- Email: [email protected]
A freshwater shrimp (Paratya compressa improvisa) as a sensitive test organism to pesticides
Environmental Pollution 59 (1989) 325-336
A Freshwater Shrimp (Paratya compressa improvisa) as a Sensitive Test Organism to Pesticides Shigehisa H a t a k e y a m a a & Y o s h i o S u g a y a b a Environmental Biology Division, b Engineering Division, The National Institute for Environmental Studies, 16-20nogawa, Tsukuba, 305 Ibaraki, Japan (Received 23 December 1988; accepted 24 February 1989) ABSTRACT The susceptibility of 2-week-old individuals of thefreshwater shrimp, Paratya compressa improvisa, to five kinds of insecticide and five kinds of herbicide was examined in comparison with that of two species of Cladocera, Daphnia magna and Moina macrocopa. The shrimp was especially sensitive to two organophosphorus insecticides. The 48-h LCso values for fenitrothion and fenthion to the shrimp were 1.15 and 1.04 pg litre-1 (mean value, n = 2), in contrast with 37.8 and35.3 #g litre- 1 in the case ofM. macrocopa, and more than 50 #g litre-~ with D. magna. The shrimp also showed the higher susceptibility to other insecticides, diazinon, carbaryl ( NAC) and BPMC, apart from D. magna to diazinon and NAC. The shrimp also showed higher susceptibility to herbicides. The 48-h LCso values of CNP, benthiocarb, oxadiazon, butachlor, and symetryne to the shrimp were two to eight times lower than those of two species of Cladocera, except for the LCso value of oxadiazon to M. macrocopa, which was very slightly higher. However, the shrimp showed a somewhat lower susceptibility to heavy metals than the two species of Cladocera, especially to copper, and to cadmium and zinc in comparison with D. magna. A bioassay using the shrimp with river water, collected from the river adjacent to the pad@field, showed clearly the high mortality of the shrimp following the aerial spraying with pesticides.
INTRODUCTION A freshwater shrimp, Paratya compressa improvisa, was reported by Kamita (1961) to be widely distributed in lakes, ponds, rivers and/or ditches in the north-east region of the main island of Japan, and large quantities o f the 325 Environ. Pollut. 0269-7491/89/$03.50 © 1989 Elsevier Science Publishers Ltd, England. Printed in Great Britain
326
Shigehisa Hatakeyama, Yoshio Sugaya
shrimp were caught for food at that time in this area. However, the present authors have only limited information on its recent distribution in Japan. This shrimp has been used as a test organism for evaluating the toxicity of river water to aquatic organisms; river water was freeze-condensed using a vacuum-freeze evaporator and used for acute toxicity tests with the shrimp, which is known to be sensitive to pesticides (Kariya et al., 1988). These authors regarded the shrimp as especially sensitive to organophosphorus insecticides, because mortality increased in water taken from the river adjacent to a rice paddy field after aerial spraying with insecticides. However, the relative susceptibility test of the shrimp and other standard test organisms, such as Daphnia magna, to pesticides has not been reported. In the present study, two species of Cladocera, Daphnia magna and Moina macrocopa, were used as reference organisms to evaluate the susceptibility of the shrimp to pesticides and heavy metals. Sensitive test organisms are especially useful when the solubility and/or amount of the test chemicals are limited. We can also reduce the period of toxocity test and amount of waste chemicals using such sensitive test organisms. Agrochemical concentrations in Japanese rivers are usually (Maru, 1985; Nakaminami et al., 1985; Imanaka et al., 1985; Iwakuma et al., 1988) much lower than the 48h LCso values for test fish (Macek & McAllister, 1970; Miyashita, 1988). However, it seems that concentrations of many insecticides and/or herbicides are still at levels very hazardous to many kinds of sensitive aquatic organism during the spraying season for pesticides. We may be able to monitor the seasonal variation in toxic effects of pesticides to aquatic organisms by bioassay using a very sensitive test organism. A bioassay was therefore conducted using P. compressa improvisa to evaluate the variation in toxicity of river water before and after aerial spraying of insecticide on a rice paddy field. Three species of euryhaline grass shrimp Palaemonetes, especially P. pugio, have often been used for toxicity tests of various chemicals (Buikema et aL, 1980). However, reports on toxicity tests using freshwater shrimps are very limited except for those using Gammarus and, rarely, the freshwater shrimp Palaemonetes kadiakensis. The present paper also describes the method of culture and some characteristics of the shrimp, P. compressa improvisa, as an organism for use in acute toxicity tests. MATERIALS AND METHODS Materials Paratya compressa improvisa The first release of offspring as post-zoeae, which grew into shrimps within 14 days, usually occurred at the age of about 4 months at a water
Freshwater shrimp as a test organism to pesticides
327
temperature of 24 + I°C. The size of reproductive females was ca. 30 mm. Each shrimp released 141 + 22 (mean _ SD, n = 20) offspring, ranging from 95 to 236 every 2 weeks in a room controlled at 24 + I°C and 16 L (2500 lux) - 8 D. New eggs were deposited within 2 days after the release of the previous brood offspring. Thirty pairs of adult shrimp were usually reared separately in 16 liters of rearing water using 20-1itre glass aquaria with a piece of the aquatic plant, Elodea densa CASP. The water used was dechlorinated tap-water, originating from Lake Kasumiguara, pH was 7.7 to 7-9 and hardness was ca. 80 mg litre- 1 a s C a C O 3. An adequate amount of a commercial fish food (granular type No. 4C kuranburu, Nihon Haigou Shiryou, Ltd), composed of 64% fish powder, 28% corn powder, 2% corn gluten and 6% other (NaC1, CaPO 4, dried yeast) was provided once a day. About half of the rearing water was exchanged with fresh water, and food residues were removed twice a week. All aquaria were checked each day and pairs that had released offspring were transferred to newly prepared aquaria. Offspring were reared for 2 weeks using the same methods as mentioned above and employed for acute toxicity tests. As the behavior of the shrimp was very rapid and its movements sharp, we used fine netting to transfer them to a test beaker. The body length at age 2 weeks was 5.98 _ 0.49 mm (n = 10). Daphnia magna (delivered indirectly from the USA EPA) and M. macrocopa, which has been cultured for 10 years in our laboratory, were reared in glass beakers (1000ml for the former and 500ml for the latter) using Chlorella sp. as a food in a room controlled at 23 +__I°C. The Chlorella sp. was cultured in a modified MC medium (Watanabe, 1960) for 2 weeks. Finally, the cultured Chlorella was washed with an artificial soft water (Tabata, 1972; Hatakeyama & Yasuno, 1981) by centrifugation (5000 rpm, 20 min, 3 times) to replace the Chlorella culture medium with the artificial soft water because the medium was toxic to the two species of test organism. The Chlorella was adjusted to a density of ca. 1"5 x l0 s cells ml-1 and preserved in a refrigerator for use as a food. Each day adults and their offspring were separated using a glass pipette, and offspring were reared for one day using the Chlorella as a food before the acute toxicity test. The age at the start of exposure was 36 _ 12 h. Test methods
One-hundred millilitre aliquots of pesticide stock solutions (5000 mg litre- 1) were prepared using more than 99% grade chemicals (Wako Pure Ltd) and 99.5% ethanol, and stored in a refrigerator throughout the experiment. Three kinds of organophosphorus insecticide, fenitrothion (dimethyl 4nitro-m-tolyl phosphorothionate), fenthion (dimethyl 4-methylthio-m-tolyl phosphorothionate) and diazinon (diethyl 2-isopropyl-4-6-pyrimidinyl
328
Shigehisa Hatakeyama, Yoshio Sugaya
phosphorothionate) and two kinds ofcarbamate insecticide, carbaryl (NAC, 1-naphthyl N-methyl-carbamate) and BPMC (2-sec-butylphenyl N-methylcarbamate) were used as test chemicals. Currently, these insecticides are widely used in our country. Five kinds of herbicide, chlornitrofen (CNP, 2,4,6-trichlorophenyl 4'nitrophenyl ether), benthiocarb (S-(4-chlorobenzyl) N,N-diethylthiocarbamate), oxidiazon (5-tert-butyl-3-(2,4-dichloro-5-isopropoxyphenyl)1,3,4-oxadiazolin-2-one), butachlor (2-chloro-2',6'-diethyl-N-butoxymethyl) acetanilide), and symetryne (2,4-bis (ethylamino)-6-methylthio-1,3,5otriazine), were used. The acute toxicity test was conducted in a room controlled at 23 _ I°C, and under a 12 h on-off fluorescent light regime. Eight 2-week old shrimps were put in 100 ml of artificial soft water in a 100 ml glass beaker using fine netting, and a stock solution of test chemicals was put into a beaker after appropriate dilution with ethanol and stirred with a glass rod to make a nominal concentration series. Finally, the total ethanol was adjusted to make 0-1% of the test solution for all the concentration series. In the case of D. magna and M. macrocopa, a concentration series of the 200-ml test solution was prepared, and each was divided into three replicate glass beakers with 30ml for D. magna and 20ml for M. macrocopa, respectively. Ten organisms were put in each beaker using a glass pipette, and mortality was checked every 24 h for 4 days, by removing any dead organisms each time. Acute toxicity tests for three heavy metals (Cd as CdC12, Cu as CuSO4 and Zn as ZnSO4) were conducted using the same methods as those described above. All tests on the three species were conducted two or three times, when the 95% confidence intervals of two LCso values did not overlap. The LCso values, associated 95% confidence intervals and regression lines of dosemortality relationships, were calculated using Finney's probit analysis method (1971) programmed for a personal computer. The 48 h LCs0 values for fenitrothion to P. compressa improvisa and M. macrocopa at different ages, 1, 3, 7, 14, 21 and 60 days for the former and 0 (12 + 12h), 1, 2, 3, 4 days for the latter, were obtained to establish the relationship between susceptibility to the insecticide and growth stage. The test was conducted using the same method except for the number of shrimps per test beaker; six individuals of 21-day-old and four indi,ciduals of 60-dayold were put in a 100 ml beaker (with five replicates), respectively.
Bioassay of the river water An organophosphorus insecticide, fenthion, was sprayed aerially over part of a paddy field near our institute together with a fungicide, edifenphos, to
329
Freshwater shrimp as a test organism to pesticides
control rice plant pests on 31 July 1988. River water (500 ml) was collected daily in a 500-ml glass vial from 2 days before to 5 days after the aerial spraying. From the fifth day, the river water was collected at intervals of several days for 3 weeks. River water (100 ml) was decanted into a 100-ml glass beaker with three or four replicates, and eight 14-day-old shrimps were put into the beaker. The mortality of the shrimp was then recorded once a day for 4 days. Ten 2-day-old M. macrocopa were exposed to the river water (30 ml) for 3 days with three replicate glass beakers.
RESULTS Insecticides
The LCso values for fenitrothion were between 0.96 and 1-35/~g litre- 1 in the case of the shrimp at the ages examined (Table 1). The values ranged from 25-8 (value for adults) to 48.7/lg litre-1 in the case of M. macrocopa. The susceptibility to fenitrothion of both species was fairly constant throughout the growth stages examined, although the susceptibility of 4-day-old M. macrocopa, which released offspring during the test period, was about twice that of 2-day-old organisms (Table 1). The shrimp showed higher susceptibility to the insecticides examined than the two species of cladocera, except for the case of D. magna and diazinon TABLE 1
LCso Values (/~g liter-
Agea in days
0 1 2 3 4 7 14 21 28 60
1) of Fenitrothion
to Paratya compressa improvisa and Moina macrocopa 48-h LCso values
Paratya cornpressa improvisa
1.10 (0.72-1.69) [2-20]b 1"00 (0-79-1'25) 1.04 (0.81-1.34) 1.27 (1.05-1.54) 0.96 (0.83-1.11) 1.35 (1.11-1.65)
Moina macrocopa 33.9 (29.7-38.7) 39.7 (34.9-45.1) 48.7 (43.6-54.4) 45.4 (41.6-49.5) 25"8 (18.7-35.6)
[3"07] [2.82] [2.17] [4.45] [3-42]
a Age, 0, 12 __ 12 h, 95% confidence intervals in parentheses. b Slope function of the regression line.
[3-59] [3.61] [4.61] [7.21] [3"62]
BENTH IOCARB
SYMETRYNE
B U TACH LOR
10
- I1--11
1
i
I
I
t
I
i
@
I[
I
".. B _
n~
100 1000 CONCENTRATION (fig 1-1)
=l
IIii
10
I
100
|
10000
II
t
,-o-- ,_D_
,-o-..o_
=
m<~',. .
magna ( - I )
with 95% confidence intervals (bars at both sides of symbols).
Fig. 1. The 48-h LCso values of five kinds of pesticide and five kinds of herbicide to P. compressa improvisa (-C)-), M. macrocopa (-[]-), and D.
n.,
_u O X A D I A Z O N
LM a
or)
CHLORNITROFEN
BPMC
CARBARYL
DIAZ INON
(J l(.,t M.I
.=
FENTHION
u)
FENITROTHION
CHEMICALS
;=p
Freshwater shrimp as a test organism to pesticides
331
(Fig. 1). The mean 48-h LCso values (n = 2) for fenitrothion and fenthion to the shrimp were 1.15 and 1.04, respectively. The shrimp was about 30 times more sensitive to both chemicals than M. macrocopa based on the 48-h LCs o values for both species. It was difficult to obtain the LCso values of fenitrothion and fenthion to D. magna, because the dose-response (mortality) relationship was not clear with the concentrations used (up to 100#g litre- 1) in the present toxicity test. The individuals which had been knocked down at the higher concentrations vibrated their limbs more intensely, although they could not swim. The ECso value (immobility as the criterion) was estimated to be about 50 #g litre-1 to provide a reference (Fig. 1). The susceptibility of the shrimp to diazinon was about 1.5 times greater than that of M. macrocopa. The shrimp was 10 and 28 times more susceptible to carbaryl and BPMC than M. macrocopa, and 10 times more susceptible to BPMC than D. rnagna. However, exceptionally, the shrimp showed somewhat higher (ca. x 2) susceptibility to diazinon and carbaryl than D. magna in the 48 h acute toxicity test.
Herbicides The 48-h LCso values for five kinds of herbicide to three species of test organism are also shown in Fig. 1. Chlornitrofen (CNP) was the most toxic to all three species. The mean 48-h LCso values (n = 2) of C N P to P. compressa improvisa, M. macrocopa and D. magna were 54.4, 413 and 424 #g litre- 1, respectively. The ratios of the shrimp's LCso value to that of the two cladocerans were 7.6 and 7.8, respectively. The shrimp was also 2.2 to 8 times more susceptible to the three herbicides, benthiocarb, oxadiazon and butachlor, than the two species of Cladocera, except for approximately the same susceptibility of M. macrocopa to oxadiazon. Symetryne was the least toxic to the three test organisms. The 48 h LCso values to the shrimp, M. macrocopa, and D. magna were ca. 14.7, 32 and more than 50 mg litre-1, respectively. The ratios of the 24-h LCs0 values of ten kinds of chemical at 48, 72 and 96 h are shown in Fig. 2 for the shrimp and M. macrocopa. In the case of the shrimp, the LCso values decreased to less than 50% of the 24-h LCso values at 48 h with almost all the pesticides (Fig. 2). In contrast, the 48 h LCso values of M. macrocopa were more than 50% of the 24-h LCso with all nine chemicals. The 96-h LCso values for N A C and BPMC to the shrimp decreased to less than 10% of the LCso values at 24 h.
Heavy metals The results of the acute toxicity test for three metals to P. compressa imorovisa and the two cladocerans are shown in Table 2 as 48-h LCso values.
o
PER
CENT 0
PER
CENT 0
O0
~" o
3O
-
•
~
0 Z
-
D
@
"r
Er~
i>
t-gr~ ?
W
,-
0
•
Y
N 0 ;[
Z
cS ./
: ,~
o/
S n~fnSnS o1.~so,~'num,fa~pl~Hns~.tla,~!~/S
~
Freshwater shrimp as a test organism to pesticides
333
TABLE 2
48-h LCso Values (#g liter- x) of Heavy Metals to Three Species of Aquatic Organisms Metals
Cd Cu Zn
P. compressa improvisa
Moina macrocopa
Daphnia magna
59-2 (39.9-87'8) [3'16] a 71.4 (63-7-80-0) [6-12] 77"5 (68.0-88.4) [4.99] 80"4 (71"9-96'4) [5"42] 26.7 (18-5-38'3) [2.04] 5"90 (5.60-6.22) [16.7] 32"6 (25.7-41.4) [2.13] 7"45 (7"13-7"77) [21"7] 879 (706-1094) [2'48] 1 320 (880-1 970) [2"36] 722 (586-899) I"2"16] 1 643 (1 186-2 277) [2'02]
28.6 (20.1-40.9) [2'37] 22.4 (15-5-32.4) [1'82] 3.53 (3.17-3.92) [6.68] 4.32 (3"77-4"96) [6'08] 115 (89-147) [2-63] 124 (96-165) [2.94]
95% confidence intervals in parentheses. Slope function of the regression line. T h e L C s o values f o r Cd, C u a n d Z n to the s h r i m p were 68.4, 29-7 a n d 801 #g l i t r e - ~ (n = 2), respectively. T h e susceptibility o f the s h r i m p to c a d m i u m was a b o u t 2.7 times less t h a n t h a t o f D . m a g n a , a n d a l m o s t the s a m e as t h a t o f M. m a c r o c o p a . T h e s h r i m p s h o w e d 4"4 a n d 7"5 times less susceptibility to c o p p e r a c c o r d i n g to the 48-h L C s o r a t i o s t h a n M . m a c r o c o p a a n d D. m a g n a . T h e TABLE 3
Mortality (%) of the Shrimp in River Waters Collected from the River Hanamuro Adjacent to Paddy Field Before and After Aerial Spraying of Insecticide (Fenthion) (1988) Time (h) o f exposure Date
July 29 30 31a August 1 2 3 4 5 8 13 18 19
24
48
72
96
0 0 100 78 (10) 71 (3) 59 4 0 4 0 0 0
0 0 100 97 (13) 92 (100) 100 21 4 4 0 0 0
0 4 100 100 (73) 100 (100) 100 96 17 8 0 0 0
6 13 100 100 100 100 100 63 33 4 12 0
Test was conducted using three or four replicate beakers each containing eight 14-day-old shrimps. Mortality of M. maerocopa is shown in parentheses. a Insecticide (fenthion) was aerially sprayed.
334
Shigehisa Hatakeyama, Yoshio Sugaya
susceptibility of the shrimp to zinc was also much less than that ofD. magna, although it was slightly higher than that of M. macrocopa (Table 2). Bioassay of the river water
There was no mortality of the shrimp after 2 days in river water collected during the 2 days prior to aerial application of the pesticides, fenthion together with edifenphos. However, the mortality of the shrimp increased to 100% after 4 days in the water during the 5 days following aerial spraying (Table 3). Subsequently, the mortality of the shrimp 96 h after being put into the river water decreased gradually to 63% and 0% on the seventh and twentieth days after the aerial spraying (Table 3). On the other hand, the mortality of M. macrocopa was observed only on the 1st and 2nd days in August. DISCUSSION As mentioned above, test organisms highly sensitive to chemicals are very useful in acute toxicity tests. The shrimp used in the present study showed fairly constant susceptibility to fenitrothion in the growth stages from 3 to 60 days old. Mortality of the control was almost always zero for the 4-day test period, because the shrimp did not show internecine struggle, although cannibalism often occurred against weak, moribund and/or dead organisms. The 14-day-old shrimps never jumped out of the test beaker, but this problem often occurred with 60-day-old shrimps. Mortality criteria were easier to check in the case of the shrimps, because the translucent body of the live shrimp changed to a dark color when observed through a binocular microscope using reflected light, and cessation of thoracic appendages and/or pleopods movement was also a good criterion for dead organisms. However, there are several disadvantages in using the shrimps as test organisms. In the case of P. compressa improvisa, the first release of offspring occurred at 4 months, in contrast to several days to 2 weeks in the case of cladocerans. The interval of offspring release was 2 weeks under the present rearing conditions. Productivity of young is lower in shrimps than in cladocerans. Handling of shrimps is difficult because they move very quickly, so we used fine netting to handle them instead of the glass pipette used for cladocerans. The susceptibility of the shrimp P. compressa improvisa to the pesticides used in the present paper was usually higher than that of the two species of cladocera. The 48-h LCso values for 2-day-old guppy Poecilia reticulata
Freshwater shrimp as a test organism to pesticides
335
(Miyashita, 1988), which has been used in our laboratory as an experimental organism for toxicity tests, are 47 to 3700 (fenitrothion) times higher than the values found with the five kinds ofinsecticide used in the present study, while the LCso values to the guppy of the five kinds of herbicide used here are 1.3 to 16.9 times those to the shrimps. The 48-h LCso values for the shrimp decreased to less than 50% of the values at 24 h, while in the case of guppy (Miyashita, 1988), even the 96-h LCso values were more than 50% of the 24-h LCs0 values except for three chemicals, carbaryl, BPMC and benthiocarb. The present result (Fig. 2) and the result cited above suggest that the high susceptibility of the shrimp to pesticides will be even more marked in chronic toxicity tests. The susceptibility of the shrimp to two kinds of organophosphorus insecticide was very high, the 48-h LCso values for fenitrothion and fenthion being around 1 #g litre- 1. Organophosphorus insecticide in the river water, such as those mentioned above, often exceeded 1 #g litre- 1 for several days or longer periods during the insecticide spraying season from June to August in agricultural fields (Iwakuma et al., 1988). Mortality of the shrimp, P. compressa improvisa, actually increased in the river water on the day the insecticide was sprayed on the paddy field, although a sensitive test fish, Tanichys albonubes, was rather insensitive in the bioassay of river water developed by Kariya (Kariya et al., 1988). These results indicate that the shrimp is especially susceptible to insecticides. However, the susceptibility of the shrimp to heavy metals (Cd, Cu, Zn) was lower than that of the two species of Cladocera, especially D. magna. These results indicate that the shrimp is a promising test organism for bioassays, especially to pesticides in the water around agricultural fields, without the need for condensing the water by freeze-evaporation (Kariya et al., 1988). Bioassays using 14-day-old shrimps of river water adjacent to the rice paddy field strongly suggest that the habitat of the shrimp may be restricted severely in the pesticide spraying seasons by pesticides such as organophosphorus insecticides. Plenty of the shrimp used to be netted as a freshwater fishery product from ponds and ditches in several districts of north-east Japan (Kamita, 1961). However, attempts to obtain shrimp catch are scarcely conducted at the present time (pers. comm.) because its habitat has decreased severely. The present acute toxicity and bioassay results suggest that there may be a causal relationship between the disappearance of the shrimp and the presence of pesticide in the ponds and/or ditches adjacent to agricultural fields. At present we can see the shrimp in the fairly restricted ponds or small streams not polluted with pesticides and/or other types of chemical. Further bioassays at several river sampling points throughout a year, including the herbicide spraying seasons, are now in progress, and the results will be published in later papers.
336
Shigehisa Hatakeyama, Yoshio Sugaya ACKNOWLEDGEMENTS
The present authors are very grateful to Mr Y. Ogaminoa, Mr K. Kawabe, and M r Y. N a k a y a m a , who provided highly competent and consistent assistance throughout the course of the experiment. We are also very grateful to Dr T. Kariya (Professor of T o h o k u University at that time), who kindly distributed the shrimp to our laboratory in 1982, and gave valuable suggestions on rearing methods.
REFERENCES Finney, D. J. (1971). Probit Analysis, Cambridge University Press, London. Buikema, A. L., Niederlehner, B. R. & Cairns, J. Jr (1980). Use of grass shrimp in toxicity tests. In Aquatic Invertebrate Bioassays. ed. A. Buikema & J. Cairns, Jr, pp. 155-73. Hatakeyama, S. & Yasuno, M. (1981). Effects of cadmium on the periodicity of parturition and brood size of Moina macrocopa (Cladocera). Environ. Pollut., 26, 111-20. Imanaka, M., Hino, S., Matsunaga, K. & Ishida, T. (1985). Oxidiazon residues in surface water and crucian carp (Carassius cuvieri) of Lake Kojima. J. Pesticide Sci., 10, 125-34. Iwakuma, T., Shiraishi, H., Nohara, S. & Takamura, K. (1988). Pesticide residues in water and sediment of the River Koise and its tributaries. Res. Rep. Natl. Inst. Environ. Stud. Jpn., 114, 73-83. (In Japanese, Abstract in English.) Kamita, T. (1961). Studies on the freshwater shrimps, prawns and crayfish of Japan. (In Japanese, Abstract in English.) Published by Sonoyama Book Ltd, Matsue, Shimane, Japan. Kariya, T., Oouchi, K., Suzuki, A., Niwa, T. & Sato, M. (1988). A new bioassay method to detect low-level toxicity of waters. Proceedings of the 4th IUBS International Symposium on Biomonitoring of the State of the Environment (Bioindicators), Tokyo, 1987. Biological Monitoring of Environmental Pollution. ed. M. Yasuno and R. B. A. Whitton. Tokai University Press, Tokyo, pp. 23-31. Macek, K. J. & McAllister, W. A. (1970). Insecticide susceptibility of some common fish family representatives. Trans. Am. Fish Soc., 59, 20-7. Maru, S. (1985). Monitoring survey of pesticides in river water within Chiba Prefecture. Ecological Chemistry, 8(3), 3-10. (In Japanese.) Miyashita, M. (1988). Tests of acute toxicities of pesticides and heavy metals to the fry guppy, Poecilia reticulata. Jap. J. Public Hlth., 35, 246-54. Nakaminami, G., Ishida, N., Kunimatsu, T. & Tsuruyama, S. (1985). The pollution of Lake Biwa, Yodo River, Chikugo River, Ariake Sea and others by diphenyl ether herbicides. Ecological Chemistry, 8(1), 3-11. (In Japanese.) Tabata, K. (1972). A proposed standard method for the TLm bioassay using the Himedaka. J. Water. Waste (Japan), 14, 1297-303. Watanabe, A. (1960). List of algal strain collections at the Institute of Applied Microbiology, University of Tokyo. J. Gen. Appl. Microbiol., 6, 283-92.
A Freshwater Shrimp (Paratya compressa improvisa) as a Sensitive Test Organism to Pesticides Shigehisa H a t a k e y a m a a & Y o s h i o S u g a y a b a Environmental Biology Division, b Engineering Division, The National Institute for Environmental Studies, 16-20nogawa, Tsukuba, 305 Ibaraki, Japan (Received 23 December 1988; accepted 24 February 1989) ABSTRACT The susceptibility of 2-week-old individuals of thefreshwater shrimp, Paratya compressa improvisa, to five kinds of insecticide and five kinds of herbicide was examined in comparison with that of two species of Cladocera, Daphnia magna and Moina macrocopa. The shrimp was especially sensitive to two organophosphorus insecticides. The 48-h LCso values for fenitrothion and fenthion to the shrimp were 1.15 and 1.04 pg litre-1 (mean value, n = 2), in contrast with 37.8 and35.3 #g litre- 1 in the case ofM. macrocopa, and more than 50 #g litre-~ with D. magna. The shrimp also showed the higher susceptibility to other insecticides, diazinon, carbaryl ( NAC) and BPMC, apart from D. magna to diazinon and NAC. The shrimp also showed higher susceptibility to herbicides. The 48-h LCso values of CNP, benthiocarb, oxadiazon, butachlor, and symetryne to the shrimp were two to eight times lower than those of two species of Cladocera, except for the LCso value of oxadiazon to M. macrocopa, which was very slightly higher. However, the shrimp showed a somewhat lower susceptibility to heavy metals than the two species of Cladocera, especially to copper, and to cadmium and zinc in comparison with D. magna. A bioassay using the shrimp with river water, collected from the river adjacent to the pad@field, showed clearly the high mortality of the shrimp following the aerial spraying with pesticides.
INTRODUCTION A freshwater shrimp, Paratya compressa improvisa, was reported by Kamita (1961) to be widely distributed in lakes, ponds, rivers and/or ditches in the north-east region of the main island of Japan, and large quantities o f the 325 Environ. Pollut. 0269-7491/89/$03.50 © 1989 Elsevier Science Publishers Ltd, England. Printed in Great Britain
326
Shigehisa Hatakeyama, Yoshio Sugaya
shrimp were caught for food at that time in this area. However, the present authors have only limited information on its recent distribution in Japan. This shrimp has been used as a test organism for evaluating the toxicity of river water to aquatic organisms; river water was freeze-condensed using a vacuum-freeze evaporator and used for acute toxicity tests with the shrimp, which is known to be sensitive to pesticides (Kariya et al., 1988). These authors regarded the shrimp as especially sensitive to organophosphorus insecticides, because mortality increased in water taken from the river adjacent to a rice paddy field after aerial spraying with insecticides. However, the relative susceptibility test of the shrimp and other standard test organisms, such as Daphnia magna, to pesticides has not been reported. In the present study, two species of Cladocera, Daphnia magna and Moina macrocopa, were used as reference organisms to evaluate the susceptibility of the shrimp to pesticides and heavy metals. Sensitive test organisms are especially useful when the solubility and/or amount of the test chemicals are limited. We can also reduce the period of toxocity test and amount of waste chemicals using such sensitive test organisms. Agrochemical concentrations in Japanese rivers are usually (Maru, 1985; Nakaminami et al., 1985; Imanaka et al., 1985; Iwakuma et al., 1988) much lower than the 48h LCso values for test fish (Macek & McAllister, 1970; Miyashita, 1988). However, it seems that concentrations of many insecticides and/or herbicides are still at levels very hazardous to many kinds of sensitive aquatic organism during the spraying season for pesticides. We may be able to monitor the seasonal variation in toxic effects of pesticides to aquatic organisms by bioassay using a very sensitive test organism. A bioassay was therefore conducted using P. compressa improvisa to evaluate the variation in toxicity of river water before and after aerial spraying of insecticide on a rice paddy field. Three species of euryhaline grass shrimp Palaemonetes, especially P. pugio, have often been used for toxicity tests of various chemicals (Buikema et aL, 1980). However, reports on toxicity tests using freshwater shrimps are very limited except for those using Gammarus and, rarely, the freshwater shrimp Palaemonetes kadiakensis. The present paper also describes the method of culture and some characteristics of the shrimp, P. compressa improvisa, as an organism for use in acute toxicity tests. MATERIALS AND METHODS Materials Paratya compressa improvisa The first release of offspring as post-zoeae, which grew into shrimps within 14 days, usually occurred at the age of about 4 months at a water
Freshwater shrimp as a test organism to pesticides
327
temperature of 24 + I°C. The size of reproductive females was ca. 30 mm. Each shrimp released 141 + 22 (mean _ SD, n = 20) offspring, ranging from 95 to 236 every 2 weeks in a room controlled at 24 + I°C and 16 L (2500 lux) - 8 D. New eggs were deposited within 2 days after the release of the previous brood offspring. Thirty pairs of adult shrimp were usually reared separately in 16 liters of rearing water using 20-1itre glass aquaria with a piece of the aquatic plant, Elodea densa CASP. The water used was dechlorinated tap-water, originating from Lake Kasumiguara, pH was 7.7 to 7-9 and hardness was ca. 80 mg litre- 1 a s C a C O 3. An adequate amount of a commercial fish food (granular type No. 4C kuranburu, Nihon Haigou Shiryou, Ltd), composed of 64% fish powder, 28% corn powder, 2% corn gluten and 6% other (NaC1, CaPO 4, dried yeast) was provided once a day. About half of the rearing water was exchanged with fresh water, and food residues were removed twice a week. All aquaria were checked each day and pairs that had released offspring were transferred to newly prepared aquaria. Offspring were reared for 2 weeks using the same methods as mentioned above and employed for acute toxicity tests. As the behavior of the shrimp was very rapid and its movements sharp, we used fine netting to transfer them to a test beaker. The body length at age 2 weeks was 5.98 _ 0.49 mm (n = 10). Daphnia magna (delivered indirectly from the USA EPA) and M. macrocopa, which has been cultured for 10 years in our laboratory, were reared in glass beakers (1000ml for the former and 500ml for the latter) using Chlorella sp. as a food in a room controlled at 23 +__I°C. The Chlorella sp. was cultured in a modified MC medium (Watanabe, 1960) for 2 weeks. Finally, the cultured Chlorella was washed with an artificial soft water (Tabata, 1972; Hatakeyama & Yasuno, 1981) by centrifugation (5000 rpm, 20 min, 3 times) to replace the Chlorella culture medium with the artificial soft water because the medium was toxic to the two species of test organism. The Chlorella was adjusted to a density of ca. 1"5 x l0 s cells ml-1 and preserved in a refrigerator for use as a food. Each day adults and their offspring were separated using a glass pipette, and offspring were reared for one day using the Chlorella as a food before the acute toxicity test. The age at the start of exposure was 36 _ 12 h. Test methods
One-hundred millilitre aliquots of pesticide stock solutions (5000 mg litre- 1) were prepared using more than 99% grade chemicals (Wako Pure Ltd) and 99.5% ethanol, and stored in a refrigerator throughout the experiment. Three kinds of organophosphorus insecticide, fenitrothion (dimethyl 4nitro-m-tolyl phosphorothionate), fenthion (dimethyl 4-methylthio-m-tolyl phosphorothionate) and diazinon (diethyl 2-isopropyl-4-6-pyrimidinyl
328
Shigehisa Hatakeyama, Yoshio Sugaya
phosphorothionate) and two kinds ofcarbamate insecticide, carbaryl (NAC, 1-naphthyl N-methyl-carbamate) and BPMC (2-sec-butylphenyl N-methylcarbamate) were used as test chemicals. Currently, these insecticides are widely used in our country. Five kinds of herbicide, chlornitrofen (CNP, 2,4,6-trichlorophenyl 4'nitrophenyl ether), benthiocarb (S-(4-chlorobenzyl) N,N-diethylthiocarbamate), oxidiazon (5-tert-butyl-3-(2,4-dichloro-5-isopropoxyphenyl)1,3,4-oxadiazolin-2-one), butachlor (2-chloro-2',6'-diethyl-N-butoxymethyl) acetanilide), and symetryne (2,4-bis (ethylamino)-6-methylthio-1,3,5otriazine), were used. The acute toxicity test was conducted in a room controlled at 23 _ I°C, and under a 12 h on-off fluorescent light regime. Eight 2-week old shrimps were put in 100 ml of artificial soft water in a 100 ml glass beaker using fine netting, and a stock solution of test chemicals was put into a beaker after appropriate dilution with ethanol and stirred with a glass rod to make a nominal concentration series. Finally, the total ethanol was adjusted to make 0-1% of the test solution for all the concentration series. In the case of D. magna and M. macrocopa, a concentration series of the 200-ml test solution was prepared, and each was divided into three replicate glass beakers with 30ml for D. magna and 20ml for M. macrocopa, respectively. Ten organisms were put in each beaker using a glass pipette, and mortality was checked every 24 h for 4 days, by removing any dead organisms each time. Acute toxicity tests for three heavy metals (Cd as CdC12, Cu as CuSO4 and Zn as ZnSO4) were conducted using the same methods as those described above. All tests on the three species were conducted two or three times, when the 95% confidence intervals of two LCso values did not overlap. The LCso values, associated 95% confidence intervals and regression lines of dosemortality relationships, were calculated using Finney's probit analysis method (1971) programmed for a personal computer. The 48 h LCs0 values for fenitrothion to P. compressa improvisa and M. macrocopa at different ages, 1, 3, 7, 14, 21 and 60 days for the former and 0 (12 + 12h), 1, 2, 3, 4 days for the latter, were obtained to establish the relationship between susceptibility to the insecticide and growth stage. The test was conducted using the same method except for the number of shrimps per test beaker; six individuals of 21-day-old and four indi,ciduals of 60-dayold were put in a 100 ml beaker (with five replicates), respectively.
Bioassay of the river water An organophosphorus insecticide, fenthion, was sprayed aerially over part of a paddy field near our institute together with a fungicide, edifenphos, to
329
Freshwater shrimp as a test organism to pesticides
control rice plant pests on 31 July 1988. River water (500 ml) was collected daily in a 500-ml glass vial from 2 days before to 5 days after the aerial spraying. From the fifth day, the river water was collected at intervals of several days for 3 weeks. River water (100 ml) was decanted into a 100-ml glass beaker with three or four replicates, and eight 14-day-old shrimps were put into the beaker. The mortality of the shrimp was then recorded once a day for 4 days. Ten 2-day-old M. macrocopa were exposed to the river water (30 ml) for 3 days with three replicate glass beakers.
RESULTS Insecticides
The LCso values for fenitrothion were between 0.96 and 1-35/~g litre- 1 in the case of the shrimp at the ages examined (Table 1). The values ranged from 25-8 (value for adults) to 48.7/lg litre-1 in the case of M. macrocopa. The susceptibility to fenitrothion of both species was fairly constant throughout the growth stages examined, although the susceptibility of 4-day-old M. macrocopa, which released offspring during the test period, was about twice that of 2-day-old organisms (Table 1). The shrimp showed higher susceptibility to the insecticides examined than the two species of cladocera, except for the case of D. magna and diazinon TABLE 1
LCso Values (/~g liter-
Agea in days
0 1 2 3 4 7 14 21 28 60
1) of Fenitrothion
to Paratya compressa improvisa and Moina macrocopa 48-h LCso values
Paratya cornpressa improvisa
1.10 (0.72-1.69) [2-20]b 1"00 (0-79-1'25) 1.04 (0.81-1.34) 1.27 (1.05-1.54) 0.96 (0.83-1.11) 1.35 (1.11-1.65)
Moina macrocopa 33.9 (29.7-38.7) 39.7 (34.9-45.1) 48.7 (43.6-54.4) 45.4 (41.6-49.5) 25"8 (18.7-35.6)
[3"07] [2.82] [2.17] [4.45] [3-42]
a Age, 0, 12 __ 12 h, 95% confidence intervals in parentheses. b Slope function of the regression line.
[3-59] [3.61] [4.61] [7.21] [3"62]
BENTH IOCARB
SYMETRYNE
B U TACH LOR
10
- I1--11
1
i
I
I
t
I
i
@
I[
I
".. B _
n~
100 1000 CONCENTRATION (fig 1-1)
=l
IIii
10
I
100
|
10000
II
t
,-o-- ,_D_
,-o-..o_
=
m<~',. .
magna ( - I )
with 95% confidence intervals (bars at both sides of symbols).
Fig. 1. The 48-h LCso values of five kinds of pesticide and five kinds of herbicide to P. compressa improvisa (-C)-), M. macrocopa (-[]-), and D.
n.,
_u O X A D I A Z O N
LM a
or)
CHLORNITROFEN
BPMC
CARBARYL
DIAZ INON
(J l(.,t M.I
.=
FENTHION
u)
FENITROTHION
CHEMICALS
;=p
Freshwater shrimp as a test organism to pesticides
331
(Fig. 1). The mean 48-h LCso values (n = 2) for fenitrothion and fenthion to the shrimp were 1.15 and 1.04, respectively. The shrimp was about 30 times more sensitive to both chemicals than M. macrocopa based on the 48-h LCs o values for both species. It was difficult to obtain the LCso values of fenitrothion and fenthion to D. magna, because the dose-response (mortality) relationship was not clear with the concentrations used (up to 100#g litre- 1) in the present toxicity test. The individuals which had been knocked down at the higher concentrations vibrated their limbs more intensely, although they could not swim. The ECso value (immobility as the criterion) was estimated to be about 50 #g litre-1 to provide a reference (Fig. 1). The susceptibility of the shrimp to diazinon was about 1.5 times greater than that of M. macrocopa. The shrimp was 10 and 28 times more susceptible to carbaryl and BPMC than M. macrocopa, and 10 times more susceptible to BPMC than D. rnagna. However, exceptionally, the shrimp showed somewhat higher (ca. x 2) susceptibility to diazinon and carbaryl than D. magna in the 48 h acute toxicity test.
Herbicides The 48-h LCso values for five kinds of herbicide to three species of test organism are also shown in Fig. 1. Chlornitrofen (CNP) was the most toxic to all three species. The mean 48-h LCso values (n = 2) of C N P to P. compressa improvisa, M. macrocopa and D. magna were 54.4, 413 and 424 #g litre- 1, respectively. The ratios of the shrimp's LCso value to that of the two cladocerans were 7.6 and 7.8, respectively. The shrimp was also 2.2 to 8 times more susceptible to the three herbicides, benthiocarb, oxadiazon and butachlor, than the two species of Cladocera, except for approximately the same susceptibility of M. macrocopa to oxadiazon. Symetryne was the least toxic to the three test organisms. The 48 h LCso values to the shrimp, M. macrocopa, and D. magna were ca. 14.7, 32 and more than 50 mg litre-1, respectively. The ratios of the 24-h LCs0 values of ten kinds of chemical at 48, 72 and 96 h are shown in Fig. 2 for the shrimp and M. macrocopa. In the case of the shrimp, the LCso values decreased to less than 50% of the 24-h LCso values at 48 h with almost all the pesticides (Fig. 2). In contrast, the 48 h LCso values of M. macrocopa were more than 50% of the 24-h LCso with all nine chemicals. The 96-h LCso values for N A C and BPMC to the shrimp decreased to less than 10% of the LCso values at 24 h.
Heavy metals The results of the acute toxicity test for three metals to P. compressa imorovisa and the two cladocerans are shown in Table 2 as 48-h LCso values.
o
PER
CENT 0
PER
CENT 0
O0
~" o
3O
-
•
~
0 Z
-
D
@
"r
Er~
i>
t-gr~ ?
W
,-
0
•
Y
N 0 ;[
Z
cS ./
: ,~
o/
S n~fnSnS o1.~so,~'num,fa~pl~Hns~.tla,~!~/S
~
Freshwater shrimp as a test organism to pesticides
333
TABLE 2
48-h LCso Values (#g liter- x) of Heavy Metals to Three Species of Aquatic Organisms Metals
Cd Cu Zn
P. compressa improvisa
Moina macrocopa
Daphnia magna
59-2 (39.9-87'8) [3'16] a 71.4 (63-7-80-0) [6-12] 77"5 (68.0-88.4) [4.99] 80"4 (71"9-96'4) [5"42] 26.7 (18-5-38'3) [2.04] 5"90 (5.60-6.22) [16.7] 32"6 (25.7-41.4) [2.13] 7"45 (7"13-7"77) [21"7] 879 (706-1094) [2'48] 1 320 (880-1 970) [2"36] 722 (586-899) I"2"16] 1 643 (1 186-2 277) [2'02]
28.6 (20.1-40.9) [2'37] 22.4 (15-5-32.4) [1'82] 3.53 (3.17-3.92) [6.68] 4.32 (3"77-4"96) [6'08] 115 (89-147) [2-63] 124 (96-165) [2.94]
95% confidence intervals in parentheses. Slope function of the regression line. T h e L C s o values f o r Cd, C u a n d Z n to the s h r i m p were 68.4, 29-7 a n d 801 #g l i t r e - ~ (n = 2), respectively. T h e susceptibility o f the s h r i m p to c a d m i u m was a b o u t 2.7 times less t h a n t h a t o f D . m a g n a , a n d a l m o s t the s a m e as t h a t o f M. m a c r o c o p a . T h e s h r i m p s h o w e d 4"4 a n d 7"5 times less susceptibility to c o p p e r a c c o r d i n g to the 48-h L C s o r a t i o s t h a n M . m a c r o c o p a a n d D. m a g n a . T h e TABLE 3
Mortality (%) of the Shrimp in River Waters Collected from the River Hanamuro Adjacent to Paddy Field Before and After Aerial Spraying of Insecticide (Fenthion) (1988) Time (h) o f exposure Date
July 29 30 31a August 1 2 3 4 5 8 13 18 19
24
48
72
96
0 0 100 78 (10) 71 (3) 59 4 0 4 0 0 0
0 0 100 97 (13) 92 (100) 100 21 4 4 0 0 0
0 4 100 100 (73) 100 (100) 100 96 17 8 0 0 0
6 13 100 100 100 100 100 63 33 4 12 0
Test was conducted using three or four replicate beakers each containing eight 14-day-old shrimps. Mortality of M. maerocopa is shown in parentheses. a Insecticide (fenthion) was aerially sprayed.
334
Shigehisa Hatakeyama, Yoshio Sugaya
susceptibility of the shrimp to zinc was also much less than that ofD. magna, although it was slightly higher than that of M. macrocopa (Table 2). Bioassay of the river water
There was no mortality of the shrimp after 2 days in river water collected during the 2 days prior to aerial application of the pesticides, fenthion together with edifenphos. However, the mortality of the shrimp increased to 100% after 4 days in the water during the 5 days following aerial spraying (Table 3). Subsequently, the mortality of the shrimp 96 h after being put into the river water decreased gradually to 63% and 0% on the seventh and twentieth days after the aerial spraying (Table 3). On the other hand, the mortality of M. macrocopa was observed only on the 1st and 2nd days in August. DISCUSSION As mentioned above, test organisms highly sensitive to chemicals are very useful in acute toxicity tests. The shrimp used in the present study showed fairly constant susceptibility to fenitrothion in the growth stages from 3 to 60 days old. Mortality of the control was almost always zero for the 4-day test period, because the shrimp did not show internecine struggle, although cannibalism often occurred against weak, moribund and/or dead organisms. The 14-day-old shrimps never jumped out of the test beaker, but this problem often occurred with 60-day-old shrimps. Mortality criteria were easier to check in the case of the shrimps, because the translucent body of the live shrimp changed to a dark color when observed through a binocular microscope using reflected light, and cessation of thoracic appendages and/or pleopods movement was also a good criterion for dead organisms. However, there are several disadvantages in using the shrimps as test organisms. In the case of P. compressa improvisa, the first release of offspring occurred at 4 months, in contrast to several days to 2 weeks in the case of cladocerans. The interval of offspring release was 2 weeks under the present rearing conditions. Productivity of young is lower in shrimps than in cladocerans. Handling of shrimps is difficult because they move very quickly, so we used fine netting to handle them instead of the glass pipette used for cladocerans. The susceptibility of the shrimp P. compressa improvisa to the pesticides used in the present paper was usually higher than that of the two species of cladocera. The 48-h LCso values for 2-day-old guppy Poecilia reticulata
Freshwater shrimp as a test organism to pesticides
335
(Miyashita, 1988), which has been used in our laboratory as an experimental organism for toxicity tests, are 47 to 3700 (fenitrothion) times higher than the values found with the five kinds ofinsecticide used in the present study, while the LCso values to the guppy of the five kinds of herbicide used here are 1.3 to 16.9 times those to the shrimps. The 48-h LCso values for the shrimp decreased to less than 50% of the values at 24 h, while in the case of guppy (Miyashita, 1988), even the 96-h LCso values were more than 50% of the 24-h LCs0 values except for three chemicals, carbaryl, BPMC and benthiocarb. The present result (Fig. 2) and the result cited above suggest that the high susceptibility of the shrimp to pesticides will be even more marked in chronic toxicity tests. The susceptibility of the shrimp to two kinds of organophosphorus insecticide was very high, the 48-h LCso values for fenitrothion and fenthion being around 1 #g litre- 1. Organophosphorus insecticide in the river water, such as those mentioned above, often exceeded 1 #g litre- 1 for several days or longer periods during the insecticide spraying season from June to August in agricultural fields (Iwakuma et al., 1988). Mortality of the shrimp, P. compressa improvisa, actually increased in the river water on the day the insecticide was sprayed on the paddy field, although a sensitive test fish, Tanichys albonubes, was rather insensitive in the bioassay of river water developed by Kariya (Kariya et al., 1988). These results indicate that the shrimp is especially susceptible to insecticides. However, the susceptibility of the shrimp to heavy metals (Cd, Cu, Zn) was lower than that of the two species of Cladocera, especially D. magna. These results indicate that the shrimp is a promising test organism for bioassays, especially to pesticides in the water around agricultural fields, without the need for condensing the water by freeze-evaporation (Kariya et al., 1988). Bioassays using 14-day-old shrimps of river water adjacent to the rice paddy field strongly suggest that the habitat of the shrimp may be restricted severely in the pesticide spraying seasons by pesticides such as organophosphorus insecticides. Plenty of the shrimp used to be netted as a freshwater fishery product from ponds and ditches in several districts of north-east Japan (Kamita, 1961). However, attempts to obtain shrimp catch are scarcely conducted at the present time (pers. comm.) because its habitat has decreased severely. The present acute toxicity and bioassay results suggest that there may be a causal relationship between the disappearance of the shrimp and the presence of pesticide in the ponds and/or ditches adjacent to agricultural fields. At present we can see the shrimp in the fairly restricted ponds or small streams not polluted with pesticides and/or other types of chemical. Further bioassays at several river sampling points throughout a year, including the herbicide spraying seasons, are now in progress, and the results will be published in later papers.
336
Shigehisa Hatakeyama, Yoshio Sugaya ACKNOWLEDGEMENTS
The present authors are very grateful to Mr Y. Ogaminoa, Mr K. Kawabe, and M r Y. N a k a y a m a , who provided highly competent and consistent assistance throughout the course of the experiment. We are also very grateful to Dr T. Kariya (Professor of T o h o k u University at that time), who kindly distributed the shrimp to our laboratory in 1982, and gave valuable suggestions on rearing methods.
REFERENCES Finney, D. J. (1971). Probit Analysis, Cambridge University Press, London. Buikema, A. L., Niederlehner, B. R. & Cairns, J. Jr (1980). Use of grass shrimp in toxicity tests. In Aquatic Invertebrate Bioassays. ed. A. Buikema & J. Cairns, Jr, pp. 155-73. Hatakeyama, S. & Yasuno, M. (1981). Effects of cadmium on the periodicity of parturition and brood size of Moina macrocopa (Cladocera). Environ. Pollut., 26, 111-20. Imanaka, M., Hino, S., Matsunaga, K. & Ishida, T. (1985). Oxidiazon residues in surface water and crucian carp (Carassius cuvieri) of Lake Kojima. J. Pesticide Sci., 10, 125-34. Iwakuma, T., Shiraishi, H., Nohara, S. & Takamura, K. (1988). Pesticide residues in water and sediment of the River Koise and its tributaries. Res. Rep. Natl. Inst. Environ. Stud. Jpn., 114, 73-83. (In Japanese, Abstract in English.) Kamita, T. (1961). Studies on the freshwater shrimps, prawns and crayfish of Japan. (In Japanese, Abstract in English.) Published by Sonoyama Book Ltd, Matsue, Shimane, Japan. Kariya, T., Oouchi, K., Suzuki, A., Niwa, T. & Sato, M. (1988). A new bioassay method to detect low-level toxicity of waters. Proceedings of the 4th IUBS International Symposium on Biomonitoring of the State of the Environment (Bioindicators), Tokyo, 1987. Biological Monitoring of Environmental Pollution. ed. M. Yasuno and R. B. A. Whitton. Tokai University Press, Tokyo, pp. 23-31. Macek, K. J. & McAllister, W. A. (1970). Insecticide susceptibility of some common fish family representatives. Trans. Am. Fish Soc., 59, 20-7. Maru, S. (1985). Monitoring survey of pesticides in river water within Chiba Prefecture. Ecological Chemistry, 8(3), 3-10. (In Japanese.) Miyashita, M. (1988). Tests of acute toxicities of pesticides and heavy metals to the fry guppy, Poecilia reticulata. Jap. J. Public Hlth., 35, 246-54. Nakaminami, G., Ishida, N., Kunimatsu, T. & Tsuruyama, S. (1985). The pollution of Lake Biwa, Yodo River, Chikugo River, Ariake Sea and others by diphenyl ether herbicides. Ecological Chemistry, 8(1), 3-11. (In Japanese.) Tabata, K. (1972). A proposed standard method for the TLm bioassay using the Himedaka. J. Water. Waste (Japan), 14, 1297-303. Watanabe, A. (1960). List of algal strain collections at the Institute of Applied Microbiology, University of Tokyo. J. Gen. Appl. Microbiol., 6, 283-92.