vvEPA
            United States
            Environmental Protection
            Agency
            The Enforcement Division
            Region V
            Chicago IL
EPA-600/9-80-026
May 1980
            Research and Development
Proceedings of the
Seminar on Biological
Monitoring and Its Use
in the NPDES Permit
Program

Held at the
Conrad Hilton Hotel
Chicago IL
October 2, 1979
   Do not remove. This document
   should be retained in the EPA
   Region 5 Library Collection.

-------
                RESEARCH REPORTING SERIES

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.

-------
                                              EPA-600/9-80-026
                                              May 1980
         PROCEEDINGS OF THE SEMINAR
          ON BIOLOGICAL MONITORING
          AND ITS USE IN THE NPDES
               PERMIT "PROGRAM

              Chicago, Illinois
               October 2, 1979
         This seminar was conducted
             in cooperation with

     The Enforcement Division, Region V
    U.S. Environmental Protection Agency
              Chicago, Illinois
CENTER FOR ENVIRONMENTAL RESEARCH INFORMATION
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
           CINCINNATI, OHIO  45268

-------
                                 PREFACE
       From the  earliest federal  water pollution  control  legislation, the
concept of protection of human health and the  environment  has  been  important
in the development of regulations to control  waste discharges  into the  waters
of the United States.   The  Clean Water  Act  of  1977 reemphasized  this concept
by requiring as a  paramount concern the control  of toxicant discharges which
may cause  damage  to  human  health or the environment.   The U.S.  EPA and the
state and other pollution control agencies are  responding by increasing their
efforts to work with  industries and municipalities to  identify, evaluate, and
control the discharge of toxicants.   Such  endeavors during  the  past  few years
have  resulted in  the  installation  of  industrial  and municipal  treatment
systems providing  major  control  of conventional  pollutants and,  directly  or
coincidentally,  some  toxic substances as well.

       Nationally, U.S. EPA  is developing effluent  guidelines  and water
quality criteria to control  the  discharge of a select  number of  toxicants  of
national  concern.   In  addition, U.S. EPA's  regional  offices  and the  states
are now working  to implement  programs  to address facilities which may  be  of
specific  local  or regional  concern  because  of  their particular  products,
processes, raw  materials,  etc.,  or to  address  the  unique  assemblage  of
industries in  a given river  or  lake basin.

       Two of the tools being  utilized to  evaluate the potential  for toxicant
discharge  are  biomonitoring  and industrial   process  evaluations.   A  limited
number of  industries  having  the potential  for the discharge of toxicants have
been  or will  be  required to conduct  special  testing,  monitoring, and  evalua-
tions  utilizing  these  tools.    The information  obtained  from these  assess-
ments, along with  national guidelines, is being  utilized in the  NPDES  permit
reissuance  process to   evaluate  the need for  additional   toxicant  controls.

       Although many  industries, the states, and  the U.S.  EPA  have  done some
biomonitoring, the  need for  increased use of  biomonitoring as  a tool  to
control toxicants  has  become  increasingly evident.  Therefore,  this  seminar
was held to clarify the methods and uses of biomonitoring  and  its application
to setting limits in  NPDES permits.   Presentations were made concerning
biological  monitoring  tests which  make use  of   freshwater and  marine  bio-
logical  systems  -  phytoplanktons,  zooplanktons,  macroinvertebrates,   fish,
bacteria,  etc. -  in  tests for  toxicity  in process and  waste discharges.  The
range of tests include static  and flow-through  bioassays,  including  tests for
bioaccumulation, Ames tests  which use bacteria to  test  for mutagenicity, and
some  rapid assessment methods such  as the fish cough response test.
                                     -11-

-------
                              TABLE OF CONTENTS


                                                                     Page

Preface	   ii

Welcoming Remarks
  John McGuire	     1

Biomonitoring of Effluents in Perspective
  Donald I. Mount 	     3

The NPDES Permit Policy as It Relates to Biomonitoring
  Sandra Gardehring 	     8

Static Test Using Fish
  Ronald Preston	   20

Static Test Using Algae
  William Miller'.	   25

Flow-Through Test Using Fish
  William Peltier 	   31

Bioconcentration Tests for Effluents
  Gilman Veith	   40

Sediment Bioassay
  Max Anderson
  Bayliss Prater	   47

Tentative Guidelines for Flow-Through Early Life Stage
Toxicity Tests with Fathead Minnows for Use in the U.S.
EPA, OTS-ORD Round Robin Test
  Donald I. Mount	   62

Effluent Guidelines Limitations and
Lethal Units
  Kenneth J. Macek	   69

Applicability of the Ames Test in Biomonitoring
  Larry Claxton	   80

Rapid Assessment Methods (Fish Cough Response, etc.)
  Robert Drummond 	   95

Test Organism Acquisition and Culturing in the Lab
  Charles Steiner	   97
                                  -iii-

-------
                             WELCOMING REMARKS
                              John McGuire
                          Regional Administrator
                            U.S.  EPA, Region V
                             Chicago, Illinois


       I would  like  to  welcome all of  you to the  seminar  on  Biological
Monitoring and  its  use  in the NPDES Permit  Program.   Your  participation  is,
of course, the  key  to the seminar.   The turnout  -  some 400  people -  suggests
that this seminar is  needed  and  that many of you are  aware  of  the importance
of  biomonitoring  in  developing  limitations  for toxic  substances in  NPDES
permits.

       Although  many industries,  the states, and the  U.S. EPA  have done  some
biomonitoring,  the  need for increased use  of  biomonitoring as a  tool to
control toxicants has become increasingly evident.   We thought  it would be
useful,  therefore,  to  hold a seminar to clarify the methods  and  uses of
biomonitoring  and  its  application  to  setting limits  in NPDES permits.
Notices of the  seminar  were  sent  to those  industries  in  Region  V which we
believed would  benefit  most  from  biomonitoring  or which  might  be  required
to do so.

       While  the  concept of  protecting  aquatic  life was  paramount  in  the
earlier Federal  Water Pollution Control Legislation, more recent legislation,
the  Clean  Water  Act,  has  emphasized  the control  of toxic substances  with
emphasis on the  aquatic  environment  and human health.   We in EPA,  the  states,
and other pollution  control  agencies  have  responded by increasing  our  efforts
to work  with  industries  and municipalities  to  surface, evaluate  and,  where
necessary,  control the discharge  of toxicants.  Industries  and  municipalities
throughout  the  country  have installed  treatment systems which  have  provided
major control  of the more common  pollutants and  coincidentally  some toxic and
persistent  chemicals.

       Nationally, EPA is  developing  effluent limitations and  water  quality
criteria to control  the  discharge of a  select number of toxicants  of  possible
national concern.   Region V, along with the states,  is now  working to imple-
ment a  program  to  address  those other toxicants  which may be of local or
regional concern  because  of unique  concentrations  of industries.   Biomoni-
toring  and industrial process  evaluations  are some of the tools  being  util-
ized to evaluate  the  potential  for toxicant discharge.  A  limited number of
                                    -1-

-------
industries are or will be required to conduct special  testing,  monitoring and
evaluations utilizing these tools.  This  information  along  with  the  national
guidelines will be utilized in the NPDES permit  reissuance process in evalua-
ting the need for additional  toxic substances control.

       In working with industries to date, a  number of  common  questions have
been  raised.   The  purpose of  this workshop  is,  in part,  to  answer  these
questions and to let you  know some  of  our thinking  concerning  biomonitoring,
process evaluation, and the need  to  place  greater emphasis  on  the control  of
toxic substances.

       I  am  particularly concerned that  efforts  to control persistent  sub-
stances be given  an extremely high priority in the Great Lakes  Basin.  I need
not recount here  the list of these substances which  have been found thus far.
We appear to be  finding  others, also.   As  U.S.  Cochairman of the Great  Lakes
Water Quality Board I am charged with the responsibility for implementing the
Water Quality Agreement between  the United States  and  Canada.

       That agreement calls for  increased  attention to  and  vigilance in the
control of toxic  and  persistent  substances in the Great  Lakes Basin.   While
we have made  and continue to make progress in that area  we  are  obligated  to
increase our efforts  and  to  ultimately control  these dischargers.   The fact
that the Great Lakes also  suffer  from  atmospheric input  of  many  contaminants
only  increases the  need to reduce  or  eliminate those dischargers which are
most controllable.  While there  is an absolute need  for  heavy, emphasis on the
Great  Lakes,  we  are  also  aware that  significant efforts are  needed in our
other major basins such as the Ohio River Basin.   To a lesser extent, efforts
are also needed in the Mississippi River Basin.

       Biomonitoring represents only  one tool for  achieving our goals.   As
Dr. Mount will discuss  later,  biomonitoring  has not  traditionally been  used
for measuring persistence.  Perhaps it  is time we  break  with tradition and  as
appropriate,  use  all  of  our  tools -  biomonitoring,  process evaluation  and
others - more fully.

       Again, I want to welcome  you here today to  this Biomonitoring  Seminar.
I would like now to turn  the  program over  to one  of EPA's most distinguished
scientists, Dr. Donald Mount.
                                     -2-

-------
                 BIOMONITORING OF EFFLUENTS  IN  PERSPECTIVE
                              Donald I.  Mount
                U.S.  EPA,  Environmental  Research  Laboratory
                             Duluth, Minnesota


       I'm not  exactly  sure of the  backgrounds  of  most  of  you, and I don't
know how much you know about biomonitoring  - maybe you  know  a  lot more than  I
do.  Biomonitoring is a term that means many things to many people  and it is
misunderstood because  of  that.   As John  said,  the thrust  of  biomonitoring
really relates principally to the protection of organisms, including man, and
I'd like to use a definition or a concept of biomonitoring in  a  very restric-
tive way  this morning  as  I make my comments  so that you'll  know what I'm
talking about.

       There are many kinds of biomonitoring,  including those in which we do
field studies to  assess the health  of a stream  or  a lake  in  relation to an
outfall.   But  I  think the  upstream-downstream  studies  of the fifties are
probably  not  useful   in most heavily  industrialized  or  urbanized  locations
simply because there are so many outfalls  in close  proximity  that one cannot
really tell the effects of one outfall  from another. And so,  although it has
a place in terms of receiving water  quality,  I  don't really  see  that as being
especially pertinent to the  problems that  I  think  we are here  to talk about
today, which are problems related to the toxic characteristics  of  particular
effluents.  So, in my comments, I'm  going to be talking more about laboratory
or  out-of-stream  monitoring as opposed to in-stream monitoring,  although  I
certainly don't  want to rule out  the use  of cages  in waste streams and
similar techniques which might be considered  in-place monitoring.

       I made a statement  in the meeting in Washington  on the  first  of August
which I thought  was  fairly obvious,  but  I find, upon discussion with other
people  since  then,  it was not really  understood.   There is really only one
way to  determine  the  toxicity  of anything  -  and that's with an organism.   I
don't believe there is any  other way to decide what toxicity  is or  how toxic
something may  be  without  the use of  organisms.   I  think by definition  it's
only a  response that  organisms give to an exposure.   There is  no electronic
instrument that I  know of  that measures toxicity  directly, although  certainly
measurements  can  be  correlated  in  many   instances  with  toxicity  measures.
                                     -3-

-------
       An important implication is  that  there.is no true  reference standard
for toxicity.   There  is  no way to know  precisely the  exact toxicity  of some
particular chemical - and  I'm  talking  now about pure chemicals as  well  as
effluents.   This means  that  the approach  to  toxicity measurements  must  be
different than  when you  measure the amount of mercury in a sample and know
precisely what  you  have  in your  standard.   How  do you know  which toxicity
measurement  is right and  which one is wrong?  They are really almost relative
to each  other.    And  maybe everyone  is  exactly  correct  for  the  conditions
under which  it  was done  - but nonetheless you will   get  different answers.
So, it is inconceivable to  me that  any  discharger who  is  concerned about the
impact of his effluent on  receiving  water or  any  regulatory  agency  who  is
concerned about  the  control  of  what  we have  rather  sloppily  been  calling
"toxics", wouldn't  use organisms  to determine  toxicity  because there  is  no
other way to  measure  it.   If we  would use  organisms,  we  would eliminate the
problems of relationships between the response  of organisms and the chemical
measurements  made with electronic  measurements.

       The most  important  reason  we have effluent control or  pollution con-
trol, be it  air, water  or whatever,  is for  the protection  of biological
organisms somewhere, including ourselves, of course.   There are a few uses of
water which may  not be directly biologically  related,  but  for  the most part,
they are. It's obvious then that the role of organisms to measure toxicity of
effluents is  certainly appropriate  and  far overdue.   If  you  go back  in the
literature -  Tarzwell, Doudoroff,  and others - were  calling for biomonitoring
in the  late  forties  and  early fifties.   Only  just now are we  becoming aware
of the value  that it has.

       Now,  let  me  insert one  other  word of  caution.   None  of us are pro-
posing that biomonitoring is a replacement for chemical measurements - that's
just clearly  not  the  case.   Both  are needed -  they are both tools that tell
us different  things about  the effluent.   And  in some cases chemical analysis
may be far better than biomonitoring  for what  we want to  know.   But  I think
that if  you  will  stop for a moment  you  will  realize that  the  kinds of chemi-
cals that we are now focusing  our  attention  upon are not BOD and suspended
solids but complex  organic chemicals.    To measure those chemicals  is a real
job.   If you think that  the  cost of purchasing the electronic  equipment  to
make these measurements  is bad,  even though  it may be $1/2 million  in some
instances, the cost of buying competent  chemists  who  can  operate that equip-
ment  and who know what  it's telling  them is  an  even greater cost.   The
capital   cost  of  the equiment  is small  compared to the  manpower to operate an
installation   like  that.    And  so, we certainly  do  need  to have cheaper and
more effective methods to use as  routine tools and save those costly analyses
for  places where we  really  need them.  So,  if we talk  in  terms,  as some
speakers  later today may, of  $450 or $500 for  a biomonitoring test, that cost
pales when compared  to the alternate costs  if we have to go  the analytical
route.   I remind you  again that  we are  not  now talking  about  measuring BOD,
which  is a cheap thing  to do.   We are talking about much  more  difficult
measurements   for  a  great  many more  chemicals.   That's  not to  say there won't
be surrogate  measures to  which  we can relate toxicity  as  we  learn more about
it,  but  at  the  present  state-of-the-art,  that doesn't seem to be possible.
                                     -4-

-------
       Some of  the  advantages of biomonitoring,  as  opposed to only  chemical
monitoring,  are that  it  probably  more  closely  approximates  the  receiving
water  conditions  than nearly  anything  we can  do, and  it  does consider  the
interactions that may occur between the components of  a  waste  stream.   Again,
for the most  part,  I'm talking about complex wastes that are made up of many
things and  vary in quantities.   This  is  the difference between working with
industrial  wastes  and classical  toxicology.    I'm not  considering  in these
comments  biomonitoring  for human health.   There  is  another line of defense
which  makes that concern  a  little  less  important  in that few  of  us drink
untreated water.   The protection of the  organisms does not have that  second
line  of  defense.   So our  discussion  more  commonly,  and  maybe  rightly,   is
focused  on  aquatic  organisms rather  than  on  the human  health  aspect even
though I think we all agree that  the health  aspect is more  important.

       While there may be  additive  or  synergistic properties in wastes there
are also  antagonistic properties  or subtractive  properties,  whatever you want
to call  them.   From  this  standpoint we get  at  the  problem of mixtures very
effectively - probably as effectively as any  way we can  approach the  question
at the present  time,  because the  state  of toxicology in  the aquatic  area,  and
I think also  in the  mammalian area, is not really advanced  far enough yet  to
know  how  to deal  with interactions.   You can measure  the  interaction of  one
mixture but there's  no  way to relate that information to another mixture.   I
suggest as  proof of  this, you  ask your physician  what he knows  about  the
interaction of  commonly used  drugs in your body  and you'll  find  out that
the state-of-the-art  is embryonic.

       A  third  distinct  advantage  is  that  we  are measuring  the  target  re-
sponse and  not  an  index of  it.   When  you measure COD,  TOC, or mercury in  a
waste, you  are  not  measuring  the same condition  as exists in the  receiving
water.  You come much closer in a biomonitoring  test.

       Let me digress to comment  on how animals  behave in  pickle jars.  There
may be differences between the way animals  respond  in glass bottles and  the
way they  respond  in  the stream,  but the  alternative  is to do it  all in  the
stream.   If any of you have done much  in the way  of in-stream work, you know
the tremendous  cost  that's  involved  and the tremendous frustration you often
feel.   It  may  grate against your  scientific  credibility to think  about
extrapolating responses  of  animals  in  bottles to real-life situations.    But
there  is  a  two-edged sword which I'm  afraid  industry, in  particular, is  not
aware  of.   If  we  can't  extrapolate with effluents, then  when registering  a
new chemical  under TOSCA,  I don't see  where  EPA has  any other choice but  to
say that you  can't extrapolate those either.   That means testing lake trout,
shovel-head catfish,  and elephants.   My point is,  either we extrapolate or we
grind to  a  halt country-wide.   I don't know  what  position  the agency's going
to take -  I don't  make policy in the agency  -  but if they  ask me,  I'm going
to say,  if  we  can't  extrapolate on  effluent  biomonitoring, then  we can't
extrapolate on  the registration  of  new chemicals either under FIFRA or under
TOSCA.   There's going to  be  some  error,  but we  all  work  with error daily.
More   often  than not,  the difference in response of  animals in the bottle  is
not because they're  responding differently  than in the  stream;  we  are just
not aware of  all  the conditions  to  which  they  are exposed.  If you look  at
                                     -5-

-------
the costs of biomonitoring  and  compare them to the probable costs of chemical
analysis, biomonitoring  is  an efficient use of resources.

       There are obviously  disadvantages.  The world's problems are not going
to be solved by a simple 96-hour or 24-hour bioassay.  Some disadvantages are
that there will be physical-chemical  changes  in  the properties of the wastes
after they are  released  into the  receiving water.   If  we  run  tests with the
whole effluent, even though we may use  receiving  water as a diluent,  we are
not going to measure all of the same chemical and physical interactions that
go on  in  a  stream.  But that  is  equally true of BOD  measurements.   It's  a
difficult and rather subjective art,  I think you'll all  have to agree, to get
from a BOD test to the DO in a  stream.  And we have a very comparable problem
when we work with whole effluents.  But nevertheless, you can  certainly gain
a good first cut  at  what may be of concern in your  waste  by use of biomoni-
toring.

       It's  difficult to measure  chronic effects,  whether they  be chronic
toxicity, carcinogenicity,  or  bioconcentration,   without going  to  rather
extensive and longer-term tests.   At the  present  time, we simply don't have
any  overnight  test  that looks all  that promising to get at the chronic
effects.   So,  you cannot expect to  go from a simple short effluent test to
a no-effect  concentration or a  safe concentration in the receiving water, and
have any  credibility.   We  are not trying to go from  an  effluent  to  a safe
receiving water  concentration with a simple  lethality  test.   It's  just
nonsense  to  think you can do that  as yet.

       As John  mentioned earlier  in  his comments, most  of our biomonitoring
to date has  not attempted  to measure persistence.   If  there  is one property
about  chemicals  on which we ought to  focus, a  property  which we  ought to
eliminate above others,   it is  persistence in  the  environment.   If you think
of previous  headlines in the  newspapers of the last ten  years and identify
the chemicals named  - methyl-mercury, DDT, Dieldrin, PCB's, Mi rex,  Kepone  -
they all   have that property in common - they are environmentally persistent.
If I were making new chemicals  or  concerned about the effects  of my discharge
on  receiving  water, I  would  clearly want to know what the  persistence of
this material is.  If we did not have persistent chemicals in  discharges, all
of our toxicity or environmental  contamination problems would  be local prob-
lems - they  would not   be  ubiquitous like we have had  in the  past.   We are
not as adept at effectively measuring persistence with simple  tests.  I think
we  are much better at  measuring toxicity,  but  measuring persistence is
doable.

       There are special considerations  in using effluent tests.  They cannot
be done only once.   Effluents  are variable  from day to day as all  who have
worked with  effluents know.  Such toxicity measurements  are only as good as
the  representativeness  of  the  sample.   If the sample  is  not  typical  of the
waste, then neither is the toxicity, and  you  cannot say much  about the
impact on the   receiving water.   Biomonitoring  tests must be cheap enough,
simple enough,  and  short enough,  so  that  they can be done on  enough samples
to  characterize the waste. What  I  have  said is  equally true  of chemical
measurements as well.    Measuring the TOC in  a waste  once  a  month  is not
                                     -6-

-------
telling us much for many wastes.   There  is  no sense in making the toxicity
measurement  any better  than the  representativeness  of the sample  used.

       One message  I hope to leave with you from this discussion this morning
is that it's  time we consider toxicity as  a property  of  wastes which may need
to be  reduced  and  which  we  should  measure.   This does  not necessarily mean
that  effluents  must  have zero toxicity.    Toxicity  is  relative and we  are
concerned  about those where  the  exposure that  will occur  in the  receiving
water is harmful.
                               DISCUSSION
Question:   I know  very little  about the testing that you  are speaking
            of  -  this biological  monitoring -  but  if we do  have to put this
            into  industry,  in  what  form do you think it will  be?  Will this
            be  a  fish pond, or are there any different types of scientists or
            chemists  that it involves?  What does it  mean?  What will  we have
            to  resolve - the topics here at issue?

Answer:      In  general terms without getting into specifics of tests which we
            are going  to  do later today, it's clear that  tests must  be
            simple,  doable  by something other  than  the PhD,  and  they must
            have  a reasonable cost.  I think if we don't achieve those things
            in  the  effluent tests,  we  have achieved very little.   I  think
            Sandy and perhaps  some  of the  other speakers  may  be in a better
            position  to  know what  form that  takes.   We've been  talking  in
            terms of tests that  can be  done offsite in most instances.

Question:   Would EPA consider it  good enough to  hire  a free-lance com-
            pany  out  to do the biomonitoring?   In Syracuse there is a  company
            that  just  runs bioassays, and we  send a lot of  our  samples
            there.

Answer:     Again,  I can't  speak  for  the agency,  but those of us in  the
            agency who  have been talking  about  these  problems  I  think  all
            recognize the  essential  roles  which consulting firms  or  consul-
            ting  laboratories must  play in this  program.   I  think whether a
            discharger goes to a consultant or  does it himself will depend in
            large part  on  the  size  of his  operation, the  kind  of staff that
            he  has, and the particular location.  But I don't  think any  of us
            at  this point are ruling out the use of consulting firms.
                                    -7-

-------
          THE NPDES PERMIT POLICY AS IT RELATES  TO  BIOMONITORING
                             Sandra Gardebring
                      Director, Enforcement Division
                            U. S.  EPA,  Region V
                             Chicago, Illinois


       Good morning.  I would like to add my personal  welcome to  that  of John
McGuire, the Regional  Administrator, and to thank you for your  enthusiastic
response to our  seminar  on biomonitoring.   The  fact  that so many of you are
here underscores the timeliness of this  topic and the value of biomonitoring
as a tool in achieving clean water.

       What are  we  really  talking about  when  we say biomonitoring?  As used
in this conference, biomonitoring, or biological monitoring tests are  studies
which make  use  of freshwater  or  marine  biological  systems - phytoplankton,
zooplankton, macroinvertebrates, fish - and in some cases bacteria - to test
for toxicity in  process  and  waste discharges.   The  range of tests includes
static  and  flow-through bioassays,  Ames  tests,  and some  rapid assessment
methods  such as  fish cough response.  Depending  on  which test  is used, the
data  generated   can  indicate  acute  or  chronic  toxicity,  carcinogenicity,
rnutagenicity,   or  can  suggest potentially  problematic   bioaccumulation  of
organic chemicals, trace  metals, or other toxicants.

       However,  it was the passage of the 1972 Act, and  its requirement of a
nationally uniform water discharge permitting  program with technology based
standards, which finally focused federal  and state resources on water pollu-
tion clean  up.    In  this  region,  we issued about 7,800 industrial  permits
with  BPT or water-quality-based requirements,  addressing  a wide range
of water pollutants.   Emphasis was often on the more common pollutants such
as suspended solids, BOD,  and pH, along  with  more persistent toxicants such
as heavy rnetals  and  PCBs.   The long  process of treatment design and instal-
lation  was  started, and  although some  dischargers   had to  be  "persuaded"
through  federal  and state enforcement  actions, overall,  our efforts  at
achieving compliance  in  this  region have  been  quite successful.   Some  75
percent of industrial dischargers- in  our six states now  meet the limitations
of their  NPDES  permits,  and  the  remainder of the major permittees,  almost
without exception, are the subject of state or  federal  enofrcement actions
aimed at bringing about their  compliance.
                                     -8-

-------
       During this  same period,  several  other things  were going  on which
brought us  to where  we are  today.    First,  the  instrument makers  and  our
friends in the laboratories combined  to  bring us  tools and techniques which
allow the  discovery and quantification of  water  pollutants at lower and lower
levels -  parts  per million, parts per billion, parts  per  trillion.   At the
same  time,  we weren't  very good  at  telling whether  the   presence  of such
pollutants - even so-called "exotics"  - at these very low levels was signifi-
cant  or not.   The "safe"  levels  in ambient  air or  water  just  weren't known
for many pollutants.

       Secondly,  both the  public and those of us whose business is pollution
control became  more and more  aware  of the  real  risks associated  with  the
unregulated flow of these  chemically  complex materials into our environment.
PBBs, PCBs, Mi rex, Si 1 vex,  Dioxin, Kepone - terms known heretofore primarily
to white-coated men  in  chemistry  labs -  became, almost literally, household
words.  The removal  of  solids  and the increase in dissolved  oxygen  in our
waterways  was a truly remarkable and wonderful effort,  but it clearly  was not
enough when chemically  complicated materials with unknown biological  effects
continued  to be discharged.

       The  state  and   federal  environmental  regulatory  agencies  began  to
respond under existing  laws; a  process which was accelerated under the teYms
of a  consent decree agreed  to by EPA and  the Natural  Resources  Defense
Council in  1976.   That agreement, which ended  an  NRDC lawsuit  related  to
toxic  effluents,  required EPA to test  for the presence  of 129  specific
toxicants   in  waste water  discharges  from 34  industrial  categories  and  to
control any  of  these priority  pollutants found at  harmful  levels via toxic
pollutant  effluent guidelines  regulations.   Certain  of the technology based
limitations and water  quality  criteria needed  to  carry out the terms of the
decree have begun to appear this year  in the  Federal Register.

       However, we simply cannot afford to wait  for or  depend solely upon the
effort to  develop categorical  limitations  for the 129 pollutants to solve our
toxic pollutant problems.

       Given that  there are nearly 50,000 chemical  compounds  in commercial
usage; given  that more  than  12,000  of them  are  suspected  toxicants, 1,500
possible  carcinogens,  and  many others considered capable  of injuring human
health and aquatic biota;  and given that  we  know little or  nothing about the
synergistic effects  of  these  compounds  at  even  low  levels,  we  must begin
addressing  the  specific problems in   our states,  whether  or  not  they  are
related to the 129 consent  decree toxicants.

       It  is  in  this effort that  biomonitoring, when  combined with  process
evaluation and chemical  analysis,  can be  a  potent  tool both for  discovering
and for limiting  the discharge of toxic substances, and that is how we intend
to use it.

       Let me provide a few more details about our approach.

       EPA authority  for requiring biomonitoring  stems from Section  308  of
the Clean  Water  Act.    The  states, in order  to  receive NPDES authority,  are
                                     -9-

-------
required to  have  similar authority.   Our review indicates that  all  of the
states in Region  V  do have similar  authority  and can require biomonitoring
and process evaluation either as  part  of a permit or as a separate indepen-
dent information-gathering  requirement.

       Perhaps some of the  most  important questions,  particularly to indus-
try,  are:   Who are the  target  industries?  Who  should  monitor?  And beyond
that, how will  the  results be interpreted  and  used?   As  I  have  noted pre-
viously, biomonitoring is  only  a tool  and  as  such  it must  be  applied with
full  knowledge  of its limitations.   It  can  be relatively expensive.   The
results are sometimes less  than  clear  and the  tests often must be repeated.

       On the other hand, biomonitoring is useful  in telling us the response
of  an  organism to a  complex  effluent.   The interaction  -  simple additive
effects and  synergisms -  of chemicals in  an  effluent and the ultimate effects
upon  aquatic life are difficult to judge  from available literature and
criteria.  Many,  if  not  most  of  the bioassays  used  to  develop criteria are
run with one, perhaps two, or rarely three substances interacting.  Informa-
tion  on  substances which  bioaccumulate  is  even more  limited and  various
biomonitoring tests  can  be used  to derive  data  to  address  these toxicant
effects.

       I am  sure  that during  the seminar you will  explore many more of the
limitations  and benefits  of biomonitoring.  I have highlighted only a few of
those just to indicate our own awareness of the need  for reasonable applica-
tion of biomonitoring  and all  other  tools.

       Returning to those questions  I posed  previously:  Who will be required
to biomonitor?  How will  they  be  selected?

       Let me outline  generally  our thinking.   Since most of  the state and
federal   interest  is   focused  on  major  dischargers,  those lists   provide  a
natural   place  to  start.    There  are between 500  and 600  industries  on the
major list  in Region  V.  Those  industries  which  might  well be  required  to
biomonitor include chemical,  pharmaceutical, pesticide,  plastics,  iron and
steel, rubber,  chlor-alkali,  paper,  and  other  similar chemical-type facili-
ties.

       In part, the level of treatment  may also play a part in our decision.
Certainly a  process  stream  receiving  treatment with carbon  filtration will be
looked upon differently  than  one  receiving  more conventional  treatment.  An
industry discharging only noncontact cooling waters  with total  process
stream recycle or deep-well  injection may  also be  viewed differently.

       Furthermore,  I  want to make  it  clear that  we do not intend to ignore
those dichargers which have been  previously classified as minor.  We believe
that there are more than  a  few of those dichargers which may contain signifi-
cant contaminants and require  closer scrutiny than they have received in the
past.   Again, a  careful  evaluation  of the  facility,  the process involved,
present  treatment  and the likelihood  of a toxicant  being in  the discharge
will be necessary before  biomonitoring  is  required.
                                     -10-

-------
       Finally, discharges to the Great Lakes or its tributaries will  receive
 emphasis  for finding  and  eliminating  or severely  limiting  those  substances
 which bioaccumulate  and may  pose  a  long  term hazard to other aquatic  life or
 higher  animals  including  man.    Our  agreement with  the Canadians, and  the
 general tendency of bioaccumulative substances to be particularly troublesome
 in the Great  Lakes,  forces us  to  place more emphasis in that area.   However,
 our  emphasis  on  the Great Lakes will  not be  to the exclusion  of other  major
 drainages  such  as  the Ohio and  Mississippi  Rivers  and their  tributaries.

       It  should  be  obvious that  the need  for  biomonitoring  will  largely
 be determined on  a  case-by-case basis.  Once  the  determination  is  made that
 biomonitoring  is  needed,  then the specific  test  or  series  of tests  must
 also  be  determined.   Much will depend  upon  the  nature of  the  contaminants
 or  suspected  contaminants.   In some cases,  "failure"  in  one test may lead
 to another.   In  a few cases where  data  presently  exists,  long term bioaccu-
 mulation  may  be required  directly.   We  estimate  at  present  that  15  to  20
 industries may  need to conduct long  term bioaccumulation  tests.    It  is
 also  estimated  that 150 to  200 industries  will  require other forms of bio-
 monitoring.

       How then will  the results  of these tests be applied?   How will permit
 limitations be developed from test results?   Let me again outline our general
 approach.   At the outset,  I  should note that state water quality  standards
 generally require that nothing  be toxic  within the  mixing  zone.   Some states
 specify that  the  concentration of  a  substance within the mixing zone  shall
 not  exceed  the  96-hour TLM for that substance.  We  interpret  these require-
 ments to  mean that  the discharge at the end  of the pipe must not  be toxic.
 In addition,  state  water quality standards  require that outside the mixing
 zone  concentrations  of substances must  not  exceed  1/10 of  the  96-hour TLM.
 These standards will  be  used to develop permit limits  for appropriate  toxi-
 cants.  In  some cases  the  toxicant  will  not  cause  a problem  at the  discharge
 or exceed 1/10  of  the 96-hour   limit at  the  edge  of the mixing  zone, but  it
 may  bioaccumulate  in aquatic  organisms  or  be a  known carcinogen, mutagen,
 or teratogen.   For  some  of  these  toxicants  we  will  have  the  information
 necessary to develop appropriate control  levels  for protecting aquatic
 organisms and human  health.    For  others,  very little  will  be known  and  we
 will   have  to work  with  the  states and  the  industry to develop  appropriate
 control  levels based on data  derived from biomonitoring,  process  evaluations,
 and toxicity reduction plans.

       At this point I would  like  to address  those  of you  in  the  audience  who
 represent industry.   Much of what I have said  so far has  been  addressed more
 to the state people  - engineers, biologists,  lawyers -  those  often intimately
 involved  in the  NPDES permit program.  In some cases the  states  are ahead  of
 us in the  biomonitoring  area.    In  other states  little if any biomonitoring
 has occurred.   In  addition  to exchanging  information with our states, part  of
the purpose  of this  seminar is  to  spread  ideas  and  information  and to let you
 know what we  are thinking  about.   For those of you who may  be most affected
by what we are saying and  proposing here, we believe  it  is to  your  advantage
not  to wait.   We  believe  it  is  to your advantage  to seize the initiative
before the states  or EPA  knock  at  your  door.
                                     -11-

-------
       Looking early  at  your facility gives  you  all the  options,  including
one  of  the  most  important elements  - time.  With no  regulatory  onus  you are
free to  investigate  and  experiment.   You  are  able  to select the  most  cost-
effective approach  and  time  with due  regard to  plant  operation and  least
interference with  union  rules,  worker scheduling, and the  other constraints
which generally translate to time and money.   Nothing is less cost-effective
and  more disruptive than  when the regulatory  agency  imposes a tight  schedule
on an industry because of concerns  generated  by  a process  totally outside of
your control.    You  also  should  utilize  biomonitoring  in evaluating  which
treatment scheme  you  would  install  to  control  conventional  or  other  toxic
substances  so  that  you can  obtain  the  greatest  incidental control of  toxi-
cants.

       For  instance,  we are just beginning  a  program to  collect  samples from
selected river basins  for a broad organic scan of fish  flesh to detect sub-
stances which  may have bioaccumulated.    These  scans tell   us whether or not
something of  concern  has been  or  is being  discharged.   Where  something  is
found,  for example,  hexachlorobenzene  (HCP),  chlordane,  PCBs or any  of
dozens  of other  compounds,  we  will  then begin to  look for probable  sources.
Knowing  the nature  of the  various industries  in  a basin  we  can usually
pinpoint likely  sources  or a probable source.   This will  lead  to inquiries
to the  industry  such  as  a  308  letter requiring  process  evaluation,  biomoni-
toring  or  additional chemical  sampling.   Those of you  who  receive  such
inquiries may  not have a  problem but  it's pretty certain  that  your  company
will  feel some anxiety and  there  will be time constraints  for your response.

       In such an instance, you  can  no longer plan the most optimum and  least
costly  path - you're  under the gun.   To avoid  that  situation  you  need  to
begin now to do  some  of  these  things on  your own.   What are  your processes?
What  chemicals go into them?  What  by-products  are  produced?   What  are the
trace contaminants?   If  one  process  seems  to  pose a  problem,  should you have
a bioassay  run or check  for bioaccumulation?   What  process modifications can
you  make to reduce or eliminate the  problem as opposed to  end-of-pipe treat-
ment?   We  are not asking you to  do everything at once.   We  are asking that
you  establish priorities,  putting a high priority on toxic  substances,
particularly those which are persistent,  which  bioaccumulate or  which  rep-
resent  a human health  hazard.   By taking the initiative you will  be  able to
establish priorities  and work  within  the  many  other constraints.   In  the
meantime, we  are  encouraging the states  to initiate fish  flesh monitoring,
biomonitoring,  process evaluation,  and  other tools of the  pollution  control
business.

       If any of  these lead to your industry  you will  be  ahead  of the game
and  probably save money  if you've already done some  of  these things  on your
own  schedule.   It should be obvious that  the direction  we have outlined  is
for  a  long  term  - we do not expect biomonitoring  or process evaluation  to
suddenly appear in all NPDES permits.  However, as these permits  are reissued
this  year and in  1980-81  we will be  stressing these areas and they will  begin
to appear more often  in permits.

       In summary, the tools we are discussing  today are  not new, but they
will  receive  increased use in the  future  particularly as  they  enable  us  to


                                     -12-

-------
identify and control  toxic substances.  We are hopeful that this seminar will
provide greater  understanding of where we  are-headed:  increased cooperation
between state  and  federal agencies and  industry;  greater  interplay  between
biologists,  engineers,  and  administrators  of  agencies  and companies;  and
finally greater  understanding and use of biomonitoring as  a tool.   These in
turn, are,  of  course,  only  steps  along the  way  to  our  ultimate goal:   fish-
able and swimmable  waters  in all  parts of this country.
                                DISCUSSION
Question:   I am  wondering  if biomonitoring will  be  acceptable to the  129
            priority pollutants.

Answer #1:   If one runs bioassays on an effluent, recognizing that an  efflu-
(Mount)      ent is  a  complex  mixture, the 129 priority pollutant assessments
            will   never  be  satisfied.    I  don't  think  I could  really  answer
            that  question for  certain,  but as much  as I  understand  the
            consent  decree,  I would think  that  biomonitoring  would  not  be
            acceptable to the 129 priority pollutants.  It  might  be useful,
            but it  seems to  me  that  the consent  decree requires certain
            levels of treatment; I  don't  think  that toxicity  is  one  of  the
            parameters by which that treatment is devised.

Answer #2: I think  that answer is generally  correct.  I  would not want
(Garde-     to say, however,  that biomonitoring will not be part of the pro-
 bring)     cess  that we will use when dealing with those particular parame-
            ters.   We  are  all kind  of feeling our way along on  the consent
            decree pollutants, and I think there will be some use of biomoni-
            toring in that process.  However, it won't be  solely  a toxicity-
            based  approach.

Question:   Along those lines,  in  our experience, we have found that  we
            are exceeding a level of toxicity for priority pollutants,  but in
            our biomonitoring  program there  is no toxicity.   What  can  you do
            about  the  inconsistency there?

Answer:     I don't  think that it's an inconsistency.  I mentioned this morn-
(Mount)     ing that  there will  be  antagonisms  in the effluents  as well  as
            the opposite, which  is what  we always hear  about -  the synergis-
            tic properties.    I  think that one ought  to try to separate
            biomonitoring  from the  specific" consent decree  program because
            they arose out  of different  contexts even though  they  are  gener-
            ally both  aimed at controlling toxicity.  We have all  felt, for a
            long  time, that when you put  mixtures  together  different  things
            happen than when  you keep  them  separate,  which  is why I  think
            biomonitoring  effluents  is so  valuable.   It  gives us a  better
            handle on  the real world impact in those situations.

Question:   What   if we  do  scan  the  river basins and  collect  data by bio-
            monitoring  on  our  effluents  although these are  not on your
                                    -13-

-------
            consent decrees.  When it comes down to the actual permit process
            and  legal  proceedings,  all  of  the  data which we  collect  before
            the  process begins becomes evidence both in our favor and in many
            cases  against us - am I  correct?

Answer:      I think that's generally correct.   All  I can suggest to you, sir,
(Garde-      is that your alternative is  simply  to wait, as I indicated, until
 bring)      you  get a 308  letter.   The 308 letter  will  have  a time schedule
            in  it, but  if you think  you  have a problem with some contaminant
            or  other  toxic material  that  you can identify  in  your effluent,
            you're better  off  to begin now.   I  wouldn't  suggest that  you
            charge out  and hire the consulting  firm this  lady mentioned
            earlier,  but  I  would suggest  that you  might want  to  come  in  and
            sit  down  with  the  state officials or  people  from my  staff  and
            talk about the kinds of  things  that  should  be looked at.

Question:    What do you  anticipate your reaction will  be, from an enforcement
            point  of view,  if  a toxicant  which  wasn't listed in the per-
            mittee's application or  control  in the permit shows up through  the
            biomonitoring?

Answer:     I  think  it's  a  little  hard   to  answer  that  question in  the
(Garde-       abstract.  The purpose of the  program that we have outlined here
 bring)      is to  identify dischargers that have toxic  materials  present in
            their  effluents.   Recognizing  the limitations  that are suggested
            through the  use of  this  word,  I think we all generally know what
            we're  talking  about.   If we  identify  the presence of a discharge
            that  has that kind  of  material  through biomonitoring,  we'll
            ultimately be  looking to some  kind  of permit limitation.   It  may
            be a  negotiated limitation;   it  may  be  one based  on a  legal
            process that  you're painfully familiar with,  but we will,  I
            suspect,  in  at least some  of  those cases,  be  looking to  permit
            limitations based on the presence of that material.

Question:   What  happens  to  these  laboratory tests  -  are these  data sub-
            ject to subpoena?

Answer:      I presume that they are.  I  know no reason why they would not be,
(Garde-      although, as I said, I  am more of an  administrator than a lawyer.
 bring)      I can't imagine that they would not be  subject to  subpoena unless
            they  are subject to some kind  of  confidentiality claim through
            one  of the provisions of the  Act.

Question:    What  if there is a  natural  toxicant in  the waters  coming into our
            plants and  that  through our  waste treatment  systems  it's  not
            removed - are we responsible  for that?

Answer #1:  Well,  I  think  it's  a legal question,  not  a biological one.   How-
(Mount)       ever, I  think that there should not, in  general, be any problem
            in measuring the increased toxicity,  if there  is  any which your
            effluent  has  as  a  result  of   natural  toxicants,  by  appropriate
            biomonitoring  tests.
                                    -14-

-------
Answer #2:  There is a specific answer to that  question.   I know that the Act
(Garde-      deals  with  that  very  specifically  within the  regulations and
 bring)     unfortunately it  has  escaped me.  I will  try to get  someone to
            get the answer for you.   (See statement hy Glen  Pratt, p. 44  --
            Ed.)-

Answer #3:  I  don't really see that the  problem  is any different, in princi-
(Mount)      pie, from other  situations  where particular  chemicals  or sub-
            stances, especially metals, are more  toxic  in  one type of receiv-
            ing  water than  in another.    I think  it's  the same situation as
            when there are properties of receiving waters which are not good
            for  the organisms  and  therefore  affect the tests  done  in those
            waters.

Answer #4:  I would just  like to comment  on an  earlier  question asked by this
(Garde-     gentleman  referring to  whether or not biomonitoring would replace
 bring)     screening  for the  129.    It is  our intent,  in the June 14 draft-
            collated permit  application  performance regulations,  not  to use
            biomonitoring as  a screen for the 129 in  some  cases.    We do,
            however, encourage that  in certain situations the permit writer
            include biomonitoring  as  part  of the  recorded  departments  in
            developing the permit  limitations.   I  just wanted to  make that
            clear and that  the NRDC  consent  decree  is  not  necessarily the
            driving force behind  the indigenous biomonitoring program.   It
            was just something that  got us started, and now that my office in
            Washington is in the  process of  posing the  consolidated permit
            application form  regulations  as well  as what we call the biomoni-
            toring policy protocol, biomonitoring can  be  used  on  a  case-by-
            case  basis   in  addition  to  greater chemical  test  condition.

Question:   What is the status of application  factors these days?  The labor-
            atory fish production  indices and all  these  other  things  that
            your  lab  has been involved with.  Is it basically a 96-hour
            median tolerance  limit  or  is  an artificial  number chosen in order
            to give safe  dosages?   Is there  any  flexibility in  this  proce-
            dure?   What's the present-day status?   Is it  going  to  be 1/10,
            1/100, or  what?

Answer:     Well,  I'm not  sure,  unless you have an hour,  that  I  can answer
(Mount)     that  question.   I  think  it's fair  to say that if you do not have
            long-term  toxicity data  on an effluent or on a chemical,  that is,
            all  you  have is  some  estimate  of legality  from a  short  test,
            there isn't   any  better  way  to predict the no-effect  level  than
            with the application factor.   Now  having said that, I should also
            say  that  there  is a lot  of  error in doing  it this  way.   As  a
            result,  there  will probably  be   some  error  in  any  prediction.
            Short of  having  data,  I  think it's  about  the only tool  that  I
            know of right  now.  Based on  comments  that  we  received  on the
            guidelines for the consent decree criteria  documents,  we are now
            thinking  along  those  lines mainly because one  of the  early
            comments that came back  consistently was that you cannot  general-
            ize among  chemicals.  We  agree with  this.  We are accepting that
                                     -15-

-------
            this is  the situation,  and we are  faced with this  question:
            You have  chemical  X and  you  have only  96-hour  lethal  data on
            it.  What is the  estimated no-effect level?   We believe  that
            the application  factor is eminently better than  guessing.   I  want
            to quibble with  some  of your terminology.   It  is not a  safety
            factor;  I think  I  heard  you mention  the word safe or  something
            like that,  but maybe  I'm misquoting  you.   In any  event, the
            application  factor,  as it is being proposed,  is  an  attempt  to get
            from the  lethal  level  to  the no-effect  level, and  I think  that's
            different from  a  safety  factor.   We  do know that one can put
            chemicals in  categories in  terms of the magnitude  and the  appli-
            cation  factor,  for  example, some are  on the order  of 1/10 or
            less, some are on the order  of 1/1000  or more, and  others fall in
            between.   So there  are some generalized differences that you can
            see among chemicals, but  when you try  to  decide whether the
            application  factor   is 1/12  or  1/11  or 1/15, you  find that the
            data are  simply not good enough to allow you to do that kind of
            precise  work.   How much  of that  is  due to experimental  error,
            problems  of  techniques and  variations  between laboratories, and
            how much  is  real?  We simply don't know.

Question:    We've heard  that we're going to have to run chemical analyses for
            specific  toxicants,  and,  based on your  comments, if we're  always
            within  limits we should  never see any negative  response of the
            biotest for  that parameter.   Isn't the  biotest kind of  redundant
            on that?

Answer:      The question  was,  if  one is within  the chemical  limits as  mea-
(Mount)     sured  by analytical  chemistry,  one  would  not expect  to see
            toxicity  in  a biomonitoring test,  and the answer to that  is no.
            I  think you  will expect to  see it from time to time, and  we do.
            The reasons  are, first of all,  that  the limitations are clearly
            not on all  chemicals in any consent  decree.   And  the  second
            point is  that  we have not yet come to  grips in the  consent  decree
            v/ork or  in any  other  work  - all  of these standards deal  really
            with the  problem of  nature.

Question:    Let's  just assume that you are  operating  a waste treatment plant
            for industry.   You  have biomonitoring set up, and you determine
            that your effluent  is toxic;  you don't  necessarily  know  what
            to attribute  it  to.  What's your next  step to correct the prob-
            lem, since that's what we're interested in doing?

Answer:      I  think chemical analysis is the next  step, along with more toxi-
(Mount)      city measurements,  and I think the way in which most people go
            about  it, if  there  are multiple  streams  making  up the effluent
            streams,  is  to  go  back  up  the line  and  see  where the toxicity
            is coming from - where most  of  it  is  coming from.   I think it is
            far cheaper  to locate  the source with  biomonitoring and make the
            identity  with chemical  analysis, than to  try  to do  it with
            chemicals all  the way.    If  you go the  analytical  route, you may
                                    -16-

-------
           measure 475 chemicals before you find the one that's causing the
           problem.   But  if toxicity is your concern then it seems obvious
           that  toxicity  ought  to  be the endpoint  in the measurements that
           you  make.  Ultimately, of course,  you'll  have to  identify it
           analytically.

Question:   Isn't  it  possible to  get a toxic  result  and  not actually have
           a toxic condition?  What if there is not enough  DO  in your  efflu-
           ent  stream,  or another condition exists that is not necessarily
           favorable to the organisms that you are using?

Answer:     I think  what  you're  referring to  is  the situation  where the
(Mount)     conditions  of  the test  are different from the conditions  of the
           receiving water.   In that  case,  I  think  the  test would be  consi-
           dered  improper.

Question:  If  all of these  companies and  all  of  these laboratories that
           are  government run  get into bioassay,  and  we  choose something
           like  rhododendron  roots  or something,  isn't  there  going  to be a
           logistic  shortage  of test  vehicles and equipment?  How long will
           it  be before you  have  enough  supply  to  meet  the demand where you
           think  this is profitable?

Answer:    I  don't think  it's  a  problem because  I think that the profit
(Mount)      motives of those  who  do  the test  are  such  that  they will stay
           well   ahead of the rate  at which the  regulatory agencies can
           implement this.   If that is not  the case  then I think it's
           obvious,  and  common  sense  dictates,  that  we have to  move at a
           pace  so that  we can keep  up  with  the demand.   But,  you will be
           surprised at how many firms already have this capability, and the
           evidence  for this  is the fact that the price  is  coming  down.  Now
           it  may not be in absolute dollars  but  it  certainly is  in real
           dollars.   There was  a  time when a  static bioassay was $800, now
           they're considerably cheaper in  a number of  places.  So  I  really
           think  that that's not  a  big problem.   We're not talking  about
           sophisticated  equipment at  this stage  nor  expensive equipment
           compared  to the  modern  analytical  chemistry lab.    I  think the
           problems in measuring 129 chemicals will be more  costly than
           this.

Question:   You  know that  the chlorination  process  causes carcinogens, and
           I  am  wondering if your biomonitoring  program will  take this
           into  consideration, maybe  change, modify, or  ban the  chlorination
           process?

Answer  #1:  I know of no   specific program addressed particularly to the
(Mount)       chlorine  problem.   I  am aware  of  the controversy  in the regula-
           tory  as well   as  the  scientific  community.   I suppose that some
           of  the bioassays that will be  run  may identify  chlorine as a
           particular  problem and  then we  would deal  with it at  the  parti-
           cular site.    I don't know that  any  of  the biomonitoring that we
           have  been talking about  here,  which is really  intended  to be
                                    -17-

-------
            fairly  site-specific, would  go  to  any sort of national  policy or
            program change  with regard to chlorination.   I  think  that issue
            exists  in the  agency,  in  state agencies,  and  in  Canada,  and is
            the subject  of  a  lot  of conversation.  There  has been some change
            in the  policy and  there may  indeed  be more,  but  at this point I
            know of no particular biomonitoring effort.

Answer #2:  I  know of a  number of  places in the country where they have gone
(Garde-     to dechlorination  in order to avoid the  problems of the toxicity
 bring)     of chlorine  itself.  Whether chlorine itself is a problem as far
            as aquatic  toxicity is  concerned  would  depend on  how  much time
            elapses  before you measure  that  toxicity,  and that's  why the
            comments  I've made about persistence are particularly important.
            But with regard  to  the problem of  forming  other  chemicals that
            may be  carcinogens or other such properties,  I think it's recog-
            nized that  there  are  some that  are formed  in  certain  kinds of
            situations.   It seems to me  that  it's  going  to be a negotiation
            between the   trade-offs, which  are what  we  gain  by chlorinating
            and what  we  lose  by chlorinating.   It is  simply a conflict of not
            being able  to do  everything  perfectly;  there  has  to be a trade-
            off.   I think that we may come to recognize, in our society as a
            whole and in  this  field  as  a  whole,  that  there  are worse villians
            running around than coliforms.

Question:    When we  think  of  other large industrial  nations,  where  do they
            stand  on the subject of biomonitoring  and  its application to
            regulations?   Can you  give  us  some  information  on  that?  Are we
            the leaders,  or are we  the  followers?   Are we  in  step  with the
            rest of the world?

Answer #1:  I  can't give you any information.   I can't  say  anything about the
(Mount)     requirements of regulatory  agencies  in other  countries.  But I do
            know that there  is  much  activity  in  Europe,  for example, in
            England  and  South Africa,  in  developing  biomonitoring  tests.

Answer #2:  The only place I know of now is over in the United Kingdom on the
(Garde-     Thames where they've  instituted  continuous  monitoring  have had
 bring)     systems  going for the  past about 6-10 years or so - very active.

Question:    What is  the  perceivable time frame as  far  as post regulations?

Answer:     In terms of  our biomonitoring requirements,  I can't give you any
(Garde-     specific number.   Certainly  as  we go  into the next  round of
 bring)     permit  issuance  and  as we look toward BAT,  we  will be starting
            down that road.   Within Region  V,  as I  said, we are starting to
            target  particular geographic  areas where  problems  may  exist.   We
            are doing that  through the  fish  flesh  analyses and  other data
            gathering that we have.   So,  it will  be,  if not  in the imme-
            diately forseeable future,  within the next year,  into the
            calendar years 80 and  81,  as  we go through the next  round of
            permit  issuance.   It's coming very soon.
                                    -18-

-------
Question:    Do you  envision biomonitoring for POTW1s eventually?

Answer #1:   I guess Glen (Pratt) should probably answer this  question  because
(Garde-     he's the pretreatment expert, but let me give you my  own personal
 bring)     perspective.   I think  that  that really depends on the extent to
            which  we are  able  to put  into  effect  an  effective  pretreatment
            program.   We  know  that  there  are  a tremendous  number of toxic
            materials  that  pass through normal  POTW systems and get out  into
            the waterways.   I  think  the way to  go  about  it  is to look  into
            some process  evaluations for  industries  that discharge  in the
            public  systems,  and I  would  say  generally for this  round at
            least,  we  will be  looking  principally  at  industrial  dischargers.
            For municipal  dischargers, our focus will  be on  getting the
            pretreatment program going.

Answer #2:   Let me  briefly add  to that.  There  might be  a  few cases where we
(Pratt)     would  be  talking  about having the  municipality  do biomonitoring
            on its  influent  rather  than  on  its  effluent.   It  would  be  used as
            part of a program  to go back and control  specific industries or
            to look for  industries  that might  need control.   I  know we are
            looking,  in several  cases,  at  having  biomonitoring  requirements
            for specific  industries  that  would  be  going  into  municipal
            systems.   In general, to emphasize what Sandy  said,  we are going
            to be  looking  in  our  normal  reissuance process  so  that  when a
            given  permit  comes  up for reissue,  particularly in  the  short-term
            permits, we're going to  be  requiring biomonitoring as part of the
            permitting  process, as  well as where we  have  become aware of
            specific  problems  within  a given  watershed.   So I think  it's
            going  to  be through both  of those channels.   As  far  as the
            question on which  countries are doing  things,  I  think  that we
            should  probably  look even closer to  home.  There  are significant
            differences  between the regions  and the states.  Our  region
            appears a  little bit behind several of  the  other regions.   And
            certainly,  some  states have  been more  able to use this in their
            regulatory  process.  So you  will see variations  in biomonitoring
            program applications over the next  year  or two.   This  is just a
            matter  of  getting a  program going with a lot  of  people involved.

Question:    Are you looking at  static testing as adequate?
Answer:
(Garde-
 bring)
 It's not the total answer,  but  we're  certainly  contemplating  us-
ing it.    I  think  we'll  hear more about that as the day wears  on.
                                     -19-

-------
                          STATIC TEST USING FISH
                              Ronald  Preston
                           U.S.  EPA,  Region III
                          Wheeling, West Virginia


       My topic this morning is on static bioassay using fish, and I know my
attempts that follow in describing static  bioassays will  be unfamiliar
to some  of you and  will  be very old hat to others.   To those to whom it is
going to be a little bit  new,  all  I can say is, well, welcome to the mystical
world of bioassay.  I say that with  some facetiousness because we frequently
run  into the  response  that  it is a  very mysterious type of test,  and some-
times we, as biologists,  have some difficulty  in accepting  that response.  I
think that  if you can  accept  a  test standard which is accepted  by many people
and  is  a very, useful  tool,  then  we don't have  any trouble  accepting  the
static bioassay test.

       The  static test  is a  nondynamic  test  consisting simply  of  taking  a
solution of  a  mixture,  placing test organisms in  it,  and  leaving  it  for  a
specified period  of time  to  observe  the  response  of  the test organisms.
Normally the mixture is not added to or taken  away from during the test,  and
once the test  is set  up it  is  left alone for the  test  period.   There is
another type of  static test  that  is  called  a  static  "renewal" test wherein,
on a specified periodic basis, a solution of the test concentration is added
to or exchanged with the  original  solution.  There are several  advantages  and
disadvantages to  the static test.  The advantages  meet  many of the criteria
that we were into  this  morning.   That  is,  it's  simple, it's cheap,  and
althpugh it requires trained technicians,  it doesn't  necessarily require  the
highly trained PhD biologist.  The disadvantages,  of  course, are that in  the
static test, the  toxicity  that might be present may change during  the test
period due to either degradation or  volatility of  the  substance, absorption
of the  toxicant,  uptake  of the toxicant by the organisms, or several  other
factors.  The  flow-through test that you  will  hear  about later takes away
those  disadvantages.   However,  the basic experimental design and  basic
requirements are  well  known  for static  tests, and I  will refer  you  to many
excellent references.   They  are all  very similar and describe  the  tests in
detail.  These are:   Standard Methods,  ASTM, and  the  EPA documents on efflu-
ent toxicity testing.ORSANCO  has a  bioassay procedure and,  of course, there
are many state agency  documents  describing static testing.
                                     -20-

-------
     When we  are considering  the  performance  of  a static  bioassay perhaps
it's worthwhile to understand  what our  objective  is in looking at effluents.
It  is  simply  this:  The basic  purpose  of the effort of  biomonitoring  at the
point  of  discharge  is  to determine  if acute toxicity to  aquatic organisms
exists as a  property  of that  discharge  and  if it  does,  then  there  may be a
potential  impact  in the receiving  water.  The  next step would  be to  try to
evaluate that  potential  impact.    In  attempting to  determine  this objective
(which might  be  termed as a  screening  test)  the static  test  serves well in
measuring this unknown property of  effluent testing.

     I think the best way I might be able to  describe  the static test, before
I refer to  some  of  the requirements  that are in all the  standard  references,
is to discuss some of the types of questions  that we have received during the
performance of  our  program  in Region  III.   We  have  been  going  on-site to
industrial   and  municipal  treatment facilities  and  running  some flow-through
and static tests.   I  think  it might be  good  to repeat some of the questions
that we  frequently  run into from  the  industrial  personnel.   Once we  get by
the question:   Why are you there?  (and  I always refer  to  the statement that I
just made earlier, i.e., we are simply trying to determine  the acute toxicity
of  that  discharge),  the next  question  that  we  ordinarily get is:   "What is
the test organism that  you're  using  in the static bioassay  and why is it the
fathead minnow?   Why  don't  you  use  some indigenous form from the  receiving
screen that's  out there from our discharge?  Why don't you use  carp or  blue
gill, or anything else?   Basically we have a real problem with trying to use
wild stock  as  the test organism.  You  don't  know anything  about  the quality
of  the  group  of fish that you might  collect,  and  you have a very difficult
time  getting  a  constant  size and age.   You don't know the  history  of the
organism, or  what  it  has been exposed  to,  and you  do not  have control  over
the  disease  problems  that  might  be  inherent with  the  strain.    There  is a
whole host of variabilities that may play a role  in  evaluating the results of
the  test,  in  addition to  the difficulties of  capturing the critters in
sufficient  numbers  to begin with.   Therefore it's  much  better  for  the  test
organism to be a fish that can be  cultured, can be  obtained  commercially, and
that can be maintained in healthy stocks.   We are  trying  to reduce all the
known  variables, so  the only variable that we would  be  measuring  is the
quality  of  the  effluent.   By maintaining  the tight  control over  the  test
organism that  is used, you  have  eliminated   another area that could be  dis-
tractive to the  results  of  your  test.    As to why  the  fathead minnow,  it can
be  said that it  is native to  the midwest and  eastern North  America,  and  it is
the white rat, so to  speak,  of the aquatic testing  world.  There are a lot of
data available  for  the toxic response of this  organism.   It  can  be cultured
in  the  lab,  and we have  been  able to  use  it successfully.    There are  other
organisms that  probably are  just  as  sufficient.    Other fish  such  as  blue
gill and channel  catfish  all  meet  the same kind of  requirements,  and in  some
situations  where there  is  a  cold water  environment, the  trout  meet  these
criteria, also.

     The next  question  that  we run into frequently is:   "We are going to use
the  fathead minnow; where do we get the fathead minnow?"  In our region, we
are  attempting  to develop a  list  of  commercial  sources  for  fathead minnow,
and, upon  request,  we would supply that list to  anyone who is attempting to
                                     -21-

-------
set up a bioassay program.  We can't guarantee the quality of the fishes that
you want from these  sources.   We have,  in our program,  attempted to make use
of these commercial  sources  as well as other  sources that  we  have available
to us,  i.e., federal  and state  hatcheries.   We  have  had  some  problems  in
utilizing the  commercial  sources,  and  we have  had  some excellent  results,
too.   The  point is:   The fathead  minnow is  available, and  the person  or
industry who receives  the fathead minnows must take extra  precautions  to  be
sure that  the  quality  of the fishes is  well  maintained.  Another question
that we  frequently get  is:   How are you sure they are  fatheads?  This brings
a chuckle from  us  too, because we  didn't really  expect that question.   Just
recently we  had  a couple  of industries bring that  one up to  us.   My only
response at this time  is:   If you feel  there may  he some question about the
credibility of the  source,  that is, the species that you are  getting from a
commercial  source,  I would simply recommend  some  local  taxonomist to look  at
a sample of  the  fish,  make  identifications, and  provide  the industry with a
statement of certification  that these  are  indeed fathead minnows.   As this
bioassay monitoring program goes down the road, regulatory agencies will have
the  opportunity  to  visit  industry   or  consulting labs  that  are performing
toxicity tests,  and we  will  have  the chance to make identifications ourselves.

     Another question that comes  up .-  maybe this one comes first  -  is:   How
much is  this  going to cost?   You've heard some  comments this  morning about
some cost figures,  and the guides that we have received from various contacts
range somewhere between $300  and  $500 for a  static test.  And  that means the
sample shipped  to a central  laboratory where the static bioassay is performed.

       I would like to  move  on to some  of the basic  requirements in perform-
ing a static bioassay.   What we are  trying to emphasize, and no doubt in some
cases it will  be overemphasized,  is the  quality  control  necessary to have a
good bioassay program.   The reason  that we need  to  emphasize  this,  I think,
is to  reduce attempts  by either  a  regulatory  agency or a  discharger who  is
having bioassay performed to  discredit  the  quality of  data  that is produced.
We are  concerned  with  several areas of  controls  in  the test.    All  of these
are described  in  the references that I mentioned  earlier,  and  I simply want
to review  them  in  this discussion  today.   We start off  with  facilities and
equipment,  and one  of  the primary  things that  is recommended,  of course,  is
to  have  an  adequate  temperature control so  you are reducing  that variable
which can influence the results.  You should have  a well-ventilated facility.
It should be shielded from outside  disturbances.   In that regard,  so you know
what  I'm talking  about  here, it has  been  suggested  in  certain  tests that
disturbances, such  as  laboratory  people walking  by highly visible test
containers, noise  from machinery  and other  apparatus, and  so  forth, can
affect fish  behavior and the  test  results.   I think this  is  probably over-
emphasized, but we  should  try to  control that  condition.   There  may  be
very specific additional  facilities  required for  maintaining a  stock of fish
for a  long  period  of time.   For instance, you may need  to have apparatus for
continuous flow to your stock  of fish in  order to  keep them healthy.  You may
need a UV sterilizer for  disease control or recirculatory system  for treating
the  water with  charcoal  filters.    The materials  that you  use  for water
transport,  the construction material of the containers  that  you hold the fish
in during either the test or  during  holding times  should be relatively  inert.
                                     -22-

-------
It has been recommended  that  316  stainless  steel,  glass, teflon, or a similar
type  of  substance  would minimize leaching,  dissolution,  or adsorption.
Another frequent item that's discussed  is  the  size  of the containers.   This
is dependent upon the size  of  the test organisms and the loading specifica-
tions, which are adequately  described  in the references.  We  do not want to
have undue consumption of oxygen in the test container; therefore you need a
certain volume of the test solution depending upon the size of the organism.

       Another area  that's important to consider  and  one  that needs to have
particular attention paid to it, if you're running frequent and intermittent
tests, is  the  cleaning  procedure that  must  be  performed  between  the static
tests that are run.   The cleaning procedure,  as  simple as  it may sound, needs
to be  specified.  Again,  we're trying  to  reduce all  of these variables down
to where  the only  variable  is the quality  of the effluent.   The cleaning
procedure recommended is  a  simple  washing  with  a detergent,  rinse,  use of a
solvent  to  remove   any  organics  that  might  remain  from  the  previous  test,
dilute acid  wash to remove  reside  metals  from the previous  test,  and then
adequate rinsing.

       When conducting the  static  test, normally more than  one dilution or
concentration is set  up, and  then  the question  arises:   Where  do you  get
your  dilution water?   Dilution  water is  acceptable if the healthy test
organisms survive in  it  throughout  an  acclimation period, that  is,  a period
of acclimatizing your fish  to  the dilution water prior to test, and if they
survive  in that dilution water during the toxicity test without showing
signs  of stress such  as discoloration or unusual  behavior.   For effluent
toxicity testing, the dilution  water should be  a representative  sample of the
receiving water  and  should  be  obtained from  a  point  as close as possible to
hut upstream  of  and  outside  of the  zone influenced  by the effluent.  It may
be practical  to  transport batches  of  water in  tanks  to  the  testing site if
the dilution water  immediately  upstream of  your  discharges is  not  acceptable.

       Pretreatment of  the dilution  water should   be  limited to  a  basic
filtration through  a nylon sieve having  2  mm  or larger holes to  remove
debris or  to  break  up large floating  or suspended solids.  The water should
be obtained from the receiving  water as close as possible  to the  time
the test starts.   It should  not be  obtained  more than 96  hours  prior to
testing.  If unacceptable, or if acceptable dilution water cannot be obtained
from  the  receiving  water, some other  uncontaminated,  well-aerated surface or
ground  water  commercially available can be  used.   The water should  have a
total  hardness, total alkalinity,  and  specific  conductance within  25 percent,
and pH within  0.2  units of  the  receiving  water at  the time  of testing.   If
a  substitute  dilution water cannot be  obtained,  reconstituted  water  may be
prepared and used for the dilutant.

       Many of  the  items that I have  mentioned  about the  loading  of test
organisms for  example,  are  described   in the  reference books  on static bio-
assay.  I refer you again  to Standard Methods and the EPA  manual.  The
loading  and static  tests and test chambers should  not exceed  0.8  g/1 at
temperatures of  20°C or  less  and  not exceed 0.4 g/1 at temperatures above
20°C.
                                    -23-

-------
       There's  always  a  question  about dissolved oxygen.   If you must aerate
to maintain  a  minimum  level,  and  this does  happen  occasionally,  the test
solution should not  be permitted to fall  below 40 percent  saturation for warm
water species,  and  60 pecent  saturation for  cold  water species.   If you do
aerate,  then the  exact  methodology  that  you use  should be detailed  in a
report.   Of  course, organisms  are  not to  be  fed during the tests and there
should be some  accompanying  chemical  and physical data  of the solutions that
go with  your report.

       In summary,  items that  should  go  in your  report  describing the static
bioassay include the following:


       •    The name of  the test method

       •    Investigator and  laboratory  and  the  date the test was conducted

       •     A  detailed  description  of  the effluent  including  its source,
            date,  and  time of collection,  composition,  physical  and chemical
            properties,  and variability

       t    The source  of the dilution water,  the date and  time  of its
            collection, its  chemical  characteristics and description of
            pretreatment

       •    Detailed  information  about  the test  organisms, including scien-
            tific  name,  length and weight, age, lifestage, source, history,
            observed diseases, treatments,  and   acclimation  procedure used

       t    Detailed  description  of the test  procedure,  test chambers,
            including  the  depth and volume of  the solution, the day the test
            was begun, the number of organisms per treatment, and the loading

       •    The definition of the adverse effect that  you measured for the
            test - was the data for immobility or whatever?

       •    A summary of general observations  of  other effects

       •    The number  and  percentage  of organisms in each test  chamber
            including  the  control  chambers  that died  or showed  the effect
            used to  measure the toxicity

       •    How the LC50  was  calculated  should be  described, and  the EPA
            may want  the  recommended procedure described in detail  and
            examples given

       •    Along  with  this  should be any other  relevant information that
            applies  to the static bioassay.
                                    -24-

-------
                          STATIC TEST USING  ALGAE
                              William Miller
               U.  S.  EPA,  Environmental  Research  Laboratory
                             Corvallis,  Oregon


       I am going to present a static test  procedure  that has been developed
by the Corvallis  Environmental Research  Laboratory.   This test procedure  is
an ecological test procedure  that  can be used to define a toxic response  to
various compounds  whether they be  a single metal, a  complex  waste, or any
combination of interacting pollutants that enter the aquatic environment.   I
would like to describe the development of this test  from  the  basic foundation
to the complex application that we  use today.

       The test procedure was initiated  in  1969  by a  joint government indus-
try  task  force on eutrophication.    The  test was developed  primarily  as  a
tool  to define the interaction of nutrients in an  aquatic environment and  to
find out  what  effect might be obtained  if you could regulate the nutrients
entering an ecosystem.

       The key requirements that  we  adopted when  we  were  developing the  algal
assay test were:


       •   It  should  be  designed  so that  technician-level personnel can
            operate it

       t    The equipment  and instrumentation should  be modest and readily
            obtainable.  This is one  area  in  which we may have  failed (since
            an electronic  particle  counter  is necessary  to  conduct the test
            accurately within an  efficient time frame)

       t    It should  be  standardized to give reproducible  results, and the
            geographic location should not affect test  results

       t    All  results should be applied with judgement  to real-world condi-
            tions.
       The test organisms that we chose for our algal  assay  test  was  Selenas-
trum capricornutum.   It  is  a  green unicellular alga that is easily  cultured
                                     -25-

-------
in the  laboratory and  is  amenable to electronic particle  counting.   The
electronic  particle counter  was incorporated to measure cell  density as
well  as mean  cell  volume  changes  within  the  cells during  the growth period.
The combination  of  the cell  density  measurements and the mean  cell  volume
changes can then be used in a calibration curve corresponding to  dry weight,
which  is the final  product used in the test evaluation.

       Two basic publications, developed  in  1971,  which form the foundation
of current  assay research  resulted  from our initial efforts:   an  inter-
laboratory precision test  of  the provisional  algal  assay procedure, and  a
standardized algal   assay  static  test.   (This static test has been further
refined (Miller, Greene,  and  Shirogama,  1973)  and  is included  in  Standard
Methods (1975)  and  is also in the ballot process  with  ASTM.)

       One of the next things  that we accomplished was to define  the effects
of nutrients in 49 different lakes throughout the United  States.   Four basic
productivity groups  were  identified:   low,  moderate, moderately high,  and
high productivity.    We were able to  classify the productivity potential of
the 49  lake  samples  sent  to us by limnologists throughout the United States
through the use of a laboratory bioassay.  In 23 of these lakes  where we had
a  wide  background  of  limnological data,  the algal  assay  test  was  able to
predict the productivity  perfectly  (Miller, Maloney, and  Greene,  1974).

       No single  chemical  test  or biological  measurement  can  be used to
define  all  the  interactions  regulating  biological   productivity   in natural
waters.   The  relationship  between algal  assay bottle tests  and  measurements
of indigenous phytoplankton  in Long Lake,  Washington,  has been  reported
(Greene _et al., 1976).  A high  correlation (r =  0.95)  between  both mm^
indigenous  p~h~ytoplankton/l   and  mg chlorophyll  a/m3  (r =  0.93)  and  mg dry
weight  S_^ capricornutum/1 was  achieved when  consideration was  given to
whether the reservoir was stratified or homothermal.

       The ratio of TSIN to Ortho-P content  in  test  waters can be used  as a
"guide" to  nutrient  limitations  in  natural  waters.    Waters  containing N:P
ratios  greater  than  11:1  may  be  considered  phosphorus limited.   Those   con-
taining N:P  ratios  less  than 11:1  can  be   considered nitrogen  limited for
algal  growth.  Confirmation of a  nitrogen or  phosphorus limitation prediction
is obtained by  analysis of  the  assay  response to  singular and  combined
nutrient (N,P) and/or chelator additions.

        In most cases,  the  trophic  status of  lakes and impoundments  is based
on their bioavaiTable nitrogen and phosphorus content.  Those waters contain-
ing greater than 0.015  mg  bioavailable P/l and  0.165 mg bioavailable N/l are
generally eutrophic.

        It  is  important to  note  that, in the  presence of adequate nitrogen
with  phosphorus  concentrations from  6 to 1,860  mg/1, 1  mg/1  of  available
phosphorus will produce 0.43 mg dry weight of  our test organism.   Also,
1  mg/1  of available nitrogen (expressed as the summation of nitrite, nitrate,
and ammonia-N) supports 0.038  mg dry weight of our test  organism.
                                    -26-

-------
       Some of  our research  indicated  that the  presence  of toxicants  pre-
vented certain waters from attaing the predicted  yield  or nutrient  limitation
status.

       Miller, Greene, and Shiroyama  (l-976b), have reported  that the  inhibi-
tion  of  specific  heavy  metals  upon  the  growth  of _S_._ capricornutum may  be
linear (0 to  100  percent)  with  the increase in  zinc content of test  waters,
but nonlinear  for the increase in  copper and  cadmium content beyond 20  and
40  percent, respectively.   These growth responses  have  established  the
sensitivity of S_._  capricornutum  to  the  bioreactive  state of  these  heavy
metals.   The  greater  than  95 percent  Ij4 algistatic  (inhibitory)  response
of the test alga  in these test waters is  similar to that  of sensitive  indi-
genous species  to  accidental  or  recent discharges of  heavy  metals  (an  algi-
cidal   response  is verified when  a  subculture  from an  algistatic  test  water
fails  to  grow in assay  medium).   However, this inhibited response does  not
necessarily reflect  the  growth  potential  of  indigenous algae  which  have
evolved from long-term chronic exposure to heavy metals.

       The study of heavy metal interaction in natural  waters  is  complicated
by  uncertainty  of the form,  concentration, and  biological reactive state of
the metal.   Thus, with  few  exceptions,  the chemically  analyzed heavy  metal
content of a test water may not  reflect the resultant biological  interactions
and productivity  in natural waters.   The  growth  response of  S.  capricornutum
to conditions of heavy metal  stress in natural  waters is in essence a  "biolo-
gical   response  model" of  complex physical and  chemical  interactions.    The
resultant  biological  response (maximum  standing  crop) is an  integration of
the combined  effects  of  solubility,  ionic strength,  metal  concentration,  and
contact time which regulate toxicity of the heavy metal to the test organism.

       The  response  of the  standard  laboratory algal  test  organism to  the
addition of Na2 EDTA,  alone and in  combination  with  nitrogen and  phosphorus,
to heavy metal-laden test waters,  has been shown to correlate (r =  0.82) with
indigenous phytoplankton standing  crop (Greene et _§!_.,  1978).  The  indigenous
phytoplankton growth  in  these waters  can  be attributed to:   1) adaptation to
their  environment;  2) natural decomposition,  and/or complexing of  the  heavy
metals by both organic and inorganic ligands;  and 3)  the presence of adequate
nutrients.

       A  case  in  point  is the Spokane  River.    The  Spokane  River received
clarified wastewater  from the Spokane Treatment  Plant  which  was  scheduled to
be  upgraded to  secondary treatment.   The City  of Spokane needed to determine
whether or  not  they should go to  phosphorus removal  as a  treatment step and
what  affect this  nutrient  reduction might have  upon  downstream productivity,
even  though the  Spokane  River is  also subject  to upstream heavy metal inflow
from  a smelter  near Wallace,  Idaho.

       We  started  by  defining the response of our laboratory  test system to
specific heavy metals within the Spokane river system.  We found that regard-
less  of the algal species'  interaction within a  natural  system,  our labora-
tory  test  could define the growth  response if  the  system was nutrient-regu-
lated.  However,  in  the  Spokane River, the indigenous organisms had 83  years
                                     -27-

-------
to develop  a  tolerance  to the heavy metals.  .Therefore, the regulation that
was imposed upon the City of  Spokane to  actually  go  to  phosphorus  (nutrient)
removal was indeed justified since the  heavy metals  in this  particular  system
were  not  solely responsible  for  the deterioration  of  the  environment.    In
other  systems  we  may find  that the heavy metals  are responsible and  there-
fore we would have to attack the heavy  metals entering the system.

       How  can  the  algal  assay  bottle  test be  used  to assess  the  impact
of a  complex  waste?    Since  the  components  responsible  for regulating the
biological  productivity  in  natural waters  are  not  necessarily  identified
through chemical  analysis,  an alternate method is  needed.   The algal assay
bottle test can be used to indicate the presence of  bioreactive components  or
growth regulators  within  the waste.   These  components  include (but are not
limited to)  nitrogen,  phosphorus, trace elements, heavy metals, herbicides,
and  pesticides.    The  significance of  the algal  test  is  that  both  growth
stimulation and growth inhibition  can be defined.

       We evaluated 23 textile waste samples representing eight manufacturing
processes by  seven assay techniques to  define  their toxic  properties.  The
bioassessment  organisms  included  freshwater and marine  algae,   Crustacea,
fish, and mammals.  A comparison of the sensitivity  of these bioassays  (Table
1) showed that  the algal  assay bottle  test,  using ^. capricornutum, was one
of the most sensitive tests used  in the  textile waste survey.  This  test not
only identified toxic wastes, but  also  those that  were stimulatory  (Shiroyama,
jt_ jil_., in preparation).

       It is  important  to consider the  following factors  when designing  an
assay  protocol to evaluate  the  environmental  impact of complex  wastes:


       §    The method of  entry  into the  receiving  water (i.e., direct
            discharge after primary, secondary, or advanced waste  treatment;
            percolation through soils;  etc.)

       •    The anticipated  final  concentration of  the complex waste  within
            the receiving water.

       t    The degree  to which the test waters  are representative  of those
            receiving the candidate complex wastes.

       In conclusion,  the algal  assay bottle test  is  a viable tool for the
study of  nutrient  limitation  and  heavy metal toxicity.   It also shows great
potential for  the  evaluation  of.complex wastes.  The validation of  the test
to define  its  sensitivity to broad classes of  industrial wastes is  of prime
importance.  The biggest stumbling block in this validation  is  the  evaluation
of the  toxic  and/or stimulatory  effects  of organic compounds.  The reasons
for this  are:   1) the  safety factor in  handling  the compounds; 2) the vola-
tility  and  insoluble  nature  of  these  compounds; 3) the  lack of  knowledge
about the mode of interactions causing the toxicity; and,  4) the  expense  of
chemical  identification  of both  the  parent  compound  and its degradation
products within the test system.
                                     -28-

-------
American Public  Health  Association.   1975.   Standard Methods for the Exami-
   nation of Hater and Waste Water.  14th  Edition,   p. 744-756.

Greene, J.  C., R. A. Soltero, W.  E.  Miller,  A.  F.  Gasperino, and T. Shiroyama
   1976.   The  relationship  of laboratory  algal  assays to  measurements  of
   indigenous  phytoplankton  in Long Lake,  Washington.   In:   Biostimulation
   and Nutrient Assessment, E. J. Middlebrooks, D.  H.  Falkenborg,  and T.  E.
   Maloney,  eds.Ann Arbor Science,  Ann Arbor, Michigan,  pp.  93-126.

	,  W.  E.  Miller,  T.   Shiroyama,  R.  A.  Soltero,  and K.  Putnam.   1978.
   Use  of  laboratory cultures  of Selenastrum, Anabaena and  the indigenous
   isolate Sphaerocystis to predict  effects  of nutrient  and zinc interactions
   upon phytoplankton growth  in  Long Lake,  Washington.   In:   International
   Symposium on Experimental  Use of  Algal Cultures in Limnology - Publication
   of the Proceedings!  Societas  International is   Limnologiae  (S.I.L.)  (Tn
   press).

Miller, W.  E., T.  E. Maloney, and J. C.  Greene.   1974.  Algal  productivity
   in 49 lake  waters as determined  by algal assays.  Water Research. 8:667-
   679.

	, J. C.  Greene, and  T.  Shiroyama.   1976a.   Application of algal assays
   to define the effects of wastewate effluents  upon algal growth in multiple
   use  river  systems.   In:    Biostimulation and  Nutrient Assessment,  E.  J.
   Middlebrooks, D. H. Falkenborg, and T. E. Maloney, eds.  Ann Arbor Science,
   Ann Arbor, Michigan,   p. 77-92.

	,  J.  C.  Greene, and T.  Shiroyama.   1976b.  Use of  algal assays  to
   define trace-element  limitation   and  heavy  metal  toxicity.    In:   Pro-
   ceedings  of the  Symposium on Terrestrial  and Aquatic Ecological Studies
   of the Northwest.  EWSC Press,  Cheney, Washington,  p. 317-325.

	, J. C. Greene,  and  T.  Shiroyama.   1978.   The Selenastrum capricornutum
   printz algal assay bottle test:  experimental design,  application and data
   interpretation  protocol.   EPA-600/9-78-018.  U.S. EPA, Corvallis, Oregon.
   126 p.

Shiroyama T.,  E. A.  Merwin,  J.C.  Greene, W. E. Miller, A.  A. Leischman, and
   H.  A.  Long.   1978.   The comparative  results of the  AAP:BT to  other
   bioassay  procedures  in  the  determination of stimulatory/inhibitory
   effects of textile wastewater effluents,   (in preparation).

U.S. Environmental  Protection Agency.   1969.   Provisional  assay procedure.
   National   Eutrophication Research  Program,  U.S. EPA,  Corvallis,  Oregon,
   62^ p.

U.S. Environmental  Protection Agency.  1971.  Algal assay procedure:  bottle
   test.   National  Eutrophication   Research  Program,  U.S.   EPA,  Corvallis,
   Oregon.  82 p.

Weiss,  C.  M.,  and R. W.  Helms.    1971.   Inter-laboratory  precision test.
  National   Eutrophication Research  Program.   U.S.  EPA, Corvallis,  Oregon
  70 p.

                                     -29-

-------
                                                   Table 1
                           COMPARATIVE RIOTEST RESPONSES FOR TEXTILE EFFLUENTS*t
Freshwater Ecology Series
Textile
Plant
A
B
C
D
E
F
G
H

J
c^ K
o L
' M
N

pa
R
S
T
U
V
M
X
Y
Z
Fathead
Mi nnow
(96-hr LC5(
% Secondary
Effluent
19.0
MATb
46.5
NAT
NAT
NAT
64.7
c

NAT
NAT
23.5
NAT
48.8

NAT
16.5
NAT
46.5
NAT
36.0
55.2
NAT
NAT
NAT
Daphnia Selenastrum
)) (48-hr EC50) (14-day EC50)
/ % Secondary % Secondary
Effluent Effluent
9.0
NAT
41.0
NAT
7.8
81.7
62.4
40% dead at
100% concen.
NAT
NAT
28.0
60.0
100% dead at
all dilutions
NAT
8.0
NSAd
NAT
12.1
9.4
6.3
NAT
NAT
42.6
11.3
	
____
	
< 2.0
	
___ _
7.8

	
	
12.0
	
< 2.0

__ _ _
8.8

	
__ —
	
1.0
._ — —
__ _..
15.5
Recommended
Interpretation
Selenastrum
20% Secondary
Effluent
%Il4 %Si4
53
—
—
__
956
—
—
92

-_
__
81
—
956

__
95

—
__
—
95
_«
— —
84

83
187
100
	
598
390
	

76
57
	 	
149
	

38
. _ _ _
382
1911
377
232
	
163
261
	
Marine Ecology Series
Sheepshead
Mi nnow
(96-hr LC50)
% Secondary
Effluent
62.0
NAT
69.5
f
NAT
NAT
NAT
f

f
NAT
NAT
f
47.5

f
f
NAT
68.0
NAT
f
37.5
NAT
f
f
Grass
Shrimp Algae
(96-hr LC50) (96-hr EC50)
% Secondary % Secondary
Effluent Effluent
21.2
NAT
12.8
f
NAT
NAT
NAT
f

f
NAT
NAT
f
26.3

f
f
NAT
34.5
NAT
f
19.6
NAT
f
f
f
g
90
f
10 to 50
85
59
f

f
77
1.7
f
2.3

9.0
f
g
70
g
94
50
g
f
f
a Sample inadvertently collected prior to settling pond.        b No acute toxicity.        c Diseased batch of
fish nullified this analysis.        d No statistical analysis because heavy solids concentration obscured the
analysis; the sample did not appear to be acutely toxic.        e 95% growth inhibition in 2% solution of secondary
effluent.        f Analysis not performed on this sample.        9 Growth inhibition < 50% in 100% solution of
secondary effluent.        * No chemical mutagen was detected by 10 microbial strains.        t No rat mortality
after 14 days due to maximum dosage of 10~5 rn^/kg body weight (LD50).  However, six samples (B, C, F, L, N, and S)
showed potential body weight effects, and sample P resulted in eye irritation.

-------
                       FLOW-THROUGH  TEST  USING  FISH
                              William Peltier
                U.S.  FPA,  Environmental  Research  Laboratory
                              Athens, Georgia


       I looked  at  the program  today and  I  noticed I was  listed as being
located in Duluth, Minnesota,  and  I  know that there are a lot of  industrial
representatives in attendance today that have plants in Region IV  and wish  I
were in Duluth, Minnesota.  This  is because  Region  IV does have an  aggressive
program, as  some  of  you  know, and,  hopefully,  today I can  pass  on some of
Region IV's experiences regarding the flow-through testing utilizing fish as
the test organism.

       There are  several  types of flow-throughs; the  first  is  a  continuous
flow-through system,  and  that's  exactly what it means.   There  is a contin-
uous flow  of the pollutant  mixed  with  a continuous flow  of dilution water
that,  after  mixing  together,  goes  to  an  aquarium  containing  fish.   This
procedure can  be  accomplished in  several ways.   One is through  the use of
metering pumps, where  prescribed volumes of  a  pollutant  and dilution water
can  be delivered to  a mixing chamber.   Another  approach  is the  use  of  a
constant water level  head box  of the  type that many  of you have seen in some
of  the research  laboratories  or  in  some of  the state mobile laboratories.
This  second  approach  uses  capillary tubes  of different  inside diameters
inserted in  head boxes  containing the  pollutant  and  dilution  water which
allows for a continuous regulated flow into  a mixing  chamber.   In  both
examples, following  the mixing,  the solution flows  into  the replicate test
tanks  containing  fish.  These are two examples  of  a continuous flow-through
system.

       The   second type  of  flow-through  system  is  the  intermittent  flow-
through system which is widely used  and is  the  type used  in Region IV.   The
intermittent flow is  patterned  after  the  famous  Mount  and  Brungs (1967)
proportional  diluter  system upon which we all have tried to  improve over the
past years.   It's like trying to  build  a better mouse  trap.  For  instance,
Region IV has  designed  and  built a total solenoid system which is described
and  pictured  in  the  EPA  1978  publication  Methods for Measuring the Acute
Toxicity of Effluents to Fish and Invertebrates.   This  system will be shown
during theslide  presentation andits  operation explained  to  those people
that aren't familiar  with this diluter  system.   In this system, flows to the
                                    -31-

-------
proportional diluter can be adjusted to allow for the filling  of the  separate
dilution  water and  effluent  chambers,  and then  when they  are  completely
filled,  a liquid  level  switch triggers  the emptying cycle.   All  chambers
empty into mixing  chambers and from there  flow  into  the aquariums  containing
the test  organisms.   There is a delay  in  the  flow of solution going to  the
aquariums  during   the  filling period.   This  delay varies anywhere from  2
minutes to 15 minutes and is  usually dictated by the desired turnover time in
the aquariums.

       The  flow-through  system has  long  been  used in the  fish  toxicology
field.   Applications  of the  system have  been  in the area of fish,  inverte-
brate,  algal,  and  bacterial  testing.    At the  present  time  in  Region  IV,
flow-through tests  are  being  used  to  determine the acute toxicity of efflu-
ents on fish species -  acute  meaning a  short-term  exposure period  of usually
96  hours  or less  in  duration.   Flow-through  tests also can  be used for  a
longer duration or  chronic testing.  Chronic testing may  include exposure to
a pollutant for a  portion  of  the fishes'  life  or may include  the entire  life
cycle.  A life cycle test usually starts  with  embryos and continues through
life  stages  such   as  larval,  juvenile,  adult,  and  second generation embryo
larval stages.  Of course this approach is time-consuming  and  very  expensive.
In  the  past,  most  of the flow-through  testing has been  conducted   using  a
single chemical as the pollutant and  determining  its  affect  on fish.   This
type  of  work  has  been  conducted  at  universities, industries  and   federal
facilities.  Flow-through  testing  has  been used to establish safe limits or
no effects limits  for chemicals such as  pesticides, toxaphene,  and endrin or
the heavy metals,  zinc, copper,  and mercury.

       The  use  of the  flow-through  system can be  taken  a  step further  and
used in multiple chemical  testing.   For  instance, with a  slight modification
of  the  diluter system,  it has  been  used  to  determine  the  interaction  of
toxaphene when mixed with malathion.   The main  thrust  in Region  IV's  toxicity
testing program is to determine the toxicity of  complex effluents  by utiliz-
ing the flow-through  system.   The  system  is very useful  when one  is dealing
with a "witch's brew" of compounds  that are being discharged into the receiv-
ing water.  The system  is  very useful  in  detecting  wide variations in efflu-
ent concentrations during testing.   In Region IV, effluents have been tested
where in excess of 60 organic  compounds were identified in  a single effluent;
however, they were not limited in the  NPDES permit.   At times, fish lethality
has occurred well   into the test and has  been attributed to a  slug  release of
effluent.   Fortunately, because a flow-through  system was  being utilized,
these slug releases were detected by the fish dying.

       The flow-through system can.also be used  to determine the bioaccumula-
tion of individual  compounds  or compounds  in effluents by fish  when  they are
exposed over  a certain period of  time to  the  pollutant.   Fish tissues  can
then  be  analyzed   for  residue accumulation.  The  test  can be  taken a  step
further by  sacrificing  a  portion of the fish for chemical analysis  and  then
flushing  the  aquariums  containing the  remaining  fish  with  uncontaminated
water for a period of time.   Following a  period  of exposure to the uncon-
taminated water,  the  remaining  fish are  sacrificed  and  the tissue  analyzed
for the  depuration of the pollutant.   Also, the  system  can  be utilized  by
                                    -32-

-------
industries  as an on-line monitor  to  allow the  permittee to  continually
evaluate the effluent for  changes  which  could  adversely affect fish in the
aquarium and  permit the  industry  to  headoff  potential  fish  kills  in the
receiving water.

       Additional  uses can  be made with slight modifications  to  the  system to
study the avoidance  of fish to various concentrations of pollutants, the fate
of pollutants such as degradation products,  or  decay rates.  As  you can see,
the  flow-through  system  offers  a wide  application  of uses in  the area  of
aquatic toxicology.

       The  following are advantages in using flow-through systems.   First, in
addition to the acute test results, the  duration  of the  exposure period can
be lengthened to  derive chronic  or  sublethal  results.   Second, the  problem
with maintaining  a  proper DO level  can  be  reduced  considerably in a flow-
through  system.   This  problem   is  not  completely  eliminated  because some
effluents  have a very  high biological  oxygen demand or chemical  oxygen
demand, and  because  of this,  it has been  necessary  to aerate  the  solutions in
the aquariums in order  to  maintain  suitable  oxygen levels.   However,  for the
most part,  in  flow-through systems, the oxygen level  is easily maintained
without  the  aid  of  aeration.   As  a  result,  larger  fish can  be tested  or
increased numbers  of fish can be tested  for  each  concentration.  Third, the
metabolic waste products from fish, which  deplete the DO, are  flushed from
the  aquariums  and do not  cause  the problems  experienced  in  static tests.
Fourth,  it  is  easier to  achieve a steady-state  chemical  concentration  in
a flow-through system  when testing  single compounds or mixtures  of the
compounds.   Fifth,  in  Region IV,  approximately 50 to 75 percent  of the
effluents were  variable  in  their  concentrations.    As  a  result,  the flow-
through  system was  useful in detecting slug discharges.  It's  the slug
discharges  that  usually cause the fish kills in  our receiving waters.

       Additional  advantages  are better control  of  the concentrations  of
the volatile pollutants and better precision.

       The  following are  some of  the disadvantages.   The  first  would  be that
this sytem  requires  a  higher degree of  technical expertise to conduct the
test.  Second,  a  greater  degree  of equipment  sophistication is required as
will be  seen from  the slide  presentation.   Third,  it requires  copious
quantities  of  dilution  water.    Fourth,  at  times  it  can require much more
space.   Finally,  the most  important  disadvantage is  cost.  After polling 187
consulting  firms  located  in Region  IV  during  August  1979,  it was established
that 27  of  them  had  the   capability for conducting  toxicological  studies.
Some firms  provided  information  on  the  cost of  various tests.  Therefore,
some cost estimates  are available at this time.   An acute flow-through test
conducted at the consultant's laboratory  on  a single compound or an effluent
ranges between $700 and  $1,000.   The lowest estimate was $400  with the
highest at  $1,500.  For conducting  acute flow-through tests in  the field on
an industrial effluent, the  consultant's cost  ranged from $5,000 to $7,000
for  an  8-10 day  study.    The  lowest  estimate  was  $3,000  and  the highest,
$15,000.
                                     -33-

-------
       At this  time,  I would  like  to  show  a  few slides.   The first  slide
(Figure  1)  shows  one of  Region  TV's  mobile toxicity  trailers  set up  at  an
industrial  site.  In the trailer there are two  solenoid flow-through systems.
During  a test, a portion of the effluent is continually pumped from the
discharge ditch into the trailer.   Dilution water is  collected daily from the
receiving water upstream  from  the discharge outfall.   The diluter  water  is
stored in a  tank and then  pumped into the trailer.

       The  second  slide (Figure 2)  shows  the  proportional diluter  solenoid
system.   The dilution  water fills  upper chambers;  when the  last chamber
fills, a liquid level switch triggers  an electrical mechanism  which opens the
solenoids,  allowing the dilution water  to  flow into mixing chambers.   Simi-
larly, the  effluent goes  into  the  small  chambers  located directly  beneath
the dilution water  chambers.  The effluent  in  the chambers also empties into
the mixing  chambers at the  same  time  the dilution  water is  emptying  into
them.

       The  third  slide (Figure 3)  shows  what   I would call  a  poor  man's
diluter  system.   It is where a minimum of three solenoids are  utilized and
the operation is based on a vacuum siphon system.  The siphon  system operates
very similarly to the total  solenoid  system; however,  it is much  less expen-
sive than the total  solenoid  system.

       At the  EPA  laboratory in  Athens, Region IV is  evaluating the Reckman
Microtox, which some of you may have read or heard about.  I want to make one
thing perfectly clear, and that  is  Region IV is evaluating the instrument for
Beckman  - we  are  not endorsing or selling  the instrument.  For the  past  18
months, we have been testing complex  effluents  to determine the 24-hour LCBOs
(lethal  concentration which  is  lethal  to  50  percent  of  the  test  organisms)
for fish and daphnids and comparing  the results with  a 5-minute EC50s (effec-
tive concentration which causes an effect on 50 percent of the  test  organisms)
obtained from the  Microtox unit.   The principle of the test is based  on
the  use  of  freeze-dried,  saltwater,  luminescent  bacteria which  are  recon-
stituted  in  a  buffer  solution.   An  aliquot of  cells are then placed  in  a
cuvette  which  in turn  is  placed into  a  chamber in order  to measure the light
output from  the bacteria.   Then the  effluent  is  added and the  light output
is measured again for either a  decrease,  increase, or  no change in the light
output.  When  testing various effluent  dilutions,  an  EC50 can be calculated.
The EC50 in  this  case  would  be  the  concentration that  would reduce the light
by 50  percent.  I  would just like to  briefly say that, from our observations
and the  limited data,  the  unit  has  some  advantages and  some disadvantages.

       The following are the advantages  of the  Microtox.    Its  evaluations are
very  rapid; it takes  about an hour to test  a  control  and five effluent
concentrations.  While  the results from the  static fish  and Daphnia test are
not available  for  24 hours,  the data from the Microtox  unit  is available  in
one  hour.   When  using  a  standard  toxicant,  the  results from the  unit are
fairly accurate and provide  good  reproducibility.  The  unit  is very compact
and  fits nicely on  a  standard  laboratory bench.   A  direct  digital  display
readout  is  standard, or a  recorder may  be attached for continuous  recording.
                                     -34-

-------
CO
en
                    Figure 1.   Mobile  toxicity  trailer set up at an industrial  site (Region IV).

-------
Figure 2.   Proportional diluter solenoid system.
                      -36-

-------
Figure 3.   Solenoid,  vacuum siphon diluter system,
                      -37-

-------
       The following  are disadvantages of the Microtox.   First,  the highest
concentration  that can  be tested  at  the present time  is 50 percent.   The
reason for this  limitation  is  the test  requires a 0.5 ml addition of bacteria
cells to  0.5 ml  of test  solution which  results in the  50  percent test solu-
tion.   Therefore, the  highest concentration that  can  be  tested is  a  50
percent  effluent concentration.   Second,  the color  of  effluents poses a
problem due to  certain  colors masking the  luminescent  output from  the  bac-
teria.  Therefore, a  reduction in  light may be due to either the pollutant's
effect on the bacteria or due to the color absorbing  the light.   Third,  some
tests have been  providing  false  negatives.   False negatives  are defined  as
where lethality  has occurred  in the tests with fish and/or invertebrates, but
the Microtox  test results have indicated  no  significant change or an increase
in the luminescence  from the bacteria.  Therefore,  if you  are using the  unit
as the sole tool  in determining an effect,  you're going to be misled in  some
cases.   So far,  between 10  and  20  percent  of  the effluents  tested  have
resulted  in  false negatives.   Fourth,  fresh  water  effluents  require  that
sodium chloride be added  to  bring  the salinity up to  20  parts  per thousand
prior to testing.  The reason for this is the bacteria  used  in the test  is a
saltwater species.   Therefore,  there  is some  question from  the regulatory
aspect  on the  validity  of the results  obtained from  testing   an  effluent
sample to  which  some compound has been  added  that  may change  the effluent
chemical  characteristics.   Representatives from Beckman  Instruments are fully
aware of all  the disadvantages and  are working to resolve  the problems.  They
have  indicated  recently that they  may have  a  breakthrough  in the color
problem,  and may  also  have a  technique  for increasing  the  effluent concen-
tration tested.    In  Region  IV,  the Microtox unit will  certainly not replace
fish  toxicity tests  in  the  regulatory  requirements;  however,  once  the prob-
lems  are resolved,  it  may be used  as  an  additional  tool  in   the aquatic
toxicologist's  tool  box  for screening effluents  for  toxicity.   Also, keep in
mind  that  the  findings  from the  evaluators testing  pure compounds  may  be
entirely different than  what Region IV has  found with the complex effluents.

       Also,  there is  one last topic  I would like  to cover briefly which
involves  an  instrument  which was  developed and  is  being  used  in  Europe  to
monitor  toxic  effluents.   Region  IV^  is presently  evaluating  this instru-
ment  which is  called TOXIGUARD and  is  manufactured  by  EUR-Control.   The
company has an office in Decatur,  Georgia.   The Europeans  are using it as an
on-line monitor  of raw  waste  before  to the wastewater  goes  into a treatment
system.   The instrument  utilizes bacteria  growing on multiple plates  with a
small continuous  flow of  wastewater  passing through  an aeration chamber and
then  through a  chamber containing  the bacteria  growing  on the  multiple
plates.   If  the  wastewater is not toxic and the bacteria are not stressed,
then  normal  biological  activity  occurs and  the  DO  in  the wastewater  is
removed.   As the wastewater exits  the chamber, it passes across a  DO probe
which monitors  the DO level.   If there is zero DO,  then the bacteria are not
stressed, and it  is  assumed that  the wastewater is  not toxic.   However,  in
the  event the DO begins to  increase and  it  approaches a pre-determined
milligrams per  liter,  an alarm bell goes  off, and the operator can divert the
    of November 1979,  Region  IV is no longer evaluating  the TOXIGUARD.
                                    -38-

-------
wastewater to a back-up lagoon so  it  won't  kill  the bacteria in the treatment
system.   We're starting to evaluate the  instrument  using complex effluents
and comparing the results with the results  obtained from acute toxicity tests
with fish.
                             LITERATURE  CITED
Mount, D. I. and W.  A.  Brungs.   1967.   A simplified dosing apparatus for fish
     toxicological  studies.   Water Res.  1:21-29.

U.S. Environmental  Protection Agency.   1975.   Methods  for measuring the acute
     toxicity of effluents  to  aquatic  organisms.   Environmental Monitoring
     and Support Laboratory, Cincinnati, Ohio.
The author is presently an aquatic biologist  with  EPA  Region  IV and is respon-
sible for the  aquatic  toxicology  program.   His office is located in Athens,
Georgia  30605,  and  he can  be contacted  at the following phone  numbers:
FTS-250-2294 or Comm. 404/546-2294.
                                     -39-

-------
                    BIOCONCENTRATION TESTS FOR EFFLUENTS
                             Gilman D. Veith
                 U.S.  EPA Environmental Research Laboratory
                            Duluth, Minnesota


       One of the objectives of biomonitoring programs for complex effluents
is to  select  the most cost-effective set of  tests  which provide reasonable
assurance that potential  hazards of  a discharge  have  been evaluated.  Stan-
dards for BOD have been  in  use  for a long  time,  yet it is obvious that many
highly hazardous  chemicals could be  discharged even  though the BOD standards
are met.

       Although  recent studies indicate  that most effluents receiving proper
secondary treatment show little acute toxicity to-  aquatic  organisms,  acute
tests are the best experience tests  to  conduct  on an  effluent after BOD has
been removed.  The next level  of concern  is the chronic toxicity of the waste
stream which  could  be determined from  the more  costly  embryo-larval  test.
Finally,  even if testing showed no  hazard from  the  above  tests, chemicals
such as PCB, hexachlorobenzene,  etc.  could  be discharged at quantities great
enough to  cause  hazardous  residues  in  fish  in  the receiving waters.   The
reason is  that  these chemicals are  extremely  bioaccumulative and  will  be
found as concentrated residues  even  when the concentration  in the water is
below a toxic level.  Consequently,  there is a need for a test for bioaccumu-
lable chemicals  and a need  for a strategy  for  using  the test  in an appro-
priate manner.

       First of  all,  let me  just  define some terms.   The bioconcentration
process is  the accumulation of  chemicals in the  body  of  an  animal  over and
above  the  ambient concentration.   In  the case of fish,  it's  simply the
amount in  the fish tissue  compared  to   the  water environment or the  food.
There's been  an  argument in the literature whether fish  bioaccumulate more
through the gills  by  respiratory  uptake and  partitioning  into the fish, or
through the food  chain.   The answer is  that  both mechanisms are important,
and  the  question  of  how important  depends  on  the environment in  a very
predictable way.   The  bioaccumulation process  is the sum of bioconcentration
and  biomagnification;  biomagnification   is  that  which  is accumulated  from
ingesting the chemical,   and  bioconcentration  is that  which  fish  can accumu-
late through the gills when placed in water with that chemical dissolved in
it.
                                    -40-

-------
       The  bioconcentration  test  is  conducted  by  placing  approximately  35
fish such as the fathead  minnow in  a large  aquarium.  The chemical or effluent
is continuously added,  and  composite samples of five fish are removed after 2,
4, 8, 16, and 32 days for  analysis.   It  is generally observed that the chemi-
cal is accumulated in a manner simulated by a first-order process which gives
a  rapid  initial uptake  followed  by  a  slower  rate  and finally  a relatively
constant residue concentration (Figure  1).  This point is  called the steady-
state  concentration  for the  exposure and the  bioconcentration factor is
calculated  by  dividing  the steady-state residue  by  the water concentration.

       Reasearch has  shown that the  ratio between  the fish  residue  and the
water  concentration  of  a  chemical,  i.e., the   bioconcentration factor,  is
constant  over  a wide  range  of  water concentrations.    This means  that  the
steady-state residue  in  fish  can be  calculated  from  the  water concentration
if  the latter  is  known.   Also,  the hioconcentration factor  for  chemicals
varies  from less than  one to  greater  than  100,000.   Chemicals which  have
small  bioconcentration  factors generally  reach  steady-state in  fish  within
the  first few  days  of exposure whereas  those  with  high  hioconcentration
factors may take several  weeks  or may never reach  a  true  steady-state con-
dition.  Figure  1 shows the uptake of hexachlorobenzene in  various age groups
of  fathead  minnows.   Although  it  appears  that  a steady-state  is reached, a
closer  examination  shows  that  the   residue  continually  increases  over the
115-day exposure.

       The research we have completed  at the Environmental Research Laboratory
-  Duluth  suggests  that  a  30-day exposure  is the longest  period of exposure
needed to assess the bioaccumulation  potential even though the actual accumu-
lation in the environment may  be slightly higher  for  the higher bioaccumulable
chemicals.   Moreover,  the test  can  be  readily adapted for  effluent  tests
using either live cages or a  single aquarium  receiving  the  effluent.  We have
conducted bioconcentration  tests concurrently  with toxicity  tests  by analyzing
surviving fish  from toxicity  tests for residues.  While the kinds of experi-
ments  give a rapid  indication  of the presence of highly bioaccumulative
chemicals,  they may not  give  the same bioconcentration factors obtained from
the longer bioconcentration test.

       Because  bioconcentration  tests on  complex effluents require  a quali-
tative  and  quantitative chemical  analyses  of residues in  the  exposed fish,
the  cost  of the tests  in  a monitoring  program  is  still  rather  substantial.
Consequently, we have developed  a  screening test  which  is based  on the use of
bioconcentration factors but  reduces  the  overall costs.  We tested chemicals
which  were  representative of  a  wide range  of chemicals and  found  that the
bioconcentration  factor could  be related  to the  n-octanol/water  partition
coefficient  as  shown  in Figure  2.   The  octanol/water partition coefficient
is  a measure of  the  fat  solubility  of  chemicals and  is generally  inversely
related to  water solubility.   Chemicals such  as  detergent builders,  solvents,
phenolics,  etc.  have  a  high  water solubility and a corresponding low biocon-
centration  factor.  Chemicals  such as the chlorinated  pesticides, PCB's,  and
brominated  flame retardants  have  a  very low water  solubility  and  a corres-
ponding high bioconcentration  factor.
                                      -41-

-------
I
.£»
ro
                                                                               • Newly hatched fry
                                                                               o 30-day fry
                                                                               A 90-day juvenile
                                                                               • Adults
                        20     30     40     50     60     70      80

                                        Exposure Time (days)
90
100    110     I2O
              Figure  1.  Accumulation of hexachlorobenzene by  fathead minnows of four age groups.

-------
CO
     "8  5
      O>
-8
CM
ro 2
     U_
     O
     GO
          0
           0
                Log BCF=0.85 Log P-0.70
                R =0.947
                N=59
• Fathead minnow
D Rainbow trout
A Bluegill
                     234567

                           Log P (n-octanol/water)
       8
             Figure 2.  Relationship between the bioconcentration factor of 54 organic chemicals
                      in fish to the log P of the chemical.

-------
       This correlation is  useful  because it will permit estimates of biocon-
centration factors for  chemicals  without  testing.   Only  if  those chemicals
which have a  large partition coefficient  are being used  in  a  plant  would a
detailed  bioconcentration  test  be justified to determine  if such chemicals
are leaking out of the  plant.   Since the  log  P can be estimated from struc-
ture, the  overall  estimate of bioaccumulation potential can be achieved at
the screening level at low  cost.   We  have found that chemicals with partition
coefficients  less  than 1000 are  unimportant  from a bioaccumulation viewpoint,
and bioconcentration tests  should be  limited  to  those chemicals most likely
to be bioaccumulative.  Since the majority of industrial chemicals belong to
this group, these  screening  relationships may substantially  reduce  the cost
of testing.

       If the effluent or  wastewater  is so  complex  that it is impossible to
list the possible chemicals, it  is still  possible to screen the effluent for
bioaccumulative chemicals to  determine if  the bioconcentration  test is
justified.   In the  EPA report   EPA-600/3-78-049  (available  from NTIS),  we
showed that the partition coefficient  of chemicals can be determined from the
retention  time  of  the  chemical  on  a  reverse-phase  liquid chromatography
column.   Consequently, the  retention  time of  the  "peaks"  on  this column can
be  used  to estimate  log P and  therefore  the bioconcentration factor of
chemicals, even though the  identity  of the  "peaks"  is not known.  Screening
for chemicals  of  high bioaccumulation potential is  translated  to screening
for chemicals which have a  long  retention time on  the reverse-phase column.
Since water is  used as  a solvent, effluent  samples can be injected  directly
onto the column or a  precolumn  and  effluent screened within  twenty minutes.
If "peaks", or  chemicals,  are detected which  have  a long  retention  time, it
is highly probable they will have a large log P and a large bioconcentration
factor.    If  none  are detected,  it  is unlikely a  bioconcentration test will
provide useful  information  concerning  the hazards of the effluent.
                               DISCUSSION
Statement
by Glen
Pratt, EPA
Chicago
Question:
Let me just clarify one question that was asked earlier today on
natural  toxicants in an industry's  intake water.   If you have a
material in your intake water and if you do not add any contami-
                              you are not  responsible for it in
                              is that if you take water from one
                              put it  into  another waterway, you
                              a lower quality than the receiving
nants to  it in your process,
your effluent.  The exception
waterway or  from  a well  and
would be liable if it were of
waterway.
In the  last  session a  question  was asked  regarding  what other
jurisdictions were doing with respect to biomonitoring.  I might
point out that  in Canada there is  a  requirement  in the regula-
tions to  have  industry  do  bioassay tests,  basically 6-hour 1X50
flow-through  tests.  Also, to  carry on with a question that was
raised this morning which addressed  the increase in the amount of
biomonitoring that's going to be  requested  by the U.S. from the
                                    -44-

-------
Answer:

Question:
Answer:
Question:
Answer:
state agency, I've got a  problem  with  lab  availability  now,  and
it may  be  a problem in the  future.   Also what kind  of quality
control  problems  will  there be  as  a result  of  the increased
workload?

Are you asking will  laboratories be available?

The problem is that now  in Illinois I don't know of more than  two
or three sources that are available  commercially  to  do biomoni-
toring.  For example,  if 30 industries have biomonitoring require-
ments within  their  permit,  how  are they going to  get the  moni-
toring accomplished?  We  will get  to  the  point where labs  start
to  spring  up, and  then  what kind  of quality assurance are  we
going to get from these  labs?

I  don't have too  much  concern with the first  problem of  lab
availability  so  long as  the regulatory  agencies don't  impose
unreasonable deadlines in doing  this.   I  think you'll  be  sur-
prised  to  find how many  labs  really are  now able to  do  these
tests, and  I can  assure you that,  if the need is there, those  who
are  willing to  do them  for a fee will also  appear.   This  is
state-of-the-art  technology  rather  than research tools  at  this
stage.   As far  as quality control  is concerned, there are a
number of round-robin tests being  performed right now to measure
the  precision of  these  various  tests.  There  are activities  by
the  Office of Toxic Substances and probably state agencies,  too,
to develop good laboratory practice manuals and offer laboratory
certification.  So I think there's a lot of activity in the area
of quality control.
We're looking
what criteria
for any given
at a lot of bioassay tests today,
is used in determining which test
waste stream?
and I'm wondering
might be suitable
It's my view that this should  be  based on an industry-by-industry,
pi ant-by-plant basis.  What we're trying to do today is to give
you the  broad range  of  tests  that  are available.   In  our work
and the  state's  work with  industries,  we  need to  look  back  at
the specific  processes   in  the plant.    This  is  the  one reason
why we believe  it's tied  with process evaluation.  Are you
looking at a  type  of compound  in the plant that you think would
tend to be a long-term bioaccumulative material?

Are you  looking  at  a  process  in a  plant that  produces  only  an
intermediate compound which is consumed  within  the  plant?  When
Don Mount  says you can't generalize, we  totally agree,  and this
is why we're  not  having  a broad  spectrum requirement that every
industry must do  test  A,  B,  or 2  from Column A and one from
                            saying that  it  should  be a pi ant-by-
                            are in an industry that  doesn't have
            Column B.   Instead,  we're
            plant evaluation.   If  you
            biologists on  its  staff, you  might want  to  get together  with
                                     -45-

-------
            either  the state agency  or  a biological  consulting firm  to  see
            which  particular  test  would  be suitable for your given facility.

Answer #2:  In the way of  clarification,  in case  it  slipped  by some  of you,
            the  bioconcentration  tests that Gil  described is  directed  at  a
            phenomenon or  an effect  that we're concerned about at  levels
            below  those  concentrations causing toxicity.   So if you  have  a
            direct toxicity problem, you  may not  be concerned with  the
            bioaccumulation  test  at  this  point.    I  think it  is  clear from
            everyone  I've talked to that  it is not prudent to require all of
            these  tests  by  any  means, but  rather to select those that  are
            most appropriate to the problem that you have.
                                    -46-

-------
                             SEDIMENT BIOASSAY
                               Max Anderson
                        Central  Regional  Laboratory
                            U.S.  FPA,  Region V
                             Chicago,  Illinois

                           Bayliss (Rock)  Prater
                 Aqua Tech Environmental  Consultants,  Inc.
                                P.O.  Box  76
                            Mel more,  Ohio 44845
Max Anderson
       The need for a sediment bioassay in Region V  became  quite  apparent  in
1976 and  1977  when  we  encountered an increase in dredging  activities in the
harbors and  streams  around the Great Lakes.   At  that time, Rock Prater was
working for the Central  Regional  Laboratory on a  special  assignment.  Through
a combined effort, we  adapted  an apparatus that was  originally described  by
Cal  Fremling of Wannona  State  College and have been  using  that as our  sedi-
ment bioassay chamber.   The chamber is depicted in Figure 1.

       The chamber  is  constructed entirely  of glass and  is  a closed  cycle
system.  An aerator forces air and water  to  circulate throughout  the system.
The sediment is deposited in the  upper part along  with the organisms that are
being tested.   The  small  cup  on  the top  left  side  is  used for  the Daphnia
bioassay.    We place other organisms within  the  chamber itself.   If  we're
using fish as  our test organisms, they are  placed in one of the larger  glass
jars.   The cost of this  particular  apparatus runs  between $30  and  $50,  so
it's not very expensive and it  works  quite well.

       We   have adapted  and enlarged this  apparatus  at  the Central  Regional
Lab for use in long-term biomagnification work that  is just getting started.
Since we're  only  using  fish  in  this  particular  study  we  find  that  we can
maintain  them  for a  longer period  in the  larger  unit.   The principle  is
exactly the  same  as  the smaller  unit  but it  is  constructed  of  a 10 gallon
aquarium  on the  top and  two 5-gallon  aquariums   underneath (Figure 2).
In  Figure  3  we have  two tandem  units together.   Figure 4 shows a  bank  of
units in our environmental  chamber.   We estimate that the cost  for setting  up
                                     -47-

-------
*****
             Figure 1.  Schematic diagram of a closed-cycle sediment
                       bioassay chamber.
                                      -48-

-------
-617-
      •(A uoiSey)  /Uo^ejoqei [euoiSay  L^UBQ at|i A'q
paq.depe SB  sn^ejedde Aesseotq ^uemipas  a[0/Co-pasoL3
                         "i


-------
I
en
o
                    Figure 3.   TWO tandem closed-cycle bioassay units (Central Regional Laboratory,
                               Region V).

-------
  I
^-1
IT)

-------
one of these unit would be about  $75  to  $100, which  is relatively inexpensive
if you're interested in a long-term biomagnification study.

       I would like now to turn the time over to Rock Prater who is going to
discuss the application  of this  sediment  bioassay  apparatus  based upon some
extensive work that he has done.
                                     -52-

-------
Rock Prater
       It never fails to amaze me  that,  after 21  years of teaching,  I still
get nervous  standing  in front of  a group.   Not  to  use this platform  as  a
business opportunity,  but as of  last  Friday,  I am no longer with Heidelberg
College; I'm with a consulting firm that's doing primarily sediment bioassay
work.

       As Donald Mount said  in his  introduction this  morning,  there  are two
accepted approaches to  determining  the  condition  of an  aquatic  ecosystem  -
animal  counts  in standing  crop  and identification of  chemical  parameters.
Although the  ecosystem  might contain  an  acceptable  set of chemical  para-
meters, there are situations in which no  animals are present.  Unfortunately,
the impact of the  sediments in such a situation is usually overlooked.

       This  particular  apparatus that we  are currently using  for sediment
bioassay is quite basic; however, I think that the data that we have  accumu-
lated  after investigating harbors on all  the major lakes, with  the exception
of Lake Ontario,  shows that  it seems to be  working quite well.  The apparatus
still  needs  refinement;  we  are  soliciting a  grant to  continue  development.
Mr. Robert Hoke, who  is  also with  Aqua  Tech,  will  be a valuable asset since
he has just finished three years  of  graduate research in this area.

       We  have  used  this  apparatus  in both  riverine  systems and  lakes.
Figure  1 shows  a  small creek  which receives effluents  from four major
industries.   One  of these,  the  "oil  refinery", was  being taken to  task  by
EPA for  thermal  pollution  of this  stream.  We took  samples  above  and below
the major outfalls  to  determine what the  bottom type was like.

       We used three organisms in this bioassay -  Hexagenia limbata,  Asellus
communis and  Daphnia.   Hexagenia  limbata is a  burrowing  mayfly which  is
especially usefulff you're  going to  look  at  the effects  of sediment.
Asellus communis is an  isopod that  lives at the  interface,  and  Daphnia  is  a
water  flea which lives within the water column.  In other studies we  usually
use fish.

       Figure 2  indicates  the toxicity  to the three major  species  that  we
used.    Proceeding  downstream from  the   control station  we get  100  percent
mortality during the  96  hours,  a small   amount of  recovery,  and  then an in-
crease.  Put  what  does  this mean  in  relationship to the refinery?   Should
they  be  forced  to  eliminate the thermal  discharge when  indeed  the sediment
types  that currently  exist  (and  we  did  profiles on these)  might prevent the
existence of  a  healthy  ecosystem?  In  addition,  the seiche effect  on  this
system would prevent  any scouring  so  that  the return of a healthy ecosystem
may take a  number  of  years  or  in  fact  may never occur.  This  is  a  typical
example of the need for a sediment  bioassay.   If we simply look at the water
column, or at  the  standing  crop of organisms, and  ignore  the  sediment  type
we may be amiss.
                                    -53-

-------
                                         Maumee Bay
                                                   	
                      Oil  Refinery (0.8 km)
                                                 Station 0.4 km-
                                                  Sediment 5
                                               Station 1.3 km-
                                                Sediment 4
Glass  Mfg. (10.9 km)
 Water Treatment Plant (4.9km)
                                 Station 6.8 km-
                                  Sediment 3
   Oil Refinery (8.7 km)
                          'Station 9.3 km-
                            Sediment 2
             Control Station 13.0 km-
                 Sediment 1
Sediment Station
and Number
      Figure 1.   Sediment stations  and  industry locations
                 on Otter Creek, Ohio  (not to scale).
                              -54-

-------





R$$&$$&2&&^^
   ^^<^£iS?$iSS«^

5   I

1   I

§   <4
                                                                            WCO
                                                                            CO tot*-
                                                                                                    O O
                                                                                                    •r- S-
                                                                                                    O) oo
                                                                                                 Q-ce:
                                                                                                     E
                                                                                                 c r- o
                                                                                                 ro CTt t—
                                                                                                    -iii

                                                                                                 O C
                                                                                                 •r— 't—
                                                                                                 £:
                                                                                                 o -a
                                                                                                    o>
                                                                                                  •> s-
                                                                                                 O
                                                                                                  ns
                                                                                                    O)
                                                                                                 CD -Q
                                                                                                 0> E
                                                                                                 i- 3
                                                                                                 O C

                                                                                                 ^- CU
                                                                                                 CU S-
                                                                                                 -I-) ro
                                                                                                    c
                                                                                                 4- O
                                                                                                 O -r-  •

                                                                                                 >> rO O)
                                                                                                 ro -Mr-
                                                                                                 00 C/) ro
                                                                                                 oo    O
                                                                                                 ro    oo
                                                                                                  CU
                                                                                                  cn
                                                                                               a>
                                                                                                    J2 OQ -t->
                                                                                                  c -o o
                                                                                                  CU C C
O CU >       s-
-M S- 00 O
•r- 3 ro
•— O    S-
ro ^C T3 CU
4->  I  CU +->

O CT> rO O
E    U
   ro •<- <+-
•P    "O O
C CD E
CU C ••- J=
O -f-    -M
S_ S- CU 3
CU 3 S- O
Q- "O ro E
                                                                                                  CD
                                                                                                  S-
                                                                                                  3
                                                                                                  en
                                                  -55-

-------
       This technique can  also  be used  in  making  management decisions.   We
are currently under contract  with  the  Corps and private industry to study a
number of  harbors.   The situation that  exists  in Indiana Harbor  is a good
example  since  it  is  generally  recognized  that  Indiana  Harbor is  a  toxic
environment.

       If the harbor is  dredged,  how do we dispose of the dredged material?
Should we  dump  it into  the open  lake,  or should we  take it  upland?   The
sediment bioassay can  be used  to help answer this question.

       In Figures 3 and  4,  it is evident that the toxicity of the sediments
decreases as we  go from  the harbor to  the open  lake.   Since the toxicity of
the dredged material  varies between locations, the disposal technique should
vary accordingly.   If the  alternative  is to dispose of  the toxic material
upland, it  still must be capped.   The most reasonable approach, then,  would
then be to  use the less  toxic material as the cap, since  it would have to be
dredged anyway.

       Another  example  is  Ashtabula  Harbor.   We did  a study  of  the  upper
layers only.   Using  this  sediment  bioassay  apparatus, we  determined  that
these upper layers  were  only moderately polluted or nonpolluted.  The results
led me to conclude  that  the dredged material should be allowed to be disposed
of in the open lake, which is what I would  normally recommend for moderately
polluted or nonpolluted  sediments.  However, EPA did not agree..  They wanted
to know if the  toxicity  changed  with  different depths of the sludge.

       So,  in the next series  of samples  we  looked at the toxicity at differ-
ent levels.  According  to  the results presented in Figure 5,  the  top  layer
of site 6  (Ash-6T) was  more toxic than the bottom layer (Ash-6B).   However,
at site 2,  the top layer was  less toxic  than  the  bottom layer.   If you must
decide how deep  to  dredge and  where to  put the dredged material, this type of
information is a necessity.   If the top 6 feet are not  very  toxic  but the
underlying  sediments are,  it  makes  sense  to  dredge to the  six foot level,
dispose of  the  nontoxic material  in the  open  lake  and   do  something  else
with the underlying material.

       Unfortunately,  Figure 5  indicates that  the  replicability is  not
as good  as we  would  like.   But anyone  who  has  done  chemical  analysis  on
sediments will  admit that replicates  can  vary considerably.

       Basically,  that  is the sediment  bioassay  program that  we are using to
determine what  we are going to  do with dredged  spoils  from harbors and ship
slips in the Great Lakes.  It is my  opinion that there will be more sediment
bioassay  work down the road.   Although it  needs refining,  it is a  very
substantial  technique.
                                     -56-

-------
100 -
                                            Pimephales promelas

                                         _ Hexagenia limbata

                                         = Daphnia magna
     Sediment IH-1
                  Sediment IH-2
                               Sediment IH-3
                                            Sediment IH-4
                                                                B
                                                         Sediment IH-5
 A  I  B
Sediment IH-6
                            Sediment Numbers  (Sites)
  Figure  3.   Percent mortality of Hexagenia limbata,  Daphnia magna, and
              Pimephales  promelas during  a 96-hour sediment bioassay of
              Indiana Harbor, Indiana,  1978.
                                     -57-

-------
«••
"5
   100-1
    90-
    80-
    70-
    60-
    50-
    40 H
    30-
    20-
    10-
H Pimephales promelas

_ Hexagenia limbata
= Daphnia magna
                                                                00
                                                                         0 0 S 000
Sediment IH-7    Sediment IH-8    Sediment IH-9   Sediment IH-11
                                                               A I B
                                                             Sediment IH-12
                               A | B
                             Sediment IH-13
                              Sediment Numbers (Sites)
     Figure 4.   Percent mortality of Hexagenia limbata,  Daphnia magna, and
                 Pimephales  promelas during  a 96-hour  sediment bioassay of
                 Indiana Harbor,  Indiana,  1978.
                                        -58-

-------
Sediment
Sites
Ash-4B
Ash-5
Ash-6T
Ash-6B
Ash Beach
  A
  0
         I   I   I   1   I   I   I   I    L   1
B (Replicate)
0
                                            P. promelas
                                            H. limbata
                                            D. magna
        i   i   I    i    i    i    i    I   I   I
       100 90  80  70 60  50  40  30  20
10   0  10  20 30  40  50  60  70  80  90 100
Mortality (%)
     Figure 5.  Percent mortality of Pimephales promelas,  Hexagenia  limbata,  and  Daphnia magna
               from a 96-hour sediment bioassay of Ashtabula  Harbor illustrating different
               toxicity for surface samples (T) and bottom sediments (B).

-------
                               DISCUSSION
Question:

Answer:
Question:


Answer:
Question:


Answer:
What is the time of the exposure?

96 hours, basically.  What  we  tried to do was develop a procedure
that you can do in a short  period of time because there are about
126  harbors  in  Region V that are proposed to  be  dredged.   It's
not  practical  to run six to eight  week tests  on 100 or  more
harbors.  So what we need  is  some sort of test that's a red flag.
I'm  not  proposing  that  this  be  used  in  lieu  of bulk
chemical analysis  and  bioassay of the elutriate.   I
should be used  in  conjunction with  them.  We've just
elutriate chemistry and bioassay argument to rest.   I
has  been working  on a  lengthy statistical  analysis using a large
data base to  see  when elutriate chemistry  and  bulk chemistry is
                                                                analysis or
                                                                m saying it
                                                                put the old
                                                                believe Bob
            important  and how they  relate
            hours?   That just seems to be
            speaking,  if you've  got very  toxic sediment, you'd  get the
            same results  in 24.
                                 to percent mortality.  Why 96
                                 accepted exposure.  Generally
If you're looking
apparatus?
                  at sediment, why do you  have  the  flow-through
You get a  high  BOD  with  sediments.   If dissolved oxygen were to
drop down to 1 part per million,  let's  say, and the critters die,
you  can't  say  it's  because  of  the  sediment toxicity.    So  the
recycling  is  important  in  maintaining  dissolved  oxygen  within a
reasonable level.   Very seldom does it drop  down  below 5 parts
per million.  Plus, if you're in  a riverine system, we can almost
duplicate the flow because you can change the calibration on the
apparatus  and  actually  have a  flow-through  system  that  might
approach the CFS's; or you  can change the flow across the pattern
in  the  sediment.   Primarily, though, we use  continuous  flow to
maintain dissolved oxygen.
Are your critters
yourself?
                  available from stock or do you  cultivate  them
                                                from the  Toledo
                                               are quite readily
The  last  study  we did,  we got  the Daphnia
Zoo.  Daphnia magna we  had Dr. Kreiger do; they
available.   Pimephales,  the cosmopolitan that was  talked  about
this morning, is  readily available here.   As  for  the quality,  we
get  ours  in  Newtown,  as long as  we're doing  governmental  work.
The  Hexagenia are called  wigglers,  and we get  them from  a
bait  shop  at $7  per  thousand.    The proprietor  thinks  I'm  the
greatest fisherman  in  the  world,   I  guess.   Asellus  communis  -
that's another story.   We maintain  a culture unit  and keep these.
The  biggest  hassle  is  in the winter; you've  got  to be prepared
for the winter.
                                    -60-

-------
Question:    About the relationship betv/een  toxicity with Daphm'a and toxicity
            with  your other test  animals -  it  appears  as  if the Daphnia was
            more  sensitive  most  of the time.  Is there  any advantage to using
            an additional  three  species  or should you  just go with  that?

Answer:     Section 404 says that Thou Shalt use three.  So you must use mul-
            tiple species,  number  one.   And  I think that we got the spectrum.
            Daphnia  is very sensitive, and  Pimephales  I think  you'd  have to
            hit  with  a  sledge hammer.   I  think the other  two  species fall
            somewhere  in  between.   Daphm'a  is very erratic; it is very tough.
            Sometimes we use clean  sand  controls and get  a 30% kill  with
            Daphnia.   Maybe  they're  eating the  sand  grains, I  don't  know.

Question:   Does your sedimentation  take  into consideration acclimation?

Answer:     Yes  it does.

Question:    How  long do you acclimate?

Answer:     Generally, 24 hours.  We've found  that  we get  a recolonization in
            the  sediment  very similar to  the  recolonization  that  existed in
            the  sediment  when we  took it.   In  other words,  we've  looked at
            the  different horizons after we put it  in and let it settle out.
                                    -61-

-------
                        TENTATIVE GUIDELINES FOR

     FLOW-THROUGH  EARLY  LIFE STAGE  TOXICITY  TESTS WITH FATHEAD MINNOWS

             FOR  USE IN THE U.S. EPA, OTS-ORD ROUND ROBIN TEST


                              Donald I. Mount
                U.S. EPA Environmental Research Laboratory
                             Duluth, Minnesota


       The following draft  protocol for an embryo-larval  test is being
evaluated for  adoption.   Users  should  bear  in mind that  changes will  be
made before  finalization.

       Tests as  described  in this  method estimate  chronic toxicities  as
measured in  life cycle  tests in  a  high  percentage of the comparisons  that
currently can be made with the existing data base.


1.           In  an  Early Life Stage Toxicity  Test with fathead minnows,
       organisms are exposed  to  toxicant  during  part of  the  embryonic  stage,
       all  of the  larval  stage,  and part of the juvenile stage.   The orga-
       nisms are examined  for statistically significant reductions in percent
       hatch, percent survival,  and weight  in order to determine  upper and
       lower chronic values.

            A lower chronic  value  is the  highest tested  concentration  (a)  in
       an acceptable  chronic  test,  (b)  which did  not cause  the occurrence
       (which was statistically significantly different from  the control  at
       the 95 percent level)  of any  specified adverse effect,  and (c) below
       which no tested concentration caused such an occurrence.

            An  upper chronic  value is the lowest tested  concentration  (a)  in
       an acceptable chronic  test,  (b) which caused the occurrence (which was
       statistically  significantly  different from the  control   at  the  95
       percent  level) of  any  specified adverse effect and (c)  above which all
       tested concentrations  caused  such an occurrence.

2.           Not enough  information  is currently available  concerning early
       life   stage tests with  fathead  minnows  to allow precise specification
                                     -62-

-------
       of details  for most  aspects of the test.   Enough such tests have been
       conducted  and enough aspects have been studied,  however,  to indicate
       that  these  guidelines are appropriate.  A prudent course of action for
       anyone  planning to  conduct  such tests would be to initially conduct a
       test  with  no  toxicant  to  gain  experience and  to  determine  if  the
       requirements  of items  10,  11, 19,  20,  25 and  26 are met  using  the
       planned water,  food,  procedures, etc.   General   information  on  such
       things  as apparatus, dilution  water,  toxicant,  randomization  of test
       chambers  and  organisms,  and  methods for  chemical  analyses,  can  be
       found in  Draft #10 of the  proposed ASTM Standard Practice for Conduc-
       ting  Acute  Toxicity Tests with  Fishes. Macroinvertebrates. and Amphib-
       ians.

3.          Tests  should  be conducted with at least five toxicant concentra-
       tions in  a geometric series and  at  least  one control  treatment.   The
       concentration of toxicant in each treatment, except  for  highest
       concentration and  the control  treatment,  should usually be 50 percent
       of that in  the next  higher  one.

4.           If  a  solvent  other than  water  is used to prepare test solutions,
       a solvent control  (at the highest solvent concentration present in any
       other  treatment)  is required in  addition  to  the  regular control,
       unless  such a  control  has  already been tested in the same water with
       the  same  species  of  fish, food,  and test  procedure and  the water
       quality has not changed significantly.  A concentration of solvent is
       acceptable only if it  is (or  has been)  shown that that concentration
       or a higher  one does not cause a difference (increase or decrease in
       any of the kinds of data specified in item 27) from control organisms
       that  is significant  at  the  95  percent  level using  a two-tailed t-test.

5.            For  each treatment  (toxicant concentration and control) there
       must  be  at  least  two replicate  test  chambers  each  containing one or
       more  embryo cups with at least 50 embryos divided equally between the
       embryo cups at the  beginning of the  test.

6.          Two  test chambers  have been  used  routinely:

       a.   Twenty fish have been tested in  a chamber which  is 16  cm x 44 cm
            x 18 cm high  with  a  16 cm x 18  cm, 40  mesh stainless steel screen
            6  cm from one end, with a water  depth of  12.8 cm  and with a flow
            rate of 190 ml/min.

       b.   Fifteen fish  have  been tested in a chamber which  is 6.5 cm x 17.5
            cm x 9.5  cm high with a  6.5 cm  x 9.5  cm,  40  mesh  stainless steel
            screen 2 cm from one end, with  a water depth  of 4.4 cm  and with a
            flow rate of  15 ml/min.

       All  of the above are  inside  dimensions.   In  both  test  chambers the
       water  depth  is controlled by  a standpipe located in the  smaller
       screened  compartment  with  the  test  solution  entering at  the other
       end of the test chamber.
                                     -63-

-------
7.            Embryo  cups  should  be glass cylinders  about 4.5 cm inside dia-
       meter and about 7 cm high  with  40 mesh  nylon  or stainless steel screen
       glued to the  bottom.   The embryo cups must  be suspended  in the test
       chamber  is  such a  way as to  ensure that the organisms are always
       submerged and  that test solution  regularly flows  into and  out of the
       cup  without agitating  the  organisms  too vigorously.   A  rocker arm
       apparatus driven by  a  2 rpm motor and having  a vertical  travel  dis-
       tance of 2.5  to 4.0 cm has been successfully used, as have self-
       starting  siphons that cause the level of solution in  the test chamber
       to rise and  fall.

8.           An  acceptable  dilution water for early  life  stage toxicity tests
       with  fathead minnows is one in which the species will survive, grow,
       and reproduce  satisfactorily.

9.           A  16-hour light  and  8-hour  dark photoperiod should be provided.
       A 15- to 30-minute  transition  period at  "lights on" and "lights off"
       may  be desirable.    Light  intensities  from  10  to 100 lumens  at the
       water surface  have been  used successfully, but the intensity should be
       about the same  for  all test chambers.   Light  should  be  provided  by
       wide-spectrum  (color  rendering  index  > 90) fluorescent  lamps.

10.          Tests should  be conducted  at  25°C.   The temperature  in  each
       test chamber  should be between  24  and  26°C  at  all times and  must
       be between  20  and  28°C at  all  times.    If the  water is  heated,  pre-
       cautions  should  be  taken  to assure  the  supersaturation  of dissolved
       gases is  avoided and  total  dissolved  gases should  he  measured at least
       once  during the test  in  the   water  entering  the  control  treatment.

11.          The dissolved oxygen concentration should be between  75 and
       100  percent saturation  at all times  in all  test  chambers.   At  no
       time  during  the test should one  test  chamber have a dissolved oxygen
       concentration  that is more  than  1.1 times the dissolved oxygen concen-
       tration occurring in another tank at  the same  time.

12.         The flow rate  of test  solution through the test chambers must  be
       great enough to maintain the dissolved oxygen concentration (see items
       11 and 22)  and  to  ensure  that  the toxicant  concentrations  are not
       decreased significantly  due to  uptake by test  organisms  and material
       on the sides and bottoms of the chambers.

13.           A test  begins  when  embryos in embryo cups are placed  in  test
       solution  and ends 32 days later.

14.         Embryos  and fish should not be  treated to  cure  or prevent disease
       or fungus before or during a test.

15.           Embryos should be obtained  from  a  fathead minnow stock culture
       maintained  at   25°C  and a  dissolved oxygen  concentration  between  75
       and  100  percent saturation with a 16-hour light  and 8-hour  dark
       photoperiod.   Frozen  adult  brine shrimp have been  successfully used  as
                                    -64-

-------
       a  food  for  adult  fathead minnows.  The most eggs have been obtained in
       a  30 cm x 60 cm x 30 cm deep chamber with a water depth of 15 cm when
       15 cm x 30  cm  quadrants are formed with stainless steel screen and one
       male, one female  and  one or two  substrates are placed in each quadrant.
       Half-round  spawning substrates with an inside diameter of 7.5 cm and a
       length  of 7.5  cm  have been used  successfully.

16.            The  afternoon  before a  test  is  to begin,  all  of the substrates
       should  be  removed  from  an appropriate  number  of tanks  in  the stock
       culture unit  and  should be  replaced  about  the  time the  lights  are
       turned  on the next morning.   Enough  (at least  three) substrates with
       embryos  on them  should be removed six  hours later and soaked in
       dilution water  for  two  hours.    For each  individual   substrate  the
       embryos should be gently separated and  removed  and  visually examined
       using a  dissecting  scope  or a  magnifying viewer.   Empty  shells  and
       undeveloped and opaque  embryos  should be  discarded.   If less than 50
       percent of  the embryos  from  a substrate appear to be  healthy  and
       fertile, all  the embryos  from  that  substrate  should  be  discarded.
       Single  embryos with no fungus or  partial  shells  attached are prefer-
       able, although embryos with some fungus or partial shells attached and
       clumps  of two or three  embryos  (with or without separation) have been
       used successfully.  An approximately equal  number  of acceptable
       embryos from  one  substrate should  be  impartially  distributed to each
       embryo  cup  and the process repeated for  at  least  two more substrates
       until the  proper  number of embryos have  been  placed in each  cup to
       give at  least  50 embryos per treatment.   The embryo cups  should be
       standing in dilution  water  when  the embryos  are being distributed and
       then the cups  should  be  randomly placed in the test chambers.

17.          Twenty-four  hours  after  they are placed in the embryo cups, the
       embryos should be  visually examined under a dissecting scope or magni-
       fying viewer and all  dead embryos  discarded.   Embryos that are alive
       but  heavily  fungused should  also be  discarded.    Forty-eight  hours
       after the start of the exposure, all dead and heavily fungused embryos
       should  be removed and the remaining healthy, fertile embryos randomly
       reduced to  the required number  (at least  30 per treatment).   If more
       than about   25 percent  of embryos  in  the control treatment  are dis-
       carded  within  the first  48 hours of the test because they are dead or
       heavily fungused, the test  should probably  be  restarted.    Each  day
       thereafter  dead embryos  should be discarded.

18.          In each treatment, when hatching is about 90 percent completed or
       48  hours after first hatch  in  that  treatment,  the live  young fish
       should   be  counted and   released into  the test  chambers.   Unhatched
       embryos should be left  in the  cups and released into the test chamber
       when they hatch.   The  range of time  to hatch in each  cup should be
       recorded.

19.          A test should  be  terminated  if the average percent hatch in any
       control  treatment is less than  50  percent or if the percent hatch in
       any control  embryo cup  is more  than 1.6 times that in another control
       embryo  cup.


                                    -65-

-------
20.           The  flow rate,  size  of the test  chamber and the amount of food
       added should be  such  that  the average weight of  the  controls at the
       end of the  test would not  be significantly greater if  only half as
       many fish were  tested.

21.           Each  test chamber  containing  live  fish  over 2 days old must be
       fed live,  newly-hatched  brine shrimp at  least  twice a  day  at least
       6  hours  apart   (or  three times  a  day  about  4  hours apart)  on days
       2 to 5 after hatch  and at  least  5 days  a week thereafter.  They must
       be fed at least once a day on all other days.  Other food may also be
       provided  in addition  to  the  above.  The amount  of  food provided to
       each chamber may be  proportional  to  the number and size of fish in the
       chamber,  but each  chamber  must  be treated  in  a  comparable manner.
       Quantifying the amount of live,  newly-hatched brine  shrimp  to be fed
       is  difficult,  but   the  fish  should not be  overfed or  underfed  too
       much. A large  buildup of food on the bottom of  the chamber is a sign
       of overfeeding.  A  sign  of  not  feeding  enough  of the  right kind of
       food is that in a sideview the abdomen does not protrude.

22.            Test chambers  should  be  cleaned  often enough  to maintain the
       dissolved oxygen concentration  (see items 11 and 12)  and  to ensure
       that the toxicant concentrations  are not decreased significantly due
       to sorption by  matter on the bottom and  sides.   In most tests if the
       organisms are not overfed too much  and  the flow rate is not too low,
       removing  debris from  the bottom  once or twice a  week  should be ade-
       quate. With some toxicants  that  promote growth of bacteria, the sides
       and bottoms should  be cleaned more often.  Debris can be removed with
       a large pipette and rubber  bulb  or by  siphoning into a  white bucket.
       The pipette or  bucket should  be  examined  to ensure that no live fish
       is discarded.

23.           Temperatures should be recorded in all test  chambers once at the
       beginning  of the test   and  once near  the middle of the test.   In
       addition, temperatures should be recorded at least hourly in one test
       chamber throughout the test.

            The  dissolved  oxygen  concentration should  be measured  in each
       treatment near  the 1st, 21st and  28th days  of the  test.

            Hardness,  pH,   alkalinity, and  acidity should be measured once a
       week in the control  treatment and once  in the highest toxicant concen-
       tration.

            The  concentration of toxicant should be measured at  least twice a
       week in each treatment.

24.            Dead fish should   be  removed  and  recorded  when  observed.   At a
       minimum,  live fish  should be  counted 11, 18, 25  and 32  days  after the
       beginning of the test.   The  fish  should  not be  fed  for the last 24
       hours prior to termination.  At  termination  the number of fish that
       are visibly (without the  use of dissecting  scope  or  magnifying viewer)
                                     -66-

-------
       grossly  abnormal  in  either swimming behavior  or physical  appearance
       should  be determined.   Also  at termination the  weight (wet, blotted
       dry)  of  each  fish  that  was  alive at the  end of the  test  should be
       determined.   'If the  fish exposed to  toxicant appear to be  edematous
       compared  to control fish, determination of dry  rather thin  wet
       weight  is desirable.

25.          A test  is  not acceptable  if  the  average  survival,  of the controls
       at the  end of the test  is  less  than  80  percent or if  survival  in any
       control chamber is less than 70 percent.

26.          A test  is  not acceptable  if  the  relative standard deviation  (RSD
       = 100 times  the  standard deviation divided  by  the mean) of the weights
       of the  fish  that  were alive at  the end of the  test in any control test
       chamber is greater than 40 percent.

27.          Data to  be statistically analyzed are:


       •    Percent  normal hatch

       •    Percent  survival  at  end  of test  (based  on fry, not embryos)

       •    Percent normal  at end  of  test  (based on  fry, not  embryos)

       •    Weights  of individual  fish that  were  alive  at end  of  test.


28.           Dichotomous data (live-dead, normal-abnormal)  should be analyzed
       using contingency tables or log  linear  techniques.   For weight data,
       the  individual fish  are used  as  the replicates  unless a  two-tailed
       F-test  indicates  that differences between  replicate  test chambers are
       not negligible.   Weight  data  may be  analyzed using Bartletts' test and
       one-way analysis  of  variance, but  to obtain information concerning the
       upper and lower  chronic  values,  Dunnett's procedure  (Steel and Torrie,
       Principles and Procedures of Statistics.   1960,  p. Ill)  should be  used
       to identify treatment  means that  are  statistically significantly
       different from the controls at the 95 percent level.


                               References
Benoit, D.  A.  and R.  W.  Carlson.  1977.  Spawning success of fathead minnows  on
       seclected artificial  substrates.  Prog. Fish-Cult. 39:67-69.

Flickinger,  S.  A.  1969.   Determination  of sexes  in the fathead minnow.
       Trans.  Amer.  Fish.  Soc.  198:526-527.

Gast, M. H. and  W.  A.  Brungs.  1973.   A procedure for  separating  eggs of the
       fathead minnow.   Prog. Fish-Cult. 35:54.
                                    -67-

-------
May, R.  C. 1970.    Feeding larval  marine  fishes in the laboratory:  a review.
       Calif. Mar.  Res.  Comm.,  California  Cooperative Oceanic Fisheries
       Investigations Report 14:76-83.
This tentative procedure  was written by Charles Stephan with the help of many
members of  the staff  of the  Environmental  Research  Laboratory  in  Duluth,
Minnesota.
                                    -68-

-------
             EFFLUENT GUIDELINES LIMITATIONS AND LETHAL UNITS
                             Kenneth J. Macek
                          EG & G, Bionomics, Inc.
                          Wareham, Massachusetts


       I've been asked to talk a little bit about the toxic unit concept.   It
seems  to  me, from  some  of the  questions that  have  been asked, that  there
still  seems  to  be  a certain amount of confusion  in the minds  of some  people
here  about  what exactly this  is all  about  and  why  you should  be  concerned
about  it.   Therefore,  I'll  take just a minute to review  a few concepts that
led me to  go through some of the thought  processes which  I  did in  preparing
this paper.

       Obviously,  the  requirement  for measuring  toxicity  is  here to stay
whether people  in  the  agency who have to  deal with the problem or  people  in
the  industry who  have to live  with  the  problem like it  or  not.   Another
thing  that's just  as  clear  is  that  the  reliance  on analytical  chemistry
measurements in the  permit program  has not adequately  addressed the question
of toxicity, primarily because constitutents not routinely monitored can, and
do, contribute to toxicity.   However,  I think it's also obvious that increas-
ing the analytical burden of the permittee is  not really  the answer, despite
the fact that you  might  feel you'd rather make  a few  more measurements than
deal  with  this  nebulous  area  called  "biomonitoring".  I  think in  the long
run,  as Dr.  Mount suggested this morning, we'll  find that the  cost and the
effort related  to  doing it  biologically  and  toxicologically  are  probably
going  to be  less  than  continuing to increase the already  complex and  signi-
ficant analytical  costs.

       But more  important are the toxicological reasons why analytical mea-
surements  aren't  the  answer.    Even if  we  did  measure  everything   that was
in  the waste,  we'd  have to know the  toxicity to  the aquatic  organisms  of
each of those constituents.   If  any of you have looked at the  proposed  water
quality criteria documents  being published under the  consent  decree, you'll
find that even for some  very,  very  common compounds  like  benzene and chloro-
form, we  don't have toxicity data for aquatic organisms.   As the chemicals  in
your waste become  more and  more unique or more and more  characteristic of a
particular industry, the less and  less  likely  it is that such  data exists.
                                      -69-

-------
       Furthermore,  even if we had such data, we would then need data on all
of the combinations of  constituents that  could  be in the wastewater and the
interactions toxicologically  that  could  occur  there.    So it's  clear that
measuring all  of  these  constituents  is not the answer - we can't  do it,  we
don't have  half of  the  toxicity  data  required,  and even  if we did, we would
still have  a long  way  to go.   So  the simplest approach  is  to  measure the
toxicity of  the whole  waste,  and that  is what  I think  this  meeting is all
about.

       I'll   spend a little bit  of  time  talking  about toxic  units although
I'm not  sure in my  own  mind where we  can  go  with it at  this  point.  Back in
1971, a group working  in the San  Francisco Bay area  led by  Esvelt and cowork-
ers at the  University  of California  completed  a  comprehensive  study of the
toxicity of municipal  wastes to  aquatic  organisms.  As a result  of these
efforts  they  proposed  measuring, monitoring, and  regulating  toxicity as we
have  historically regulated conventional  pollutants  such as  solids, BOD and
COD.  In its most simplistic  form, they proposed to  measure toxicity, estab-
lish  a discharge  limitation  in the form  of  a mass  emission  rate,  and apply
treatment technology  designed to  "remove toxicity"  in order  to  meet  the
reasonably established  limitation.

       I  guess I better go into a little  detail here because this toxic unit
concept  can  be confusing  if  you try  to  make  it anything more  than it is.
All it represents  is a  simple  alternative  way of expressing an LC50.

       Basically what they proposed was that LC50 values, expressed  normally
as  mass/volume or  volume/volume  concentrations,   be  translated  into  what
would represent a relative toxicity "concentration" which could be  expressed
as  toxic units.  The  concept of toxic  units  is well  established in the
literature  at  this  time.   In  its  simplest form, if you have a constituent
like  copper  whose LC50  to some organism is  2 yg/1,  and  you  have 10 yg/1 of
copper in your  waste, the copper is said  to contribute 5 toxic units to the
toxicity of  the waste.   When you have an effluent,  for which the  effective
concentration of the "constituent" is  100 percent, the number of toxic units
in the effluent is simply the  LC50 for that effluent  to some organism  divided
by  100.   The quotient  thus represents  a  unitless measure of toxicity arbi-
trarily  called  the  toxicity  "concentration" and expressed as  toxic units.
Unfortunately, the  use  of the term concentration can be  misleading as the
calculation yields toxic units, a measure independent of any specific volume
-  it's  2 per  liter,  2 per gallon,  2 per million  gallons,  whatever.  They
further  propose, with respect to the  mass emission  rate, that the  product of
the  toxicity concentration  (in  toxic  units per  some  unit  volume)  in the
wastewater times the flow rate of the wastewater  in  the  same unit  volume per
unit  time  would yield  a  mass emission rate in toxic units  per unit time.

       The authors recognized  that,  if toxicity  was  to be considered  in water
quality  management  in a manner similar to conventional   pollutants, they had
to evaluate the effectiveness  of  conventional and advanced  treatment technolo-
gies  to  reduce toxicity.  They also recognized that,  since  a toxic  unit was  a
test-specific  parameter, the  tests  had to be fairly standardized  and  repro-
ducible.  This latter point is particularly important if, in  fact, the
                                     -70-

-------
concept of toxic  units  ever gets into the NPDES program.   I'll  elaborate on
that in a few minutes.

       I'd like to  give  you just a brief outline of  the  results of Esvelt's
program.  They worked primarily  with  five municipal  waste effluents,  and the
mean LC50 value of  all  samples  tested ranged from  40 to 60 percent effluent.
These  LC50  values translate to  1.5 to 2.5  toxic  units.    Furthermore,  they
found that the mean  relative  standard  deviation  derived from repeating tests
24 times  on  the  same source of  effluent  was about +20 percent.   Neither of
the  above findings should be particularly startling, as  relatively low
toxicity and relatively  little variability in the toxicity of municipal waste
would not be unexpected.

       They went  on  to  point out that when  they evaluated  the mass emission
rate for  a  single  effluent tested daily for  eight consecutive  days,  they
found that,  although the  absolute LC50  value  and the absolute flow rates could
vary by as much  as  100  percent,  the mean mass emission rate for toxicity had
a relative standard  deviation of  only  15 percent.

       Unfortunately they didn't go on to say whether this was coincidence or
whether they had  some fundamental  basis for assuming there is a real relation-
ship here.  I certainly don't know  what the  relationship would be, and I sus-
pect it's more coincidence.

       They went  on to look at the  reproducibility of these toxicity tests in
their  program, and  they  found that the  replication  of results was excellent
with a relative  standard  deviation of only 5  percent.   Now that's a lot
better  than  I've  ever done, and I  suspect it's a  lot better than most other
people  in the room have done.  Again,  I don't know what the reasons for those
numbers  are,  but in  my opinion  I  believe  a  more representative  number of
reproducibility would be something on  the  order of 15  to 20 percent.

       They  also went on  to look  at whether  one could  treat toxicity to
reduce  it.   They looked at biological treatment as  you  would  in a municipal
waste  system, and they found  that even for those wastes that were relatively
non-toxic to  begin  with (1  to 2 toxic  units),  biological  treatment   reduced
what little toxicity was  there  by more than 75 percent.

       Now that's all  I'm  going to  say  about the toxic  unit  concept.  It's
just another way of expressing the  toxicity which one measures by a bioassay.
I'm  a  little concerned about  using it in  a mass emission sense because,
even in this limited  program,  they  showed  that the same waste  could vary
by jf 20 percent.   If you believe that the reproducibility of the bioassay is
+_  20  percent,  then you've  got a propagation  of errors there  that makes it
very hard to live with a single mass emission rate number.

       Now  I'd  like to  tell  you  about  some studies that we  conducted for
EPA  to further  evaluate this  toxic unit concept  and how it  might apply to
industrial waste  treatment  facilities  and  NPDES programs.
                                     -71-

-------
       Based  on  the results of  the  study by  Esvelt  and his coworkers,  and
motivated  by  a  desire  to  control  toxicity of  wastewaters  discharged  into
surface waters, the EPA was concerned  with  answering  several  questions  about
this  potential  approach.   The questions to  be answered  in our  program  were
concerned with (1) the efficacy of treatment systems  for  removing toxicity in
industrial wastes;  (2) the  variability of the toxicity  of both  the  influent
to and the effluent from industrial  waste treatment systems;  (3)  the  required
frequency of  sampling for enforcement  and compliance  monitoring;  and  (4)  the
effect of species selection for monitoring.   EG&G,   Bionomics,   using  mobile
laboratory facilities,  visited  seven industrial  sites (comprised of  one
organic chemical  facility,  five pesticide  facilities  -  selected because  of
their anticipated relatively high toxicity,  and one pulp  and  paper facility).
The  sites were  also selected because  they represented various  types  of
treatment technologies  essentially equivalent to  BPT, and in some cases
technologies more advanced  than  BPT.   In order to investigate some  of  those
areas of  interest  to  the  agency, we conducted 24-hour static toxicity  tests
with fathead minnows using batch  wastewater  samples collected  twice a  day  for
10  consecutive  days.   Also,  in  order to investigate  the effect of  species
selection, we concurrently tested Daphnia magna  on  alternate  days.   And  I'd
like now to show you some  of that data.

       The first data set is the  results, expressed  in toxic  units,  of  tests
with fathead minnows at  a  pulp mill  of International  Paper Company in George-
town, South Carolina (Table  1).  There are several  things to  note about  these
data.  First  is  the relatively low toxicity  of the influent to  the treatment
system.   Second, and perhaps  more   important, is  the relatively low  varia-
bility in the toxicity of  the influent to the treatment system.   The  third is
the  complete  removal  of  toxicity by the  treatment  system.   Lastly,  we find
that  the  comparability  of  the results from  the  two species tested seems
reasonable.  Now this  is obviously a  pretty clean plant.

       The next data set is from  an  organic  chemicals plant,  Union Carbide's
plant in  Charleston, West Virginia  (Table 2).    Significant aspects  of  these
data  are  the  slightly higher  toxicity of the  influent;  the occasional  order
of  magnitude  excursions  in  toxicity  of the  influent; the complete  removal
of toxicity  as  a result  of  treatment  even   in the face  of these relatively
large  excursions;  and again  the comparability  of  the  daphnid  data to  the
fathead minnow and  the ability of either species to  detect the  rather  large
excursions in the  toxicity in the  influent.   In this  case the treatment
system did not entirely  remove the toxicity,  although  I must  caution  you that
measurements of toxic units  less  than  unity are tentative to say the  least.
You have  to extrapolate when you don't have  a  complete 100 percent mortality
or mortality above 50 percent  to try to  get  at these  numbers, and I  wouldn't
put a lot stock  in toxic units  less than unity.

       The last set of data is from  a  pesticide  manufacturing plant  operated
by Diamond  Shamrock in Green  Bayou,  Texas  (Table 3).   This plant is  charac-
terized by  a  relatively high  variability in the toxicity of the  influent  to
the treatment system, with  excursions  exceeding  two  orders of magnitude;  on
the other  hand you  get  absolutely zero variability in the effluent from  the
treatment system.  So there  is a rather remarkably consistent degree  of
                                    -72-

-------
                            Table 1



TOXIC UNIT CONCENTRATION IN THE EFFLUENT TO.  AND EFFLUENT FROM
THE INTERNATIONAL PAPER COMPANY INDUSTRIAL WASTE TREATMENT PLANT

Sample
AM-1
PM-1
AM-2
PM-2
AM-3
PM-3
AM-4
PM-4
AM-5
PM-5
AM-6
PM-6
AM-7
PM-7
AM-8
PM-8
AM-9
PM-9
AM-10
PM-10
Mean
(S.D.)
GEORGETOWN

Minnow
, SOUTH CAROLINA


Influent Effluent
1.9
2.9
2.4
2.9
2.9
2.4
2.4
1.9
1.9
2.4
2.4
4.3
2.4
2.4
4.3
4.8
2.4
2.9
3.7
>8.3
2.8
0.8
0
0
0
0
0
0
0
-
0
0
0
0
0
0
0
0
0
0
0
0



Daphnid
Influent Effluent
<1.5 0
2.0 0
-
-
1.8 0
1.2 0
-
-
1.3 0
1.4 0
-
-
1.3 0
<1.5 0
-
-
2.4 0
2.4 0
-
-
1.7
0.5
                               -73-

-------
                            Table 2



TOXIC UNIT CONCENTRATION IN THE EFFLUENT TO,  AND  EFFLUENT  FROM

THE UNION CARBIDE
INDUSTRIAL
CHARLESTON, WEST



Minnow
Sample
AM-1
PM-1
AM-2
PM-2
AM-3
PM-3
AM-4
PM-4
AM-5
PM-5
AM-6
PM-6
AM-7
PM-7
AM-8
PM-8
AM-9
PM-9
AM-10
PM-10
Mean
(S.D.)
Influent
6.7
10.5
4.3
5.6
6.7
8.3
4.0
6.7
3.6
3.6
4.5
7.7
50.0
6.3
13.3
8.3
4.5
38.5
3.6
5.6
10.2
11.7
Effluent
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0


WASTE TREATMENT PLANT
VIRGINIA

Daphnid
Influent Effluent
4.8 0
5.9 0
-
-
3.8 0
4.8 0
-
-
5.3 0
7.1 0.6
-
-
55.6 0
7.7 0.6
-
-
2.0 0.6
21.3 0.8
-
-
11.8
15.4
                               -74-

-------
                            Table 3



TOXIC UNIT CONCENTRATION IN THE EFFLUENT TO.  AND  EFFLUENT  FROM

THE DIAMOND SHAMROCK
INDUSTRIAL WASTE
TREATMENT PLANT
GREEN BAYOU, TEXAS



Minnow
Sample
AM-1
PM-1
AM-2
PM-2
AM-3
PM-3
AM-4
PM-4
AM- 5
PM-5
AM-6
PM-6
AM-7
PM-7
AM-8
PM-8
AM-9
PM-9
AM-10
PM-10
Mean
(S.D.)
Influent
>400
794
37
952
19
20
47
>2000
12
19
400
24
30
24
1785
1785
19
24
>2000
588
373
645
Effluent
1.2
1.2
1.2
1.2
1.1
1.1
1.2
1.1
1.2
1.1
1.1
1.1
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
0.05

Daphnid
Influent Effluent
535 3.1
>1700 2.5
-
-
-
-
-
-
26 2.5
42 2.5
-
-
96 2.5
71 3.1
-
-
24 2.5
28 2.5
-
-
133 2.6
199 0.25
                               -75-

-------
removal  of  toxicity even  in  the face of  unusually high excursions.   Also,
these data again suggest that if you had to decide whether to use one species
or the other,  I  don't  really think  it makes much difference.   Certainly from
an economy  of scale,  the Daphnia  would  seem to  be a much  more attractive
organism.

       Now very quickly some other data sets were very similar.  For example,
the treatment  system at  a  pesticide plant  operated by Monsanto in Muscatine,
Iowa  removed  an average  of 96  to  98 percent of  the toxicity of  the waste
going into the system.   And the mean toxic  unit concentration of the effluent
for the  treatment  system for fatheads was  1.5 +_ 0.3, and for daphids  it was
1.3 +_ 0.2 - essentially the same  numbers.

       The treatment  system at Monsanto's  pesticide plant in Luling,  LA
removed  92  percent of  the toxicity,  leaving an  average  of  about  6.0 toxic
units.

       Now,  what can we  do with  all this  information?  I'd  like to take just
a  few minutes to  give you my  thoughts.    Let me preface these  comments  by
emphasizing the fact that the following thoughts  are my own  and should not be
construed as  representing  in  anyway the  conclusions or  philosophy  of  the
agency, nor any of  the  companies  participating in the project.

       It seems to  me  that data on the toxicity  and variability  of influent
to a  treatment system  should  be  considered by the industry  in the same light
as any engineering  data for the design and  operation of  a  treatment system.
Similarly, such data should  be of interest to the agency only  as it relates
to an  assessment  of the  technological  capability to remove  toxicity  - that
is, is it feasible, is  it practical, is it  economical, and so forth.

       It further seems to me  that,  from the  standpoint of regulating
toxicity, it  is the data regarding  the  toxicity of the effluent  from  the
treatment system which  should  be of prime consideration.  With that in mind
I'd like to take a  look at a  summary  of these data (Table 4).  These are the
mean number of toxic units in  the  plant  effluents which  we  evaluated to both
fathead  minnow  and  daphnid.   The important  thing  to note here is  the rela-
tive  standard  deviation of the  mean.   You  can  see  it  runs  as  high  as  26%
in one case  and  as low as 4 percent  in another,  but that's  fairly uncharac-
teristic.  A  value between 10 and  20 percent is a  good  number.   These data
suggest that the toxicity  of  discharges from treatment  systems  does not vary
dramatically when the  system  is  up  and  functioning and  that  approximately 95
percent of the time a bioassay should produce results within 25 to 50 percent
of a  mean value,  assuming a level of reproducibility  in bioassays  of  20
percent.   This observation relates  to the  toxic  unit limitation.   It appears
that any  toxicity  limitation,  or "lethality baseline",  included  in  a  permit
would  of necessity have  to  be incorporated in  the form of  a  "range"  of
acceptable values.    I  don't see how  you  could  pick a single number and say
"thou shalt not ever violate it".  The data further  suggests that perhaps one
could expect slightly  less variability using  daphnids, although  I'm not sure
those differences are  significant.   I personally  believe  that the utility of
toxicity tests lies in  providing data for  establishing  the  permit conditions
                                    -76-

-------
                            Table 4
MEAN (RELATIVE STANDARD DEVIATION. %) TOXIC UNIT CONCENTRATION
IN THE EFFLUENT FROM INDUSTRIAL WASTE TREATMENT SYSTEMS


PLANT
Georgetown, SC
Charlestown, WV
Muscatine, IA
Green Bayou, TX
Kansas City, MO
Luling, LA
Laporie, TX
TYPE
Paper
Organ ic s
Pesticide
Pesticide
Pesticide
Pesticide
Pesticide
MINNOW
0
0
1.5 (20%)
1.2 (4%)
3.8 (26%)
6.0 (23%)
2.4 (12%)
DAPHNID
0
0
1.3 (15%)
2.6 (10%)
4.4 (9%)
3.2 (22%)
-
                            -77-

-------
relative to toxicity. However,  I have  serious  questions  about  the  utility of
such tests  for  monitoring  compliance with those permit  conditions.   Now let
me go on to explain this last statement.

       I think  that bioassays  could  provide  significant information on  the
efficacy of treatment systems for removing toxicity.  Clearly they can provide
information on  what I call  the "residual  acute toxicity"  of  the  wastewater
and on normal variability  in toxicity  of  a  typical discharge remaining  after
treatment, thus providing relevant  information regarding  dilution required in
a stream to meet  water  quality standards.  I think a program  similar to  the
ones we've described here would provide data on the small and random fluctua-
tions in effluent quality  associated with the  regular operation  of an indus-
trial  plant and its effluent control  equipment, thus  defining a range of LC50
values  which might be considered "normal".  I think  if you  couple this
approach with the embryo-larval test  that  Dr.  Mount described  earlier,  one
could even use bioassays to evaluate  the  potential  hazard associated with  the
long term  chronic  effects of  "residual  toxicity"  of  a  typical   discharge
after treatment on receiving water  organisms.

       I  strongly urge  both the  agency and  industry  to  utilize  reliable
toxicity  testing  with  aquatic organisms,  where  appropriate  for  toxicolo-
gically characterizing a waste  stream  for the  purpose of establishing permit
conditions with  regard  to  toxicity,  and the  required degree of  control  of
toxicity.

       However,  I do have some problem with  the concept of requiring industry
to use bioassays  as a monitoring  tool.  An effective monitoring tool should
be one which has  the  capacity to detect  most  or all  excessive pollutant dis-
charges to  the  environment.  Now  clearly bioassays  have the capacity  to  do
this as  demonstrated by some  of the  data  we've just looked at.    However,
the monitoring system should produce  information on three critical  variables:
the frequency,  intensity,  and  duration  of  significantly elevated  pollutant
discharges.  Now although bioassays have  the ability  to detect the  intensity,
they don't tell  us much about the  frequency or the duration.   I  believe that
bioassays  could  provide   such  information  if utilized  frequently  enough.
However, the  costs  of using bioassays with sufficient frequency (real-time,
hourly, or daily) would  clearly be prohibitive, and I think  such  monitoring
could be more effectively  done  using a surrogate physical  or  chemical  para-
meter for which changes can be  correlated with, or indicative of,  changes in
toxicity.

       For example,  there is some data that  suggests  that with certain wastes
there is a  straight  correlation between  the pH of the waste as  it  comes  out
of the treatment  system  and toxicity.   There may be a future  in using  some-
thing like a Microtox system which  can be correlated  to toxicity  as a routine
monitoring tool.  Now the  reason I say this  is because of the frequency with
which  I  think  we'd  have  to do bioassays  to  effectively monitor  effluent
toxicity.  I  base this  observation on a recently published  study  on a  paper
mill  in British Columbia by Nemetz and Drechsler.  This  represents  the most
complete and comprehensive  analysis of sampling design  in effluent  monitoring
which I've ever seen.  In  it they  concluded that  "composite sampling even at
frequent  intervals...performs  the  ironic  function  of  apparently  measuring


                                     -78-

-------
elevated  pollutant  releases without detecting  them."  As  a  result,  they
conclude that  composite  sampling fails  to provide  either  of the  essential
functions   of  an  effective monitoring  system:  (1)  a  high  representational
accuracy of the effluent profile; or  (2)  the  detection of most  or all  pollu-
tant  releases  with  potential  serious  environmental consequences.   They
recommend  that various considerations be  given to  the cessation  of composite
sampling  in  effluent monitoring  programs  and their  replacement  by frequent
grab  sampling  and  sequential  analysis.   How  frequent  is  the .question.   They
found that with  a  paper mill  where they  were measuring  a parameter having a
mean  of X, with  an  N of  120,000,  and a  relative standard deviation  of  50
percent (similar  to  bioassay data)  that  the duration  of 95 percent of the
"spills" was  less  than  4  hours.   As I  interpret their data,  they further
estimated   that  if  one took instantaneous  grab  samples every day,  one  would
still  not  detect 99  out of  every  100 "spills".   Clearly  then,  monitoring
using bioassays  is not  a feasible method  of  "monitoring" toxicity.   I'm not
suggesting that  bioassays  have  no  place  during  the  life of a  permit.   I'm
clearly suggesting that  their  place  is  in  the  establishment of  the  permit
conditions.  But in  addition, if one has  evidence to suggest that, or expect
that, the  quality  of the  discharge  may be significantly different from the
typical discharge, either  due to changes  in  the  inputs to  the  system  or  in
the  operation  of  the system,  it might be  prudent  to  monitor  the toxicity  of
the  discharges using bioassays  at these times.  However, I see little utility
in  arbitrarily  predetermined,  periodic  (e.g., quarterly  or annual) toxicity
testing except to  reassure everyone that  the system  is  working  as it should
on the day and at the time  the sample  was  taken.
                                     -79-

-------
              APPLICABILITY OF THE AMES TEST IN BIOMONITORING
                              Larry Claxton
                U.S.  EPA,  Environmental Toxicology Division
                  Research Triangle Park, North Carolina


       Most of  the  things that you've  heard  so far  deal  with assays that
detect mortality, and/or morbidity in different test organisms.  This  assay
that we'll be describing  detects changes that occur in  hereditary  material.
In other  words,  the  question  that  we'll  be asking  is:  Do the chemicals in
effluents  that  are  tested alone,  when  reaching hereditary material,  cause
changes  in the  hereditary material  that can  be  inherited?  Do  chemical
effluents  have  the  capability of causing  hereditary  diseases  or  if  inter-
actions occur in what are called somatic  cells of the  body,  cells other than
the  gonadal  tissue  cells,  could  this  interaction possibly  cause  cancer?

       The test that we'll be describing today  is  commonly called, and most
people know  of  it as, the Ames test. It's used for  a variety of reasons.
There are  several reasons why people might like to use  it.  It is  rapid and
inexpensive.    It is  a  first  level  indication  of  genotoxicity rather than
mortality or morbidity,  and we can  use it to set priorities  in  further  geno-
toxicity testing.  Something that's  very  important  to  many of you is  that we
can use this test coupled with chemistry,  technology,  and  a  variety of  other
things in order  to direct  the  development of technology and to help  those who
are  in  technology development to  know  what  types  of active  compounds are
coming out of the system.

       There are  two  assumptions  I  think that you  should  be  aware of that
this test  is based  upon.   First,  we are assuming,  and  this  is  generally
accepted  in the  scientific  world  today, that the structure of  DNA, not
chromosomes, but DNA structure,  is the  same for all organisms,  except
for  some  viruses.   Second,   gene  mutations  that  occur  at the DNA level
will  occur generally by the  same molecular mechanisms and therefore be
inherited  in  the same way.   There are  some  debatable  issues  within  that.

       The Ames  test  is  a test run  with  a bacterium called Salmonella typhi-
murium.  Figure  1 is  an outline of how the test is run.  It's a technically
simple test to  perform; however, it gets more complicated when you get into
the laboratory.    Essentially  what we do  is we have a  soft agar overlay that
is melted.   Into this overlay  we  add the bacteria indicator  organism, the
                                    -80-

-------
chemical  or substance or effluent that is to be tested, and we, on occasions,
will also add  in  a  mammalian  activation  system.   We refer to this activation
system as 59 because it  is  a 9,OOOG  supernatant microsome-containing fraction.
This is  used  to  activate  promutagens  and procarcinogens to  an  active  form.
After these are combined together into the agar overlay and overlaid onto the
plate,  they are incubated  for 48 hours.   These are derived bacterial  strains
with a mutation within  the histidine  locus.   They mutate "in reverse"  to the
wild type, which means that they  can grow  in the selective media and when that
reversion  occurs  we get   a response.   We  plate  approximately  10^  bacteria
per plate; therefore, on a control  plate  we'll  find a few spontaneous  mutant
colonies that  will  grow.   If our substance  is mutagenic and  has  no  killing,
we'll  see  increasing numbers  of  colonies with increasing  dosage.   So  there-
fore,  we  tested a variety  of  doses; we usually do  this  in our laboratory in
triplicate at  each  dose.   We  run the proper positive  and negative  controls
and usually, if we have enough material,  replicate the experiment.  So  we get
a fairly large amount of data coming out of this.

       Bruce Ames  in developing  this  system  recognized,  as  most geneticists
do  today,  that there is a variety  of gene mutations that can  occur;  there-
fore,  he  developed  five different  tester strains of bacteria.   Each  tester
strain has a specific type of mutation incorporated and other characteristics
that are  important  to  the people conducting the  test.   But  generally,  for a
complete screening, at  least these  five  strains and sometimes other strains,
if you're in a research  mode, could  be  used.

       There are some major variables  that can occur within a system.   We are
maintaining a  biological organism over time,  frozen, and  yet even with that
there are some natural  variations within  subcultures.   This  makes the  use of
positive and  negative controls, the checks that Ames  describes for  his
strains,  mandatory  for essentially  every test.  There  are also areas of
variation within  the mammalian activation  system,  but the greatest variation
that's brought  into the test  is usually due to  the physical  and  chemical
characteristics of the  substance  that's to be tested.

       One of  the topics  I  was  asked to present is the equipment that's
generally needed.   Table 1 is  a list of this equipment.   A laminar flow
biological  safety hood  is needed  because  we do  use known  carcinogens  and
mutagens  which are  hazardous  to humans  and  the  safety factors  within  the
laboratory are very stringent.   Much  of the  remaining  list  also  involves
safety.   We allow no mouth pipetting,  the area  has to be properly exhausted,
and so  forth.   This can  be  quite  expensive.  These are optional types of
equipment that  you can use which would  increase  the use of your manpower,
and there is other common biological equipment that  might be needed.
                                    -81-

-------
               S-9 MICROSOMES
00
ro
i
         BACTERIA
      CHEMICAL
            SOFT AGAR OVERLAY
        CONTROL
                                   PLATE WITH
                                    OVERLAY
                                       RESULTS:
INCREASING DOSAGE	

(MUTAGENIC WITH NO KILLING)
                                                         INCUBATE
                                                        48 HOURS, 37°C
                Figure 1. Schematic diagram of the plate incorporation (Ames) test.

-------
                                  Table  1

                                 EQUIPMENT
        •    Laminar Flow Biological  Safety  Hood

        t    Automatic Pi pets

        •    Properly Exhausted Incubators and  Work Areas

        •    (Automatic Colony Counter)

        •    (Automatic Plate Pourer)

	•    Other Equipment Common to Microbiological Laboratory	


       Two  things  that  I  forgot  to  add onto  this list  are very important.
Most  people  prepare  their own  mammalian microsomal  activation,  and so
if  you do that, you'll have  to plan  also on having  a rodent colony and
biochemical laboratory backup to check those preparations.

       When  it  gets to  you,  you're  probably not interested  as  much  in the
first  part as in  how we  interpret  the data that's  coming  out  of  the Ames
test.   I'd like  to run  very quickly through some aspects that are generally
accepted with the  Ames test.   A positive response (Table 2)  is usually easy
to  ascertain  because  positive responses are  usually  dramatic in  the vast
majority of cases  that we've dealt with.  A positive dose is defined in the
following  manner:   There's at  least  a  2.5  fold increase in  revertant colony
counts  over the  spontaneous or  control  level.   When you  plot  the  dose re-
sponse  from the  zero  dose  up  to your  highest dose, there's usually  a very
regular type of dose  response  curve.  As  I  mentioned, we always run specific
controls for these particular strains.  We expect  the proper  response  in each
of  those controls.  Another thing  that  we can  do  in the  system  is that after
we  find  some mutants, in order to make sure that we're not getting a false
positive, we can check these mutant colonies to make sure that they truly are
mutants.   So, this is what we would define as a  definite  positive response.

                                  Table 2

                             DATA INTERPRETATION


Positive Response

        t    At  Least 2.5 x Increase

        •    "Regular" Dose Response  Curve

        t    Expected Response for all Controls

	t    Check for True Reversion	
                                      __

-------
       One of  the somewhat  more difficult jobs,  however, is  assigning  a
negative response.  Table 3  gives the  criteria that we  use  in  our  laboratory
for assigning  a  negative  response.   This  may vary a little bit  according to
whom you  talk to, but  this is pretty acceptable  now to most  people.  We
expect, within our laboratory,  at least five doses  at  1/2-log  intervals.  So
we're  exploring  quite  a dose range  when  we do the test.   The  top dose we
expect to be either toxic or greater than  or equal  to  5 mg per  plate.   This
amount is  chosen because  it's  physically  about the largest amount that you
could  normally put into the test.   We also expect  the  proper  responses  from
all controls,  of course.    We  expect  the  doses to  be  replicated  within the
experiment or  we expect replicate  experiments  -  either would  be acceptable.
And we expect the major  tester strains  to be used and  the  proper  steril-
ity controls to  be completed.   If  you  find that all of these  components are
completed  and  you have no  dose  response   -- there's no activity  other  than
spontaneous — you can  he confident in saying that the Ames test  indication
is negative.


                                 Table 3

                             DATA INTERPRETATION
Assigning A Negative Response:

        •    At Least 5 Doses At  Half-Log  Invervals

        t    Top Dose is Toxic  or Fquals or Greater than 5 mg/Plate

        t    Fxpected Responses from  all the Proper Controls

        •    At Least Replicate Dose  Responses and/or Experiments

        •    Tested in 5 Major  Tester Strains

        t	Proper Sterility Controls	
       That  leaves  a  little bit of data that's  left  with  a  questionable re-
 sponse  (Table  4).   Sometimes you'll see an  increased  response  with  dose but
 it  doesn't  ever reach the 2.5 fold  level.   Sometimes  the  response  will
 approach that  point.  Because of the toxicity that tends to mask the response
 after  that  point,  you  can't force the  response to go  past  a  2.5  fold in-
 crease.  So  this  leaves you  with a  questionable  result.  This uncertainty can
 usually be  removed by replicate  experiments.   Sometimes  you will  see at
 certain  doses  that, indeed, you have a count that's  greater than  a  2.5 fold
 increase,  but  the  data is  very irregular.   We  would generally call  this  a
 negative,  qualify,  and explain it.
                                     -84-

-------
                                  Table 4

                            DATA INTERPRETATION
Questionable Responses:

        1.  Increase Response With Dose, But Never Exceeds
            Spontaneous Level by 2.5 Fold

        2.  Irregular Response Curve With One or More Responses
	Greater Than 2.5 x Spontaneous	


       There  are  many  different  ways,  and  Table  5 lists  some  of the  ways,
that  you  may see  the data  summarized.   When you  get  out  the data from  one
experiment  in the  Ames test, you're looking  at  many strains  of  bacteria done
in  triplicate  at  five  doses or  more each.  Then  with a  whole series  of
controls, you're  looking at over 200 pieces  of  data in one test; and we  can
run  three  to  four tests a  week  in one laboratory  with just  two  technicians.
So there's  a  vast  amount of data,  and  people, including us, tend to  summarize
data  in a variety  of  ways.   So there are a variety of ways  to summarize  data;
however, we tend to  be very careful in comparing  data  by using  only  specific
activity or a statistical modeling curve.


                                  Table 5

                            DATA INTERPRETATION
Methods of Data Summary:

        1.  Fold Increase or Mutagenic Index

        2.  Net Response (In Revertants Per Plate)

        3.  Specific Activity

        4.  Specific Activity Fold Increase

        5.  Specific Response Activity

        6.  Maximum Response

	7.  Statistical Modeling Curve	


       For those who  are  involved in water effluents, there  are some associ-
ated techniques (Table  6) that  turn  out  to be very important:  concentration,
extraction, and  fractionation.   There are a  variety of  concentration  proce-
dures, and most of you are probably quite familiar with these.  And then there

                                      -85-

-------
are extraction procedures.  Within water samples from effluent  sites you  also
find  particulates,  and this is where  extraction  can become very  important.
Particulate matter generally interferes with the test.   Particles make  plates
very  hard to  read, to count,  and so  forth.  So generally what we do  is
extract the organics off the particulate matter by different methods and  then
place  the  extract into the test  system.   Fractionation can be important  if
you're going  into technology  development  and want to know what compounds  or
what types of compounds are causing the activity.


                                  Table 6

                           ASSOCIATED  TECHNOLOGIES
1.  Concentration

        A.  Adsorbants (XAD)
        B.  Lyophilization
        C.  Reverse Osmosis
        D.  Filtration (Particulate)
        E.  Extraction

2.  Extraction

        A.  Solvent Systems
        B.  Sonication
        C.  Physiological

3.  Fractionation
       There is a  variety  of  other tests that I'll just review very quickly
that are essentially modifications  of the Ames test - the original Ames test.
One is a simple spot test (Figure 2)  where the bacteria and activation system
are first put on a plate, the  substance  is spotted  in the center, and you get
either a negative or a positive in  which you  had a  zone of killing and a zone
of mutation  and  outside of that another zone of  spontaneous  mutation.   For
many of the things that you'll be  dealing  with, this will not work too well.
Many of the  organics  that  you test will not diffuse  readily in agar, and so
the spot test  has limited applicability  when you  concentrate the organics.

       There's  a  liquid suspension test (Figure  3)  which is important
because it not only  gives  you a revertant number,  but  it also gives you the
survival number.   For a geneticist  this can turn out to be  very important
because he's looking  at  the number of hits on a particular  gene in a parti-
cular organism.   Very seldom, unless you  are in  a research mode,  would you
want to use  this  test.   There are  a couple of occasions, however, if you're
working with some specific types of compounds such as nitrosamines, when you
may want to go to a test such  as  this.
                                     -86-

-------
           S-9 MICROSOMES
                 ft	BACTERIA
                                                    PIPET WITH CHEMICAL

                                                    DISC
        SOFT AGAR OVERLAY
                                   PLATE WITH OVERLAY
00
-J
I
        INCUBATE 48 HOURS, 37°C
                          RESULTS:
                                                    ZONE OF KILLING


                                                    ZONE OF MUTATION


                                                    ZONE OF SPONTANEOUS

                                                            MUTATION
           CONTROL
NEGATIVE
POSITIVE
                        Figure 2. Schematic diagram of the spot test.

-------
00
CO
                 BACTERIA
                    SOLVENT
            CONTROL
                                 S-9
                          CONTROL
                          INCUBATE
I
                MUTANT PLATES
                               CHEMICAL IN SOLVENT
                          TEST

1
                                                              SERIAL DILUTION
H      SOFT AGAR       H


 I   OVERLAY TO PLATE   I

^3              i          I
          INCUBATE
                                     CONTROL
                        SURVIVAL PLATES
                       Figure 3.  Schematic diagram of the liquid suspension test.

-------
       Something  that's  very similar but more rapid than the  suspension  test
is the preincubation test  (Figure  4)  which is just a  shortened  modification
of the suspension test.   It is also  good  for testing compounds such  as  the
nitrosamines.

       A  modification that we've developed to the  spot test is the  well  test
(Figure 5) in which we can test two chemicals - a solvent control and a posi-
tive control  -  with  and  without  activation on one plate.

       I've run through  quite a bit very rapidly - it's really about a 2 hour
lecture what I've tried to  do  in about  20  minutes  here.   I  brought  along two
articles  that  you  might be  interested  in.   One  of them describes  the  Ames
test in more detail  - some  of  the  pitfalls and some of the uses of the Ames
test.  The other is a  much  more  general  and lay-like publication  that  des-
cribes short  term  tests for  carcinogens,  mutagens, and other genotoxic
agents.
                               DISCUSSION
Question:    I  am  a  plant  engineer of a  medium sized  steel  plant  on the shore
            of Lake  Erie at  Lorraine.   I  am impressed by  the enthusiasm
            and great mental capacity devoted to the  goal  here of preserving
            the long term existence of  aquatic organisms.   From  where I sit,
            I am also interested  in  preserving the existence  of  an organism.
            The organism  I  think  of is  my plant.  It  employs about  7,000  to
            8,000 people;  about  30  to 40  percent   of  the members of the
            community depend on its employment.   And  gentlemen, that organism
            finds itself in  a heavily  polluted environment.    What this
            pollution is  and  who  the polluters are of this  environment that
            makes  it so difficult for these  organisms  to survive,  I leave  to
            everybody's  imagination here.   I can take some  inflation,  taxa-
            tion,  and  here  comes  the 30-day regulation.   I  was one  of  the
            unfortunate  cosignatories of  a  $500,000  order a  few months ago,
            and that money  is going to give  us  nothing  but a book  or maybe
            two books  that  went  to  a consulting firm to  tell   us  just what
            this plant, this  medium sized steel  plant,  would have  to  do  to
            comply with BAT.   Now we've been through  BPT,  we've  been through
            NPDES,  we've  been through  air  problems,  and  now we  are facing
            BAT.   I  want  to be sure that now that you've  invited  me to this
            conference,  when  I  get  back I ge.t the right message to my plant
            management  and  to  the corporate  staff that  we  work  with, and  it
            poses  a  little  question  to  me.   In a few  months  we  are going  to
            get out  a few nice little books which will say this is what we'll
            offer  to the  regulating agencies as  our  BAT approach.   Now,  my
            question to you gentlemen is, in developing your emerging science
            of the effects of toxicity on aquatic organisms, are  you going to
            give us  your  virginity  - the industry your  virginity  - when RAT
            time  and negotiations  and  decisions come around to  land  on
            something that meets the long term requirements economically.   My
                                     -89-

-------
    BACTERIA
ID
O
                S-9
                 i
CHEMICAL
                            ^INCUBATE
                               INCUBATE
           1
                               RESULTS:
f
SOFTAGAR
                                                     OVERLAY
                                                  PLATE WITH OVERLAY
                  CONTROL
       INCREASING DOSAGE
                     Figure 4. Schematic diagram of the preincubation test.

-------
                S-9 MICROSOMES
      SOLVENT
      CONTROL
   CHEMICAL
      A
>£>
         POSITIVE
         CONTROL
            NEGATIVE
            CONTROL
           CHEMICAL A-
CHEMICAL   POSITIVE
    B      WITH
           ACTIVATION
               PLATE WITH WELLS
J
                                             BACTERIA IN
                                             SOFT AGAR
        PLATE WITH WELLS AND OVERLAY
                                         RESULTS:
POSITIVE
CONTROL
                                                        CHEMICAL B:
                                                        POSITIVE
                                                        WITHOUT
                                                        ACTIVATION
                                             t
                           Figure 5.  Schematic diagram of the well  test.

-------
            second question is this:   Water is  an  essential  medium that
            supports  life.   We  started  by chemical  analysis,  we've gone  to
            toxicity  of aquatic organisms.  Now there  are organisms that live
            in air.   Next  year  are we  going to  be  analyzing  the  effect  of
            toxicity  of air on the  variety of organisms  such as  butterflies,
            mosquitos, sparrows, etc.?   Maybe you gentlemen are missing a  big
            bet.    I  would  appreciate  a reply to both  of these  questions.

Answer:      I'm  not  sure  I  picked  up  the question.   I heard  the comment.

Question:    You know  what we are talking  about  as far as  BAT is  concerned.
            Are you going to spring an  air problem on us  just as  you sprung
            the water problem on us?

Answer:     The  main reason  we  had  the session now and  sponsored  a session
            about  two years  ago called  "Industry Takes Initiative" was  to
            encourage industry to  take a look at the  processes  and to avoid
            as much  as  possible end  of the line treatment  by looking  at
            process  modification  and  changing the  way things  are made.

            You're saying you know  what  may come  out  in  BAT or what may come
            out  of guidelines in Washington.   But we're  trying  to say that,
            although  there  may  be  some  numbers  that  come  out of  the guide-
            lines  in Washington, the national numbers will not  cover every-
            thing. The specific problems at each facility must be  considered
            individually.   For  instance, at a steel  plant, the  general type
            of toxicants  is known  - such  as  polynuclear  hydrocarbons from a
            coke  operation.  In  fact,  Congress and  our agency  are already
            changing  the  requirements in many industries  where  they have,  you
            might  say,  backed off  in the BAT area  and have gone to Best
            Conventional  Treatment.   And I agree with them because they  are
            eliminating  the need  to  install expensive pollution control
            equipment that  really has little effect for the cost.

                 The  purpose of  this  session  is to get you, the  industry, to
            start  considering the  toxicity testing described here  today.   If
            you  look at  processes  in your own  plant  and  current  treatment
            technology, then  combine that with some toxicity testing, you  can
            determine and  install  the treatment technology that will elimi-
            nate  some as  yet unregulated toxicants in  addition to  "standard"
            toxicants and put yourself ahead of the game.  But  you  must first
            evaluate  your individual  process  to  predict  future requirements.
            So I  think that's an answer to the  first  question  as far as that
            goes.   And as far as the second question goes, I don't  know about
            measuring butterflies, but we are in fact  significantly concerned
            about  air emissions  to the  extent that some  get into  the water.
                                   don't think  that there would  be that much
                                   for  instance, municipal  sludge treatment
                                   beautiful  example  of delaying  the  intro-
Now in your facility I
of a  problem,  but say
and incineration  is  a
            duction of a  toxicant to the receiving waters.  By incinerating a
            toxicant removed  at a treatment plant, you may just volatilize it
                                    -92-

-------
            so  it  can fall  back  into the water.   It's estimated,  I  think,
            that 80 to  90 percent of the PCBs going  into  Lake  Frie  and Lake
            Michigan are  from  the air not from the water.   You  need  to look
            at what you're doing  with your waste,  and if you're incinerating
            something, are you  really  destroying  it or are  you  only  volati-
            lizing  it?   And  so you may need to be concerned about  air emis-
            sions.   And certainly the RCRA requirements  I think fit  very
            strongly into this.

Question:   Excuse  me.   I  want  to go back to  that.   He says that  we should
            look into  it but back  with  Mr.  Macek  -  he didn't  say  how fre-
            quently we  had to test.   Like  with your  wastewater system,  how
            often  do  we have to  run a bioassay?  Daily,  weekly, monthly,
            bimonthly, are you going to set up  some  sort of system?

Answer:     There have  been several  people  here today  who have said  you
            cannot  set  a time and  say  if you sample  it  every  day or every
            week or every  month  that you will  thereby  resolve  your  problem.
            You're going to  have  to  know, you're  going to  have  to learn what
            your system is and learn how frequently you have to test in order
            to get the answer.  In some systems it may turn out that there is
            just no change over  any reasonable amount of  time,  so  you could
            sample on a  very infrequent  basis.  But  there  are  other  systems
            that change on an hourly basis,  so  there is really no way you can
            set up a sampling program whereby you  sample infrequently  and not
            miss certain  things  that will  happen.   You  would   have to  have
            almost a continuous test in some cases  or find some other way to
            measure to tie it down.

Answer:     Just to further  clarify. It would be  very easy to  come  out  and
            say that every industry will run a 24-hour test.  Every industry
            will run  a  96-hour test.   And, this  regulatory-wise,  would  be
            much easier.   Maybe  that's what you  all  wanted to hear  us  say
            today.   But  we're  trying to  say,  which is I think  a little more
            difficult  for  both  of us, that  we and  the  states   want to  work
            with you  and for you to go ahead on  your own  to  look at your
            specific facility.   You may have  only one product  line,  but  we
            have plants where there are several  thousand  products  in  one
            plant.   We  would much  rather work with you to try to gear  the
            requirements or  gear  the testing  to  your  specific  operation,  or
            have the states v/ork  with you if you're going  ahead on  your own,
            or have your consultant work  with you.   We're not trying to avoid
            setting numbers;  it's just that there's  so much variability from
            one type  of  industry  to another  that  I think  it  would  be  unfair
            to you for us  to  propose  to  the  region an absolute  set  of guide-
            lines.   And  so I  hope that we're  not  sounding  like  we're  putting
            you off by failing to  give you specific numbers without  going to
            a  particular class.

Answer:     What  Ken's saying is we're taking  the  hard way.   It is  much more
            difficult  from our standpoint to work to try  to  sit down  and
                                    -93-

-------
figure out which tests  should  be run on  what  frequency,  at what
facility  and how  to  use the  results  than it is  to say  every
discharger in the  Region  shall  have  to run one 96-hour  test  or
one 24-hour test on  10  percent effluent or  100  percent  effluent
and if he  gets  this  score he is thereby  failed or  passed.   That
is an easy way - an easy approach which does not yield an answer.
And we just don't need any more  nonanswers.
                         -94-

-------
          RAPID ASSESSMENT METHODS  (FISH COUGH RESPONSE,  FTC.)


                             Robert  Drummond
               U.S. EPA, Environmental Research Laboratory
                            Duluth,  Minnesota


       Rapid assessment methods are  not a new  thing as people  have been
researching this area  for several years.   For example, there  is  an EPA
publication called  Short-Term Tests for Health  and Ecological  Effects.
A number  of tests are  outlined in this publication such as  fresh water
algal  assay bottle tests;  acute fish toxicity  tests;  subchronic embryo-
larval  tests;  chronic  fish toxicity  tests  with  flag fish;  fish   avoidance
tests; acute invertebrate toxicity  tests; subchronic invertebrate tox-
icity  tests; acute  plant  toxicity  tests; and  acute  static  and acute  flow-
through tests  with marine fish.   Some  of the  endpoints of  these  tests,
their  strengths,  weaknesses,  status of  development,  application,  and some
idea of the  cost of the test and people to contact are also discussed.

       Other recent  publications  are concerned primarily  with fish  as a tool
for biomonitoring effluents.   Some of these  papers are:  A  Computerized
System for Monitoring Fish Activity   (Virginia   Polytechnic   Institute,  Vir-
ginia) ;   A Continued Biological Monitoring System to Predict Chronic  Effects
of Toxicants (Virginia Polytechnic Institute.  Virginia); Fish  Locomotor'
Behavior Patterns  as a Monitoring Tool  (Morgan,  South  Africa);  Fish Activity
Monitors  (Water Research Center.  England); Ventilatory Patterns  of  Fish  Using
a Microcomputer Monitoring System (U.S.  Army MedicalCenter,Maryland);  An
Automated Biological Monitoring Facility for Rapid Assessment  of  Industrial
Effluents(VirginiaPolytechnic  Institute,Virginia);Bluegill Respiratory
Activity~and Prediction  of Chronic Toxicity  (Proctor and   Gamble, Cincinnati,
Ohio); and   Biomonitoring with  Fish:An Aid to  Industrial  Effluent and
Surface Water Quality Control (Morgan,  South  Africa).Our particular  labora-
tory(U.S.EPA,Environmental  Research Laboratory-Duluth), has  been involved
in studies that are summarized in publications titled Procedures  for Measuring
Cough Rates  of  Fish  and  The Fish Cough Response:  A Method for Evaluating the
Quality of Treated Complex Effluents.

       In  order for behavioral  responses of  fish to be  useful  for biomoni-
toring effluents  the  test  procedure must be  inexpensive,  simple  enough to
be carried out  by non-technical  personnel  in  a  short  period  of  time,  and the
endpoint  being  measured  must have some relevance to  the  animal's  ability to
                                     -95-

-------
live  and function  properly.   Of the various  behavioral  endpoints  being
considered, five appear to  be  closely allied  to the animal's well-being and
meet  these  other criteria.  They are:   1) reproductive  habits,  2) feeding
patterns/habits, 3) fear responses, 4) the  righting  reflex,  and 5) respira-
tion.  In a further breakdown,  respiratory tests, in my opinion, are the most
applicable and  advantageous as an  endpoint  for monitoring  effluents  for a
variety of reasons.

       The term  respiration as used  here  is  an overall  term  denoting gill
purge activity (cough  rates), ventilation rate/amplitude, oxygen consumption,
or any combination of the above.  Advantages of using fish respiration as an
endpoint become clear when  one  reviews the  abundance of literature that has
been published showing that numerous toxicants/effluents affect fish respira-
tion rates.   Fish act as  an integrator of toxicity - synergistic or antago-
nistic  effects  of  an effluent  will  be reflected in terms of changes in
respiration rates.   These  changes  occur  within minutes or  hours  at lethal
concentrations.   At sublethal  concentrations  it may take 1  to  3  days  for a
change to  appear.   Thus,  the  response appears quickly.   Another advantage
is that  the  response  is graded to  the concentration.  The  intensity of the
response can be used as a  predictor of adverse effects.  Lastly, the test is
adaptable to both flow-through  and recirculating bioassays. The recirculating
bioassay approach has  a  big advantage in that  hundreds  of effluent samples
(5 to 8  gallons) can  be shipped to a central  point  for  testing.   The small
volume of  effluent  required and  speed  at  which respiratory changes appear
make this an  attractive approach  for  screening  out those effluents that are
of concern and need  further testing.

       We are now at a point where  it is no longer necessary to observe fish
directly to tabulate changes in  respiratory rates.  These data can be collec-
ted, tabulated,  and  analyzed automatically using computers or other automatic
data  processing  devices.    The  automated  approach allows more data  to  be
collected and processed  accurately at  an affordable cost.
                                    -96-

-------
            TEST ORGANISM ACQUISITION AND CULTURING  IN THE LAB


                              Charles Steiner
                        Central  Regional  Laboratory
                            U.S.  EPA, Region V
                             Chicago, Illinois


       Good afternoon.   The purpose of  this discussion  is  to cover some of
the  major considerations for  culturing  or acquiring  organisms.    For this
discussion, I'll use the working definitions that culturing organisms is the
rearing  of these  test organisms  for the  entire 1ifecycle, that  is,  the
entire reproductive cycle  within the  lab, while acquiring organisms is
receiving them from an outside source and holding within the  lab for a dura-
tion of time.   The purpose of both  techniques is  to  provide test organisms in
which  as  many  variables  as  possible are controlled  with the ideal situation
being that the only variable between the  organisms is the test variable.  The
achievable goal, on  the  other hand,  is  to  provide  test  organisms which are
healthy,   relatively free of  pollutants,  of  known age,  and which are physio-
logically representative of the  species.  Two of the  general points to
consider  in culturing  and holding  organisms are  that  the holding or rearing
facilities should be of  a  size  which is adequate for the number and species
of organism being utilized and the  water  supply to the culture unit should be
of high  quality  and  be delivered through inert  piping.  This  is  not to say
that one  must  have a  pristine water  source for  a culture  unit.   There are
successful units now operating which  use deep  well  water,  river water, lake
water, and even chlorinated  tap water.

       Many factors produce  stress  in organisms in the lab.  Stress-producing
factors should be avoided.  The organisms should be shielded from any excess
noise.   The loading capacity  of the facility should  not be  exceeded,  and
proper diet  should be maintained  at all  times.   Also,  one  must  consider
that  for some  organisms special  environmental  requirements  are  necessary
such as flowing water or  fluctuating water levels.

       As was  mentioned  earlier  this morning, the  recommended construction
materials for  toxicity testing  are teflon,  glass, and  number 316 stainless
steel.   Fortunately, holding facilities  are not  so  limited  in the choice of
materials.   For many  years, concrete  raceways  and troughs  were  used  for
holding  fish  within  the  lab.   These facilities have the  disadvantage that
they are  permanent and do not allow flexibility within  the  lab.   Recently,
however,  fiberglass tanks are more  in line  with  what  is needed because they
are movable, durable,  and  versatile.  And  additionally, these  units can be
                                     -97-

-------
rigged  to  be  self-cleaning,  which is  a real  plus  for  anyone  that has  to
work with the animals.

       Table 1  is  a list of  the recommended test species and test  tempera-
tures.  It includes also the salt water organisms, which I am not addressing.
I would like to say that there are occasions when an organism would be used
for a  test  that is  not  included in this list.   In  the case of  an  effluent
toxicity test  sometimes  it  may be desirable to use  a sensitive  species that
is indigenous to the receiving water and/or  is commercially or recreationally
important to the receiving strain.

       Let us  now move to the  test  organisms themselves.   The test  organisms
can be broken  into  two groups:   the fish and the  invertebrates.   Fish may be
obtained from three different sources.   They may either  be reared in the lab,
obtained from  a hatchery, or  they may  be collected from the wild population.
                                 Table 1


	RECOMMENDED SPECIES AND TEST TEMPERATURES	
Species                                               Test Temperature (°C)a

Freshwater

   Vertebrates

      Coho salmon, Oncorhynchus kisutch                        12
      Rainbow trout, Sal mo gai?dneri                           12
      Brook trout, Salvelinus fontinalis                       12

      Goldfish, Carassius auratus                              22
      Fathead minnow, Pimephales promelas                      22
      Channel catfish, Ictalurus punctatus                     22
      Bluegill, Lepomis macrocHTrus                            22

   Invertebrates9

      Daphnids, Daphm'a magna or D^_ pulex                      17
      Amphipods, Gammarus lacustris, G. fasciatus,             17
                  or G^ pseudolimnaeus                          17
      Crayfish, Orconectes sp., Cambarus sp., Procambarus      22
                sp., or Pacifastacus leniusculus               22
      Stoneflies, Pteronarcys sp.                              12
      Mayflies, Baetis sp. or Ephemerella sp. Hexagenia        17
                limbata or H^ bilineata                        22
      Midges, Chironomus sp.                                   22
                                      -98-

-------
                                 Table 1


	RECOMMENDED SPECIES AND TEST TEMPERATURES  (Continued)^
Species                                               Test Temperature  (°C)a

Marine and Estuarine

   Vertebrates

      Sheepshead minnow, Cyprinodon variegatus                 22
      Mummichog, Fundulus heteroclitus                         22
      Longnose killifish, Fundulus sTmilis                     22
      Silverside, Menidia sp.                                  22
      Threespine stickleback, Casterosteus aculeatus           22
      Pinfish, Lagodon rhomboides                              22
      Spot, Leiostomus xanthurus                               22
      Shiner perch, Cymatogaster aggregata                     12
      Pacific staghorn sculpin. Leptocottus armatus            12
      Sanddab, Citharichthys stigmaeus                         12
      Flounder. Paralichthys dentatus, P. lethostigma          22
      English sole. Parophrys vetulus                          12

Marine and estuarine

   Invertebrates9

      Shrimp, Penaeus setiferus, P. duorarum, or               22
              P. aztecus
      Grass shrimp, Palaemonetes sp.                           22
      Shrimp, Crangon sp.                                      22
      Oceanic shrimp, Pandalus jordani                         12
      Blue crab, Callinectes sapidus                           22
      Dungeness crab, Cancer magister                          12
      Mysid shrimp, Mysidopsis sp., Neomysis sp.               22
      Atlantic oyster, Crassostrea virginica                   22
      Pacific oyster, Crassostrea  gigas	20	


aFreshwater  amphipods,  daphnids,  and midge  larvae and  shrimp should  be
  cultured  and tested at  the recommended  test  temperature.   Other  inverte-
  brates  should  be  held  and tested within 5°C of the temperature of the water
  from  which  they were obtained.   If  the recommended test  temperature is not
  within  this  range, they should  be tested at the temperature from the series
  7,  12,  17,  22,  and 27°C that is closest to the recommended test  temperature
  and is  within the allowed  range.

  Material  obtained  from:   Methods for Measuring the Acute Toxicity of Efflu-
  ents to Aquatic Organisms.U.S.EnvironmentalProtectionAgency,1978.
                                     -99-

-------
       I'd like to address the collection  from  the wild  population  first.   As
was mentioned this morning,  I  would  put collection from the wild  population
as the last resort in the case of fish,  for the reasons  that the  life  history
is not  known, the fish  population may be diseased, the  population  may  be
hybridized, or it may  already be contaminated  with pollutants.  Fish  reared
in the laboratory, on the other  hand, present  a real  source and  a  continuous
source of fish for most  laboratories.  As  I  stated  earlier, the  water  supply
should be  of high quality,  but  is not limited  to  a source such  as  a deep
well.  An example is our own  laboratory;  our only source of water  is  chlori-
nated tap water  supply.   What we did was  take the  hot  and cold water taps,
put on a mixing  valve to regulate  the temperature and send the water  through
an activated  carbon  filter.   The  carbon   filter is  capable  of removing  the
total residual chlorine down to less  than  five  parts per billion  and the unit
itself has a  capacity  of 5,000 gallons  a  day,  which is a lot more than most
of you would ever anticipate using.

       The advantages  to maintaining a culture are many.  As  I mentioned
earlier,  it  provides a  continuous supply  of fish  of all  stages.   This  in-
cludes eggs and allows for testing  of the  more  sensitive life stages,  a topic
which was covered earlier.  The  life history of the fish from  a  culture unit
is documented, and the  fish remain relatively stress-free.  But I think  the
most  important  aspect  of having a rearing unit is that the  fish are main-
tained in a  disease-free state.   As  some  of you well know, there  is  nothing
more frustrating than starting a bioassay  and  having your control  fish start
dying off because of a  disease problem.   However,  along with the  advantages
of the culture unit, there are  some disadvantages.  I feel  that these  are
minor and probably are far outweighed by the advantages.  The first disadvan-
tage is that  there is a  moderate initial  cost  for  setting  up a culture unit.
It would be hard for  me to describe the  entire  unit  we've got at  the lab,  but
before we moved  to our new facility, we  set up  an  entire  culture unit in  a
room for about $700.   And the second  disadvantage of the culture  unit  is that
there is  a need  for continuous  manpower  supply to maintain the  organisms.

       The  second  source of  fish  I  mentioned was  hatchery fish.   I would
suggest that  if you start looking for hatchery  fish, one thing  to consider is
that the hatchery selected should be  one which  promotes  the inbreeding of  the
parent stock to maintain the stock, rather than to  replenish the  parent stock
from an outside  source.   Therefore I believe  a hatchery should  be picked in
which the hatchery fish are several generations removed  from the  wild  popula-
tion to assure a better supply of test organisms.

       There  are  some  advantages to  receiving  fish from a hatchery.   One,  a
lot  of  you may  be  limited in space, and a rearing  unit  does take up some
space.  But  by the same token,  a  holding  unit will also  take  up  some floor
space.  Usually one can receive more  species of fish from a hatchery than  one
can rear in the  lab.   The second advantage over a  culture  unit  is  that there
is less time  spent  in maintaining the stock.  You  only need someone  to work
on the fish  when you have them  in the lab.  There  are  several  disadvantages
to receiving  hatchery fish.   The  desired size or  age  of  fish that you  may
want may not  be available at the time you  need  to do your testing.  Transport
from the hatchery  to the lab causes  stress on the  organisms,  and  also there
                                     -100-

-------
has to be  some  arrangement  made for transport of the fish.  Sometimes long-
term  planning  is required  to  allow the  hatchery to  know the  species  and
the number of fish that will be needed. But  I think the most important point
here  is that  fish received from an outside  source have  the  potential  to be
diseased.   Therefore,  hatchery-reared fish must first be quarantined at least
10 days, preferably 21, before used as test organisms.  Additionally, it may
be advisable to  give  the  fish  a prophylactic disease treatment upon arrival
in the lab.  At  this  point  I'd  like to comment that on the disease treatment
in hatchery  fish, rather  the  disease prevention  of fish received  from an
outside source.   Table 2  includes  the recommended  prophylactic  and thera-
peutic disease treatments, and Figure 1 is a copy of the procedure we use at
Central  Regional Lab.  And I'd  like to emphasize that  any fish  that we
receive from an  outside  source are given a  prophylactic  disease treatment.

       One thing I  tried  to do  was  find some  sources of  hatchery  fish.
Commonly,  we go  to  federal and  state hatcheries.   Unfortunately,  these
facilities are not available to everyone in the audience.  And so, in digging
around, I did find some  information.   One possible contact for getting test
fish   would  be the American Fish Farmer  Federation.   Their administrative
offices are  in   Lonok, Arkansas,  and  they  can  put  you  in touch  with bait
dealers in your  area.  Another source  may be to  obtain  a copy of the Annual
Buyer's Guide published  by the  magazine  Commercial Fish Farmer  where they
have  listed all  their  members under  the type of services provided.

       The maintenance of  any  fish  for testing  should  follow  the strictest
quality control   practices  and  a  log  should  be  kept  on all  the  aspects of
rearing or maintaining these test  organisms.    I had thought  about talking
about the  actual  maintenance practices,  but  I  don't think that  time  would
allow me to cover all  the  points.   I  took Table  3 from  a publication by the
Sport  Fish  and   Wildlife  Service  that showed  the organic contamination of
fish  food.   This is  something that should be  considered  in  your background
information.
                                    -101-

-------
                                 Table 2
                 RECOMMENDED PROPHYLACTIC AND THERAPEUTIC
                     TREATMENTS FOR FRESHWATER FISH*
Disease
Chemical
Concentration
   (mg/1)
Duration ofc
 Treatment
External
Bacteria
Oxytetracycline hydrochloride
(water soluble)
Procaine Penicillin G in
25& 30-60 min
(3ml /I 00 gal) 48-72 hrs
               Dihydrostreptomycin sulfate
               solution (Franklin Lab,  Denver,  CO)

               Benzalkonium chloride
               (HYAMINE 1622R)
                                      l_2b
                 30-60 min


Monogenetic
trematodes
fungi, and
external
protozoa^




Parasitic
copepjods
Nitrofurazone (water mix)
Neomycin sulfate
Formalin plus zinc-free
malachite green oxalate

Formalin

Potassium permanganate
Sodium chloride

DEXONR (35% Active Ingredient)
Trichlorfon
(MASOTENR)
3-5b
25
25
0.1

150-250

2-6
15000-30000
2000-4000
20
0.25b

30-60 min
30-60 min
1-2 hrs


30-60 min

30-60 min
5-10 min dip
(e)
30-60 min
Continuous^

aThis table  indicates  the  order of preference  of  treatments that have been
 reported to be  effective,  but  their  efficacy against diseases and toxicity
 to  fish  may be  altered by  temperature  or water  quality.   Caution:  test
 treatments on small  lots of  fish before making  large-scale applications.
 Fish should not be treated  the  first  day  they are in  the facility.

 Material  obtained  from:   Methods  for Measuring the Acute Toxicity of Efflu-
 ents to Aquatic Organisms.U.S.EnvironmentalProtectionAgency,1978.
                                    -102-

-------
C.  DISEASE PREVENTION TREATMENT.

    All  fish  shall  be  given preventive  treatment  for disease or  parasites
    upon receipt from the  hatchery.   Treat the fish in the holding  tank  in
    the  following day-by-day sequence  and do not stop aeration at  any time.

Day 1:  Two to three days before the fish arrive, fill the holding  tank with
        nonchlorinated  water to the  5-gallon  level  and mark the level  per-
        manently with waterproof marker.   Add 45 more  gallons  of nonchlori-
        nated  water and mark the  50-gallon level.   Fill  three storage con-
        tainers with  nonchlorinated  water  and bring  to  room  temperature.
        Aerate the water continuously.   (Caution:   Be sure to dechlorinate
        with a carbon filter if  water contains chlorine).

Day 3:  Fish arrival.  Without opening, float the plastic shipping bag (water
        and fish) in the  tank  and acclimate to  tank  temperature  for 1-2
        hours  before releasing the fish to tank water.

Day 4:  (a) Add 5 grams  ACS grade  potassium  permanganate  (KMn04) to 450
            milliliters distilled  or  deionized  water.   When  dissolved dilute
            up to 500 milliliters.   This is a  1%  solution  of KMn04.   This
            solution is  used on days 4 and  5,  and  any remaining  solution  is
            then discarded.

       (b)   Add 190  mill il Hers  of the  KMn04 solution  to the holding
            tank containing 50 gallons of water.   (Caution:   Stir while
            adding  the  MNn04 to achieve a  uniform solution.)   Hold the fish
            in the KMn04  solution for 1/2  hour.  Do not exceed the 1/2-hour
            limit,  but  continue  the treatment to that  time  even  if the fish
            begin  to surface or  otherwise seem stressed.

       (c)   After 25 minutes of  treatment, begin to drain the holding
            tank to  the  5-gallon  mark.   At 30 minutes add water   from the
            storage  container to bring the tank volume to 50  gallons.

       (d)   Add to  the  holding tank  1-2  milliliters of  0.1  N^ sodium
            thiosulfate  (HazS&z'SHz®)  solution, or  until  the water  turns
            yellow-brown.    (To  prepare  0.1 N_  sodium thiosulfate  solution
            dissolve 25  grams   sodium  thiosulfate   in  volumetric  flask and
            dilute  to 1  liter   with  distilled  or deionized  water.)   Color
            change  is slow,  so  do not  be concerned  if the water   does not
            become  yellow-brown  immediately.   Refill  storage  container with
            non-chlorinated  water.     Be sure water is at  room temperature
            before  it is used on day 5.


    Figure 1.   Disease-prevention procedure used at Central  Regional
                Laboratory,  U.S. EPA Region V, Chicago.
                                    -103-

-------
Day 5:  Repeat steps (b),  (c),  and (d) from  day 4.  It is  not necessary
        to  refill the storage containers on this day.

Day 6:  (a)  Add 4,750  milligrams  of tetracycline (the material  from  nine-
            teen 250-milligram  capsules) to  500 milliliters  of warm tap
            water.   Shake  until  dissolved.   Pour this solution into the
            holding  tank  and  mix  gently.   Leave  the  fish in  this  solution.
            Fish are fed for the  first  time on this day.

       (b)   Fill  storage containers  with  effluent-receiving  water for
            acclimation (Section  II,  D).

Day 7:  Feed fish and clean bottom of tank with siphon.  Aerate water.	
Material  obtained  from:   ORSANCO 24-Hour Bioassay.   Ohio River Water Sanita
tion Commission, 1974.

     Figure 1.  Disease-prevention procedure  used at Central Regional
               Laboratory, U.S.  EPA Region V, Chicago (continued).
                                    -104-

-------
                             Table 3
 SELECTED  CONTAMINANTS  DETECTED  IN  DIETS ANALYZED  INTERMITTENTLY
             AT  FPRL  FROM AUGUST, 1972 - AUGUST, 1973
Residues (ug/g)


Diets
Oregon Moist
Glenco
Clark
Silver Cup
(2 samples)
EWOS
(3 samples)
h- »
S Colorado State Diets
1 (9 samples)
BSFW Hatchery Diets
(2 samples)
Purina Catfish Chow


DDTa PCBb
0.06 0.30
0.11 0.30
0.11 0.20
0.08- 0.20-
0.17 0.32
0.15- 0.20-
0.39 0.30
0.11- 0.10-
0.84 2.80
0.12- 0.20-
0.19 0.30
-
Hexchl oro-
benzene
(HCB)
_d
-
-
mm
0.06
0.008-
0.046
_
0.003
—

-

Total
Organo-

chlorine Phthalate0
Dieldrin Endrin
-
-
-
• *•
0.01
0.01-
0.02 0.01
0.01-
0.30 0.01
.m .
0.01
0.01
Content
0.36
0.41
0.31
0.37
0.555
0.361-
0.766
0.213-
3.953
0.32-
0.50
0.01
esters
-
-
-
_
-
•V
3.0
^
-
_
-
-
Reference Research Diet - <0.1 - - - <0.1
(4 samples)
Minimum Detection
Limits
a All DDT analogs.
b All polychlorinated
0 Di-n-butyl phthalate
d None detected.
Table Obtained From:


0.005 0.1

biphenyls, but usual
and di-2-ethylhexyl



0.0001



0.01 0.01






0.1-
0.5

ly AroclorR 1254 and 1260.
phthalate.

Acquisition and Culture of Research


Fish: Rainbow Trout,


Fathead Minnows,


Channel
Catfish and BluegiTTs.U.S.  Bureau of Sport Fisheries and Wildlife,  1975.

-------
       The last topic  that  I was going  to  talk on was  invertebrates.  And I
saved this  for last  because there's not  much, information  I  can  give you.
With the exception of Daphnia, invertebrates are difficult to maintain in the
laboratory for  any extended  length  of  time  because  they have  very precise
environmental   requirements.   For  the most  part,  individuals  seeking  these
invertebrates  are  required  to develop their  own  sources -  that  means  going
out and scouring  the  streams  in  your area.   The reason for this is that, for
the most part, invertebrates do not ship well, and it's hard to maintain good
quality up until  the  test  time.   The exception in the invertebrate field are
the Daphnia.   Daphnia can  be obtained from many sources.   If  one is looking
to  start  his  own  Daphnia  culture,  small numbers  of animals can  be obtained
from  several  biological  supply  houses.   In  most  cases it's best  to let the
people know that  you  are looking for a  particular species  and  what they are
to be used for so  that you can guarantee proper identification.  On the other
hand, if one  is  looking  for large  numbers  of Daphnia,  I know of at least one
supply facility  in Minneapolis,  Minnesota,  Aquatic Life,  that  shipped  units
of  Daphnia  between  12,000  and  15,000   in  number  for  around  $50, including
the air freight.

       In summary,  I  would  like  to  say  that  this  has  been a general discus-
sion  on  rearing  and holding,  and  I  would  reemphasize  that  good  quality
control  practices  are needed  to maintain  the  test  organisms.    If  this  is
done, it makes the whole  job of running bioassay tests  a lot  simpler.
                                    -106-

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 REPORT NO.

        PPA
                                                            3. RECIPIENT'S ACCESSION NO.
TITLE AND SUBTITLE
       Proceedings of  the Seminar on Biological
       Monitoring and  its use in the NPDES  Permit
       Program
                                                            5. REPORT DATE
                                                               Mav 1983  ("Issuing Date!
                                                            6. PERFORMING ORGANIZATION CODE
 AUTHOR(S)
        Timothy O'Mara
        Orville E.  Macomber, Editors
                                                            8. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING ORGANIZATION NAME AND ADDRESS
        Center  for  Environmental Research  Information
        Office  of Research and Development
        U.S. Environmental Protection Agency
        Cincinnati,  OH   45268
                                                            10. PROGRAM ELEMENT NO.
                                                           11.c6NtRACf,rGRANf NO.


                                                               Inhouse
 2. SPONSORING AGENCY NAME AND ADDRESS
        Center for  Environmental Research  Information
        Office of Research and Development
        U.S. Environmental Protection Agency
        Cincinnati,  OH   45268
                                                           13. TYPE OF REPORT AND PERIOD COVERED
                                                               10/2/79	
                                                           14. SPONSORING AGENCY CODE


                                                               EPA/600/00
 5. SUPPLEMENTARY NOTES
                           seminar was conducted in cooperation with  the Enforcement and
        Surveillance  $  Analysis Divisions US  EPA Region V, Chicago,  IL.
        Contact:  Orville Macomber  f513l 684-7394 - _-_
 6. A8STRAC
             This  seminar brought together  representatives from industry,  state and
        local governments and biological monitoring consulting firms  to present the
        U.S. Environmental Protection Agency,  Region V perspective  on biomonitoring
        requirements  for the National Pollutant Discharge Elimination System (NPDES)
        permit activities.

             A limited number of industries have the potential for  the discharge of
        toxicants  and they will be required to conduct special testing, monitoring
        and evaluation of their effluents by  utilizing biological testing techniques.
        "This seminar  was held to clarify the  methods and uses of biomonitoring and
        its application to setting limits in  NPDES permits.  Presentations were made
        concerning biological monitoring phytoplanktons, zooplanktons, macroinver-
        tebrates,  fish, bacteria, etc.  The range of tests include  static and flow-
        through bioassays, including tests  for bioaccumulation, Ames  tests for
        mutagenicity, and some rapid assessment methods such as the fish cough
        response test.
17.
                                 KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lOENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
18. DISTRIBUTION STATEMENT

        Release  to  Public
                                              19. SECURITY CLASS (ThisReport/
                                                Unclassified
                                                                           21. NO. OF PAGES
111
                                              20. SECURITY CLASS (Thispage)
                                                Unclassified
                                                                          22. PRICE
EPA Perm 2220-1 (R«v. 4-77)   PREVIOUS EDITION is OBSOLET
                                             E107
                                                                  US GOVERNMENT PRINTING OFFICE 1980-657-146/5673

-------