substrate and 0.15 ml of 20% KOH in the center well
for CO2 absorption.  The final fluid volume was 3.2 ml.

     The first successful application of this enrichment technique in this
laboratory involved isolation of a PCB-degrading Pseudomonas sp. from the ac-
tivated sludge.  The development of the technique for rapid isolation of lipo-
philic compound-degrading microorganisms was explored by using this bacterium
and the commercial PCB mixture (Aroclor 1221).   Initially,  an attempt was made
to incorporate the PCB into the growth medium by dissolving it in an appropri-
ate solvent such as acetone.  As expected, the major portion of the added PCB
separated out as soon as the acetone solution was added.  No significant deg-
radation of PCB was noticed in the growth medium in spite of microbial growth.
Further investigation revealed that most sewage microorganisms could utilize
acetone as a source of carbon and energy for growth.  Microorganisms in acti-
vated sludge were capable of rapidly oxidizing acetone or acetate (Fig. 1) ,
thus, it is inadvisable to use acetone as solvent in the enrichment technique
since it may serve as an alternate carbon source.

     When ultrasonically treated PCB (as emulsion)  was added to the mineral
medium containing sodium ligninsulfonate as an emulsion  stabilizer, signifi-
cant degradation of PCB (34%)  was detected within 9 d.  Further incubation to
11 and 14 d resulted in 71% and 99% biodegradation of the PCB, respectively
(Fig. 2).  From this culture,  a PCB-degrading Pseudomonas sp. 7509 capable of
degrading PCB from concentration of 300 mg liter"1  to a level of less than
1 mg liter"1 within 48 h was isolated.   However,  the growth rate of the
pseudomonad was greatly retarded if non-emulsified PCB was  employed in the
growth medium.  Examination by phase contrast microscopy of cultures grown


with unemulsified PCS revealed that the cells were intimately associated with
the dispersion of the PCB droplets in the medium (Fig.  3).   Most of  the PCB
droplets were surrounded by the bacterial cells indicating  the possible in-
volvement of bacteria in dispersion of PCB droplets in  the  growth medium.  Ex-
amination with interference phase contrast microscopy showed that the  cells
actually attached themselves to the surface of PCB droplets, suggesting that
microbial degradation of PCB could be occurring at the  PCB-water interface
 (Fig. 4).  Similar observations were also noticed when  this enrichment tech-
nique was applied to isolation of other lipophilic compound-degrading  micro-
organisms such as a p-cresol-degrading bacterium (Fig.  5).   In a medium with
PCB as the only carbon and energy source, pseudomonad 7509  gave a near-linear
cell yield as a function of concentration of PCB emulsion  (Fig.  6).
       300 -i
                             TIME  IN MINUTES
 Figure 1.   Oxidation of acetate  and  acetone by activated sludge,
            -•— Endogenous;  -•— Acetate; -A— Acetone


     6 60

Day 0

                           O 60-
                           DC 50-

                           8 40-
                                                                Day 9


O 60-

                       Day 11


                           8 40-
Figure 2.   FID gas  chromatograms  of n-hexane extracts  of Aroclor 1221 at zero
             time, after  9 days, 11 days  and  14 days incubation  with  activated

Figure 3.  Micrographs of the culture broth taken at different growth stages
           showing the dissociation of PCB droplets.  A = early log phase;
           B = middle log phase.  Phase contrast at 400X.

Figure 4.  Interference phase micrograph showing the bacterial growth at the
           PCB-water interface.   Interference phase contrast at 1000X.
Figure 5.   Interference phase micrograph showing the growth of a p-cresol
           degrading bacterium at the p-cresol-water interface (1000X) .

Figure 6.  The  growth yield  of Pseudomonas  sp. 7509 as a function of Aroclor
           1221 concentration in  the  growth medium.
     The rate of PCB degradation by bacteria was also followed by the rate of
oxygen consumption  (2).  The  relationship between oxygen consumption and the
concentration of PCB in the flask, i.e., the amount of oxygen used by the
cells, was a direct function  of the concentration of PCB (Fig. 7).  For all
PCB concentrations employed,  there was no lag period for oxygen utilization,
suggesting that the bacterial cells oxidized PCB immediately, i.e., the cells
grown by this enrichment method possessed vitality sufficient to attack the
lipophilic compound.  The UV-VIS spectrum of sodium ligninsulfonate recovered
from the culture broth showed little difference from the control, indicating
that sodium ligninsulfonate was not degraded in the process and hence did not
provide an alternate carbon and energy source for the cells.

     Investigators studying biodegradation of lipophilic substances in the
aquatic environment have been confronted with the problem of finding a means
to dissolve or disperse such substances in growth media.  A number of enrich-
ment methods have been used for isolation of PCB-degrading bacteria, including
dissolving of PCB in acetone (6, 8), ethanol  (7) and diethyl ether  (3).  All of
these methods have their drawbacks.  Acetone and ethanol are good substrates
for microorganisms and provide alternate carbon sources for bacterial growth.

Diethyl ether is immiscible with aqueous media.  In addition, since PCS cannot
be kept.in tire growth media at concentrations above their solubilities by the
aforementioned techniques, this means that bacterial cells have little oppor-
tunity to contact the PCB; consequently, the likelihood of enhancing PCB-
degrading microorganisms is decreased.  Finally, the use of organic carbon-
rich nutrients such as glucose and peptone in the enrichment medium  (7) is of
questionable value in that numerous different microorganisms can readily uti-
lize such nutrients.  The bacteria which predominate in such growth media need
not necessarily be the lipophilic degrading ones.
           500 n
—I	1—
 4        6
 Aroclor 1221
Figure 7.  Oxygen uptake as a function of the Aroclor 1221 concentrations in
           the flask.
     The enrichment technique described here avoids most of the problems en-
countered with previous methods.  The production of a fine emulsion of the
lipophilic substance through ultrasonic means provides a much larger lipophile
water interfacial area, thus allowing the microorganisms to overcome the sub-
strate-limiting aspect which may determine subsequent bacterial growth rate.
Sodium ligninsulfonate is used rather than commercial emulsifiers due to its
recalcitrance to microbial action.  Therefore, most microorgani'sms isolated
by this technique are very active in degradation of the lipophilic compounds.

     However, this enrichment technique is useful only when the tes-t substance
can be metabolized by the microorganism in question.  Certain lipophilic

substances, such as DDT, are only degraded by cometabolism (4).  However, if
necessary, a cometabolite can be applied in the presently described enrichment
technique.  The latter can be used reliably to obtain some lipophilic sub-
stance-degrading microorganisms from the environment.
                              LITERATURE CITED

1.  Aaronson, S.  1970.  Experimental microbial ecology.  Academic Press, New

2.  Ahmed, M. , and D. D. Focht.  1973.  Degradation of polychlorinated bi-
     phenyls by two species of Achromobacter.  Can. J. Microbiol. 19:47-52.

3.  Baxter, R. A., P. E. Gilbert, R. A. Lidgett, J. H. Mainprize, and H. A.
     Vodden.  1975.  The degradation of polychlorinated biphenyls by micro-
     organisms.  Sci. Total Environ. 4:53-61.

4.  Focht, D. D., and M. Alexander.  1971.  Aerobic cometabolism of DDT ana-
     logues by Hydrogenomonas sp.  J. Agric. Food Chem. 19:20-22.

5.  Liu, D.  1973.  Microbial degradation of crude oil and the various hydro-
     carbon derivatives, p. 95-104.  In D. G. Ahearn and S. P- Meyers (eds.),
     The microbial degradation of oil pollutants.  Center for Wetland Re-
     sources, Louisiana State University, Baton Rouge.

6.  Sayler, G. S., M. Shon, and R. R. Colwell.  1977.  Growth of an estuarine
     Pseudomonas sp. on polychlorinated biphenyl.  Microbial Ecol. 3:241-255.

7.  Tucker, E. S., V. W. Saeger, and 0. Hicks.  1975.  Activated sludge pri-
     mary biodegradation of polychlorinated biphenyls.  Bull. Environ.
     Contamin. Toxicol. 14:705-713.

8.  Wong, P. T. S., and K. L. E. Kaiser.  1975.  Bacterial degradation of
     polychlorinated biphenyls.  II. Rate studies.  Bull. Environ. Contamin.
     Toxicol. 13:249-256.

                       CONTINENTAL SHELF WATERS

                      A.  E. Maccubbin and Howard Kator
                   Department  of Microbiology-Pathology
                   Virginia  Institute of Marine Science
                         Gloucester  Point, VA 23062

                Bacterial populations  indigenous to surface  (1 m)
           waters of the Middle  Atlantic Continental Shelf were
           sampled at seasonal intervals to determine the abundance
           and distribution of petroleum-degrading  (HC), chitinoclas-
           tic, and "total" heterotrophic  (HET) bacteria.  Simulta-
           neously,  unweathered  South  Louisiana crude oil was added
           to aliquots of 1-m water  samples to evaluate rates and
           patterns of saturated paraffin degradation on shipboard
           under controlled incubation conditions.  Degradation was
           examined under selected nutrient and temperature regimes
           in both closed flasks and prototype "open" or continuous
           dilution systems.  HET bacterial levels generally de-
           creased with distance from  land.  HC bacteria were most
           abundant in the coastal boundary layer.  Changes in values
           of HC/HET tended to be directly related to the abundance
           of petroleum-degrading bacteria.  Degradation of saturated
           paraffins in closed flasks  characteristically resulted in
           removal of n-alkanes  and  isoprenoids at rates related to
           season, sampling location,  and nutrient regime.

     Coincident with proposed leasing  of tracts for oil exploration on the
Middle Atlantic Continental Shelf  (MACS), we were given the opportunity to as-
sess the seasonal distribution,  abundance and relative in-vitro petroleum-
degrading capabilities of marine bacteria.  Of particular interest were how
viable counts of petroleum-degrading bacteria and the ratio of petroleum-
degrading to "total" heterotrophic bacteria varied with season and proximity
to land.  Previous authors (3, 6, 9, 10, 11, 16, 17) have indicated that in areas
of increased levels of pollution by  petroleum or related compounds, the viable
counts of petroleum-degrading bacteria were elevated.  Similarly, increases in
the value of the ratio of petroleum-degrading to "total" heterotrophic bacte-
ria have been observed under these conditions  (6, 15).

     Although viable counts of petroleum-degrading bacteria and/or increase in
the ratio of this group to "total" heterotrophs may be indicative of the pres-
ence of petroleum (or allied compounds), these values yield no direct informa-
tion concerning the potential of bacteria to degrade petroleum.  Therefore, a
sequence of closed flask "batch" degradation experiments were performed to as-
sess the influence of season, inorganic enrichment, and water type on on in-
vitro degradation of South Louisiana crude oil.  Finally, to overcome several
cultural limitations inherent in closed flask systems, we developed and per-
formed a series of preliminary studies with an "open" or continuous dilution
system for simulation of microbial weathering of petroleum at sea.
                            MATERIALS AND METHODS

     Surface water samples were collected during the four biological seasons
at selected stations  (Fig. 1) at 1-2 m depth using Niskin sterile bag samp-
lers.  Precautions were taken to prevent contamination by discharge from the
vessel.  Water samples were immediately processed for enumeration of "total"
heterotrophic and petroleum-degrading bacteria.  Appropriate dilutions or con-
-centr-atlons—on—membrane filters were prepared prior to
Figure  1.   Sampling stations,  Middle Atlantic continental shelf waters.


     Seawater inocula or appropriate dilutions or concentrates thereof were
enumerated for "total" heterotrophic bacteria (HET)' using a three-tube MPN
technique  (7) in a heterotrophic medium (HM) modified after ZoBell ' s 2216E to
reduce inorganic precipitate formation.  This medium consisted of 1.0 g/liter
peptone, 0.5 g/liter yeast extract, 0.01 g/liter ferric citrate, 0.1 g/liter
sodium glycerol phosphate, and 1000 ml of aged seawater (final pH of 7.8 after
sterilization) .  Petroleum-degrading bacteria (HC)  were also enumerated using
a three-tube MPN technique (5)  employing a minimal salts enriched seawater
(ESWB) containing 1.0 g/liter (NHiJ 2SOtt and 0.1 g/liter KaHPCH in 1000 ml of
aged seawater.  Following inoculation, approximately 1% sterile unweathered
South Louisiana crude oil (SLCO) was added as the carbon source.  Petroleum
was sterilized by filtration through a 0.4 y polycarbonate membrane.  All
media were autoclaved at 121°C for 15 min.

     HM tubes were incubated at 20-22°C for two weeks and examined at weekly
intervals.  ESWB + petroleum (ESWBP) tubes were incubated at similar tempera-
tures on a rotary shaker platform  (140 rpm) for one month and examined bi-
weekly.  HM tubes were scored positive when turbid;  ESWBP tubes were scored
positive when turbid or if the oil showed obvious signs of degradation with
associated cell debris, or a combination of both.  MPN values were calculated
using standard tables  (1) .


     Four replicate sets of pre-washed flasks (250  ml, Erlenmeyer)  received
the following treatments:  (a)  sterile control — 100 ml sterile seawater plus
100 yl sterile unweathered SLCO; (b) non-enriched — 100 ml seawater plus  100
yl sterile unweathered SLCO;  (c) enriched — 100 ml seawater plus 100 yl sterile
unweathered SLCO plus 1 ml of a nutrient solution containing 1.0 g/liter
          and 0.1 g/liter KaHPO^;  (d) oil-free control — 100 ml seawater.
     Flasks were loosely covered with foil and incubated on a rotary shaker
platform (120 rpm)  at a selected seawater temperature representative of that
season.  At selected intervals, one randomly selected flask from the oil-free
control, non-enriched, and enriched treatments were enumerated for HET and HC
viable counts.  After enumeration, the non-enriched and enriched flasks and a
sterile control flask were "pickled" with methylene chloride for subsequent
analysis of residual petroleum.


     Three replicate groups of 500 ml Erlenmeyer flasks were arranged to re-
ceive the following treatments:  (a) sterile control — 50 ml sterile seawater
plus a low concentration of SLCO (0.01-0.1%, v/v) ;  (b) inoculated — 50 ml
freshly collected "ambient" seawater plus a low concentration of unweathered
SLCO (0.01-0.1%, v/v); (c) oil-free control — 50 ml  freshly collected "ambient"

     Following the  initial inoculation,  flasks were incubated on an Orbit


Environ-Shaker  18  (Lab-Line Instruments, Inc.) at 100 rpm and average seasonal
ambient temperature.   During incubation, flask contents were diluted by  adding
seawater at  the rate  of 1 ml/h.  Seawater was pumped using a multi-channel
peristaltic  pump (Desaga, Brinkman Instruments) from two reservoirs.  One res-
ervoir contained freshly collected "ambient" seawater and the other contained
seawater + 1% formalin.  Reservoirs were replenished at regular intervals
(Fig. 2).  At various incubation intervals, representative flasks for each ex-
perimental treatment  were randomly harvested and HET and HC enumerated.  Glass
distilled methylene chloride was then added to the inoculated and sterile con-
trol flasks  as  a "pickling" agent in preparation for analysis of the residual
 Incubation at Seasonal
 Temperature with Rotation
  at lOOrpm.
+ 1%
freshly collected

Figure  2.   Schematic of prototype ''open" or continuous dilution system.


      "Pickled" flasks were acidified (pH 3.0-4.0) with HCl and extracted with
two  25-ml portions of methylene chloride.  Methylene chloride phases were  com-
bined and washed with 50 ml of acidified distilled water  (pH 4.0) containing
3% NaCl.  After phase separation, the organic phase was dried by passage
through anhydrous sodium sulfate and collected in a clean, solvent-washed,
pre-weighed evaporation flask.  Solvent was then removed by aspi-ration  at  40°C
using a McLeod gauge to instantly detect complete solvent removal.  The ex-
tracted residue was weighed and dissolved in 1 ml of hexane.


     Extracts were fractionated by column chromatography.   Glass columns  (1 cm
diameter) packed to a height of 17.5 cm with activated silca gel  (heated at
235°C for 16 h) were pre-washed with four (4) 10-ml portions of glass dis-
tilled hexane.  Extracts were then placed on columns in 1  ml hexane and eluted
with 17 ml hexane.  The first 5 ml of hexane eluting were discarded and the
next 13 ml hexane, designated the H2 fraction and containing saturates, were
saved.  Columns were next eluted with 30 ml of a hexane/ben/ene (40/60 v/v)
mixture and an aromatic-containing fraction collected.  Fractions were dis-
solved in small volumes of hexane and stored at -4°C for subsequent analysis.


     Analysis of H2 fractions was performed using a Tracor 560 gas chromato-
graph equipped with dual flame ionization detectors and a  Grob capillary in-
jection system.  Samples (2-2.5 yl) were injected with the  split closed and the
oven temperature programmed to 240°C at 4°/min with a final hold at 240°C of
10 min.  Glass capillary columns  (20 m x 0.33 mm i.d.) coated with 0.2% SE-52
were used.  Injector and detector temperatures were set at 240°C and 325°C,
respectively.  Carrier gas (He) flow through the column was 2-3 ml after

     Chromatograms were evaluated for degradation expressed as loss of normal
paraffins (nCio-nC2s, inclusive).  Identification of n-paraffins was by com-
parison of retention times with authentic standards.  Changes in n-paraffin
peak areas were expressed as the ratio of each n-paraffin  to the naturally oc-
curring isoprenoid hydrocarbon pristane (2,6,10,14-tetramethylpentadecane) .
Peak areas and retention times were recorded electronically.  A conservative
estimate of the absolute concentration of pristane/unit weight crude oil was
determined by the technique of standard addition (8).  n-Paraffin losses in
non-enriched and enriched experimental flasks were compared to sterile con-
trols to adjust for evaporation during incubation.   On the basis of complete
workup of replicate blanks and controls, n-paraffin summation losses smaller
than 10% of the control summation values were not considered significant.


     ATP concentrations were measured by the method of Strickland and Parsons
(12).  Concentrations of dissolved organic carbon,  nitrates, phosphates, dis-
solved aliphatic and aromatic hydrocarbons were obtained from relevant chemi-
cal and biological benchmark data (14, 15) .

     Figure 1 is of the (MACS)  study area and indicates the locations of 1-m
surface water stations.  Station Cl was located east of Atlantic City in the
coastal boundary layer.  Stations Dl, N3, and E3 were situated in inner and
outer shelf waters.  F2 was located in the shelf break region (100-m contour)
and Jl was located in slope water.  Seasonal variations in water type were


noted at station F2 which was frequently located in a frontal zone caused by
mixing of shelf and slope water.  Water types at Station Jl were sometimes
characterized as eddies from the Gulf Stream.

     Results from enumeration of viable counts of petroleum-degrading and
"total" heterotrophic bacteria are shown in Figures 3 and 4 and Tables 1 and 2.
Generally, values of heterotrophs decreased with increased distance from land,
with counts falling from about 105 bacterial units/ml to 102 bacterial units/
ml.  Geometric means values  (Tables  3 and 4) for all transect HET counts did
not vary much with season.  Considered on a station basis, however, geometric
mean values tended to decrease directly with distance from land.

          TRANSECT DURING 1975-1976
Log Log
Station HC/ml HET/ml
NO 3
Fall 1975
• -0






























HC = Petroleum-degrading bacteria;  HET = Heterotrophic bacteria; DOC = dis-
solved organic carbon; ALI = Dissolved aliphatic hydrocarbons.

                                                     N = Bacterial units/ml

                                                        Total Heterotrophic Bacteria
                                                        Petroleum Degrading Bacteria
                  Cl   Dl   N3   E3
Figure 3.  Viable counts of heterotrophic and petroleum-degrading  marine bac-
            teria in surface seawater  samples  (1  m)  collected during 1975-1976.
                                               ~2 J   Cl   Dl   N3   E3   F2
                                                 Jl  Jl  J  JL
                                          Cl   Dl   N3   E3   F2  /J I
                                                          N = Bacterial units / ml

                                                          D TotaI Heterotrophic Bacteria
                                                          • Petroleum Degrading Bacteria
                      Cl   Dl   N3  E3   F2
Figure  4.   Viable  counts of  heterotrophic and petroleum-degrading  marine bac-
            teria in surface  seawater  samples  (1  m)  collected during 1976-1977.

Fall 1976
Winter 1977
oo F2
Spring 1977
Summer 1977
Log Log
HC/ml HET/ml















NO 3















 HC  =  Petroleum-degrading bacteria;  HET  =  Heterotrophic bacteria; DOC = Dissolved organic carbon;
 ALI = Aliphatic  hydrocarbons;  N.A.  =  Not  available.

          SAMPLES (1975-1976)
log HC/ml
log HET/ml
log HC/ml
log HET/ml







Spring 19


76 Summer 19
F2 Jl
-0.1±1.2 -1.1±1
2.6±0.7 2.6±0

          SAMPLES  (1976-1977)
By season
Fall 1976






.4 11

1977 Spring 1977 Summer 1977
By station
0.210.3 0
4.111.0 2

E3 F2
.2+0.4 O.U0.8
.9+0.7 2.610.5


     Viable counts of petroleum-degrading bacteria exhibited similar trends as
counts decreased from inshore maxima of 10s bacterial units/ml to less than
3.2 x 10"3 bacterial units/ml at Jl during the fall of 1975.  Seasonally,
viable counts tended to be lowest during the winter inshore and highest during
the summer.  Viable counts along the transect tended to be related to water
types.  Thus, station Cl, which generally exhibited the greatest populations
of HC, was located in the coastal boundary layer.  Stations Dl, N3, and E3,
which either exhibited lower or the lowest counts as one progressed seaward,
were located in inner-outer shelf waters.  Station F2 occasionally exhibited
elevated values relative to E3 and Jl.  This station was frequently located
in a convergence zone produced by the mixing of shelf and slope waters.  Up-
welling, associated with the convergence, apparently produced elevated bio-
logical activity owing to elevated inorganic nutrient levels.

     Geometric mean and individual station count data indicated that the win-
ter of 1977 was characterized by the lowest viable counts of petroleum-degrad-
ing bacteria.  Generally, stations and seasons exhibiting counts with the

 largest standard deviations were those most strongly affected by changes  in
 climate or hydrodynamic processes,  i.e.,  least stable.   Ratios of petroleum-
 degrading to "total"  heterotrophs varied  directly with  viable counts  of petro-
 leum-degrading bacteria.

      Physical and chemical parameters measured during this  study are  presented
 in  Tables 1 and 2.   Surface water temperatures generally increased with dis-
 tance from land with the winter seasons illustrating the strongest temperature
 gradient.  This was especially so during  1977  when the  temperature at Cl  was
 -0.5°C and pancake  ice was present.   Salinity  also increased  with distance
 from  land.   Levels  of nitrates and phosphates  did not vary  with recognizable
 patterns with station but appeared to be  greatest during the  winter of 1977
 when  phytoplankton  activity would be expected  to  be a minimum.   As previously
 mentioned,  elevated levels of nutrients were sometimes  observed at the shelf
 break station F2.   Slope water at Jl and  intrusions of  slope  water further  in-
 shore generally resulted in the lowest values  of  inorganic  nutrients.

      Linear regression correlation coefficients were calculated for bacterial
 populations against various parameters (Tables 5  and 6).  Significant  positive
 correlation (a = 0.05)  was observed between petroleum-degrading bacteria  and
 heterotrophic bacteria during 1975-76 and 1976-77.   A significant (a = 0.05)
 positive correlation  between petroleum-degrading  bacteria and the ratio of
 petroleum-degrading bacteria to "total" heterotrophs was  noted.   ATP  concen-
 trations, measured  only during 1976-77, were significantly  (a = 0.05)  corre-
 lated with "total"  heterotrophic bacterial counts.   Other significant  (a  =
 0.05)  positive correlations were noted between petroleum-degrading bacteria
 and dissolved aliphatic hydrocarbons (1976-77)  and between  phosphate  concen-
 trations "and viable bacterial counts (1975-76).

             HC          HET         ALI         DOC         NOa          POi*
**Significant at a = 0.01; HC—Petroleum-degrading bacteria; HET—Heterotro-
phic bacteria; ALI—Dissolved aliphatic hydrocarbons; DOC—Dissolved organic
     Using closed flask systems, certain trends in viable counts of petroleum-
degrading and "total" heterotrophic bacteria were observed with regularity.
An example of these is shown in Figure 5.  In all experimental treatments,
levels of petroleum-degrading and heterotrophic bacteria increased from ini-
tial values with counts in enriched flasks being greater than either inocu-
lated or control flasks.  Elevated viable counts were maintained throughout
the incubation period.  The value of the ratio of petroleum-degrading to


heterotrophic  bacteria increased  above initial values  in all flasks within 3 d
and was generally larger in flasks with oil.

           TERS SAMPLED DURING 1976-1977

NO 3
*Significant at a = 0.05; **Significant at a = 0.01;  HC~Petroleum-degrading
bacteria;  HET—Heterotrophic bacteria;  ALI—Dissolved aliphatic hydrocarbons;
DOC—Dissolved organic carbon.
    Station E3, Water Inoculum
                                                    N  Bacteriol units / ml seawater

                                                    Petroleum degrading   D Heterotrophic
                                                    |  Inoculum
                                                    H  Control
                                                    (U  inoculated
                                                    E23  Enriched
Figure  5.   Typical changes observed  in viable counts of heterotrophic and
            petroleum-degrading marine  bacteria for closed  flask petroleum
            degradation studies.

     Degradation was assessed by disappearance of selected n-paraffins.  Sig-
nificant degradation, i.e., greater than 10% of the control summation of n-
paraffins, was noted only in enriched flasks.  In all cases, similar patterns
of degradation, i.e., utilization of n-paraffins of short chain length prior
to n-paraffins of longer chain lengths and isoprenoids, were observed.  Rates
of degradation and extent varied with season and inoculum source.  Average
rates of degradation were calculated based on the number of days of incubation
required for degradation of 50% of the n-paraffins (Table?).  The latter were
degraded most rapidly during the fall and slowest during the winter.  Extent
of degradation was defined as the maximum percentage of selected n-paraffins
degraded during the incubation period (Table 8).  Fall exhibited the greatest
extent of degradation and winter the smallest.  Despite the highest incubation
temperature, experiments with summer inocula yielded relatively low rates and
extent of degradation.  With respect to inoculum source, stations Cl and N3
exhibited the greatest relative rates and extent of degradation.  A consistent
relationship between viable counts of petroleum-degrading bacteria and extent
of degradation was not evident.






Temperature (
Fall 1976

°C) 15
Winter 1977

Spring 1977

Summer 1977

aBased on weight loss in treated flask as compared to control; ^Inoculated =
seawater + oil, Enriched = seawater + oil + nutrient amendment; NR = Degrada-
tion never reached 50%.

     Results of viable count enumerations for the continuous dilution experi-
ments are shown in Figure 6.  Trends similar to those noted for closed flask
systems were observed.  Viable counts of petroleum-degrading and heterotrophic
bacteria increased in-both oil-free and oil-treated flasks.  However, after

this initial- increase,  viable counts of petroleum-degrading bacteria in oil-
free flasks decreased to  lower values than in oil-treated flasks.  A relative
lag in viable counts of petroleum-degrading bacteria  in oil-treated flasks
during the winter  (6°C) was  observed.  Heterotrophic  bacterial counts remained
elevated in all flasks.   Preliminary analysis of residual extracted oil proved
disappointing as losses due  to evaporation in the controls were equivalent to
losses in the inoculated  flasks.
Fall 1976

X 57.1
Winter 1977

Spring 1977

Summer 1977


 Degradation  as  %  =  100
                          1  -
Z yg n-alkanes  (nCio-nC2s),  Enriched Flask]

Z jjig n-alkanes  (nCio-nC25),  Control Flask
             -2 J
                  0   3   6    9   12

                    DAYS OF INCUBATION
                                                 N= Bacterial units /ml

                                                 Petroleum degrading I 1 Heterotrophic
                                                 | Inoculum
                                                 0 Oil free control
                                                 E3 Inoculated
Figure 6.
           Changes  in  viable counts of heterotrophic and  petroleum-degrading
           marine bacteria in "open" or continuous dilution petroleum degra-
           dation experiments.


     The use of numbers of petroleum-degrading bacteria or the ratio of petro-
leum-degrading to "total" heterotrophic bacteria has been proposed as an indi-
cator of the presence of petroleum or related products in aquatic environments
(15).  Laboratory studies have substantiated that in closed flask and simu-
lated "open" systems, enrichment of petroleum-degrading bacteria results in
elevation of the value of the ratio when hydrocarbons are present and adequate
nutrients provided for their metabolism.  Field data presented in this paper,
obtained from seasonal monitoring of bacterial populations from surface waters
of the MACS east of New Jersey, revealed that viable counts of petroleum-
degrading bacteria (and the ratio of these to heterotrophs) did not consis-
tently correlate with concentrations of dissolved aliphatic hydrocarbons, dis-
solved organic carbon, particulate organic carbon, nitrates, phosphates and
ATP.  Although a significant  (a = 0.05) correlation of petroleum-degrading
bacteria with aliphatic hydrocarbons was observed for various seasons or for a
single year  (1976-1977), it is difficult to believe that a functional relation-
ship underlies this response.  More likely it is a reflection of correlation
with some common variable.  Although other investigators have reported sig-
nificant quantitative relationships between petroleum-degrading bacterial
counts and pollutant hydrocarbons  (3, 15), the concentrations noted have gener-
ally been 1,000-10,000 times larger than those measured for most surface water
MACS stations.  Buckley et al. (1976) found no consistent relationship of
viable counts with petroleum  (as determined by fluorescence) concentrations in
non-polluted estuarine waters.

     Levels of petroleum-degrading bacteria appeared to be related to water
type.  Thus, the largest viable counts were usually measured in the coastal
boundary layer, a water type characterized by proximity to seasonal marina ac-
tivities and generally the highest levels of nutrients, organic carbon and
aliphatic hydrocarbons.  Lower values of petroleum-degrading bacteria were
usually found for the remaining inner  (Dl, N3) and outer (E3)  shelf stations.
During six seasons relatively elevated viable counts were measured in the vi-
cinity of the shelf break  (F2).  Here shelf and slope water mixed and the re-
sulting convergence with associated upwelling would appear to stimulate bio-
logical activities.  Finally, the lowest viable counts measured usually were
found beyond the shelf break seaward of the frontal zone (Jl)  in "oceanic" or
Gulf Stream eddy water.  Simple correlations of viable bacterial counts with
physical-chemical data may not be possible considering the dynamic nature of
the MACS shelf-slope hydrographies.  Considering the uncertainties associated
with a data base limited to several samples/station each season, the precision
of the chemical analyses, and the large standard deviations implicit in the
MPN technique utilized, the lack of consistent linear correlations was not
surprising.  However, despite these difficulties, basic trends in the data
were observed and indicated that petroleum-degrading bacteria tended to de-
crease with distance from shore with variations on this pattern being due to
interactions of shelf-slope water types.

     Although petroleum-degrading bacteria could be detected at all stations,
absolute numbers of bacteria were not always indicative of the in-vitro degra-
dation potential in closed flask experiments.  With nutrient enrichment all
stations could supply an inoculum with the capacity to degrade some proportion

of the n-paraffins present.  Those stations exhibiting the greatest aliphatic
degradation potential may have been pre-adapted through selection of those
genera capable of degrading petroleum.  This adaptability, however, was not
consistently reflected in viable surface water counts.

     This research was performed under contracts No. 08550-CTS-42 and No.
AA550-CT6-62 with the Bureau of Land Management, U.S. Department of Interior.
We wish to thank S. Blanton, C. Booth, J. A. Lessard, and D. Lister for tech-
nical assistance.

                              LITERATURE CITED

1.  American Public Health Association.  1976.  Standard methods for the ex-
     amination of water and wastewater.  14th ed.  American Public Health As-
     sociation, Inc., Washington, D.C.

2.  Atlas, R. M., and R. Bartha.  1973.  Fate and effects of polluting petro-
     leum in the marine environment.  Residue Rev. 49:49-85.

3.  Atlas, R. M., and R. Bartha.  1973.  Abundance, distribution, and oil bio-
     degradation potential of microorganisms in Raritan Bay.  Environ. Pollut.

4.  Buckley, E. N., R. B. Jonas, and F. K. Pfaender.  1976.  Characterization
     of microbial isolates from an estuarine ecosystem:  relationship of hy-
     drocarbon utilization to ambient hydrocarbon concentrations.  Appl.
     Environ. Microbiol. 32:232-237.

5.  Gunkel, W.  1968.  Bacteriological investigations of polluted sediments
     from the Cornish coast following the Torrey Canyon disaster, p. 151-158.
     In E. B. Colwell  (ed.), The biological effects of oil pollution on lit-
     toral communities.

6.  Kator, H., and R. Herwig.  1977.  Microbial responses after two experimen-
     tal oil spills in an Eastern coastal plain estuarine ecosystem, p. 517-
     522.  In Proc. API-EPA-USCG Oil Spill Conf., New Orleans, La.

7.  Lewin, R. A.  1974.  Enumeration of bacteria in seawater.  Int. Rev.
     Hydrobiol. 59:611-619.

8.  McNair, H. M., and E. J. Bonnelli.  1969.  Basic gas chromatography.  Con-
     solidated Printers, Berkeley, California.

9.  Oppenheimer, C. H., W. Gunkel, and G. Gassmann.  1977.  Microorganisms
     and hydrocarbons in the North Sea during July-August 1975, p. 592-609.
     In Proc. API-EPA-USCG Oil Spill Conf., New Orleans, La.

10.  Pritchard, P. H., and T. J. Starr.  1975.  The microbial degradation of
      oil in continuous culture.  Office of Naval Research, Task No. NR 133-
      070.  Annual Report No. 2.

11.  Seki, H.  1976.  Method for estimating the decomposition of hexadecane in
      the marine environment.  Appl. Environ. Microbiol. 31:439-441.

12.  Strickland, J. D. H., and T. R. Parsons.  1972.  A practical handbook of
      seawater analysis.  Bull. 167, Fish. Res. Bd. Canada, Ottawa.

13.  Virginia Institute of Marine Science.  1978.  Middle Atlantic Outer Conti-
      nental Shelf environmental studies.  Vol. II, Chemical and biological
      benchmark studies.  Prepared by VIMS, Gloucester Point, Virginia, under
      contract No. 08550-CT-5-42 with the Bureau of Land Management, U.S.  De-
      partment of Interior  (in press).

14.  Virginia Institute of Marine Science.  1977.  Middle Atlantic Outer Conti-
      nental Shelf environmental studies.  Vol. II, Chemical and biological
      benchmark studies.  Prepared by VIMS, Gloucester Point, Virginia, under
      contract No. 08550-CT-5-42 with the Bureau of Land Management, U.S.  De-
      partment of Interior.  1450 pp.

15.  Walker, J. D., and R. R. Colwell.  1976.  Enumeration of petroleum-
      degrading microorganisms.  Appl. Environ. Microbiol. 31:198-207.


                     Steve  A.  Orndorff  and  Rita  R. Colwell
                         Department of Microbiology
                            University  of Maryland
                            College Park, MD  20742

                 Kepone-resistant bacteria were  isolated  from the
            James River,  Baltimore Harbor, and Upper  Chesapeake Bay
            where their numbers  reflected the degree  of anthropogenic
            contamination at the sampling sites.   Gram negative bac-
            teria,  of which the  majority were Pseudomonas spp., rep-
            resented approximately 99%  of the Kepone-resistant bac-
            teria capable of growth on  100 ppm Kepone agar.   Screen-
            ing of nine Kepone-resistant isolates  from Baileys Creek
            revealed the presence of large molecular  weight plasmids,
            and curing one of the isolates with  mitomycin C rendered
            it sensitive to Kepone.  A  9-chloro-Kepone derivative, ob-
            served by electron capture  gas-liquid  chromatography, was
            found to be associated with microbial  activity in Kepone-
            enriched Baileys Creek sediment.
      Within the past ten years the James  River  and  Chesapeake  Bay  have  been
contaminated with Kepone, a chlorinated insecticide  which  is  a  chemically and
physically stable compound.  Surveys of these  regions  have revealed the  pres-
ence of Kepone in finfish taken from areas beyond  the  James River and  Chesa-
peake Bay.  Like many other chlorinated insecticides,  Kepone  has been  found to
induce toxic effects in higher organisms,  e.g.,  scoliosis  in  fish  (11).

      Until recently, research on Kepone concerned invertebrate toxicity (12,
14, 20), effects on soil microorganisms (12)  and  bactericidal  effects (4).
Block  (4, 5), finding Kepone to be very effective in  inhibiting  growth  of Gram
positive organisms at concentrations as low as 8 ppm,  proposed  the  use of Ke-
pone as an antimicrobial agent.  Other, more recent  work by Bourquin et  al.
(6) and Brown et al. (7) has demonstrated  toxic  effects  of Kepone on estuarine
microorganisms.  There is a lack of information  on biodegradation,  especially
potential degradation of Kepone by bacteria in estuarine and  freshwater  envir-
onments .

     The purpose of this study was to determine the distribution of Kepone-
resistant bacteria in the James River and Upper Chesapeake Bay, and to study
the physiology and taxonomy of these bacteria.  Identification of the genetic
basis of Kepdne resistance in the bacterial isolates was also of interest, as
was the degradative potential of bacterial isolates obtained from enrichment

                            MATERIALS AND METHODS

     Samples of water and sediment from  stations in the James River, Baileys
Creek, Baltimore Harbor and Eastern Bay  of the Chesapeake Bay were collected
over a period of 18 months.  Five stations in the James River, including
Baileys Creek, were sampled at established intervals  (Fig. 1) .  Water samples
were collected aspetically at a one-meter depth using a Niskin sampler (Gen-
eral Oceanics, Miami, Fla.) and sediment samples with a Ponar Grab  (Wildco,
Saginaw, Mich.).
 Figure  1.   Sampling sites  on the James  River.   The  station designations are:
            3728N,  Bermuda  Hundred;  3728L,  Baileys Creek;  3715O,  Sandy Point;
            3713X,  Jamestown;  6525K,  Newport News.


     Samples were inoculated into yeast extract-minimal salts broth containing
100 ppm Kepone to obtain Kepone-resistant bacteria.  Enumeration of total,
viable heterotrophs and of Kepone-reistant bacteria was accomplished using the
spread plate method.  UBYE agar (19),  a medium described elsewhere and con-
taining appropriate salts, was employed with Kepone in acetone added as a sur-
face film to the solidified agar to give a final concentration (w/v) of 100
ppm.  Kepone-resistant, facultative anaerobes were enumerated using identical
media incubated in Gas-Pak containers  (Baltimore Biological Laboratory) for
one week.  PCB-resistant organisms were enumerated on UBYE agar to which 100
mg/liter Aroclor 1254 had been added.   Oil-degrading organisms were enumerated
on Oil Agar #2 (24).  Total coliforms  were determined following Standard Meth-
ods published by the American Public Health Association (21).

     Kepone-resistant organisms were isolated and maintained on media contain-
ing 0.1% yeast extract, 0.1% proteose  peptone, 0.02% K2HPOi+,  0.02% KH2POi+,
0.02% NaN03, 0.02% (NHiJaHPOit and 2% Bacto-agar, with 0.1 ml of a 10 mg/ml
Kepone acetone solution added as a surface film.  An incubation temperature of
30°C was used for both enrichment and isolation.


     Antibiotic resistance to Kanamycin, Streptomycin, Chloramphenicol, Tetra-
cycline, Gentamicin, and Ampicillin was determined using the Kirby-Bauer
method  (3).  Pseudomonas aeruginosa ATCC 27853 was employed as a control.  Re-
sistance to heavy metals, i.e., HgCl2, CdCl2, ZnCl2, NiCl2«6H2O,  CrCl3'6H20,
and the pesticides lindane, aldrin, methoxychlor, DDT and Mirex were deter-
mined by monitoring turbidity in a broth medium containing 0.1% yeast extract
and 0.1% proteose peptone to which these substances had been added.  The tubes
were examined after incubation for 4 d at 30°C.  Heavy metals were added to
the medium as deionized, distilled water solutions, at the following (w/v)
concentrations:  ZnCl2, 50 ppm; HgCl2, 10 ppm; CdCl2, 5 ppm;  NiCl2'6H20, 10
ppm; and CrCl3«6H2O, 10 ppm.  The pesticides were dissolved in acetone and
added in 10 yl aliquots to broth tubes to yield a 10 ppm (w/v) concentration.
Oil degradation was tested following the method of Mills et al. (17), using
MMC broth containing 0.1 ml Arabian light crude oil.


     All strains were subjected to a testing regime of ''core" characteristics,
as defined by Colwell and Wiebe (10).   Nine of the isolates from Baileys Creek
that were Kepone-resistant were screened for plasmids by electrophoresis in
0.7% agarose gels, developed with ethidium bromide and visualized under long-
wave ultraviolet light  (16).  Plasmid size was estimated by comparison with
two marker plasmids harbored by Escherichia coli J53  (reference strains re-
ceived from Dr. Esther Lederberg); S-a, containing a 25 mega dalton plasmid,
and strain J53KP4, containing a 34 mega dalton plasmid as reported by S. Fal-
kow.  Curing was carried out in an overnight BHI broth culture containing 0.1
yg/ml mitomycin C.


     Utilization of Kepone as a sole carbon source was determined by inocula-
tion of a minimal salts broth, supplemented with Kepone in concentrations of
1, 10, 100, 500, and 1000 ppm, with 0.1 ml of washed log-phase cells.  Inocu-
lated media was incubated at 30°C, in the dark, for one, six and 12 months.

     Microbial degradation of Kepone was examined using homogenized sediment
collected from the top 30 cm of the bottom of the James River at a site lo-
cated at the mouth of Baileys Creek.  One set of three 500-ml Erlenmeyer flasks
received 200 g.w.w.  (gram wet weight) each of autoclaved, air-dried sediment,
and a second set received an equal amount of non-sterile sediment.  A third set
of flasks containing non-sterile sediment were not amended with Kepone.  All
flasks were brought to equal moisture content, i.e., each flask of sterile,
air-dried sediment received distilled water to achieve a moisture content
equivalent to the in-situ, non-sterile sediment, with deionized, distilled
water "spiked" with 30 yg analytical Kepone/100 g.w.w. sediment and incubated
in the dark at 30°C for 8 weeks.  Concentration of background Kepone in the
homogenate prior to enrichment was determined by extraction of triplicate 100
g samples, followed by gas-liquid chromatography of the extracts.

     Extraction of sediment for Kepone analysis was carried out using over-
night Soxhlet extraction of air-dried, sieved samples with 200 ml of 1%
methanol-benzene.  The solvent extract was concentrated to ca 5 ml on a rotary
flash-evaporator  (Biichi, Brinkmann Instruments, Westbury, New York)  and cleaned
on a column of 6.0 g activated Florisil PR, 60/100 mesh, topped with 2.0 g an-
hydrous sodium sulfate.  After the column was prewetted with 1% methanol-4%
benzene-hexane, the Kepone was eluted in three fractions, the first eluting
with 1% methanol-4% benzene-2% acetonitrile in hexane.  The three fractions
were combined and concentrated to ca 2.0 ml on a rotary flash-evaporator and
added to a calibrated vial.  The concentrate was evaporated to dryness under
a stream of nitrogen at 40°C and resuspended in 1% methanol-benzene for gas-
liquid chromatography.  Solvent controls were analyzed as described above.
Extraction efficiency for spiked sediment samples yielded recoveries of 85.6-
97.1%.  Values reported are corrected for extraction efficiency.

     Gas-liquid chromatographic analysis was performed on a Shimadzu GC-4BM
equipped with a 63Ni electron capture detector and a 3 m x 3 mm column of 1.5%
OV-17/1.95% QF-1 on 80/100 Chromosorb WHP.  Column temperature was held iso-
thermal at 230°C with a N2 flow rate of 30 ml/min, and detector temperature of
250°C.  Kepone concentration was determined from the calibration curve for
peak-area measurements of Kepone  (8, 18, 22).  Kepone standards were run after
every third sample.


     Technical grade Kepone was donated by Allied Chemical Corporation, Mor-
ristown, N.J., and determined to be 88.6% pure.  Analytical grade Kepone was
obtained by recrystallization in toluene, and its purity confirmed by GLC
analysis.  All solvents were pesticide-grade quality  (Burdick and Jackson,
Muskegon, Michigan).

                           RESULTS AND DISCUSSION

     Microbial populations in the Chesapeake Bay and James River water and
sediment were enumerated during seven cruises between 1976 and 1977.  Three of
these cruises, conducted in 1977, are reported here.

     Numbers of Kepone-resistant bacteria were found to decrease with distance
downstream from Hopewell, Virginia, the major impact site of Kepone contamina-
tion.  The percentage of total viable, aerobic bacteria in the water column,
that were Kepone-resistant, fluctuated but remained constant in the sediment.
The number of total coliforms was, in general, high in both water and sedi-
ment, notably at Baileys Creek in the James River where the numbers of Kepone-
resistant bacteria were significantly large (Table 1).

     It was possible to compare the numbers of Kepone-resistant bacteria in
Colgate Creek and Bear Creek in Baltimore Harbor, and at the Chester River and
Eastern Bay stations, with results for samples collected in Baileys Creek and
Bermuda Hundred in the James River.  Water and sediment counts for both of the
James River stations were significantly higher than for the Upper Chesapeake
Bay stations.  In general, bacterial counts for samples collected at the un-
polluted stations, i.e., Chester River and Eastern Bay, were lower than for
samples collected in Baltimore Harbor.  The latter were less than the counts
obtained at the James River stations.  These relationships were strongly evi-
dent for the sediment samples (Table 2).

     The total numbers of facultative anaerobes resistant to Kepone were lar-
gest at the 20 to 30 cm depth in sediment cores taken at the Windmill Point
channel at Baileys Creek.  However, the total number of facultative anaerobes
that were Kepone resistant amounted to less than 2% of the total microbial
populations  (Table 3).

     The bacteria comprising the microbial populations in the Upper Chesapeake
Bay did not demonstrate a marked capability for growth in the presence of
chlorinated hydrocarbons, in general, compared with a significant capability
for petroleum degradation.  Less than 2% of the bacteria comprising the total
viable aerobic heterotrophic bacterial populations in the samples collected in
Baltimore Harbor were resistant to 100 ppm Kepone and less than 10% were re-
sistant to 100 ppm Aroclor 1254.  In contrast, ca 16% of the bacteria present
in the water column and 37% of those in the sediment,  as enumerated by plate
counts on oil agar, were capable of degrading petroleum (Table 4)

     In order to study the taxonomy and physiology of Kepone-resistant bacte-
ria, colonies were transferred from UBYE agar medium containing 100 ppm Kepone
to fresh UBYE-Kepone agar for purification.  Enrichment for Kepone-resistant
bacteria in Baileys Creek sediment and water,  using flasks to which had been
added 1 mg/100 g.w.w. or ml Kepone, was made by successive transfers to liquid
medium containing Kepone, whereupon all cultures, including isolates from the
plate counts, were tested for growth in 1000 ppm Kepone broth.  All of the
strains were capable of growth in the medium containing 1000 ppm Kepone.
Therefore,  all strains were concluded to be resistant to Kepone.

Baileys Creek-
Sandy Point-
Newport News-
coli forms
TVC Keponer % Keponer MPN/100 ml TVC
1.2xl06 5.7xl03 0.48 1.6xl02 9.7xl06
8.4xl06 3.4xl05 6.1 1.7x10^ 8.2xl06
6.9xl03 2.4xl03 35 7.0X101 3.6xl06
5.6xl03 8.6xl02 15 3.4X101 4.1xl06
5.9xl03 5.6xl02 9.5 8.0 4.3xl05
r r
Kepone % Kepone
3.6xl06 37
5.9xl05 7.2
2.9xl05 8.2
l.OxlO5 2.4
2.3xl04 5.3

coli forms
MPN/100 ml
2.5 x 105
7.0 x 103
2.5x 10 **
aMean plate counts from three sampling expeditions, 1977.
rDenotes resistance.

Bear Creek

7 x
3 x
1 x
7 x
4 x

10 k
10 **
10 3
 Determined from mean plate counts for three sampling expeditions.
heterotrophic aerobes
(CFU/g.w.w. x 104)
facultative anaerobes
(CFU/g.w.w. x 102)
Ratio of
(CFU/g.w.w. x 10~3)
          MORE HARBOR

Top water
Bottom water
Oil Agar

     Taxonomic analysis of 88 of 300 isolates revealed approximately 99% to
be Gram negative, oxidative rods.  Further testing of 50 of the isolates to
date revealed 36 to be Pseudomonas spp., 6 Aeromonas hydrophila, 5 Flavobacte-
rium spp., and 3 Acinetobacter spp., using "core" characteristics for identi-
fication.  Many of the 88 strains were lipolytic and proteolytic, but not  amy-
lolytic.  A majority were denitrifiers and hydrocarbonoclastic organisms and
a smaller number were proteolytic strains  (Table 5).


  Source      Lipolysis  ,  ,  n   .   Proteolysis  Denitrification   ,    ^ .  .
                         hydrolysis         *                      degradation
James River





aValues represent percentage of all Kepone-resistant isolates from James River
or Baltimore Harbor that possess the characteristics listed.

     Six isolates from Baileys Creek, all Pseudomonas spp., were found to pos-
sess one or two large plasmids, 25 or 45 million daltons in size, by agarose
gel electrophoresis.  Plasmids were not detected in three of the isolates
tested.  The nine strains were all subjected to a set of physiological tests,
including resistance to antibiotics, heavy metals, and pesticides.  One of the
plasmid-carrying strains was found to be resistant to ampicillin, all of the
pesticides and heavy metals tested and able to degrade oil.  Curing the strain
by mitomycin C rendered it sensitive to Kepone, but did not alter its resis-
tance patterns to antibiotics, heavy metals or pesticides.  None of the phy-
siological characteristics of the cured isolate, when retested after curing,
differed from those of the parental strain.  The evidence to day suggests that
in this isolate, the plasmid only conveys resistance to Kepone.

     In degradation studies employing Kepone-resistant strains,  no turbidity
or increase in cell numbers was detected in any broth cultures when Kepone was
provided as the sole source of carbon.  Growth was obtained if a supplemental
carbon source was provided.

     In another study fresh samples of Baileys Creek sediment were placed in
flasks with 30 yg/100 g.w.w. Kepone and incubated for 8 weeks at 30°C in the
dark.  No other nutrients were added.  Gas-liquid chromatographic analyses of
Kepone-enriched sediments to detect degradation were run in triplicate and
corrected for background Kepone contamination, which amounted to 2.2 ppm Ke-
pone.  Analysis of the Kepone-enriched and non-amended control flasks revealed
only Kepone to be present  (Fig. 2).

     The non-sterile enriched flasks, however, exhibited a major peak, in
addition to the Kepone peak  (Fig. 3).  Analytical evaluation of Kepone and
Mirex photoproducts by Alley et al. (1, 2), Ivie et al.  (15) and Carlson (9)
indicated that the behavior of the monohydro forms of Kepone was the same as
that for the secondary peak noted in this study.  The concentration of the Ke-
pone derivative was less than 0.1% of the total Kepone added, as determined by
peak-area comparison with calibration curves of Kepone.  A similar derivative
has also been reported by Dr. James Smith of Allied Chemical Corporation (per-
sonal communication), who analyzed soil heavily contaminated with Kepone.
Confirmation of the identity of the derivative by mass spectrometry is in pro-
gress  (23), with the analysis being done in collaboration with Dr. Henry Fales
of the National Institutes of Health.


8       12

Figure 2.  Electron capture detector GLC traciny

   of sterile (chemical control) Kepone-enriched


                                      8       12

                                 Figure 3.  Electron capture detector GLC tracing

                                    of non-sterile (biological treatment) Kepone-
                                    enriched flasks.

     Thus, by enumeration of the various physiological groups of microorga-
nisms found in the Chesapeake Bay and James River, it is concluded that micro-
bial resistance to Kepone can be demonstrated for -several genera of estuarine
and freshwater bacteria.  This resistance occurs predominantly in Gram nega-
tive bacteria, particularly Pseudomonas spp., which appear tp be physiologi-
cally diverse.  Microbial strains resistant to Kepone were widely distributed
throughout the Upper Chesapeake Bay and James River, but were present in lar-
gest numbers in the James River and Baltimore Harbor where pollution is a
problem and the total viable aerobic heterotrophic bacterial populations are
very large, compared to other parts of the Bay.  The percentage of Kepone-
resistant bacteria comprising the total, viable, aerobic heterotrophic micro-
bial populations in the James River remained constant in the sediment, ca 6%,
but fluctuated between 1-35% in the water column.  This situation possibly
arises from the presence of given bacterial species present in relatively
stable numbers in the sediment, with fluctuations in the overlying water in
response to abiotic or anthropogenic factors.  Despite the fact that a very
large percentage of the microbial population in Baltimore Harbor was able to
degrade petroleum, relatively few of the strains were resistant to Aroclor
1254 and Kepone, i.e., 1-10% of the total population.

     L. A. McNicol  (Department of Microbiology, University of Maryland) ana-
lyzed and identified the Kepone  resistance plasmids.  We thank D. C. Wolf for
his advice and  review  of this manuscript.  We also thank the Captain and crew
of the R/V Aquarius and the  R/V  Orion of the Chesapeake Biological Laboratory
for their assistance in collection of the samples.
                               LITERATURE  CITED

 1.  Alley, E. G., D. A.  Dollar,  B.  R.  Layton, and J. P. Minyard, Jr.  1973.
     Photochemistry of Mirex.  J. Agric.  Food Chem. 121:138-139.

 2.  Alley, E. G., B. R.  Layton,  and J. P. Minyard, Jr.  1974.  Identification
     of the photoproducts  of the insecticides Mirex and Kepone.  J. Agric.
     Food Chem.  22:442-445.

 3.  Bauer, A. W., W. M.  M. Kirby, J. C. Sherris, and M. Twick.  1966.  Anti-
     biotic susceptibility testing  by  a standardized single disk method.
     Am. J. Clin. Pathol.  45:493-496.

 4.  Block, S. S.  1968.  Antibacterial and  antifungal activity of Kepone de-
     rivatives.  Dev. Ind. Microbiol.  9:430-436.

 5.  Block, S. S., and M. M. Darley.  1969.   Kepone as bactericides.  Allied
     Chemical Corp., U.S.  3,448,194 (Cl.  313-424; A61k).

 6.  Bourquin, A. W., S.  K. Alexander,  H.  K.  Sp-idel, J. E. Mann, and J. F.
     Fair.  1972.  Microbial interactions with  cyclodiene pesticides.  Dev.
     Ind. Microbiol. 13:264-276.


 7.   Brown,  L.  R.,  G.  W.  Childers,  and D.  D.  Vaishnav.   1974.   Effects of Mirex
      and related  compounds on estuarine microorganisms.   Dev.  Ind. Microbiol.

 8.   Burke,  J.  A.,  and W.  Holswade.   1965.   Gas  chromatographic column for pes-
      ticide residue analysis;  retention time and response data.  J.  Assoc.
      Off. Agric.  Chem.  49:374-385.

 9.   Carlson,  D. A., K.  D. Konyha,  W.  B. Wheeler,  G.  P.  Marshall,  and R.  G.
      Zaylskie. 1976.  Mirex in the environment:   its  degradation to Kepone
      and related  compounds.  Science  194:939-941.

10.   Colwell,  R. R., and W. J.  Wiebe.   1970.   "Core"  characteristics  for  use in
      classifying  aerobic, heterotrophic bacteria by  numerical  taxonomy.   Bull.
      Georgia Acad. Sci.  28:165-185.

11.   Couch,  J.  A.,  J.  T.  Winstead,  and L.  R.  Goodman.   1977.  Kepone-induced
      scoliosis and its histological consequences in  fish.  Science 197:585-587.

12.   de la Cruz, A. A.,  and S.  M. Naqui.  1973.   Mirex  incorporation  in the en-
      vironment :   uptake in aquatic organisms and effects on the rates of pho-
      tosynthesis  and respiration.   Arch.  Environ.  Contam. Toxicol. 1:255-264.

13.   Gawaad, A. A., M. Hammad,  and  F.  H. El-Gayar.  1973.  Effect of  some soil
      insecticides on soil microorganisms.   Agrokem.  Talajtan 22:161-168.

14.   Hansen, D. J., L. R.  Goodman,  and A.  J.  Wilson,  Jr.   1977.  Kepone chronic
      effects on embryo,  fry, juvenile, and adult sheepshead minnows  (Cyprino-
      don variegatus) .  Chesapeake  Sci. 18:227-232.

15.   Ivie, G.  W.,  H. W.  Dorough, and E. G.  Alley.   1974.   Photodecomposition of
      Mirex on silica gel  chromatoplates exposed to natural and artificial
      light.  J. Agric.  Food Chem.  22:933-935.

16.   Meyers, J. A., D. Sanchez,  L.  P.  Elwell, and S.  Falkow.  1976.  Simple
      agarose gel  electrophoretic method for the identification and characteri-
      zation of plasmid deoxyribonucleic acid.   J.  Bacteriol. 127:1529-1537.

17.   Mills,  A.  L.,  C.  Breuil, and R. R. Colwell.  1978.   Enumeration  of degrad-
      ing marine and estuarine microorganisms by the  Most Probable Number
      Method.   Can. J. Microbiol. 24:552-557.

18.   Moseman,  R. F., H.  L. Crist, T. R. Edgerton,  and M.  K. Ward.   1977.   Elec-
      tron capture gas chromatographic determination  of Kepone  residues in en-
      vironmental  samples.  Arch. Environ.  Contam.  Toxicol. 6:221-231.

19.   Sayler, G. S., J. D.  Nelson, Jr., A.  Justice,  and  R. R.  Colwell.  1975.
      Distribution and significance of fecal indicator  organisms in the Upper
      Chesapeake Bay.   Appl. Microbiol. 30:625-638.

20.   Schimmel,  S.  C.,  and A. J.  Wilson, Jr.  1977.  Acute toxicity of Kepone to
      four estuarine animals.  Chesapeake  Sci.  18:224-227.

21.  Standard Methods for the Examination of Water and Wastewater.   1971.
      American Public Health Association, Washington, D.C.

22.  Thompson, J. F.  1968.  Retento-Rule, a device for rapid peak identifica-
      tion of chlorinated pesticides.  J. Gas Chromat. 6:560-562.

23.  Uk, S., C. M. Himel, and T. F. Dirks.  1972.  Mass spectral pattern of
      Mirex and of Kepone and its application to residue analysis.   Bull.
      Environ. Contam. Toxicol. 7:207-214.

24.  Walker, J. D., and R. R. Colwell.  1975.  Factors affecting enumeration
      and isolation of actinomycetes from Chesapeake Bay-  Mar. Biol.


                             CONTRIBUTED SESSION

TIEDJE:  I would like to compliment Lilly Young and associates.   I thought
that was an outstanding advancement in the understanding of anaerobic degrada-
tion of aromatic compounds.   One thing that was very interesting to me is the
apparent degradation of the aromatic ring by  the primary organism without the
production of hydrogen.  In other words, you  do not need the hydrogen sink to
simulate the degradation.  Is that a correct  interpretation?

YOUNG:  We did not monitor hydrogen in our systems,  because it is utilized
very rapidly.  In that last slide where we inhibited the system,  we were sur-
prised to find that an active electron sink was not needed for benzoate to be
utilized.  It doesn't seem that ferulate-utilizing  organisms need that elec-
tron sink for continual ferulic acid decomposition.   I  would think the next
experiment to do would be to look for an increase in hydrogen in those experi-
ments .

TIEDJE:  It would mean also that the organism should be easily isolated.

YOUNG:  Yes, in terms of separating the system, the ferulate-utilizing orga-
nism should be very easy to pull out.

TIEDJE:  Have you gone through the thermodynamic calculations on whether it's
feasible to have degradation without a hydrogen sink?

YOUNG:  No.  We gathered some thermodynamic information on ferulate,  but our
data are incomplete.

COONEY:  Dr. Young, what was the solid phase  or the sediment phase,  say, in
your cultures?

YOUNG:  We have sodium sulfide as a reducing  agent  with iron as  ferric chlor-
ide; so it's a ferrous sulfide precipitate in our systems which  helps keep the
system reduced.

DiGERONIMO:  Dr. Young, I would like to go along with Dr. Tiedje and congratu-
late you on the work.  Maybe I'm reading your stoichiometry wrong,  but regard-
ing your anaerobic processes, could you tell  me what was your cell yield or
your efficiency?

YOUNG:  We estimate that for most anaerobic processes,  it runs between 1% and
10%.  We haven't tried doing cell yield yet.   The major problem  is separating
the organisms from all the materials we have  in that system.   The lower level


of COa may indicate that some of that material is going into cell yield, but
we don't have that information.

DiGERONIMO:  Based on your stoichiometry, was your balance within 10%?

YOUNG:  in the stoichiometry it was  in  terms of total degradation to COa and
methane, so we didn't have a cell yield in that stoichiometry.

DiGERONIMO:  Was that one to one, that's my question.  The gas formation and
disappearance of the product seemed to be one to one.  The carbon is accounted
for either as CO2 or methane.  It doesn't look as if any of the organisms are
using any part of the substrate.

YOUNG:  I said, that in our actual measured value, the agreement is not as
good, and perhaps that can be accounted for as cell yield.

CHAPMAN:  Dr. Young, I'd like to endorse what other people have said this
evening about the excellent nature of your presentation.  I'm delighted to see
certain structural relationships between isovaleric acid and ferulic acid.  It
seems almost as if the three-carbon side chain of ferulic acid attached to
that C-l position of the ring is the basis for the origin of isovaleric acid.
Is it true to say that you only see isovaleric acid when it's derived from
ferulic acid, and that you fail to see it when vanillic acid is used as a
source of carbon in your experimentation?

YOUNG:  We don't have that information.  We have used vanillic acid, vanillic
enrichments, but we don't have that information as yet.

CHAPMAN:  If that is the case then there is clearly a lot of exciting new bio-
chemistry to be discovered in the process.  Thank you.

COONEY:  I have a question for Steve Orndorff, please.  Does your definition
of Kepone resistance mean tolerance to 1000 ppm?

ORNDORFF:  We initially screened organisms that were resistant at the 100 ppm
level.  In subsequent testing we found that 100 ppm seemed to be a threshold
concentration, in other words, organisms that were resistant at that level
would be resistant at nearly any higher levels.  Thus we used resistance to
100 ppm as our criterion.

COONEY:  So it is 100 ppm or one hundredth of one percent?

ORNDORFF:  Yes, but all organisms were found to be resistant to 1000 ppm

COONEY:  What was the alternate carbon source or sources that you added in
order to get degradation or transformation of Kepone?

ORNDORFF:  In the pure culture studies we have done, which I did not discuss
here, glucose was used.  In testing resistance at 1000 ppm,  0.1% yeast extract
or proteose peptone was used.

BOURQUIN:  I was just reviewing some of our data.  When we first read your ab-
stract we were pleasantly surprised because your data were in agreement with
our laboratory results.  So I was surprised when you said your data conflicted
with our results.  As I recall, we showed that sensitive organisms showed amy-
lolytic and lipolytic activity which is, I think, what you showed.

ORNDORFF:  In the physiological testing, I did not screen the Kepone-sensitive
bacteria.  The results I presented were only for Kepone-resistant bacteria.
The James River isolates primarily were not amylolytic, but in the Baltimore
Harbor, approximately 40% of the Kepone-resistant isolates were amylolytic.
That was the main point where our data disagreed with your results.

BOURQUIN:  If, in our studies, amylolytic organisms were shown to be more sen-
sitive, and you demonstrated fewer resistant amylolytic organisms in a river
having been exposed to Kepone, then I think it is reasonable to assume the
sensitive organisms may have been eliminated.  I interpreted that as in agree-
ment with your data although indirect evidence.

MACALADY:  I have a question on Kepone-resistant bacteria.  I was really sur-
prised by one number on your slides that you commented on briefly, but I would
like to ask another question on it.  It seems rather amazing to me that in
sediment in the James River you found such a remarkably higher percentage of
Kepone-resistant bacteria, in an area where any Kepone was deposited.

ORNDORFF:  Yes, there is a problem in hydrology, at the Bermuda 100 station
which is about 3 miles upstream.  A lot of the Kepone was carried upstream.
We don't know how, but there is quite a bit of tidal action in that area.  It
is essentially fresh water, but it is tidal, with 10-foot tides in that area.
Some of the Kepone could have washed back to that area.  In addition,  there is
quite a bit of coliform contamination in that area., and subsequently coliforms
are quite resistant to Kepone also.  In fact, essentially all of the gram-
negative bacteria tested appear to be resistant.

BETTINGER:  You said the plasmid related to Kepone resistance also appears to
be related to heavy metal and antibiotic resistance.  You indicated a rather
wide spectrum antibiotic profile, and that you could cure the Kepone resis-
tance but retain its antibiotic, heavy-metal resistance.  Is this the same
kind of plasmid that Dr. Colwell referred to this morning that could pick up
lactose-fermentable ability?  I'm curious how you can cure just the Kepone re-
sistance but keep the heavy metal and antibiotic resistance.

ORNDORFF:  The Kepone-resistant strain did contain two plasmids.  It is quite
possible that the other plasmid carried the antibiotic resistance factors, and
we cured one of the two plasmids.  In subsequent studies we have another iso-
late which contains only one plasmid and it does demonstrate, in addition to
Kepone resistance, resistance to some heavy metals and one or two antibiotics
and pesticides.  When we cured this strain it also lost its resistance to Ke-
pone.  Then, when we tested it for reversion frequency, we found it to be ap-
proximately 1.6 x 10~12.  I think that's about the same as for one gene.  We
don't have confirmatory evidence, but it's possible that one plasmid could
mediate some factors which granted resistance to Kepone.

BETTINGER:  What was the curing agent?

ORNDORFF:  Mitomycin C.

MACALADY:  The James River is also famous for another sort of incident that
involves chlorination.  I'm surprised that you didn't mention anything about
testing for residual chlorine levels at the time you were doing other tests.
Can you comment on that?

ORNDORFF:  Actually, we never thought of doing that.  That's a good point.
The James River and Baltimore Harbor both receive quite a lot of contamination
from many sources.  Pesticides as well as industrial effluents in the Richmond
and Baltimore areas are contributing to those waters.

JOHNSON:  In this Kepone problem, what really is the relevance of using high
concentrations of these chemicals?  These numbers do not correlate with the
amount of Kepone you find in the environment.  In other words, we do not de-
tect 100 or 1000 ppm in the sediment.

ORNDORFF:  We were trying to exert as much pressure as possible without elimi-
nating the entire population, and are attempting to select for those organisms
which are highly resistant, and possibly capable of degradation.

JOHNSON:  When you were talking about different physiological groups, their
presence or non-presence, what concentrations did you use in making those
tests?  What types of test were used?

ORNDORFF:  I just used the standard procedures for hydrolysis of starch,  Tween
80, lipid.  The tests were done employing the "core" characteristics of our
laboratory and include standard microbiological methods.  I did not run those
tests with Kepone in the test media.

BAUGHMAN:  I have one question I'd like to ask on Kepone, following up on
something Dr. Johnson is pursuing.  As I recall, the solubility of Kepone at
the pH you are working at is about 60 ppb.

ORNDORFF:  It's insoluble in water.

BAUGHMAN:  How insoluble is that?

ORNDORFF:  It is only soluble to about 0.1 ppb.

BAUGHMAN:  This brings up the question that concerned us in trying to look at
degradation of highly insoluble compounds.  These materials are strongly
sorbed by microorganisms.  Since you cannot get the aqueous phase concentra-
tions to exceed the solubility, I'm wondering how you benefit by increasing
the amount of material in that system greatly above the apparent solubility.
Do you increase the availability of the materials to the organism?  I think it
can be shown fairly clearly that you do not.

ORNDORFF:  Well, it's true that these chlorinated insecticides, especially Ke-
pone, will be bound up tightly by any organic matter present.  They can pre-


cipitate out with salts.  Some investigators have gotten around, or try to get
around the problem by suspending the insecticides in acetone, or sorbing to
glass beads, or some other structure which can be suspended in a broth medium.
There is probably quite a bit of sorption to the bacterial cell.  This cannot
be accounted for in our experiemnts.  However, we did try to maximize the
availability of the compound for the bacteria.

BAUGHMAN;  I think perhaps it's a point worth considering by those of us con-
cerned about these studies.  I often wonder what we do to the physiology of
the organism when we really sock it to them at levels of 1000 or 10,000-fold
greater than the aqueous phase.

T. JOHNSON:  We're (NMFS) quite interested in the effect of chemicals on fish.
Many times microbiologists (and this is one microbiologist who does the same
thing) neglect the organisms that we are charged to protect.  For example, DDT
is considered relatively insoluble in water, yet roughly to 1-5 micrograms per
liter  (1-5 ppb) is the LCso value for a good percentage of our freshwater game
fish.  So we should at least consider low concentrations, because they are
very important to the fishery resource.  When we work at very high concentra-
tions, the relevance of the data to the fishery resources is questionable.

AHEARN:  One possible consideration in relation to the effect of water insol-
uble compounds on microorganisms would be the microcosm effect in an oil or
lipid layer.  The compound could be available to microorganisms in that micro-
cosm at a rather high concentration.

COLWELL:  I feel compelled to speak to the point concerning the concentration
of Kepone used in our studies.  What is neglected in the discussion is that
the effect we are testing is that which alters the growth of the organism in
the presence of Kepone.  In Steve's work, he used several concentrations, in-
cluding a 1000 ppm and lower concentrations.  In the case of one system em-
ployed, when the bacterium was cured of the plasmid, it was no longer capable
of growth on the Kepone.  Biodegradation aspects of this work should be empha-
sized.  The concern about the concentration of Kepone used is understandable,
but in effect this was, in the case of our sediment studies, to negate binding
of the Kepone by providing an excess; that is, to have available Kepone in

ORNDORFF:  Everyone has put too much emphasis on the 1000 ppm concentration.
This was only one test, carried out in a broth culture,  merely to see if the
organisms remained viable at that concentration.  None of the other tests were
done using a 1000 ppm concentration of Kepone.

RIBBONS:  I wonder if you could tell me the evidence for the plasmid?

ORNDORFF:  Yes, it was separated by agarose gel electrophoresis, developed
with ethidium bromide and visualized under long-wave UV.

RIBBONS:  You showed two plasmids.   Did you show the absence of plasmids with
the cured strains?

ORNDORFF:  Yes, we did.

RIBBONS:  And did you transform back into a cured  strain?

ORNDORFF:  That's very difficult with Pseudomonas.  We have not been able to
do that yet.

RIBBONS:  Do I infer, then, that the cured strain  had antibiotic resistance
and heavy-metal resistance which was not plasmid-borne?

ORNDORFF:  In one isolate, yes.  But it is possible that this may have been on
another plasmid which could have possibly integrated into the cell chromosome.

BOURQUIN:  Dr. Cerniglia, you were talking about degradation of aromatic hy-
drocarbons by fungi.  Do you think that fungi actually compete with bacteria
to any great extent in the environment?

CERNIGLIA:  As far as aromatic hydrocarbons are concerned, I don't think that
there is a study being done to look at the fate of these compounds with re-
spect to fungal degradation.  I don't think we actually know.   That's one of
the things I've been trying to emphasize in our task group discussion.   Are
fungi playing a role in degradation of these types of compounds?  We know that
they do in pure cultures.

BOURQUIN:  What about partial metabolism of some of the polynuclear aromatics?
Do you think fungi might partially metabolize compounds and actually form,
say, arene oxides or something that could accumulate?

CERNIGLIA^  From our study it seems it is very likely.

AHEARN:  A number of questions have been voiced during break.   I will indicate
these questions and perhaps we can have open discussion on these.   I think
this is directed to Dr. Liu.  It has to do with the disappearance of the PCB
from the culture system due to absorption rather than metabolism.   Perhaps  Dr.
Liu can discuss this point.  The other question was a point of clarification
in relation to the Kepone.  Whether or not there was a metabolite demonstrated
in this study, whether or not there was growth.

ORNDORFF:  Since that sediment experiment, we've done pure culture work.
We've been able to identify the same metabolite in pure culture broth using
glucose as a carbon-energy source, so that it appears as if it is a cometa-
bolic effect.  In fact, I even have the GLC tracing of it if you would like to
see it.  Several pure cultures, including isolates from the James River,
Bailey's Creek, and the sewage treatment plant at Hopewell,  Virginia,  were
examined.  All of these isolates yielded chromatograms after growth in a Kepone
medium that appear basically the same as this one.   The major  peak that you see
to the left of the Kepone is absent in all of the controls.

GARNAS:  Can you tell us what kind of confirmational evidence  you have that
this metabolite is the monohydro-Kepone?

ORNDORFF:  There are problems in measuring the derivative.   It appears to be
a monohydro derivative.  The concentrations are extremely low  and are barely
amenable to gas-liquid chromatographic analysis.  Cells have been grown in


batch culture and as much of the product as possible has been recovered.  Mass
spectral analyses of these extracts are in progress.

GARNAS:  I am very happy that you came here to give the paper.  We've completed
a detailed study on the fate, movement, and possible biodegradation of Kepone
in various estuarine systems.  To date, we haven't seen any evidence of degra-
dation, and we specifically looked for the mono- and dichlorinated products.
We used 14C materials to get a very acceptable budget analysis on our system.
So we are quite interested in seeing a little more confirmation of this peak.

OKNDORFF:  As I said in the talk, Dr. James Smith at Allied Chemical has ana-
lyzed samples taken from the Kepone dump at Hopewill.  He and his colleagues
found monohydro derivatives.

GARNAS:  It is speculated that perhaps the compound was in the original syn-
thesis batches that were dumped.

ORNDORFF:  No, the synthesis of Kepone is a clean reaction, a very efficient

GARNAS:  I will differ with you strongly on that point.

ORNDORFF:  From what I've been told by Allied Chemical, they claim it is a
very efficient reaction.  They claim the monohydro derivative they've found in
the soil samples was not present before dumping.

GARNAS:  They did find some of the dechlorinated products in some aquatic spe-
cies also.  But where you find residues, it doesn't necessarily mean that they
were produced in that site.  It is speculated, and there is good evidence to
back it up, that the discharging of Kepone was not in a pure state.   It was
distributed with other materials including hexachlorocyclopentadiene and the
residual from batch synthesis.  The compound is not synthesized pure.  There
is a certain amount of impurity in it.  It is very difficult to obtain 99.9%
pure analysis standards to work with.  So I submit that the materials that
were probably found in those sewage treatment sediments could have been intro-
duced in the original disposal of material.

ORNDORFF:  That may be true.  Nonetheless, in our pure culture studies, there
was no monohydro derivative in the starting material.

GARNAS:  I agree with you that the gas chromatographic analyses are quite dif-
ficult.  I think it is very important to keep in mind that in electron capture
GC, cleanup conditions become very critical and proper controls become very
critical.  Peaks are due to the electron-capturing nature of the molecule; and
unless you have a good standard available, peak size is no indication of the
level of conversion, if it is happening.  I still submit that it's a very
critical issue, and I think it requires substantiation.

BOURQUIN:  I have a question concerning the pure culture studies.  Have you
ever looked at another chlorinated compound to see if there is dechlorination
other than with Kepone?  To see if there is a non-specific reaction for a cage
structure or something like, for example, can you show dechlorination of mirex?


ORNDORFF:  No, we haven't, but we are looking at the precursors used in the
Kepone synthesis—hexachlorocyclopentadiene and dicyclopentadiene.

BOURQUIN:  I don't know if I misunderstood you, but I thought you said your
product was 1/10 of 1% of what was added?

ORNDORFF:  That's correct.

BOURQUIN:  That seems extremely small.

DiGERONIMO:  A question for Dr. Pritchard or Dr. Bourquin about Kepone.  Do
you think not seeing Kepone degradation in your systems could be accounted for
because of the enrichment period in James River sediments could cause orga-
nisms to evolve or mutate and they have been selected, whereas you people are
working with a rather fresh sample?

BOURQUIN:  We are using James River sediment in our systems.

DiGERONIMO:  That answered my question.  I have to bring it up because from
our work in the sewage system, we found that responsible organisms were in
very low numbers.  They probably could be missed depending on your sample size.
This may be a reflection of the complexity of one microcosm showing something
vs another microcosm.

BOURQUIN:  It could be we have missed or not selected for a specific organism.
I think the whole question is very interesting.  However, I'm wondering about
the 0.1%.  What does it mean in terms of degradation?  Is it significant in
the environment, if it is a product?  Something we have found in our studies
with Kepone, but haven't reported yet, is another product in our systems.   It
amounts to 10-20% of the Kepone.  Other people at Fish Pesticides Laboratory,
Dave Stalling and his group, have reported a similar product in fish of up to
40 or 50% of the Kepone added.  These are not dechlorinated products.   They
are less polar compounds and they are probably conjugated with lipoprotein.
This is not confirmed, because mass spectral analyses have not been performed.

DiGERONIMO:  You commented on a 0.1% product.  Just to plug cometabolism,  ac-
tually 0.1% of the compound for a cometabolic system would probably be a pre-
dictable  amount based on cell mass.  Because you are not expecting any in-
crease in cell mass; therefore, from our experience, you wouldn't expect a
total turnover or any appreciable amount of turnover of a compound.   Secondly,
I have to make one more comment about the contamination of compounds.   When we
first started out, we were using 2-3-6 trichlorobenzoic acid.  We were finding
really fantastic results until we purified it.  Then we found these results
disappeared, even though we were assured 99% purity; it wasn't.  So a lot of
the data we did accumulate on the 2-3-6 and a lot of the previous data became
suspect because of the purity of the compound.

BOURQUIN:  Just one other thing on looking for Kepone metabolites.  In all
systems used to study Kepone at our laboratory, there have been GG searches as
well as radioisotope searches for metabolites.  In the toxicology laboratory
there have been any number of studies using flowing water bioassays, with
fish, crustaceans and oysters.  All of these samples were checked for


monohydro- and dihydro-Kepone as well as any other product.  They never really
found any products.  In our systems we've run continuous-flow degradation sys-
tems and a number of the eco-core experiments in which we have used fresh es-
tuarine sediments and water from the local Range Point salt marsh previously
uncontaminated with any type of pesticides that we know of, and using James
River sediments.  We have looked for degradation in systems under anaerobic
conditions (purged with nitrogen),  and in systems with additional growth sub-
strates like naphthalene or glucose added to increase microbial activities or
to increase aromatic hydrocarbon-degrading activity within the sediment or
water system.  We've looked at flowing water systems using 10-gallon aquaria
where   C-Kepone was added to either James River or Range Point sediments (two
different systems).  Here again we  did find the conjugate product in associa-
tion with decomposing lug worms.  In all cases where we find this other pro-
duct, it has always been in anaerobic conditions, but in no case have we found
any other type of metabolite.

COLWELL:  Have you ever done any pure culture studies?

BOURQUIN:  Yes, we have.  We've got a Corynebacterium-like organism that decom-
poses camphor as a sole carbon source.  We were looking at this organism as a
possible source for a fortuitous dechlorination reaction.  We don't have any
real positive evidence for that either.

COLWELL:  Have you studied in pure  culture any organism that exhibited any
Kepone resistance?

BOURQUIN:  Yes, meaning the organism grows on Kepone agar plates;  there are
any number of those.  We didn't do  any culture studies that would show any-
thing like metabolism of Kepone.

GIBSON:  I'm surprised to find myself with an unexpected interest in Kepone.
First of all, I would like to ask if there any idea of the mechanism of the
toxicity of this compound.

ORNDORFF:  In studies using Tween-80, we found that in the presence of 10 ppm
that Kepone was toxic to the cells.  So there may be some sort of permeability
factor that enables the organisms to be resistant to Kepone.   Since these are
very lipophilic compounds they may  be bound up in the LPS of the cells, we
don't know.  We can only guess at that.

GIBSON:  Just looking at that structure, you would predict that no organism in
its right mind would look at it.  If you look at it chemically, as Peter Chap-
man pointed out, it is a chloromethyl ketone which will react with sulfhydryl
groups, and will probably be a very toxic compound.  I was wondering,  when an
organism becomes resistant, do you  think it's due to biotransformation?

ORNDORFF:  It could be.  Your guess is as good as mine.  I don't know.

GIBSON:  You said the two things went hand in hand, that if it lost the ability
to transform, it also becomes Kepone sensitive?

ORNDORFF:  In the  isolates tested  in the plasmid  screening, those isolates
that were cured were no  longer resistant to Kepone.  But we are still in the
process of testing whether these cured  isolates have any ability to transform
the molecule.  I suspect that they will not be able to.

GIBSON:  I was a little  worried about that number of reversion, one in 10"12,
did you say?

ORNDORFF:  1.6 x 10~12,  approximately the same as for one gene in one isolate.

GIBSON:  Did you say you couldn't  transmit the plasmid back?

ORNDORFF:  We haven't done the experiments yet but we are working on it.  It's
very difficult with pseudomonads,  for obvious reasons.  They are not like the
Escherichia coli system.

GIBSON:  I guess I don't understand how you got the number,- then.

ORNDORFF:  This was the  reversion  frequency of the resistant organism, spon-
taneous reversion  frequency.

GIBSON:  For George Baughman or Al Bourquin,  when you are looking at a really
insoluble substance like this, how can you have Kepone plates at levels that
are way above the  solubility level and be able to say certain concentrations
are toxic?

ORNDORFF:  This is a real problem.  Testing resistance on agar plates is really
a very non-specific test and doesn't reveal much  information.   We  screened all
isolates on a 100  ppm Kepone agar, then picked each individual colony and re-
tested it in broth culture, which  is a better method for testing for resis-
tance to Kepone than agar, for obvious reasons.

GIBSON:  So you still don't know what your concentrations are?

ORNDORFF:  That's  true.  Kepone usually precipitates out if you've got high
salts in a broth,  or it  will bind  up with any of  the other organic matter,
whatever may be in the broth.

BOURQUIN:  In addition to agar plate toxicity testing,  we are confirming our
results using O2 uptake  studies on the oxygraph.

GIBSON:  If it's taken this long just to get rid of one chlorine and there's a
heck of a lot of chlorine still on that molecule, are you ready to predict how
long that molecule is going to stay in the James River?


BETTINGER:  Just a comment concerning agar plates and insoluble materials.   I
don't know if the  analogy can be drawn,  but in the SalmoneJla-microsome assay,
many times we look at compounds which precipitate but still show dose response.
We're kind of at a loss  to explain that.  I don't know if that's really rele-
vant, it's just an observation.


GIBSON:  That really worries me,  because in the Ames test for a compound to
exert its effect, it has to be electrophilic.   The most reactive compounds
that are known are very, very unstable and react with any nucleophilic site
they contact.  So if you put a compound in that precipitates, plus all the
hazards they have to go through before it exerts its mutagenic effect, that's
the compound that scares the heck out of me.

BOURQUIN:  Dave, in answer to your question on high concentrations, in our
toxicity tests, we were looking at the sensitivity of organisms to Kepone at
concentrations of .02, 0.2, and 2 ppm.  These  are suspensions, and you are
right, you really don't know the concentration.  It's not like an antibiotic
test where you have diffusion through the agar.  What we are showing in those
tests is a positive or negative result.  From  there, any confirmation on tox-
icity is backed up by oxygen uptake study, where a low concentration of Kepone
is added in solvent, to a whole cell suspension and additional substrate.
What we've done from there then is to back calculate to get a proportionate
ratio of Kepone concentration to a measured number of bacteria from an envir-
onmental sample.

GIBSON:  I realize that these experiments are  very difficult; we work with in-
soluble substrates, too.  They are a real pain.  Is it possible to take   C-
Kepone and just add it to the bacteria and find out how much is adsorbed to
the cells?  Does that actually happen?

BOURQUIN:  Sure.  You mean in pure culture?


BOURQUIN:  It could be done very easily.  We normally add it in acetone to the
medium, but it is rapidly adsorbed to the cells.  In fact,  I think it is 90%
or more of the radioactivity adsorbed in the cells almost immediately in these
high density experiments.

BAUGHMAN:  I'm not sure what relevance it has  to the metabolite question, but
you might bear in mind something that I think  is often overlooked in much en-
vironmental work.  First, there are very few organic compounds you can obtain
more pure than probably 99.5%; 99% would be very good; and you can almost al-
ways expect impurity levels on the order of the 0.1% or better.  Now, clearly,
if you have a compound with a solubility of 0.1 ppm and if you add 100 ppm of
that material to a system and if that material you add has 0.1% of impurity,
you have an equal amount of impurities potentially in solution to what you
have in the material you are testing.  And we've seen examples where this can
play all manner of hazard with the kinds of measurements that are made in en-
vironmental work.  For example, in the measurement of octanol/water partition
coefficients.  There is a tremendous motivation to work with labelled mate-
rials for environmental studies of any kind.   Labelled materials are particu-
larly noted for being impure, and since you are looking often at very hydro-
phobic materials, the impurities are likely to be more polar so you can get
orders of magnitude error from that sort of thing.

RAYMOND:  Let's switch gears a little bit here, Carl, in the metabolism of
aromatics by fungi.  Do these compounds conjugate readily like they do in


mammalian systems?  Or do the intermediates that you see?

CERNIGLIA:  In mammalian systems, naphthalene metabolism has been shown by
Jerina.  Conjugated metabolites have been reported.  I have presented evidence
looking at soluble organic metabolites, but, yes indeed, the aqueous fractions
do contain conjugates.

RAYMOND:  And so that in contrast to microbial or bacterial systems where you,
I think, probably see less conjugation because of the type of materials, you
would not expect it really to be competitive then, would you?  At least as far
as total degradation?

CERNIGLIA:  Yes, I would not.

IVANOVICI:  My question is addressed to Dr. Maccubbin.  Did you check the mi-
crobial populations for viability and if so, how did you do it?  I don't re-
call what you said.  I was wondering what sort of viability test you used.

MACCUBBIN:  The only reference I made to viability was the viable count ob-
tained by enumerating bacteria on a growth medium and assuming that growth
represents viability.

IVANOVICI:  The other question relates to your measurements of ATP.   ATP is
used by some people to determine biomass.  Is that why you measured ATP or did
you measure ATP concentrations to get an idea of some other aspect of the

MACCUBBIN:  It was measured in the reference of biomass, not stress.

IVANOVICI:  Are you going to be doing any more work in this area?


AHEARN:  We still have the question that had been posed earlier about the pos-
sibility of absorption of the PCBs in the culture system accounting,  perhaps,
for part of its disappearance.  Do we have any comments in relation to that
area to the person who posed that question or to Dr. Liu?

SAEGER:  The mono- and dichlorobiphenyls are quite readily degradable and Aro-
clor® 1221 is basically mono- and dichlorobiphenyls.

GIBSON:  We find exactly the same and that work has been published in the lit-
erature.  Although some of the trace amounts of the higher chlorinated bi-
phenyls that are in 1221 still stay behind.  We haven't been able to demon-
strate any degradation of those.


Chairperson, R. L.  Raymond


                               Jack  R. Plimmer
                  USDA Science and Education Administration
                              Federal Research
                   Beltsville Agricultural Research Center
                            Beltsville, MD 20705

                This review of the physical and chemical factors that
           affect  the distribution and degradation of pesticides in
           aquatic environments treats processes such as volatiliza-
           tion, partitioning, hydrolysis, oxidation, reduction, and
           photochemical reactions.  Reactions in sterilized and non-
           sterilized soils are discussed and compared.

     At a symposium on "Nonbiological Transport and Transformation of Pollu-
tants orf Land and  Water" held at the National Bureau of Standards in 1976,  in
a paper entitled "Oxidation and Reduction of Pesticides in Soils- and Sedi-
ments," I discussed some of the difficulties in drawing a clear distinction
between reactions  that are solely chemical in origin and reactions that accom-
pany biological processes.  The proceedings (29)  provide a useful starting
point for a discussion of physical and chemical factors that affect distribu-
tion and degradation of pollutants.

                              PHYSICAL FACTORS

     Without microbial or chemical degradation, the concentration of a pollu-
tant in an aquatic system will be reduced by physical factors,  like dilution,
adsorption, or volatilization.  These processes in turn will be affected by
the physical properties of the molecule, particularly water solubility,  vapor
pressure, and polarity.  Insoluble or adsorbed pollutants will  be transported
or deposited.  Dissolved materials also are subject to transport, adsorption
and subsequent deposition.  Substantial effort has been devoted to the study
and modeling of such processes.  A model developed by Crawford  and Bailey (4)
to describe movement of pesticides and nutrients in agricultural lands took
into account factors that are important in nonpoint pollution:   movement of
pollutants in runoff over land with subsequent percolation of dissolved mate-
rials to ground water aquifers, the adsorption of pollutants on soil particles
followed by erosion and movement of the soil as sediment, and the erosion of
pollutant particles (4).


     Adsorption and desorption are important because they affect the chemical
reactivity and biological availability of pollutants.  There has been much
recent discussion on the fate of bound residues and the question of their fur-
ther significance as pollutants (14).

     The loss of pollutants by volatilization from water bodies was studied by
Mackay and collaborators (19).  If the pollutant concentration in water and in
the atmosphere and Henry's law data are known,  approximate values for the mass
transfer coefficients,  KL (the liquid phase mass transfer coefficient)  and Kg
(the vapor phase mass transfer coefficient)  can be calculated.  A mass  flux
rate can then be calculated from these values to within a factor of about five.
This information may be sufficient to predict whether volatilization repre-
sents a major pathway of loss for a given pollutant.  However, several  compli-
cations limit the value of this treatment as a tool for prediction of environ-
mental fate.  The surface layer of organic matter normally present on natural
water bodies, the process of spray formation and transfer by bursting bubbles,
and th