STUDY
                          1 OF FECflL
  U. S. ENVIRONMENTAL PROTECTION AGENCY
ARCTIC ENVIRONMENTAL RESEARCH LABORATORY
          COLLEGE, ALASKA 99701

-------
              PRELIMINARY STUDY

COMPARATIVE WINTER SURVIVAL OF FECAL BACTERIA

            IN A SUBARCTIC RIVER


                     by


             Dale J.  Van  Donsel

              Ronald  C. Gordon

           Charlotte  V. Davenport
            WORKING PAPER NO.  28
    U.  S.  Environmental  Protection  Agency
   Arctic  Environmental  Research  Laboratory
               College,  Alaska
           Associate  Laboratory  of
   National  Environmental  Research  Center
              Corvallis, Oregon
     Office of Research and  Development
                  May  1974

-------
A Working Paper presents  results of investigations which are, to some
extent, limited or incomplete.  Therefore, conclusions or recommendations,
expressed or implied,  are tentative.  Mention of commercial products or
services does not constitute endorsement.

-------
                                  m

                                ABSTRACT

     A preliminary investigation of Salmonella survival  in a subarctic
Alaskan river was conducted during March 1973, when there was total  ice
cover and 0C water temperature.  Since most of the domestic pollution
entered the river from one source and there were no additional  sources
downstream, it was possible to examine survival  without  interruption for
seven days flow time.   Six Salmonella serotypes  were isolated,  and
salmonellae were still recoverable both quantitatively and qualitatively
after seven days flow time in the presence of 330 total  coliforms/100 ml
and 60 fecal coliforms/100 ml.

-------
                           TABLE OF CONTENTS
                                                                   PAGE
INTRODUCTION
MATERIALS AND METHODS                                                4

RESULTS AND DISCUSSION                                               12

CONCLUSIONS                                                          19

REFERENCES                                                           20

-------
                            LIST OF FIGURES

NUMBER                                                              PAGE
           Map of the lower Tanana River showing the location
           of sample stations                                         5

           Gasoline powered ice auger being used to cut through
           ice at each sample station                                 6

           Samples for quantitative analysis of salmonellae
           and fecal indicator bacteria being dipped from ice
           hole                                                       7

           Setting up tripod to support line with attached
           qualitative Salmonella sampler and station marker
           flag                                                       9

           Method of retrieving line with attached qualitative
           Salmonella sampler                                        10

-------
                                 VI
                          LIST OF TABLES


NUMBER                                                            PAGE
           Salmonellae/Liter and  Fecal  Indicator Bacteria/
           100 ml  of Sample Obtained  at Each Sample Station
           During  the March 21, 1973.  Sample Run                     13

           Salmonellae/Liter and  Fecal  Indicator Bacteria/
           100 ml  of Sample Obtained  at Each Sample Station
           During  the March 27, 1973  Sample Run                     14

           Salmonella Serotypes Isolated at Each Sample
           Station from Samples Used  for Quantitative  and
           Qualitative Analysis                                     16

           Most Frequently Isolated Salmonella Serotypes
           in the  United States During  1972 (4)                ,     17

-------
                              INTRODUCTION

     Decreasing the suspending medium temperature has shown to increase
survival time of enteric bacteria and viruses (1, 9, 11, 16, 17, 19, 20).
Survival time appears to increase continuously as the temperature decreases,
so the longest survival would be expected when the temperature is at or near
0C.  However, most of the survival studies described in these reports were
conducted in the laboratory using pure cultures, rather than directly in
rivers receiving the enteric microorganisms discharged in effluent from
waste treatment systems.  Thus, information to fill  the practical need for
survival time parameters, as stated by Berg et al.  (3) in 1966* still re-
mains to be obtained.  It has also been stated that coliforms must be quan-
titatively related to the presence of enteric pathogenic bacteria if coliform
enumeration is to continue as a useful criterion of health risk from polluted
waters (17).
     Increased enteric microogranism survival at low suspending medium
temperatures becomes more significant when viewed in relation to current
treated wastewater disinfection practice.  Recent studies (12, 13) have
demonstrated that chlorine disinfection at low effluent temperatures may
not effectively remove the fecal indicator bacteria before the effluent
is released into the receiving water.  Therefore, disinfection under these
circumstances may not provide an adequate barrier to the spread of disease
caused by enteric pathogenic bacteria.  There is also very little chance
that enteric viruses are being removed in the disinfection process.
     In an effort to obtain low temperature survival time data, a survival
study was conducted in a subarctic Alaskan river (11).  This study demon-
strated that a significant number of total coliforms, fecal coliforms
and enterococci remained viable after seven days flow time when the water

-------
 temperature was essentially 0C under total ice cover.  The fecal  coliform
 results were  compared with temperate climate winter data  (2), and  the  percent
 survival  was  found  to be approximately five times greater at 0C than  at the
-warmer temperatures.  That study did not include salmonellae, so quantita-
 tive  survival  relationships are not available.
      In 1911,  Ruediger  (20) demonstrated that salmonellae survived in  much
 greater numbers for a much longer period in an ice covered river than  during
 warm  weather.  He used  pure culture suspensions of the bacteria in celloidin
 sacs  which he placed in the river and sampled at various time intervals.
 More  recently, Gallagher et al. (9) found salmonellae to be extremely  per-
 sistent in the Red  River under ice cover.  They isolated two Salmonella
 serotypes 73  miles  and  four days travel time downstream from the outfall
 source.   In addition, salmonellae were consistently isolated from  river
 water when there were 1000 or more fecal col iforms/100 ml (9), and were
 occasionally  isolated when the fecal coliforms ranged from 4 to 20/100 ml
 (5, 6, 8, 9,  21).
      Approximately  36 percent  of all surface waters examined throughout  the
 world have been found to contain enteric viruses (1).  Even though enteric
 viruses have  been implicated as the etiological agent in numerous  diseases,
 infectious hepatitis virus is  the only one for which definite epidemiological
 evidence  has  been obtained showing water as the mode of transmission (17).
 However,  the  extended survival of viruses in water makes this a potential
 route for viral disease transmission  (1).
      The  need for survival time data and quantitative relationships in
 arctic and subarctic rivers, where the water temperature  is 0C for about
 six months each year, cannot be overemphasized since low temperatures
 enhance enteric microorganism  survival causing an  increased potential

-------
health hazard downstream.  As a result, this preliminary investigation was
conducted to determine whether salmonellae could be isolated at the same
stations where fecal indicator bacteria had previously been isolated and
to establish, if possible, a quantitative relationship under low tempera-
ture conditions.

-------
                          MATERIALS AND  METHODS

Selection of River to be  Studied
     The reach of the Tanana  River studied  is  shown  in  Figure 1.   The descrip-
tion of the river and reasons for selecting this  particular river for study
have been described previously (11).

Sample Station Selection  and  Sampling  Schedule

     Nine sample stations were selected  (Figure 1).   T-100 through T-800
are essentially in the same locations  as previously  described (11).  Station
T-900 was selected to obtain  background  data upstream from all  known sources
of domestic pollution.
     Two sampling trips were  conducted during  this preliminary study.  All
stations were visited on  March 21, 1973, for the  purpose of collecting
samples for quantitative  fecal indicator bacteria and Salmonella enumera-
tion, and placing long term qualitative  Salmonella samplers (Moore swabs).
A second trip was made on March 27, 1973, to obtain  anot&er set of samples
for quantitative analysis, and to retrieve  the swabs.

Sampling Techniques

     A gasoline powered ice auger was  used  to  cut through the 1-1/2 to 4
feet of ice encountered at each sample station (Figure 2).  Ice chips were
skimmed from the water surface, and the  samples for  quantitative analysis
collected by dipping from the water surface (Figure  3).  Samples for fecal
indicator bacteria enumeration were collected  in  sterile, wide mouthed,
screw capped, 1-liter, polypropylene  containers.  These containers were
transported in ice chests to  minimize  freezing problems.

-------
                                                                                                                       '!"*
    8*.
64*10
            TANANA RIVER
              LOWIR IECTIOH
                                                                                                                                 tn
    Figure  1.   Map of the lower  Tanana River showing  the location of  sample stations

-------
                                                                                                               .-.
Figure 2.   Gasoline powered ice auger being used to cut through ice at each sample station

-------
Figure 3.   Samples  for  quantitative  analysis of salmonellae and fecal indicator bacteria being dipped
           from  ice hole

-------
                                   8

     For Salmonella quantitation, samples were collected in  5-gallon  Cubi-
tal ners.  There appeared to be  little or no ice formation  in  these  large
volume samples.  Qualitative Salmonella samples were obtained with  the  gauze
swabs as described by  Spion (22).  After the samples for quantitative analysis
were obtained, the line with the gauze swabs attached was  lowered through  the
hole in the ice, and attached to a tripod for support while  the  hole  was
freezing closed (Figure 4).  The tripod also supported a flag which marked
the sample station.  In order to retrieve the gauze swabs  after  a week  in  the
water, it was necessary to cut  another hole in the ice just  downstream  from
the first hole.  A steel rod with a hook on the end was used  to  catch the
line and pull it up through the new hole (Figure 5).  The  guaze  swabs were
placed in 18-ounce Whirl-Pak bags, and transported to the  laboratory  in ice
chests to minimize freezing.

Fecal Indicator Bacteria Enumeration
     Total coliforms,  fecal coliforms and fecal streptococci  were enumera-
ted by the membrane filter method (13).  All sample volumes were filtered
in triplicate, and volumes up to 200 ml were used in an effort to obtain
numbers of colonies on each filter in the statistically valid range.

Quantitative and Qualitative Salmonella Enumeration

     The presence of salmonellae was quantitatively determined by the most
probable number (MPN)  procedure.  The diatomaceous earth method  of  Hammar-
strom and Ljutov (14)  was used  to concentrate the water samples.  Sterile  dia-
tomaceous earth (0.5 gm/1) was  added to the water.  For convenience the membrane
filters were removed from sterile, disposable, membrane filter holders  (Fal-
con Plastics, number 7103) and  replaced with filter pads of  the  type  accom-
panying membrane filters. A slurry of 0.5 gm diatomaceous earth was

-------


                                      ,

Figure 4.   Setting  up  tripod  to  support  line with  attached qualitative Salmonella sampler and station
           marker flag

-------
Figure 5.   Method of retrieving  line with  attached  qualitative Salmonella sampler

-------
                                    11
spread on the filter pad, and 12 liters of sample was vacuum filtered.
To remove substances that caused erratic salmonellae isolation, the diato-
maceous earth plug on the pad was washed five times with sterile buffered
distilled water and three times with Dulcitol Selenite Enrichment Medium
(DSE) (18).  The concentrate plug was then removed from the filter pad and
homogenized in a small amount of DSE.  Additional DSE was added to bring the
total concentrate volume to 120 ml (100:1 concentration of the original
sample).  The concentrate was then thoroughly mixed, and one 50-ml, five
10-ml, five 1-rnl and five 0.1-ml portions were immediately planted into
the MPN series in DSE and incubated at 41.5C.  These portions were equi-
valent to 5,000, 1,000, 100 and 10 ml volumes of the original sample.
After 24 and 48 hours of incubation, material from the MPN tubes was
streaked on Hektoen Enteric Agar (HE) and Xylose Lysine Desoxycholate
Agar (XLD), and incubated for 24 hours at 35C.
     For the qualitative determination of salmonellae, each gauze swab
was divided into three equal portions.  One portion was placed in a flask
of DSE and two in Tetrathionate Broth.  Incubation was at 41.5C and streaks
were made on HE and XLD at 24 and 48 hours with incubation at 35C.
     Suspect Salmonella colonies on HE and XLD were picked to Triple Sugar
Iron Agar (TSIA) slants.  Colonies producing reactions indicative of sal-
monellae in TSIA tubes were further screened for urease, indoTe and lysine
decarboxylase production before being identified by the serological methods
of Edwards and Ewing (7).  Additional biochemical tests were performed only
if warranted by questionable screening or serological results.

-------
                                   12
                           RESULTS  AND  DISCUSSION

     The objectives of this limited preliminary investigation  were:  [1] to
determine whether salmonellae could be  isolated as many days flow time
downstream from any source of domestic  pollution as previously isolated
fecal indicator bacteria  (11), and; [2]  to establish whether or not  a quan-
titative relationship exists between salmonellae and fecal  indicator bacteria.
Because this quantitative relationship  was to  be based on  the  number of
bacteria/unit volume of river water, discharge measurement were not  necessary
at any station.
     The results presented in Tables 1  and 2 show that the numbers of fecal
indicator bacteria decreased progressively downstream from the T-700 to the
T-100 station in a manner comparable with earlier results  (11).  The T-700
station (Figure 1) is located downstream from  the last source  of domestic
pollution while T-900 is  upstream from  any source.  The fecal  coliform
numbers were lower than expected from the previous study.   However,  the
m-Fecal Coliform Broth Base (BBL lot #204625)  may have been a  contributing
factor.  This medium gave a final reaction pH  of 7.2, which is within the
accepted 7.2-7.6 pH range.  On further  checking, it was found  that the broth
base had a pH of about 6.4 before rosalic acid addition and heating.  This
is 0.5 to 0.7 pH units lower than obtained with other broth base .lots from
the same manufacturer.  Fecal coliform  recovery from unchlorinated primary
sewage effluent was examined using  the  above mentioned broth base lot, and
recovery was compared directly with two other  lots from the same manufacturer.
The results indicated that the broth base used during this study recovered
only 48-60 percent as many fecal coliforms as  the other two lots.  This
suggested that the fecal  coliform numbers reported in Tables 1 and 2 may
be considerably lower than the actual number/unit volume  present in  the
river.

-------
                                            13
                                         TABLE 1
                 Salmonellae/Liter and Fecal Indicator Bacteria/100 ml  of
                          Sample Obtained at Each Sample Station
                           During the March 21, 1973 Sample Run
Salmonellae
Station
No.
T-900
T-800
T-700
T-600
T-500
T-400
T-300
T-200
T-100
Total
Col i forms
(No./lOO ml)
2300
9600
1700
920
920
370
410
320
Fecal
Col i forms
(No./lOO ml)*
130t
1700
89t
160
110
64
36
34
Fecal
Streptococci
(No./lOO ml)
<0.5
43
130
9t
15t
15t
9t
lit
13t
No. /liter
MPN
0.1#
0.1
<0.1#
0.1
0.1
0.1
0.8
1.7
<0.1#
95% Confidence
Limits
No. /liter

<0.05 0.4
<0.05 0.4
<0.05 0.4
<0.05 0.4
0.2 1.9
0.5 4.7

* These numbers are probably 48-60 percent of the actual  number; see text for explanation.

 No colonies on any filter when triplicate filters were  examined,

# No salmonellae isolated from any water volume examined.

t Average of less than 20 colonies per filter when triplicate filters were examined.

-------
                                               14
                                           TABLE 2

                    Salmonellae/Liter  and Fecal Indicator Bacteria/100 ml of
                             Sample Obtained at Each Sample Station
                              During the March 27, 1973 Sample Run
SaTmonellae
Station
No.
T-900
T-800
T-700
T-600
T-500
T-400
T-300
T-200
T-100
Total
Col i forms
(No./lOO ml)
	 **
3600t
6600
1600
2300
960
470
630
330
Fecal
Col i forms
(No./lOO ml)*
__**
480
170t
64t
260
120
43
63t
60t
Fecal
Streptococci
(No./lOO ml)
_**
73
50
26
26
12
10
21
6t
No. /liter
MPN
	 **
<0.1#
0.3
0.1
0.1
0.3
0.1
0.1
0.3
95% Confidence
Limits
No. /liter


<0.05 0.8
<0.05 0.4
<0.05 0.4
<0.05 0.8
<0.05 0.4
<0.05 0.4
<0.05 0.8
 * These numbers are probably 48-60  percent of the actual number; see  text  for  explanation.

** No samples obtained from this  station because of open water.

 # No salmonellae isolated from any  water volume examined.

 t Average of less than 20 colonies  per filter when triplicate filters were examined.

-------
                                   15
     Salmonellae were quantitatively measurable in samples from the T-800
through T-100 stations during one or both sampling trips (Tables 1  and 2).
Since these were the same stations at which fecal  indicator bacteria were
also found, it is apparent that low indicator bacteria numbers do not pre-
clude quantitative measurement of salmonellae.  The most probable number
(MPN) of salmonellae present at the various sample stations suggests that
the numbers of these bacteria do not decrease with flow time.  However, the
95 percent confidence limits of the MPN indicate that the salmonellae may
be decreasing in a manner similar to the indicator bacteria.  A much more
detailed study must be conducted before relative survival rates and quan-
titative relationships between salmonellae and fecal  indicator bacteria
can be established.
     When Salmonella isolates from both quantitative and qualitative samples
were examined, six serotypes were found (Table 3).  Four serotypes were
found in the quantitative samples and five were obtained from the swabs.
S. thompson was the only serotype isolated at every station from T-700 to
T-100 and was the most frequent isolate from quantitative samples.   Thus.
SN, thompson was probably present in much larger numbers than the other
salmonellae.  Insufficient data are available to ascertain the numerical
relationship of the other isolates, or to determine if survival character-
istics vary among the serotypes.
     The 10 most frequently isolated Salmonella serotypes from human and
nonhuman sources during 1972 are shown in Table 4.  Three of the serotypes
found in the Tanana River were among those most frequently isolated from
human sources and three from nonhuman sources.  S. typhimurium and S. derby
appear under both headings, while S. thompson was ranked among those most
frequently isolated from human sources and S. senftenberg from nonhuman
sources.

-------
                                  16
                              TABLE  3
       Salmonella Serotypes  Isolated  at  Each  Sample  Station From


Station
Number
T-Qnn
T-800
1 / UU
T-600

1 OUU


T-200
T-100


Samples Used for Quantitative and Qualitative
Quantitative Samples
March 21, 1973 March 27, -1973

derby None

thompson group .B
non-motile

LI lUIHJJo Ul 1 Li lUli i|p oU * 1
fhnmnc nn "t" hnmncnn

thompson thompson
None thompson
senftenberg


Analysis

Qualitative
Samples
	 **
derby
	 	 *

thompson
senftenberg
typh.imurium
derby
	 *

	 *
	 *
thompson
senftenberg
thompson
paratyphi B
typhimurium

** No samples obtained from this  station  because of open water.
 * Swab not recovered.

-------
                                17
                             TABLE  4

Rank
1
2
3
4
5
6
7
8
9
10
in the United States During

Human
typhimurium*
newport
enteritidis
infantis
hei del berg
saint-paul
thompson*
derby*
oranienburg
javiana
1972 C4)
Source
Nonhuman
typhimurium*
oranienburg
senftenberg*
saint-paul
newport
anatum
montevideo
eimsbuettel
derby*
hei del berg
*Isolated from Tanana River.

-------
                                   18
     The results presented in  Table  3  indicate  that  the  method  used to re-
cover the gauze swabs was not  entirely satisfactory  since  the swabs were
not recovered from four stations.  The problem  was apparently caused by the
insulating snow layer having been  removed  from  the ice around the hole through
which the line and attached swabs  were lowered.   During  the ensuing week, a
cone of ice developed below the  lower  edge of the ice cover and encased the
line on which the swabs were attached.   The cone  extended  deeper than could
be reached with the hook so it was impossible to  retrieve  the swabs.
Efforts are currently in progress  to develop a  means of  overcoming this
serious problem.
     Circumstances did not permit  the  examination of enteric viruses during
this preliminary investigation.   In  view of the available  information about
extended virus survival at low water temperatures (1), isolation and identi-
fication of enteric viruses should constitute a major effort during any
detailed enteric microorganism survival  study.  Field methods for concen-
tration of viruses from large  volumes  of water  are available (15), and one
or more of these should be applicable  under low temperature conditions,
making a virus survival study  feasible.

-------
                                   19
                               CONCLUSIONS

1.   Salmonella serotypes survive for at least seven days flow time down-
stream from the last source of domestic pollution when the water temperature
is 0C and there is total ice cover.
2.   After seven days flow time, a variety of Salmonella serotypes are still
recoverable both quantitatively and qualitatively when total and fecal coli-
forms are present in low numbers.

3.   The results from this preliminary study were too few to determine the
extent to which fecal indicator bacteria numbers would indicate Salmonella
density.  The results do suggest that a survival relationship may exist.
However, a more detailed study is necessary to assess the quantitative
relationship.

4.   Enteric virus survival should also be examined under the same low
temperature conditions.

-------
                                   20
                              REFERENCES


 1.   Akin,  W.  E.,  W.  H.  Benton, and W.  F.  Hill.   1971.   Enteric Viruses in
     Ground and Surface  Waters:  A Review  of Their  Occurrence and Survival.
     p.  59-70.   .In. V.  Snoeyink and V. Griffin  (ed.),  Proceedings of the 13th
     Water  Quality Conference, Virus and Water Quality:   Occurrence and
     Control.

 2.   Ballentine, R.  K. and  F. W. Kittrell.  1968.   Observation of Fecal
     Coliforms in  Several Recent Stream Pollution Studies.   In Proceedings
     of the Symposium on Fecal Coliform Bacteria  in Water and Wastewater.
     Burear of Sanitary  Engineering, California State Department of Public
     Health.

 3.   Berg,  G.,  P.  V.  Scarpino, and D. Berman.  1966.   Survival of Bacteria
     and Viruses in  Natural Waters.  Federal Water  Pollution Control Admin-
     istration, Cincinnati.

 4.   Center for Disease  Control.  1973.  Salmonella Surveillance Annual
     Summary.   Department of Health, Education and  Welfare,  Atlanta.

 5.   Dutka, B.  J.   1973. Coliforms Are an Inadequate Index  of Water Quality.
     J.  Environmental  Health.  36_:39-46.

 6.   Dutka, B.  J.  and J. B. Bell.  1973.   Isolation of Salmonellae from
     Moderately Polluted Waters.  J. Water Pollution  Control Federation.
     45_:316-324.

 7.   Edwards,  P. R.  and  W.  H. Ewing.  1972.  Identification  of Enterobac-
     teriaceae, 3rd Ed.   Burgess Publishing Company,  Minneapolis.

 8.   Fair,  J.  F. and S.  M.  Morrison.  1967.  Recovery of Bacterial Patho-
     gens from High Quality Surface Water.  J. Water  Resources Research.
     3_: 799-803.

 9.   Gallagher, T. P., J. E. Hagan, N.  A.  Thomas, and D. F.  Spino.  1965.
     Report on Pollution of Interstate  Waters of  the  Red River of the
     North  (Minnesota-North Dakota).  Department  of Health,  Education and
     Welfare,  Cincinnati.

10.   Geldreich, E. E.  and N. A. Clarke.  1971.  The Coliform Test:  A Cri-
     terion for the Viral Safety of Water,  p. 103-109.   In_ V. Snoeyink
     and V. Griffin (ed.).  Proceedings  of  the 13th  Water Quality Conference,
     Virus  and Water Quality:  Occurrence  and Control.

11.   Gordon,  R. C.  1972.   Winter Survival  of Fecal Indicator Bacteria in a
     Subarctic Alaskan River, Pub. No.  EPA-R2-72-01.   Environmental Pro-
     tection  Agency, College, Alaska.

12.   Gordon,  R. C. and C. V. Davenport.  1973.  Batch Disinfection of
     Treated  Wastewater  With Chlorine at Less Than  1C,  Pub. No. EPA-660/2-
     73-005.   Environmental Protection  Agency, College,  Alaska

-------
                                              21


           13.   Gordon,  R.  C.,  C.  V.  Davenport,  and B.  H.  Reid.   1973.   Chlorine
                Disinfection of Treated Wastewater in a Baffled  Contact Chamber at
                <1C,  Working Paper No. 21.   Environmental  Protection Agency, College,
                Alaska.

           14.   Hammarstrom, E. and V.  Ljutov.   1954.  Concentration Technique for
                Demonstrating Small Amounts  of  Bacteria in  Tap Water.  J. Acta Patho-
                logica et Microbiologica Scandinavica.   35_:365-369.

           15.   Hill,  W. F., E. W. Akin, and W.  F. Hill, Jr.   1971.   Detection of
                Viruses  in  Water:   A Review  of  Methods  and  Application, p. 17-42.
                _In_ V.  Snoeyink  and V. Griffin (ed.), Proceedings of the 13th Water
                Quality Conference, Virus and Water Quality:   Occurrence and Control.

           16.   McFeters, G. A. and D.  G. Stuart.   1972.  Survival  of Coliform Bacteria
                in Natural  Waters:  Field and Laboratory Studies with Membrane-Filter
                Chambers.  J. Applied Microbiology.  24_:805-811.
                                                  s
           17.   Mechalas, B. J., K. K.  Hekimian, L. A.  Schinazi, and R. H. Dudley.
                1972.   Water Quality Criteria Data Book, Vol. 4, An Investigation
                into Recreational  Water Quality, Pub. No.  18040 DAZ 04/72.  Environ-
                mental Protection  Agency, Washington, D.C.

           18.   Raj, J.   1966.   Enrichment Medium for Selection  of Salmonella from
                Fish Homogenate.  J.  Applied Microbiology.  14:12-20.

           19.   Rudolfs, W., L. L. Falk, and R.  A. Ragotzkie.  1950.  Literature
                Review on the Occurrence and Survival of Enteric, Pathogenic, and
                Relative Organisms in Soil,  Water, Sewage,  and Sludges, and on Vege-
                tation.   J. Sewage and  Industrial  Wastes.   22_: 1261-1281.

           20.   Ruediger, G. F.  1911.   Studies  on the  Self-Purification of Streams.
                J. American Public Health Association.   1_:411-417.

           21.   Smith, R. J., R. M. Twedt, and  L.  K. Flanigan.  1973.  Relationships
                of Indicator and Pathogenic  Bacteria in Stream Waters.   J. Water
                Pollution Control  Federation.  45_: 1736-1745.

           22.   Spino, D. F.  1966.  Elevated-Temperature Technique for the Isolation
                of Salmonella From Streams.   J.  Applied Microbiology.  14:591-596.

           23.   Van Donsel, D- J.  and E. E.  Geldreich.   1971.  Relationships of
                Salmonellae to  Fecal  Coliforms  in Bottom Sediments.   J. Water Research.
                5:1079-1087.
* U. S. GOVERNMENT PRINTING OFFICE: 1974-697-595 /54  REGION 10

-------