I t H" H ‘ W“ x H M I W ’ H HI H *l I l W l. THE FRESH WATER {st352. AS A mammm; MGNWQE a? mmm CQificfét-{WATEQM :2»; A LGTIC EEWERONMEW Thesis for {'56: Dogma 3f M. 5. MICHIGAN STATE UHEVER’SETY James W. Bedfoa'd 2.967 JHESW LIBRARY Michigan State University 4“ _. ._ -fiHH . __._... ~<——.—— ABSTRACT THE FRESH WATER MUSSEL AS A BIOLOGICAL MONITOR OF PESTICIDE CONCENTRATIONS IN A LOTIC ENVIRONMENT by James W. Bedford Fresh water mussels were introduced into the Red Cedar River at six different locations and analyzed for pesticide content following different lengths of time in the river. DDT and its metabolites, TDE and DDE, were found in all mus- sels analyzed. The concentration of DDT and its metabolites in the introduced mussels increased significantly in a down- stream direction and increased significantly with time before leveling off. Methoxychlor was found in mussels introduced into the lower sections of the study area. Aldrin was found in all mussels on two dates of retrieval from the river but was not found before or after these dates. Mussels collected from the upper portion of the study area and analyzed for pesticide content contained small con— centrations of DDT and its metabolites but there was no sig- nificant difference between species in pesticide content. THE FRESH WATER MUSSEL AS A BIOLOGICAL MONITOR OF PESTICIDE CONCENTRATIONS IN A LOTIC ENVIRONMENT BY James W. Bedford A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1967 L /// '/ / .,-/I \/ ,l -— ACKNOWLEDGEMENTS I wish to thank Dr. E. W. Roelofs for his invaluable guidance during the course of the study and in the prepa- ration of the manuscript. I also wish to thank Drs. T. W. Porter and M. J. Zabik for their suggestions and criticisms during the course of the project. I wish to thank my wife Kathel for her constant encouragement and for typing the manuscript. During this study I was supported by Federal Water Pollution Control Administration Training Grants ITI-WP- 109—01 and 5TI-WP—109-02. ii INTRODUCTION . . STUDY AREA . . . METHODS . . . . TABLE OF RESULTS AND DISCUSSION . . . SUMMARY . . . . LITERATURE CITED CONTENTS iii Page 15 51 55 LIST OF TABLES TABLE Page 1. Concentrations of pesticides in eight mussels, (L. siliquoidea) collected from the Cass River, Tuscola Co., Michigan and the 99% confidence limits around the totals(Y) of DDT and its metabolites, TDE and DDE. . . . . . . . . . . . 14 2. Concentrations of DDT and its metabolites in L. siliquoidea removed on 50 June 1966 from those placed in the Red Cedar River on 16 June 1966. . . . . . . . . . . . . . . . . . . . . . 15 5. Concentrations of DDT and its metabolites in g, siliquoidea removed on 28 July 1966 from those placed in the Red Cedar River on 16 June 1966. . . . . . . . . . . . . . . . . . . . . . 19 4. Concentrations of DDT and its metabolites in .L- siliquoidea removed on 25 August 1966 from those placed in the Red Cedar River on 16 June 1966. . . . . . . . . . . . . . . . . . . . . . 20 5. a) Results of an analysis of variance for the significance of observed differences in the total concentration of DDT and its metabolites in L, siliquoidea with respect to length of time and location in the Red Cedar River. . . . 22 b) Results of new multiple range tests for the significant differences in the total concentra— tion of DDT and its metabolites in_§. sili— guoidea with respect to length of time and location in the Red Cedar River . . . . . . . . 22 6. Concentrations of methoxychlor in L. sili- guoidea removed from the Red Cedar River at three different times from those placed in the river on 16 June 1966 . . . . . . . . . . . . . 24 iv LIST OF TABLES — Continued TABLE 7. 10. 11. 12. 15. a) Results of an analysis of variance for the significance of observed differences in the concentration of methoxychlor in L. siliquoidea with respect to length of time and location in the Red Cedar River . . . . . . . . . . . . . . b) Results of new multiple range test for the significant difference in the concentration of methoxychlor in L. siliquoidea with respect to length of time in the Red Cedar River . . . . . Concentrations of pesticides in six mussels, 'A. grandis, collected from the Red Cedar River, above Station I, and the 99% confidence limits around the totals(Y) of DDT and its metabolites, TDE and DDE . C . . . C C O O C O . O . . O . . Concentrations of DDT and its metabolites in ‘A. grandis removed on 8 August 1966 from those placed in the Red Cedar River on 25 July 1966 and in remains of mussels that died during this period. . . . . . . . . . . . . . . . . . . . . Concentrations of DDT and its metabolites in 'A. grandis removed on 5 September 1966 from those placed in the Red Cedar River on 25 July 1966. . . . . . . . . . . . . . . . . . . . . . Concentrations of DDT and its metabolites in .A. grandis removed on 5 October 1966 from those placed in the Red Cedar River on 25 July 1966 . a) Results of an analysis of variance for the significance of observed differences in the total concentration of DDT and its metabolites in A, grandis with re5pect to length of time and location in the Red Cedar River . . . . . . b) Results of new multiple range test for the significant difference in the total concentrak tion of DDT and its metabolites in A, grandis with respect to length of time in the Red Cedar River . . . . . . . . . . . . . . . . . . . . . Concentrations of methoxychlor in A. grandis removed from Station III at three different times from those placed there on 25 July 1966 . V Page 25 25 27 28 29 51 52 52 35 LIST OF TABLES - Continued TABLE 14. 15. 16. 17. 18. 19. Results of an analysis of variance for the sig- nificance of observed differences in the con- centration of methoxychlor in A. grandis with respect to length of time in the Red Cedar River . . . . . . . . . . . . . . . . . . . . . Concentrations of aldrin in L. siliquoidea re— moved on 28 July and A. grandis removed on 8 August from those placed in the Red Cedar River on 16 June 1966 and 25 July 1966, respectively. a) Results of an analysis of variance for the significance of observed differences in the concentration of aldrin in L. siliquoidea with respect to location in the Red Cedar River. . . b) Results of new multiple range test for the significant difference in the concentration of aldrin in L. siliquoidea with respect to loca- tion in the Red Cedar River . . . . . . . . . . a) Results of an analysis of variance for the significance of observed differences in the con- centration of aldrin in A. grandis with respect to location in the Red Cedar River. . . . . . . b) Results of new multiple range test for the significant difference in the concentration of aldrin in A. grandis with respect to location in the Red Cedar River. . . . . . . . . . . . . Concentrations of pesticides in mussels col- lected from the Red Cedar River at and above Station I during the Summer of 1966 . . . . . . a) Results of an analysis of variance for the significance of observed differences in the total concentration of DDT and its metabolites between different species of mussels, collected on 16 June 1966 . . . . . . . . . . . . . . . . b) Results of an analysis of variance for the significance of observed differences in the total concentration of DDT and its metabolites between different species of mussels collected on 25 August 1966 . . . . . . . . . . . . . . . vi Page 56 38 41 41 42 42 44 45 45 LIST OF TABLES - Continued TABLE Page 20. a) Results of analysis of variance for the sig- nificance of observed differences in the total concentration of DDT and its metabolites in the native mussels with respect to time of year . . 49 b) Results of new multiple range test for the significant differences in the total concen- tration of DDT and its metabolites in native mussels with respect to the time of year. . . . 49 vii LIST OF FIGURES FIGURE Page 1. Map of study area. . . . . . . . . . . . . . 4 2. Attachment method used to restrict movement of mussels . . . . . . . . . . . . . . . . . 9 5."Pen". method used to restrict movement of mussels. O O O O O O O O O O C O O O O O O O 9 4. Mean concentrations of DDT plus its metabo- lites in L, siliquoidea removed on three dif- ferent dates from those placed in the Red Cedar River on 16 June 1966. . . . . . . . . 17 5. Mean concentrations of aldrin in L. sili- guoidea and A. grandis at different loca- tions in the Red Cedar River . . . . . . . . 4O 6. Mean concentrations of DDT and its metabo- lites in native mussels collected from the Red Cedar River. . . . . . . . . . . . . . . 47 viii INTRODUCTION At the present time a large proportion of our surface bodies of water are contaminated with pesticides and other chemicals (Faust, 1964). Small concentrations of pesticides often go unnoticed as they result in no gross effect on the body of water. Yet, these small amounts may be detrimental to the aquatic environment as they can result in subtle changes such as a decrease in growth rate of aquatic organ- isms. Butler (1966) found that very low concentrations of pesticides resulted in a large decrease in the shell growth of oysters. Also these minute quantities of pesticides may, through synergistic action with other pollutants, result in reduced water quality and a less desirable aquatic biota. Thus, an effective way of measuring the presence of pesticides in the aquatic environment is needed. In a river the pesticide levels are in a continuous state of flux at any one location. They are entering the river via surface runoff, leaching, domestic and industrial waste dis— posal, and aerial drift. They are being absorbed and released by the stream bottom and organisms in the stream and are constantly being moved downstream with the flow of the river. Their identities are also being changed by physical, chemical and biological breakdown. Hence it is very difficult to get a true picture of the pesticide contamination of a stream without analyzing large quantities of water over a long period of time. The author suggests that the use of an aquatic organism could be an efficient means of monitoring pesticide contami- nation. Such an organism would have to be capable of concen- trating the minute quantities present in the water to a large degree and to reach some sort of equilibrium with the concen— tration of pesticide in the water. The principal objective of this study was to determine if the fresh water mussel, a filter feeder, possessed these capabilities and thus serve as a suitable monitor of pesticide contamination of our sur- face waters. STUDY AREA The Red Cedar River is a warm water stream located in south—central Michigan. It flows through farmland and wood- lots, several small towns, and through a large university campus before emptying into the Grand River at Lansing, Michigan. The river is further described by Linton and Ball (1965) and King and Ball (1967). Six locations on the Red Cedar were chosen for the intro- duction of fresh water mussels (Figure 1). Station I is located in the cleanest section of the river (Linton and Ball, 1965) and the bottom is principally sand, with gravel and larger rocks also present. Station II lies at the upstream edge of the Michigan State University campus and below a large suburban area. The river here is impounded by a small dam on the university campus and is very sluggish; consequently, the bottom is usually covered with silt and decaying leaves and other detritus. Under this layer, however, is fine sand and the leaves and silt were cleared before placing the mussels in the river. Station III lies below most of the campus and above the outlet of the old East Lansing sewage treatment plant which was replaced by a new plant further downstream in October, 1965. The bottom is principally sand and gravel but is covered in the summer with a large bed of Potamogeton crispus. Station IV is located about 500 meters downstream 5 .Am3ouum ma Umucocfiampv coauodpouucfl memSE mo mQOHumooH mcHBOQm Hm>wm umpmu pom mnu mo ass .a musmflh Hamm .U .M >9 mm: 9.44 - «emu «I 3455.. o... E amt x .M V 1. “M V . N 0 M 0 m. I 0 M 3 v 1 u. 0 Z 8 “u . 0 M 4 >2 .. mom} 6 m >_m M o 1 .3 o 3 NVOO '83 8330 0 8 mi. Ohm2<finz3 S AU V M O M \v 40 z mozmx.. 025.24.. w¥ '— '- "I . _,_ ._ .. m " "h~"-.-1—-.,."" "‘ -‘_.... __ , _,. .r,——~-, ‘ .. “f... ‘_ _ “5.0- ' - ' .r- ‘ " t-""" ' ‘ ""’v-v—-. q'vyrl-vd' " - _.. . . ~ . ~ A- ‘ - _. “4 . .—¢~¢ ~ - ' ' - '- 1~ , . 1‘ I". 1‘ ' I V V 'l. “ “' -O‘—-l.-—-., ,nvr ”Aw-“urns mg.“ Q. ——----~ ~.... _- _.. -. 'WI‘LHum. fl" .- ...-.- .. .- 7 I‘m“... .‘ I. . . an. _'I 3.1.. ~“"' - w- ..._- -vPcLh' ‘1'.“ , __ .7 __ a. _._ . _>_ ,g-.- i _ "- 0.. , F "7 'W‘“ ' ‘ _ 1 ~ \ " 7 '“..—— __ . __-—-‘.»*—_.—11"" .7—7 . . (db!- 4%- f ~ .Hm17._;r.-.- W .-—’--I .“nw-fi" “‘ ”M if ,- 4-— ._ .— “H‘Juq m—v‘---_- ‘... v.- .< ' ' a. - - . _ \ _ A - -. mm..- -_-.- ' ~- » *- ”Mun. .. n ~.. M ' i"' " -I\J-‘C~A_“_.'.d~ --‘ ' 1 >\~.~__- __ ‘_ , '--.- .~_. '--~ - O 7 7 .d . , 1 ‘ v ' '7 ~ “up. ..——u "' "' 7-.“".NlP-I ‘. ‘,. . a" .. . - - ._ ,. ‘ flaw...“ , Mr- W. . . -' -- ~ * . ’-—-’ “an .. "4......“ in“ -—... .___ " ‘W a. - .- . a...» 1:1- ” ‘ " . qr,“ .1..._l‘_.> , "- '17? ,. _-’ - ' -., _ ‘ x , y - ”aw-1., dc...- ‘ ' 3‘ ' *'--- - - .- "‘ - ranp.‘ . a~m » -, L , - -- ‘ ' - A _ ‘ __, .n A —r - . n “-1“ . .. _.1r..- m ' " -- ~ ; -...M __ ,_ 2- -.- rs ,, _ . .-.....- - .1 - i! o ‘7" - - ,1- :- -v'.-~.. _ r— _, -. __ I~~- » - - n”, .. _ « V ' _... " V --- -..~~.. ,____,_. .-—-.-.- , A -_‘. Figure 5 10 depth of 6 cm., leaving a rim extending 4 cm. above the bottom (Figure 5). The mussels were then placed in these "pens" at each location. This method proved to be the superior one as the tape used to attach the line to the clam deteriorated fairly rapidly in the river. Also, the lines often became tangled and debris tended to collect on them. As soon as possible after collecting, while still living, the mussel was removed from its shell and allowed to drain for a few minutes. The mussel was weighed to the nearest milligram on a Mettler Balance (Model H16) and then placed in a beaker of dry ice. While laying on the dry ice the mussel was cut into smaller pieces so that it could be ground in a blender. The diced, frozen mussel was placed in a blender jar which was also frozen. Enough dry ice was added to fill the blender_jar to the blade level and the mussel was ground to a fine powder at high speed on a Waring blender with ex- plosion proof motor (G. E. Model 5BA60VL22). The frozen powder was transferred to an Erlenmeyer flask and 500 ml. of a 2:1 hexane/acetone mixture was added. The extraction mixture was allowed to stand overnight so that all the remaining carbon dioxide was driven off. The samples were shaken for 50 minutes on a Burrell wrist-action shaker at a setting of five. The mixture was filtered and the super— natant washed twice with 500 ml. of a 10% NaCl solution to remove the acetone. 11 The pesticide residues present in the extract were par- titioned into acetonitrile by extracting the hexane with three 50 ml. portions of acetonitrile that had been previous- ly saturated with hexane. The acetonitrile extracts were combined and 50 ml. of hexane was added to it. The residues were repartitioned back into the hexane by removing the acetonitrile with 10% NaCl solution. The hexane extract was concentrated, over a steambath in a Kuderna-Danish apparatus, to a volume of less than 25 ml. for introduction into a clean- up column. Pyrex columns, 2x50 cm., and fitted with a fritted glass disk, were packed with 10 grams of a 5:1 mixture of Florasil to Celite. A layer of anhydrous sodium sulfate was added both before and after the addition of the Florasil/Celite mixture to the column. The Florasil, which was received acti- vated at 12000F from Floridin, Inc., was deactivated with approximately 5% water. The mixture was calibrated before use to insure conformation to the elution procedure used. The column was prewetted with hexane and the concentrated extract was placed on the column. Each sample was eluted with 500 ml. of hexane and then the sample was reconcentrated to a volume of 10 ml. These extraction and clean-up procedures generally follow those recommended by Shell (1965) except for several modifications. A Beckman G. C. 4 chromatograph equipped with a discharge electron capture detector was used for the analysis. It was 12 fitted with a 6 foot x 1/16 inch Pyrex column packed with 5% D. C. 11 on Gas Chrome Q and was operated at a column temperature of 2000C and 50 ml./minute helium flow. Standards were injected at the beginning of each run, after each 10 samples, and at the end of the run. The identities of the ‘pesticides found were confirmed using columns packed with 2§% QF 1 on acid base washed Chromosorb W and 2.5% S. E. 50 on Gas Chrome RP. Quantitations were based on peak height and the concentrations were based on the wet weight of the mussel. RESULTS AND DISCUSSION In the first experiment 79 fresh water mussels (Lampsilis siliquoidea) were collected from the Cass River and from these eight specimens were randomly selected and analyzed for pesticide content (Table 1). Only DDT and its metabolites, TDE and DDE, were found and they were present in very small concentrations. Confidence limits were com- puted at the 1% level for the total DDT plus metabolites, in the mussels (Table 1). The remaining mussels were divided into five groups of 14 mussels and placed in the river at Stations I to V. After a period of two weeks, three mussels were collected from each station and analyzed. The mussels collected from Stations II, III, and IV contained significantly greater amounts of DDT and its metabolites than the controls while those col- lected from Station I contained amounts within the confidence limits of the controls (Table 2). All mussels placed at Station V died, and no analyses could be made since their tissue had completely decomposed and only empty shells re- mained. The level of pesticide in the mussels was observed to increase in a downstream direction with an especially large increase between Stations II and III (Figure 4). 15 14 00. A 10000.0 m.m.m.5000.000 0000.0 0000.0 5500.0 0000.0 5000.0 .m0000>0 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 0000.0 0500.0 0000.0 0000.0 000.00 .5 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 5500.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 5000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 0000.0 0500.0 5000.0 0000.0 000.00 .0 0000.0 0000.0 0000.0 0500.0 0000.0 000.50 .0 000somxonumz va0muoe 15000000 Asmmvmoe Asmmveoo A.0500000003 .man U20 woe .mmuHHOQmqu mu0 Ucm Ban mo ANV wHMpou mnu pcsoum 000500 moco©0mcoo Ram may paw ammHQUHZ ..00 0000058 .um>0m mmmO onu 800m pmuomaaoo AmmUHOJGHHHm amv 00mmmDE uzmflm :0 mmUHUHummm mo mQOHumnucmucoo .0 00909 15 Table 2. Concentrations of DDT and its metabolites in L, siliquoidea removed on 30 June 1966 from those placed in the Red Cedar River on 16 June 1966. Station Weight(g) DDT(ppm) TDE(ppm) DDE(ppm) Total I 34.674 0.0060 0.0050 0.0043 0.0153 42.442* 0.0247 0.0092 0.0039 0.0378 44.540* 0.0101 0.0078 0.0042 0.0221 Average: 0.0136 0.0073 0.0041 0.0250 II 35.017* 0.0274 0.0257 0.0069 0.0600 42.518 0.0470 0.0381 0.0073 0.0924 26.303* 0.0426 0.0354 0.0089 0.0869 Average: 0.0390 0.0331 0.0077 0.0798 III 51.669 0.0822 0.2032 0.0217 0.3071 50.683* 0.0779 0.2230 0.0178 0.3187 33.145* 0.1192 0.1795 0.0347 0.3334 Average: 0.0931 0.2019 0.0247 0.3197 IV 42.209 0.0734 0.2345 0.0249 0.3328 43.143* 0.0916 0.2712 0.0290 0.3918 33.730* 0.0542 0.1972 0.0187 0.2701 Average: 0.0731 0.2343 0.0242 0.3316 V All mussels dead, only empty shells remaining * Specimens randomly selected for statistical analysis. 16 Figure 4. Mean concentrations of DDT plus its metabolites in L. siliquoidea removed on three different dates from those placed in the Red Cedar River on 16 June 1966. ppm DDT + TDE + DDE, 17 .805 .75+ .704 .65‘ .60“ .55“ .50‘ .45“ .35“ .30“ .25" .20n .15~ .10“ Removed on 30 June 1966 ""‘ Removed on 28 July 1966 ——”—‘ Removed on 25 August 1966 Station III IV 18 After a six-week period in the river the mussels showed still further increases in pesticide content at all four locations (Table 3). Again the levels increased in a down- stream direction with the largest increase occurring between Stations II and III (Figure 4). At this time the mussels from Station I also contained levels of DDT and metabolites above the upper confidence limit of the controls. The mussels removed from the river after ten weeks con- tained approximately the same amount of DDT and its metabolites as those taken from the river after six weeks except those at Station I which decreased considerably (Table 4). As pre— viously the pesticide levels in the mussels increased in a downstream direction with the largest increase again between Stations II and III (Figure 4). A two-way analysis of variance (from Li, 1964) with replication, was run to determine the significance of these observed differences. Since equal samples of three were not obtained due to the escaping of mussels in the field and loss during analysis, the number of replications was reduced to two. Where there were three replications, two were selected using a table of random numbers. Also, since all the remain- ing mussels at Station I had escaped after ten weeks, two specimens of a closely related species, Lampsilis ventricosa, collected at this time and location, were substituted in the analysis. This substitution was justified by the fact that no difference in pesticide content was found between species 19 Table 3. Concentrations of DDT and its metabolites in .E- siliquoidea removed on 28 July 1966 from those placed in the Red Cedar River on 16 June 1966. Station Weight(g) DDT(ppm) TDE(ppm) DDE(ppm) Total I 27.727 0.1410 0.0306 0.0108 0.1824 23.548 Sample lost during analysis 18.679 0.1520 0.0401 0.0107 0.2028 Average: 0.1465 0.0354 0.0108 0.1926 II 28.554 0.1544 0.0900 0.0256 0.2699 25.674 Sample lost during analysis 26.661 0.1256 0.0686 0.0146 0.2088 Average: 0.1400 0.0793 0.0201 0.2394 III 34.056* 0.1987 0.4257 0.0749 0.6993 30.382 0.1267 0.2571 0.0421 0.4259 27.138* 0.2321 0.4496 0.0829 0.7646 Average: 0.1858 0.3775 0.0666 0.6299 IV 27.327* 0.2708 0.5709 0.1025 0.9442 29.161 0.2051 0.4972 0.0823 0.7846 29.165* 0.2349 0.4972 0.0943 0.8264 Average: 0.2369 0.5218 0.0930 0.8517 9(- Specimens randomly selected for statistical analysis. 20 Table 4. Concentrations of DDT and its metabolites in .L. siliquoidea removed on 25 August 1966 from those placed in the Red Cedar River on 16 June 1966. Station Weight(g) DDT(ppm) TDE(ppm) DDE(ppm) Total I 46.132? 0.0058 0.0063 0.0032 0.0153 62.634T 0.0160 0.0053 0.0062 0.0275 Average: 0.0109 0.0058 0.0047 0.0214 II 24.845 0.1529 0.0962 0.0262 0.2753 45.758 0.1368 0.0916 0.0260 0.2544 Average: 0.1448 0.0939 0.0261 0.2648 III 33.637 0.2066 0.2898 0.0642 0.5606 38.114 0.2910 0.4158 0.1026 0.8094 Average: 0.2488 0.3528 0.0834 0.6850 IV 32.202* 0.2668 0.4816 0.1090 0.8574 21.161 0.1597 0.3223 0.0534 0.5354 43.436* 0.1568 0.2878 0.0580 0.5026 Average: 0.1994 0.3639 0.0735 0.6318 *- Specimens randomly selected for statistical analysis. 1. I I o LampSIlis ventricosa. 21 of mussels collected from the same location and at the same time (see page 43). The results of this statistical analysis (Table 5a) show that there was a highly significant difference in concentration of DDT and its metabolites with respect to location and with respect to length of time in the river. Also it is to be noted that there is no significant inter- action between time and location; that is,the mussels did not concentrate pesticides faster at one station than another. Duncan's (1955) new multiple range test was used to fur- ther investigate these differences. The results of these tests show that most of the difference with respect to location is due to the large increase between stations II and III and that the increase between two and six weeks accounted for most of the significant change in pesticide content with time (Table 5b). During the time that the mussels were in the river the amount of DDT present in the water ranged from trace amounts to 0.06 ppm. with the concentration usually under 105mk> (Zabik, M. J., personal communication). The concentration in the water tended to increase in a downstream direction but no increase as dramatic as occurred in the mussels was found. Zabik also found concentrations in the suspended matter from 1 ppm to 50 ppm (dry weight) but this only added a part per trillion or less to the total concentration in the water. 22 Table 5a. Results of an analysis of variance for the signifi- cance of observed differences in the total concen- tration of DDT and its metabolites in L. siliquoidea with respect to length of time and location in the Red Cedar River. Source SS DF MS Fexp. F.995 F.95 Location 1.3682 3 0.4561 45.610 7.226 Time 0.4355 2 0.2178 21.780 8.510 Interaction 0.1498 0.0250 2.500 2.996 Pooled Error 0.1198 12 0.0100 Total 2.0733 23 Table 5b. Results of new multiple range tests for the signifi— cant differences in the total concentration of DDT and its metabolites in L, siliquoidea with respect to length of time and location in the Red Cedar River. Stations I II III IV Means 0.0813 0.1925 0.5810 0.6321 Weeks 1 10 6 Means 0.4128 0.5123 0.1901 23 While these fresh water mussels concentrate DDT and its metabolites at levels much above the concentrations in the water, they appear to be considerably less efficient at this than oysters. Butler (1966) reported that oysters exposed to 1 ppm of DDT for twelve days contained from 14 to 20 ppm of DDT and its metabolites upon analysis. The reason for the death of all mussels placed in the river at Station V was not determined as the mussel tissue had decomposed and all that remained were empty shells. Another chlorinated hydrocarbon insecticide, methoxy- chlor, was found in all mussels examined from Stations III and IV (Table 6). No methoxychlor was detected in the con- trol mussels and only one mussel of those placed in the river at Stations I and II was found to contain methoxychlor. The concentration of methoxychlor in the mussels at both Stations III and IV increased from the two-week exposure level to a high after six weeks; but after ten weeks they dropped back to the two-week exposure levels. A two way analysis of variance with replication using Yates (1934) method of weighted squares of means was run to determine if there were Significant differences between locations and length of time in the river. No significant difference was found between stations and there was no significant interaction between time and location but there was a significant difference be- tween lengths of time in the river (Table 7a). The difference in time was analyzed further using Duncan's (1957) multiple 24 Table 6. Concentrations of methoxychlor in L. siliquoidea removed from the Red Cedar River at three different times from those placed in the river on 16 June 1966. Methoxychlor(ppm) Station 30 June 28 July 25 August I 0.0000 0.0000 0.0000 0.0471 0.0000 0.0000 0.0000 Average: 0.0157 0.0000 0.0000 II 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Average: 0.0000 0.0000 0.0000 III 0.2052 0.1468 0.0823 0.0552 0.0510 0.1233 0.0965 0.1953 Average: 0.1190 0.1310 0.1028 IV 0.0817 0.2616 0.1031 0.0695 0.1920 0.0865 0.0593 0.2126 0.0552 Average: 0.0702 0.2221 0.0816 25 Table 7a. Results of an analysis of variance for the signifi- cance of observed differences in the concentration of methoxychlor in L. siliquoidea with respect to length of time and location in the Red Cedar River. Source SS DF MS Fexp. F.995 F.95 Location 0.0003 1 0.0003 0.135 4.844 Time 0.0352 2 0.0176 7.001 8.912 3.982 Interaction 0.0162 2 0.0081 3.229 3.982 Pooled Error 0.0276 11 0.0025 Total 0.0793 16 Table 7b. Results of new multiple range test for the signifi- cant difference in the concentration of methoxy- chlor in L. siliquoidea with respect to length of time in the Red Cedar River.’ ' Weeks 10 Means 0.0922 0.0946 0.1765 26 range test for heteroscedastic means. The results of this test show that the mussels contained significantly greater quantities of methoxychlor after six weeks than they did after two and after 10 weeks in the river (Table 7b). Another species of mussel, Anodonta grandis, was intro- duced into the river on 25 July 1966; approximately halfway through the first experiment with L. siliquoidea. The prin- cipal reasons for this second experiment were to make a second attempt at introducing mussels below the sewage treatment plant and to confirm the results obtained with L. siliquoidea. This species was collected from the Red Cedar, 300 to 500 meters upstream from Station I. Six of these were random— ly selected and immediately analyzed for pesticide content. The results of these analyses along with the computed 1% confidence limits are presented in Table 8. The remaining mussels were placed in the river in groups of 15 at Stations II, III, and VI. As in the previous experi- ment three mussels were collected and analyzed from each station after periods of two, six and ten weeks in the river. After two weeks only the mussels at Station III were found to contain DDT and its metabolites at levels above the upper confidence limit of the controls (Table 9). The remains of four specimens found dead at Station VI during this period were also analyzed and the results are presented in Table 9. The mussels collected after six weeks exposure showed a sharp increase at Station II and a slight decrease at Station III (Table 10). All mussels remaining at Station VI were found 27 00. A “0550.0 w m 0 0000.000 0000.0 0N00.0 0000.0 0000.0 0000.0 "mmmnm>¢ 0000.0 0000.0 0000.0 0000.0 0N00.0 000.00 .0 0000.0 0000.0 0000.0 0NN0.0 N000.0 000.0N .0 0000.0 0000.0 0000.0 00N0.0 0000.0 000.00 .0 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 .0 0000.0 0000.0 0000.0 0000.0 0000.0 N00.00 .0 000£o>xonuwz ANVHMuOB Afimmvmnn Aemmvmna Afimmvaoa A.m80vu£0003 .moo 0:0 woe .00000onmume 000 0:0 son 00 vam0muou ms» pcsoum mu0E00 mocm©0mcooIR00 wnu 0cm .0 c00umum w>o£w .um>0m Hmvmu 00m map 5000 Umuom0000 .m0ucmnm .0 .mammmsfi x00 :0 me0o0ummm mo mGOHMmHucmucoo .0 w0£me 28 Table 9. Concentrations of DDT and its metabolites in A, grandis removed on 8 August 1966 from those placed in the Red Cedar River on 25 July 1966 and in remains of mussels that died during this period. Station Weighhg) DDT(ppm) TDE(ppm) DDE(ppm) Total II 50.954 0.0857 0.0591 0.0079 0.1527 19.626 0.1157 0.0571 0.0125 0.1855 19.912 0.1004 0.0527 0.0115 0.1647 Average: 0.1006 0.0496 0.0106 0.1609 III 11.557 0.1456 0.2600 0.0442 0.4498 25.541 0.1427 0.2846 0.0514 0.4817 22.726 0.1520 0.2640 0.0488 0.4448 Average: 0.1401 0.2695 0.0491 0.4588 VI 22.051 0.0765 0.0556 0.0204 0.1505 27.502 0.0218 0.0564 0.0155 0.0755 29.611 0.0506 0.0456 0.0181 0.1144 Average: 0.0496 0.0452 0.0179 0.1127 Found dead on 28 July 1966 12.654 0.2090 0.1124 0.0249 0.5465 15.478 0.1855 0.0782 0.0226 0.2845 Found dead on 8 August 1966 10.815 0.1110 0.1054 0.0505 0.2469 16.521 0.0757 0.0625 0.0224 0.1604 29 Table 10. Concentrations of DDT and its metabolites in A, grandis removed on 5 September 1966 from those placed in the Red Cedar River on 25 July 1966. Station Weight(gm.) DDT(ppm) TDE(ppm) DDE(ppm) Total II 45.716 0.2695 0.0711 0.0206 0.5610 54.822 0.5492 0.0775 0.0559 0.4606 12.458 0.4158 0.0805 0.0168 0.5129 Average: 0.5448 0.0765 0.0258 0.4449 III 15.077 0.1174 0.2102 0.0598 0.5674 16.125 0.1185 0.1910 0.0572 0.5467 17.089 0.1915 0.2645 0.0585 0.5144 Average: 0.1424 0.2219 0.0452 0.4095 50 dead with empty shells. After ten weeks exposure the pesti- cide levels in the mussels at Station II returned to the levels found there after two weeks exposure, while those at Station III dropped slightly again (Table 11). The significance of the observed differences in DDT content between stations and time periods was determined us- ing Yates (1954) method of weighted squares of means. A highly significant difference between locations and time periods plus a significant interaction between the two was found (Table 12a). Duncan's (1957) multiple range test for heteroscedastic means was used to further determine differ: ences in time and the results are presented in Table 12b. The large increase in DDT and its metabolites at Station II after six weeks appears to account for both the significant differences in time and the difference in the rate of concen— tration between the two stations. This second experiment supports the results of the first in that the mussels concentrated DDT and its metabolites only up to a certain level as if in equilibrium with the environ- ment. Also, a highly significant difference between the pesticide concentration at Station II and at Station III was again recorded. .5- grandis did differ from L. siliquoidea in rate of uptake, as it reached its plateau after two weeks while ;. siliquoidea did not reach its plateau until sometime after two weeks since they contained significantly greater quantities after six weeks than after two weeks. They also Table 11. Concentrations of DDT and its metabolites in 51 A, grandis removed on 2 October 1966 from those placed in the Red Cedar River on 25 July 1966. Station Weight(gm.) DDT(ppm) TDE(ppm) DDE(ppm) Total II 22.624 0.0765 0.0546 0.0155 0.1262 28.549 0.1145 0.0524 0.0120 0.1787 27.075 0.1259 0.0550 0.0126 0.1955 Average: 0.1055 0.0475 0.0155 0.1661 III 28.088 0.0662 0.1780 0.0580 0.2822 21.545 0.0956 0.2990 0.0455 0.4579 Average: 0.0799 0.2585 0.0417 0.5591 52 Table 12a. Results of an analysis of variance for the signifi- cance of observed differences in the total concen- tration of DDT and its metabolites in A. grandis with respect to length of time and location in the Red Cedar River. Source SS DF MS Fexp. F 995 F.95 Location 0.0961 1 0.0961 25.424 12.226 Time 0.0792 2 0.0596 9.651 8.912 Interaction 0.0866 2 0.0455 10.551 8.912 Error 0.0451 11 0.0041 Total 0.5072 16 Table 12b. Results of new multiple range test for the signifi- cant difference in the total concentration of DDT and its metabolites in A, grandis with respect to length of time in the Red Cedar River. Weeks 10 2 6 Means 0.2824 0.5098 0.4272 55 did not reach as high a concentration of pesticide as did ‘A. siliguoidea. This is probably due to the fact that they were in the river at a later part of the summer, after most spraying programs were completed, and were not exposed to as high a pesticide concentration. This explanation is veri- fied by the results of Zabik (personal communication), which showed a general decline in the pesticide content in the water as the summer progressed. The explanation for the high concentration of DDT and its metabolites found in A, grandis at Station II after six weeks appears to lie in the proportion of DDT to the sum of DDT and its metabolites. Unmetabolized DDT made up over 75% of the total at Station II while downstream at Station III, where the mussels contained about the same total amount, less than 55% of the total was unmetabolized DDT. For all other mussels of both species, which contained over 0.5 ppm total DDT and metabolites, DDT also made up less than 55% of the total. Therefore, it is hypothesized that the mussels were exposed to large concentration of DDT just before collection. Of the mussels placed in the river at Station VI, only half survived for two weeks and those that did appeared to be in rather poor condition. None of the mussels at this station were ever observed to be actively siphoning water. This fact helps explain the lower than expected amount of DDT and its metabolites found in these mussels. It also sheds some light on the possible cause of death of the mussels. It is well 54 known that mussels can close up and stop feeding for fairly long periods of time when harmful substances such as toxic materials or large concentrations of suspended matter are present in the water (Loosenoff and Engle, 1947 and Wilbur and Yonge, 1966). Also, it has been established that when the dissolved oxygen is low the mussels siphon much larger than normal quantities of water (Prosser and Brown, 1961). Thus, it is concluded that these mussels probably died from some toxic substance or combination of toxic substances in the water or lack of food resulting from the large decrease in amount of siphoning caused by the toxic materials. The reason for the higher concentration of pesticide in the dead mussels analyzed is due to the loss of tissue through partial decomposition, as live mussels with the same size shell weighed considerably more. As was the case with A. siliquoidea, no methoxychlor was found at and above Station II in A. grandis (Table 15). But, while A. siliquoidea at Station III and below contained methoxychlor on all collecting dates, only those A, grandis which were in the river for at least six weeks contained methoxychlor, except for one individual after two weeks. As with DDT and its metabolites, the level of methoxychlor did not reach as high a concentration in A, grandis as it did in A, siliquoidea. No difference was found.between amount of methoxychlor contained after six weeks and after ten weeks, using a one way analysis of variance (Table 14). 55 Table 15. Concentrations of methoxychlor in A, grandis re- moved from Station III at three different times from those placed there on 25 July 1966. Date Methoxychlor(ppm) 8 August 1966 0.0000 0.0575 0.0000 Average: 0.0191 5 September 1966 0.0665 0.0874 0.0825 Average: 0.0787 5 October 1966 0.0576 0.0975 Average: 0.0775 56 Table 14. Results of an analysis of variance for the signifi- cance of observed differences in the concentration of methoxychlor in A, grandis with respect to length of time in the Red Cedar River. Source SS DF MS Fexp. F.995 F.95 Time 0.000002 1 0.000002 0.006 6.608 Within 0.001051 4 0.000544 Total 0.001055 5 57 Aldrin was found in all A. siliquoidea collected on 28 July 1966 and in all A. grandis collected on 8 August 1966 but was never detected at an other time (Table 15). The concentration of aldrin in A. siliquoidea decreased in a down- stream direction while the concentrations in A, grandis in- creased in a downstream direction (Figure 5). A one way analysis of variance was run to determine the significance of the difference in aldrin concentration between locations for each species. The results of these analyses show a highly significant difference for each species (Table 16a and 17a). Further investigation of these analyses with Duncan's (1957) multiple range test for heteroscedastic means showed that in the case of A. siliquoidea the significant difference was principally due to the large difference between Station I and the others, while the much greater concentration at Station VI mainly accounted for the difference in A. grandis (Table 16b and 17b). In water samples taken every two weeks, the sample taken on 28 July 1966 contained 19-20 ppb aldrin at Stations I and II, 15-14 ppb at Stations III and IV and less than 4 ppb at Station VI (Zabik, personal communication). No aldrin had been found in the water previously and none was found after this date. It appears from these results that a quantity of aldrin entered the river upstream from the study area and traveled through it with the flow of the river. 58 Table 15. Concentrations of aldrin in A, siliquoidea removed on 28 July and A. grandis removed on 8 August from those placed in the Red Cedar River on 16 June 1966 and 25 July 1966, respectively. Aldrin(ppm) Station A, siliquoidea A, grandis I 1.4426 2.1414 Average 1.7920 II 0.8580 0.5556 0.5701 1.5146 1.4565 Average 0.7140 1.0555 III 0.5964 2.1256 0.4772 1.8648 0.5501 1.9517 Average 0.4097 1.9718 IV 0.1006 0.0591 0.5772 Average 0.1725 VI 4.5591 5.1654 5.5771 Average 5.6952 59 Figure 5. Mean concentrations of aldrin in A, siliquoidea and A, grandis at different locations in the Red Cedar River. 40 5.75“ siliquoidea, 28 July 1966 . grandis, 8 August 1966 Hi I II III Station VI 41 Table 16a. Results of an analysis of variance for the signifi- cance of observed differences in the concentration of aldrin in A. siliquoidea with respect to loca- tion in the Red Cedar River. Source SS DF MS Fexp. F.995 F.95 Location 5.5755 5 1.1915 18.529 8.717 Within 0.5899 6 0.6499 Total 5.9654 9 Table 16b. Results of new multiple range test for the signifi— cant difference in the concentration of aldrin in A, siliquoidea with respect to location in the Red Cedar River. Stations I II III IV Means 0.1725 1.7920 0.7140 0.4079 42 Table 17a. Results of an analysis of variance for the signifi— cance of observed differences in the concentration of aldrin in A, grandis with respect to location in the Red Cedar River. Source SS DF MS Fexp. F.995 .95 Location 10.9008 2 5.4504 17.841 11.042 Within 1.8550 6 0.5055 Total 12.7558 8 Table 17b. Results of new multiple range test for the signifi— cant difference in the concentration of aldrin in A, grandis with respect to location in the Red Cedar River. Stations II III VI Means 1.0555 1.9718 5.6952 45 It is a mystery to the writer why dieldrin, the epoxide of aldrin, was not detected as aldrin is fairly readily converted to dieldrin in aqueous solution by microorganisms (Lichtenstein and Schulz, 1960). Samples were checked with two different column packings and an ultra violet spectrum was run. The results confirmed the presence of aldrin but no dieldrin was detected. Aldrin has been found to have a great effect on the shell deposition in oysters at low concentrations (Butler, 1966). It is therefore possible that aldrin could have been one of the contributing factors to the demise of the mussels at Station VI. At Station I and other locations further upstream mus- sels are still fairly abundant in the river. Several species of these native mussels were collected during the summer and analyzed for pesticide content. The results of these analyses are presented in Table 18. A one way analysis of variance was run on the mussels collected on 50 June and 25 August 1966 to determine if there was a difference between species in pesticide content. No difference was found be- tween species for either date (Tables 19a and b). It was also observed that the concentration of pesticide in the native mussels generally declined during the summer as was observed in the introduced mussels (Figure 6). A one way analysis of variance was run to investigate the signifi- cance of this decline. A highly significant difference was 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0moo0uucm> 000000200 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000000 000005000 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000000 0uao0oa0 000000 00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 050000000 0000000 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0moo0nucm> 000000900 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 000:00u0e 00000090004 4 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000000 0uco0oc< 4 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0500050 msu0£mouum 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000000 000005000 00:0 00 0000.0 0000.0 0000.0 0000.0 0000.0 000.0 800.0 0000.0 0000.0 0000 0000.0 0o» .00 a 3 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0>00W 000000050 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 00000:“ msu0saouum 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000.0 0000.0 0000.0 0000.0 0000.0 000.00 0000000 0uco0oc< 0:50 00 AEQQVHO0£omxonumz .00u09 Afimmvmnn AaflmvmnB,AEmflvBQQ 5000:0003 0000000 0000 000 um um>0m 00000 000 .0000 mo 00E850 050 020050 H 0000000 0>onm 0:0 Eoum 0000000oo 000005E :0 0000000000 00 mCO0umuucmocoo .00 00000 45 Table 19a. Results of an analysis of variance for the signifi- cance of observed differences in the total concen- tration of DDT and its metabolites between differ- ent species of mussels collected on 16 June 1966. Source SS DF MS Fexp. F 995 F.95 Species 0.0014 5 0.0005 0.255 4.547 Within 0.0072 4 0.0018 Total 0.0086 7 Table 19b. Results of an analysis of variance for the signifi— cance of observed differences in the total concen- tration of DDT and its metabolites between differ- ent species of mussels collected on 25 August, 1966. Source SS DF MS Fexp. F.995 F.95 Species 0.000007 1 0.000007 0.1555 6.608 Within 0.000185 4 0.000046 Total 0.000190 5 46 Figure 6. Mean concentrations of DDT and its metabo— lites in native mussels collected from the Red Cedar River. ppm DDT plus metabolites, .16 .15 .14 .13 .12 .11 .10 .09 .08 .07 .06 .05 .04 .05 .02 .01 47 r 16 June 50 June Date '— 25 August 48 found between dates (Table 20a) and upon further investiga- tion with Duncan's (1957) multiple range test for hetero- scedastic means it was found that the mussels collected on 16 June 1966 contained a significantly greater amount of DDT and its metabolites than those collected at later dates, which did not differ from each other (Table 20b). During this study the bottom muds contained from less than 0.1 ppm up to 10 ppm of DDT and its metabolites but no methoxychlor was found (Zabik, personal communication). Generally the mussels contained lower concentrations of pesticide than the bottom muds from the same location in the river. The existing invertebrate fauna which I examined, principally Tubificidae and Chironomidae, were found to con— tain higher concentrations of pesticides than the mussels, but these results were based on very small sample sizes. Recently work has been done with oysters involving their use as a biological monitor of pesticide levels in the marine environment (U. S. Fish and Wildlife Service, 1964). From the results of the experiments conducted by the writer it seems that fresh water mussels would make excellent monitors of pesticide concentrations in the fresh water environment. They concentrate pesticides to levels many times greater than found in the water and, as was the case with methoxychlor, may concentrate pesticides which would have gone undetected in the water. Mussels, in comparison with other aquatic organ- isms, are especially well adapted as monitors because they 49 Table 20a. Results of an analysis of variance for the signifi- cance of observed differences in the total concen- tration of DDT and its metabolites in native mussels with respect to time of year. Source SS DF MS Fexp. F.995 F.95 Dates 0.0616 2 0.0308 55.340 6.891 Within 0.0110 19 0.0006 Total 0.0726 21 Table 20b. Results of new multiple range test for the signifi— cant differences in the total concentration of DDT and its metabolites in native mussels with respect to time of year. Dates 16 June 50 June 25 August Means 0.1474 0.0498 0.0280 50 feed by filtering large quantities of water, move very little, and have a long life span (up to 20 years). Galtsoff (1928) found that adult oysters, three to four inches long siphoned up to 5,000 milliliters per hour when the water temperature was 25°C and siphoned, on the average, 20 hours a day at a temperature range of 15-2200. Bovjerg (1957) reported that the mean movement of Q. siliquoidea when well-fed was only 2.5 meters per week and when not fed the mean movement ranged from 5.4 to 6.7 meters per week. Miller §£_al, (1966) found that the decrease in residue levels in mussels is not as rapid as in fish and that very few metabolites were found and thus he concluded that the mussels have a slower rate of metabolism of pesticides. Thus mussels would yield more of a "history" of pesticide contamination than fish as well as indicating very recent changes as occurred with A. grandis after six weeks at Station II. Also, since Miller's work was with two organophosPhate insecticides, diazonium and parathion, the mussels value as a monitor is not limited to chlorinated hydrocarbon insecticides. SUMMARY Fresh water mussels, Lampsilis siliquoidea and Anodonta grandis, were introduced into the Red Cedar River at six different locations and analyzed for pesticide content after different lengths of time in the river. DDT and its metabolites, TDE and DDE, were found in all mussels analyzed. The amount of DDT and its metabolites in the mussels placed in the Red Cedar River was significantly greater than the controls after two weeks at Station II and lower stations. The amount of DDT and its metabolites found in the intro- duced mussels increased significantly in a downstream direction. The amount of DDT and its metabolites found in the intro— duced mussels increased significantly with time at first and then leveled off. Methoxychlor was found in most of the mussels placed in the river at Station III and below but was found in only two specimens at Station II and above. 51 52 7. Aldrin was found in all mussels on two dates of retrieval from the river but was not found before or after these dates. 8. Mussels of several species were collected from the Red Cedar River in the vicinity of Station I and analyzed for pesticide content. Very small concentrations of DDT and its metabolites were found and there was no significant difference between species. LITERATURE CITED Boss, K. J. 1964. Unionidae of the Red Cedar river, Michigan. Nautilis 77: 117-118. Bovjerg, R. V. 1957. Feeding related to mussel activity. Proc. Iowa Acad. Sci. 64:650-655. Butler, P. A. 1966. The problem of pesticides in estuaries. Amer. Fish. Soc. Spec. Pub. No. 5. 154p. Duncan, D. B. 1955. Multiple range and multiple F-tests. Biometrics 11:1-42. . 1957. Multiple range tests for correlated and heteroscedastic means. Biometrics 15:164-176. . Faust, S. D. 1964. Pollution of the water environment by organic pesticides. Clinical Pharm. and Therap. 5:677-686. Galtsoff, P. S. 1926. New methods to measure the rate of flow produced by the gills of oyster and other mollusks. Science 65:255-254. Jensen, A. L. 1966. Stream water quality as related to urbanization of its watershed. M. S. Thesis. Michigan State Univ., East Lansing. 156p. King, D. L. and R. C. Ball. 1967. Comparative energetics of a polluted stream. Limnol. Oceanog. 12:27-55. Li, J. C. R. 1964. Statistical Inference I. Edwards Brothers, Inc., Ann Arbor, Michigan. 658p. Lichtenstein, E. P. and K. R. Schulz. 1960. Epoxidation of aldrin and heptachlor in soils as influenced by auto- claving, moisture and soil types. Jour. Econ. Ent. 55:192-197. Linton, K. J. and R. C. Ball. 1965. A study of the fish populations in a warm-water stream. Michigan Quart. Bull. 48:255-285. Loosanoff, V. L. and J. B. Engle. 1947. Effect of different concentrations of micro—organisms on the feeding of oysters. U. S. Fish & Wildl. Serv. Fish. Bull. 51:51-57. 55 54 Miller, C. W., B. M. Zuckerman, and A. J. Charig. 1966. Water translocation of diazinon-C14 and parathion-S35 off a model cranberry bog and subsequent occurrence in fish and mussels. Trans. Amer. Fish. Soc. 95:545-549. Prosser, C. L. and F. A. Brown, Jr. 1961. Comparative animal phySIOIng, 2nd ed. W. B. Saunders, Co., Philadelphia, 688p. Shell Modesto Method Series. 1964. Determination of chlori- nated pesticide residues in water, soil, crops, and animal products. Shell Development Co., Agriculture Research Division. 12p. United States Dept. of Interior. 1964. The effects of pesti— cides on fish and wildlife. Circular 226. Wilbur, K. M. and C. M. Yonge. 1966. Physiology of Mollusca, Vol. II. Academic Press, New York. 645p. Yates, F. 1954. The analysis of multiple classifications with unequal numbers in different subclasses. J. Am. Stat. Assoc. 29:51-66. MICHIGAN STATE UNIVERSITY LIBRARIES 3 1193 03082 9935