Acute Marine Sediment Toxicity: A Potential New Test with the AmphipodGammarus locusta

ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY

40, 81—87 (1998)

ENVIRONMENTAL RESEARCH, SECTION B ARTICLE NO.

ES981646

Acute Marine Sediment Toxicity: A Potential New Test with the Amphipod Gammarus locusta F. O. Costa, A. D. Correia, and M. H. Costa Dep. de Ciencias e Eng. do Ambiente, FCT/Universidade Nova de Lisboa, 2825 Monte de Caparica, Portugal Received August 28, 1996

composition is affected in the contaminated areas, but because there are no data on the toxicity of those sediments, it is not known whether this is related to contamination or other disturbing factors (Costa, 1989; Quintino et al., 1995). Sediment bioassays have proven to be a powerful tool in studying sediment-related toxicity and are recommended along with other methodologies to obtain information on the ecological impact of contaminated sediments (Chapman and Long, 1983; Luoma and Carter, 1993). They can also be used to determine the LC values of toxicants in laborat50 ory-contaminated sediments, from which information for sediment quality criteria may be derived (Bolton et al., 1985, in ASTM, 1992). The former approach recently underwent a major development, in the United States, where standardized methodologies for sediment toxicity testing are available for a series of freshwater species (ASTM, 1993) and for five species of marine and estuarine amphipods (ASTM, 1992). Furthermore, bioassays are currently the only way to assess the potential toxicity of field-contaminated sediments (SETAC—Europe, 1993). However, few tests have been developed with European species. For testing using marine sediments the SETAC—Europe guide (1993) recommends the use of amphipods, ‘‘preferably Corophium volutator or other locally available amphipods, particularly where routine, standard test methodologies already exist.’’ In the Portuguese coastal systems, namely, the Sado estuary, C. volutator is not available in large numbers; therefore, a new test was developed with the indigenous amphipod Gammarus locusta (Ribau and Costa, 1994; Correia et al., 1995; Costa et al., 1996). The studies performed include the setting up of a culturing system and assessment of amphipod sensitivity to selected noncontaminant factors, to contaminants, and to some field-contaminated sediments. The integration of the results from this series of tests culminated in the definition of an experimental protocol to conduct acute sediment toxicity test with the amphipod Gammarus locusta. This article summarizes the information concerning the sensitivity of this amphipod and discusses the implications to the sediment toxicity testing.

Although amphipod toxicity tests have been successfully used in the United States to assess coastal sediment toxicity, few tests have been developed with European species. The authors have been working with the amphipod Gammarus locusta, a widely spread species along European coastal areas that is particularly abundant in the Portuguese Sado estuary. This amphipod fulfills the most important requirements of a test species. It can be easily reproduced in laboratory and it is tolerant to a broad range of sediment types. A series of tests demonstrated its sensitivity to copper and c-hexachlorocyclohexane (lindane) in the sediment (LC50 5 6.8 mg Cu/dry kg, 0.9% total volatile solids; LC50 5 60.5 lg HCH/dry kg, 2% total volatile solids) and to some heavily contaminated field sediments. After assessment of the species sensitivity to several noncontaminant variables, an experimental protocol was designed to conduct acute sediment toxicity tests that are briefly described. Proposed is a 10-day static toxicity test at 15°C and 33–34& salinity, with laboratory-produced juveniles and mortality as the endpoint. General assay performance is identical to the American Society for Testing and Materials (ASTM) standard for sediment toxicity tests with marine and estuarine amphipods. The results previously obtained revealed a strong potential for this amphipod to be used in toxicological testing. Considering the wide geographic distribution of this species and its amenability for culturing, it may be an alternative or complementary test for ecotoxicological studies in other European coastal systems where the existing tests cannot be applied or do not offer a definitive solution. ( 1998 Academic Press

INTRODUCTION

The Sado estuary, a major estuarine system located on the western coast of Portugal, has been exposed to multiple anthropogenic impacts, including urban, agricultural, and industrial contamination and pollution. Sediment heavy metal contamination is particularly high in some areas, where up to 240.5 lg g~1 of copper has been measured (Cortesa8 o and Vale, 1993; Caeiro et al., 1994). Benthic community surveys indicated that sediment macrofaunal 81

0147-6513/98 $25.00 Copyright ( 1998 by Academic Press All rights of reproduction in any form reserved.

82

COSTA, CORREIA, AND COSTA

METHOD DEVELOPMENT

Test Species Gammarus locusta (L., 1758) is a free-living epibenthic amphipod with a wide geographic distribution, from the North Atlantic coastal areas to the Mediterranean Sea (Chevreux and Fage, 1925). It has been found in numerous systems of the Portuguese coast, in almost any substrate from sand to silt and also in rocky substrates associated with macroalgae (predominantly Chaetomorpha sp., Enteromorpha sp., Fucus spp., and ºlva spp.), and occurs on the medium to low intertidal zone at salinities higher than 17& (Marques, 1989). It feeds on macroalgae, especially ºlva spp. and Enteromorpha sp., and on sediment detritus. In test chambers it roams the sediment surface, picking up pieces of organic material or scraping the surface of mineral particles, probably for microflora. In muddy sediments, G. locusta eventually burrows into the sediment surface. Cannibalism may occur when there is a victim in disadvantage (personal observations). The collection side is located on the lower part of the south margin of the Sado estuary (38°27@ N, 08°43@ W) free of direct exposition to contaminated effluents that are confined to the north margin. The level of toxicants in the sediments is low and thus it was assumed to be a nontoxic reference site for the toxicity tests. The water temperature and salinity range from 12 to 22°C and 30 to 36&, respectively (Costa, 1989). The surface sediment, which is used as culture and negative control sediment, is a fine to medium sand with 1—2% total volatile solids (TVS). Numerous stones are dispersed all over the area and are used by the macroalgae as a substrate on which to develop. The amphipods are collected by collecting those macroalgae, especially ºlva spp., during low tide. G. locusta can be found there throughout the year, but is particularly abundant during the late spring, when densities up to 1200 individuals per square meter have been recorded (Correia et al., 1995). If the macroalgae are sparse or not present at all, G. locusta can still be found under the stones, although in low densities and with a quite specific and patchy distribution (personal observations).

rigida and, eventually, Enteromorpha sp. With only two microcosms (&18 liters), containing about 200 adults each, a large number of animals are continuously available for testing. General Test Design and Procedure The general design of this test is based largely on the standard guides of ASTM (1992) and SETAC—Europe (1993). The tests are conducted at 15$1°C and 33—34& salinity and under a 12-h photoperiod. These conditions were selected according to the range of the values of the parameters annually registered in the Sado estuary. The test is performed in static systems for 96 h or 10 days for water or sediment exposures, respectively. The experimental procedure for the sediment toxicity tests is outlined in Fig. 1, and a brief description is provided below. With the exception of the addition of contaminants

Rearing and Breeding of the Test Species G. locusta are cultured at room temperature in plastic microcosms with 0.45-lm-filtered seawater of 33—34& salinity and a sediment layer up to 1 cm. To provide shelter and mimic the natural environment, small stones are placed in the microcosms. Twice a week the water is sieved through a battery of screens of decreasing mesh size (1500, 1000, 475, and 250 lm), distributing the amphipods into four size classes (adults, subadults, juveniles, and newborns). These are allocated to their respective size class microcosms. Food consists mainly of the macroalgae ºlva lactuca and ºlva

FIG. 1. Summary of the method used to conduct acute sediment toxicity tests with Gammarus locusta. (1) Only for sediments to be spiked in the laboratory. (2) For negative control, reference sediment and field-contaminated sediments.

MARINE AMPHIPOD TEST FOR ACUTE SEDIMENT TOXICITY

to the test sediment, the procedure is followed for every sediment used in the assay, including control sediment and field-contaminated sediments. The tests are conducted with juveniles (2—4 mm length class) produced in laboratory. At least 24 h before the beginning of the test a substock of animals are isolated from the cultures and acclimated to the assay water temperature with unlimited food conditions. Twenty amphipods are randomly allocated to the test vessels, which may have 1- or 2-liter capacity (see discussion under Organism Density). Three to five replicates per test concentration or test sediment are used. When the experiment is finished the contents of the test chambers are sieved through a 475-lm screen and the number of amphipods is recorded as alive or dead. The test sediment is collected on the same site as the culturing sediment. The sediments are sieved through a 1500-lm screen, to remove macrofauna, and stored at 4°C for a maximum of 72 h before initiation of the test. The spiking procedure is the one described by Swartz et al. (1985) with minor modifications. The contaminants are directly added to the sediment to achieve the required nominal concentrations (expressed on a dry weight basis). This is followed by 15 min of mechanical mixing. The sediments are then allowed to equilibrate overnight (15—20 h) at 4°C. Samples for chemical analysis are collected prior to sediment placement in the beakers. The sediment layer is about 1 cm. Seawater is then added gently, to minimize sediment resuspension, and after aeration is provided, the sediment—overlying water system is allowed to equilibrate overnight. The assay starts the next day when the amphipods are allocated to the beakers. Temperature and aeration are monitored daily. Controls. Several controls may be needed. These include a negative control, a reference control sediment, a positive control, and a solvent control. Detailed definitions and explanations about the application of these control treatments are available in ASTM (1992) and SETAC—Europe (1993). Response Criteria The response criterion is survival. Dead animals can be identified by their discoloration, absence of pleopod movements, or lack of response to an external mechanical stimulus. Missing animals are assumed to have died and decomposed. This assumption was verified for the amphipod Rhepoxynius abronius by Swartz et al. (1985), and is likely to be true for other amphipod tests. Negative Control Survival/Test Acceptability For tests conducted in 2-liter plastic beakers, control survival averaged 94% in four assays (range, 98—90%). For

83

the tests performed in 1-liter glass beakers, control survival averaged 92% (range, 90—95%). Overall mean control survival was 93% for seven tests, which is within the common range of control survival for the majority of acute sediment toxicity tests with amphipods (ASTM, 1992). Based on this set of data a minimum of 90% control survival is proposed to accept the test as ‘‘regular.’’ If a lower average survival is obtained, and this is a consistent response among replicates, it is advisable to suspect insufficient ‘‘health condition’’ of the test animals. SENSITIVITY TO NONCONTAMINANT VARIABLES

Identification and quantification of the relationship between mortality and natural environmental variables are a priority in the development of any toxicological assay, and should be conducted before responses can be ascribed to contaminant effects (Buikema and Benfield, 1979; Swartz et al., 1985; DeWitt et al., 1988; ASTM, 1992; Thomas, 1993). Among the noncontaminant variables, sediment features, temperature, salinity, and organism density were suspected to be critical in G. locusta sediment toxicity testing. Sediment Properties (Granulometry and Organic Matter) Effects of natural sediment features on the survival of G. locusta were examined for a 10-day exposure to a range of sediment types from 0.5 to 100% fine fraction corresponding to 1.7 and 17.0% TVS, respectively (Costa et al., 1996). Sediment particle size and sediment organic matter content (expressed as percentage TVS) could not be manipulated independently due to the inherent technique difficulties, already underlined by DeWitt et al. (1988). The sediments were prepared by mixing the required proportion of the fine fraction of a clean organically enriched sediment with the control sediment. Consequently by manipulating the sediment particle size a range of percentage TVS was obtained. No differences were found in G. locusta survival among the five types of sediment tested. These results are in agreement with the field-observed tolerance of this amphipod to a wide range of sediment types, from sand to silt (Marques, 1989). Temperature and Salinity A 10-day sediment test was conducted in which G. locusta was submitted to a series of overlying water salinities and temperatures. The saline solutions were prepared by diluting the water from the reference site with double-distilled water to obtain 50, 25, 12.5, and 6.25% solutions of the field water. No gradual adaptation of the animals to salinity was performed. They were transposed directly from 100% of the initial salinity to the different treatments. The temperature was gradually achieved in the test chambers.

84

COSTA, CORREIA, AND COSTA

TABLE 1 G. locusta Survival for Combination of Temperatures and Salinities on a 10-Day Assay Mean amphipod survival (%$SD) Water salinity (&) 2.5 5 9.5 19 34

15°C

20°C

25°C

0 7$6 40$10 97$6 93$6

0 7$12 47$6 77$6 90$10

0 0 0 67$6 90$10

The results revealed that survival was high (590%) at a salinity of 34&, at any of the temperatures tested (15, 20, and 25°C). However, for a salinity of 19&, survival decreased to 77 and 67% at 20 and 25°C respectively, but was still high at 15°C. Below 19&, survival decreased, with no surviving amphipods at 2.5& (Table 1). The tolerance limit of the species should thus be close to 19&. These results are consistent with the known distribution of the species on the Portuguese coast, where it was not found below 17& salinity. The application of this test is consequently confined to sediments from the lower portions of the estuaries with high salinity or from the coastal zone. However, this experiment did not allow a gradual adaptation of the amphipods to lower salinities, and it is not known whether with a gradual adaptation period this species would have a higher survival at salinities lower than 19&. Organism Density The effect of organism density is one of the factors to consider when developing toxicity tests (Buikema and Benfield, 1979; DeWitt et al., 1992). Information on cannibal habits in amphipods, especially in Gammarus spp. (Sexton, 1928; Borgmann et al., 1989), raised the suspicion that this behavior, coupled with crowding, would be a critical factor in G. locusta assays. It is known that Gammarus spp. is an alga and deposit feeder, and cannibalism has been sporadically observed in culture, although no animal has been observed eating the dead bodies of its conspecifics. The first toxicity tests with G. locusta were conducted in 2-liter plastic containers containing 20 amphipods each, leading to a density of about 14 amphipods dm~2. Recent tests with organic contaminants and field-contaminated sediments were conducted in 1-liter glass beakers, with the same 20 amphipods per replicate, which increased the density to about 25 animals dm~2. To verify if control survival in sediment tests was affected by the organism density alteration, a simple test was conducted. The effects of

starving on control survival were also ascertained in the same assay. Four different treatments were performed using the two vessel types, with and without feeding, in quadruplicate. Food consisted of ºlva lactuca disks previously weighed. Mean control survival did not differ between the two types of beakers, whether with or without feeding, indicating that population density did not interfere with the test results. Subsequent experiments confirmed these results. On some occasions, survival was abnormally low in the control treatments, but consistently low in both vessel types. With respect to the starving factor, survival in the presence of food was substantially higher in both cases, with 100% survival for fed animals against 90% for the non-fed animals. Moreover, the average weight at the end of only 10 days was several times higher for the fed animals. Thus, apparently the 10% control mortality was due mainly to starving. It is probable that starving stimulates cannibalism, but there is no information that can clarify this. These results question again whether starving may contribute to higher sensitivity to contaminants during acute tests. This might be the case for other test species, but this factor has rarely been analyzed (Luoma and Ho, 1993), probably because of the ‘‘consensus’’ that acute tests should be performed without feeding. Nevertheless, the absence of food does not preclude a high level of control survival in G. locusta acute tests. SENSITIVITY OF G. locusta TO CONTAMINANT VARIABLES

This section summarizes what was found concerning G. locusta sensitivity to toxicants. Some results of the sensitivity to heavy metals presented here have been described by Costa et al. (1996). The results in this section are valid for 15$1°C and 33—34& salinity and for the sediments types tested (Table 2). TABLE 2 LC50 Values for G. locusta Exposed to Metals in Seawater and to Spiked Sediments

Toxicity test

LC 50 (95% confidence interval)

0.85$0.4 mg Cd liter~1 a (mean of two tests) 96 h copper in seawater 0.3 mg Cu liter~1 b (0.17—0.52) 10 days copper in sediment (0.9% TVS) 6.8 mg/dry kgb (5.8—8.3) 10 days lindane in the sediment (2.2% TVS) 60.5 lg HCH/dry kga (49.1—77.5) 96 h cadmium in seawater

a Nominal LC . 50 b From Costa et al. (1996).

85

MARINE AMPHIPOD TEST FOR ACUTE SEDIMENT TOXICITY

Cadmium and Copper in Seawater The toxicity of cadmium in seawater was assessed as a reference toxicant control, to establish whether the sensitivity of the test animals was consistent among the experiments. The cadimum LC values were 0.6 mg liter~1 (95% 50 confidence interval: 0.43—0.82 mg liter~1) and 1.1 mg liter~1 (95% confidence interval: 0.55—1.64 mg liter~1) at 15°C and 33& salinity. These values differ by almost twofold, but variations in sensitivity to toxicants, namely, cadmium, have been observed in other amphipods. For the amphipod R. abronius, eventually the most frequently used amphipod, variations up to fivefold in the cadmium LC have been 50 measured (Hichey and Roper, 1992). Differences in the sensitivity to contaminants between amphipod species have also been examined and are usually higher than differences within the same species (Hong and Reish, 1987, in Hichey and Roper, 1992; Meador et al., 1993). Data on amphipod exposure to cadmium in water at 15°C and 28& salinity indicate a range of LC values from 50 0.79 to 11.4 mg liter~1 (DeWitt et al., 1992). The G. locusta LC values are in the lower part of this range, indicating 50 that this amphipod is apparently a sensitive species. However, these considerations rely on the comparison of tests with a 5& disparity in their salinities, and this parameter is known to affect cadmium toxicity (De Lisle and Roberts, 1988). The acute toxicity of copper in seawater was also determined and the 96-h LC of G. locusta was 0.3 mg liter~1 50 (95% confidence interval: 0.17—0.5 mg liter~1). The lowest concentration at which survival differed from control survival (LOEC) was 50 lg Cu liter~1. Copper in the Sediment The 10-day LC of copper-spiked sediments (0.9% TVS) 50 was 6.8 mg/dry kg (95% confidence interval: 5.8—8.3 mg/dry kg) (Costa et al., 1996). The concentrations of copper measured in the water phase were low. The highest value was 20 lg liter~1, which is still less than half of the LOEC in the water-only exposure (50 lg Cu liter~1). This indicates that the amphipods responded to the sediment contamination and did so in a dose-dependent manner. Hexachlorocyclohexane in the Sediment c-Hexachlorocyclohexane (c-HCH), also referred to as lindane, is one of compounds suspected of having been used as a pesticide in agricultural activities in the Sado estuary. Because of this and the fact that there is available information about the toxicity of this pesticide to the amphipod Corophium volutator (Guerra et al., 1995), lindane was selected for use in assessing G. locusta sensitivity to organic compounds in the sediment. The nominal 10-day LC was 50

TABLE 3 G. locusta Survival after 10-Day Exposure to Field Sediments Collected near Effluents from the Agriculture–Fertilizer Industry (Site 1) and Pulp Mill Industry (Site 2) Surface sediment source Negative control Reference control Site 1 Site 2

% TVS

Mean G. locusta survival (%$SD)

1.5 4.6 3.6 13.2

90$4.8 97$2.7 40$7.1 94$4.2

60.5 lg/dry kg (95% confidence interval: 49.1—77.5 lg/dry kg), for a sediment with 2.2% TVS. Field-Contaminated Sediments The sensitivity of G. locusta to field sediments was evaluated for intertidal surface sediments close to industrial effluents in the north margin of Sado estuary. Two sites were selected in the vicinity of effluents from the (1) agriculture— fertilizer and (2) pulp mill industries. These sediments are known to be contaminated with a mixture of toxicants, especially heavy metals (Caeiro et al., 1994; unpublished data), but the contribution of these industrial effluents to this contamination is not known. G. locusta did not tolerate sediments from site 1 for which survival was very low (40%) compared with control and reference control survival. However, survival registered for sediments from site 2 did not differ from survival in the negative control and reference sediment (Table 3). Results obtained on the sediments from site 1 corroborate the information available on the macrobenthic communities in the vicinity of these industrial effluents. The benthic communities at site 1, where sediment was acutely toxic to G. locusta, are quite poor. However, the absence of acute sediment toxicity at site 2 does not combine with the even poorer diversity of its benthic communities. Previous benthic community surveys revealed a significant decrease in number and abundance of taxa from unpolluted areas of the south margin to the north margin of Sado estuary, which particularly evident in the proximity of pulp mill industry effluents (Quintino et al., 1995; Mucha and Costa, 1996). The sediments near those effluents have a very high organic content, which is responsible for its reduced conditions and probably for low contaminant availability. These results confirm the fact that sediment toxicity tests should not be the sole criterion to determine the health of benthic ecosystems. DISCUSSION

According to the SETAC guide (SETAC—Europe, 1993) the requirements for a sediment test species are sensitivity,

86

COSTA, CORREIA, AND COSTA

ecological relevance, amenability, database, sediment tolerance, and method availability. Sensitivity The output of a sediment toxicity test is a function of the test organism’s sensitivity and the degree of its exposure to contamination. To ensure that exposure is high it is common to use infaunal organisms in sediment tests. However, although an infaunal organism is undoubtedly more exposed to sediment contaminants, it is not necessarily more sensitive to toxicants than an epibenthic species. Amphipods exhibit a high degree of habitat diversity and niche requirements (Thomas, 1993), from free burrowing to tube building or epibenthic, and they have all been used to assess sediment toxicity. One freshwater epibenthic amphipod, Hyalella azteca, has been found to be one of the freshwater species most sensitive to sediment toxicants (Burton, 1991). Epibenthic species may also be useful for bioassay batteries where species with distinct ways of exposure to contamination are required (Meador et al., 1993; SETAC—Europe, 1993). Current results indicate that G. locusta responded in a dose-dependent manner to sediment contamination in acute exposures. However, this is unique information, since no published data concerning G. locusta sensitivity to toxicants were found. Definitive conclusions about G. locusta sensitivity to sediment contaminants comparatively with other species can be drawn only by simultaneous experiments with the same sediment. Previous experiments with G. locusta showed that the 10-day LC for copper in the sediment may increase up 50 to eight times in a sediment with a two times higher percentage TVS (Costa et al., 1996). Ecological Relevance Amphipods are ecologically important organisms, being one of the major components in biomass and diversity of marine (Oakden et al., 1984; Thomas, 1993) and freshwater (Borgmann and Munawar, 1989) systems worldwide. They are a primary food source for fish and voracious feeders of animal, plant, and detrital material (Borgmann and Munawar, 1989; Pennak, 1989, in Burton, 1991). The information on the ecological importance of G. locusta is minimal and circumstantial, but it is likely that it shares the ecological importance of the amphipod group. It is broadly distributed along European coastal areas, as already mentioned, and seems to play an important role in the Sado estuary, at least with respect to its abundance (Correia et al., 1995). Amenability This species has an excellent amenability for culturing and laboratory testing, which seems to be a common at-

tribute of amphipods of the family Gammaridae (Sexton, 1928; Burton, 1991). Culturing is advantageous in toxicity testing since animals are readily available and life cycles are well known and can be controlled (Luoma and Ho, 1993). G. locusta life cycle is reproduced by a simple system developed in the laboratory and a large number of individuals are available in sufficient quantities for toxicity tests. The animals are easy to handle, requiring rudimentary and relatively inexpensive facilities and equipment available in the most biological laboratories. Database The lack of information about G. locusta is probably the major handicap of this test. Some biological data exist (Sexton, 1928; Stock 1967; Jazdzewski, 1973, in Nelson, 1980), but no toxicological data have been found. Sediment Tolerance The tolerance of G. locusta to a broad range of sediment types is one of its interesting qualities that is critical for testing the toxicity of estuarine sediments. Modification of the sediment features, within the tested range, will not affect survival in acute assays with G. locusta. Nevertheless, a control reference sediment must be run when the tested sediments are outside that range. Method Availability This acute test follows, as far as possible, the guidelines from ASTM (1992) and SETAC—Europe (1993). The general test procedure is identical to ASTM (1992) standards, and consequently the method can be readily adopted by any laboratory. Further, its design allows the performance of comparisons with other species or the inclusion in bioassay batteries. CONCLUSION

Few standardized sediment toxicity tests using European species are available. The existing tests can hardly be used in every situation because test organisms are not available or are not ecologically relevant for the natural system being investigated. These limitations led to the development of a test for Portuguese coastal systems, particularly for the Sado estuary, using the indigenous species G. locusta. This species is ecologically relevant in the Sado estuary, is known to be sensitive to sediment contamination, is tolerant to a broad spectrum of sediment types, and has excellent amenability for experimentation. The fact that the design of the test was based largely on existing guidelines makes it suitable for comparisons. The sum of these attributes defines a great potential to be explored in ecotoxicological research.

MARINE AMPHIPOD TEST FOR ACUTE SEDIMENT TOXICITY

87

Considering the wide geographic distribution of this species and its amenability for culturing, it may be an alternative or complementary test for ecotoxicological studies in other European coastal systems where the existing tests cannot be applied or do not offer a definitive solution.

DeWitt, T. H., Distworth, G. R., and Swartz, R. C., (1988). Effects of natural sediment features on survival of the phoxocephalid amphipod, Rhepoxynius abronius. Mar. Environ. Res. 25, 99—124.

ACKNOWLEDGMENTS

Guerra, M. Gaudeˆncio, M. J., Castro, O., and Vale, C. (1995). Toxicidade de lindano e PCB 118 para o anfı´ pode Corophium volutator. In Resumos do 1° Congresso lbe´ rico sobre ContaminamaJ o e ¹oxicologia Ambiental, P-41. Universidade de Coimbra, Coimbra.

The work of F. O. Costa and A. D. Correia was supported by the PRAXIS XXI Program, Grants BM/2390/94 and BM/2391/94, respectively.

REFERENCES American Society for Testing and Materials (ASTM) (1992). Standard guide for conducting 10-day static sediment toxicity tests with marine and estuarine amphipods. In Annual Book of AS¹M Standards, ¼ater and Environmental ¹echnology, Vol. 11.04, E1367-90. ASTM, Philadelphia. American Society for Testing and Materials (ASTM) (1993). Standard guide for conducting sediment toxicity tests with freshwater invertebrates. In Annual Book of AS¹M Standards, ¼ater and Environmental ¹echnology, Vol. 11.04, E1383-93. American Society for Testing and Materials, Philadelphia. Borgmann, U., and Munawar, M. (1989). A new standardized sediment bioassay protocol using the amphipod Hyalella azteca (Saussure). Hydrobiologia 188/189, 425—431. Borgmann, U., Ralph, K. M., and Norwood, W. P. (1989). Toxicity test procedures for Hyalella azteca, and chronic toxicity of cadmium and pentachlorophenol to H. azteca, Gammarus fasciatus and Daphnia magna. Arch. Environ. Contam. ¹oxicol. 18, 756—764. Buikema, A. L., Jr., and Benfield, E. F. (1979). Use of macroinvertebrate life history information in toxicity tests. J. Fish. Res. Board Can. 36, 321—328. Burton, G. A., Jr. (1991). Assessing the toxicity of freshwater sediments. Environ. ¹oxicol. Chem. 10, 1585—1627. Caeiro, S., Ribau, M., and Costa, M. H. (1994) Contaminaia8 o orgaˆnica e metais pesados em sedimentos do Estua´rio do Sado. In 4a ConfereL ncia Nacional sobre a Qualidade do Ambiente. 6/8 Abril 1994, Lisbon, pp. 133—143. Chapman, P. M., and Long, E. R. (1983). The use of bioassays as part of a comprehensive approach to marine pollution assessment. Mar. Pollut. Bull. 14, 81—84. Chevreux, E., and Fage, L. (1925). Amphipodes. Faune Fr. 9, pp. 1—488. Correia A. D., Costa, F. O., and Costa, M. H. (1995) Gammarus locusta (Amphipoda) como indicador biolo´gico: Dinaˆmica da populaia8 o e metodologias de manutenia8 o da espe´cie em laborato´rio. In Resumos do 1° Congresso lbe´ rico sobre ContaminamaJ o e ¹oxicologia Ambiental, P-40. Universidade de Coimbra, Coimbra. Cortesa8 o, C., and Vale C. (1993). Metals in the sediments of Sado estuary, Portugal. In Abstracts of the First SE¹AC ¼orld Congress, Ecotoxicology and Environmental Chemistry—A Global Perspective, ¸isbon, p. 185. Costa, M. H. (1989). Macrofuna beL ntica no infralitoral do estua´ rio do Sado: »ariabilidade e interacmoJ es. Tese de Doutoramento, FCT/UNL. Costa, F. O., Correia, A. D., and Costa, M. H. (1996). Sensitivity of a marine amphipod to noncontaminant variables and to copper in the sediment. E¨ cologie 27(4), 249—276. De Lisle, P. F., and Roberts, M. H., Jr. (1988). The effects of salinity on cadmium toxicity to the estuarine mysid Mysidopsis bahia: Role of chemical speciation. Aquat. ¹oxicol. 12, 357—370.

DeWitt, T. H., Redmond, M. S., Sewall, J. E., and Swartz, R. C. (1992). Development of a Chronic Sediment ¹oxicity ¹est for Marine Benthic Amphipods. U.S. Environmental Protection Agency for the Chesapeake Bay Program, Newport.

Hichey, C. W., and Ropper, D. S. (1992) Acute toxicity of cadmium to two species of infaunal marine amphipodes (tube-dwelling and burrowing) from New Zealand. Bull. Environ. Contam. ¹oxicol. 49, 165—170. Luoma S. M., and Carter, J. L. (1993). Understanding the toxicity of contaminants in sediments: Beyond the bioassay-based paradigm. Environ. ¹oxicol. Chem. 12, 793—796. Luoma, S. N., and Ho, K. T. (1993). Appropriate uses of marine and estuarine sediment bioassays. In Handbook of Ecotoxicology (P. Calow, Ed.), pp. 193—226. Blackwell Scientific, Oxford. Marques, J. C. S. (1989). Amphipoda (Crustacea) bento´ nicos da costa portuguesa: Estudo taxono´ mico, ecolo& gico e biogeogra´ fico. Tese de Doutoramento, Faculdade de Cieˆncias e Tecnologia da Universidade de Coimbra. Meador, J. P., Varanasi, U., and Krone, C. A. (1993). Differential sensitivity of marine infaunal amphipods to tributyltin. Mar. Biol. 116, 231—239. Mucha, A. P., and Costa, M. H. (1996). Importaˆncia das comunidades macrozoobeˆnticas nos processo de transfereˆncia de carbono e nutrientes em sedimentos estuarinos. In » ConfereL ncia Nacional sobre a Qualidade do Ambiente, pp. 1383—1394, Universidade de Aveiro, Aveiro. Nelson, W. G. (1980). Reproductive patterns of gammaridean amphipods. Sarsia 65, 61—71. Oakden, J. M., Oliver, J. S., and Flegal, A. R. (1984). EDTA chelation and zinc antagonism with cadmium in sediment: Effects on the behavior and mortality of two infaunal amphipods. Mar. Biol. 84, 125—130. Quintino, V., Picado, A. M., Rodrigues, A. M., Mendonia, E., Costa, M. H., Costa, M. B., Lindgaard-Jorgensen, P., and Pearson, T. H. (1995). Sediment toxicity—Infaunal community structure in a southern european estuary. Neth. J. Aquat. Ecol. 29(3/4), 427—436. Ribau, M., and Costa, M. H. (1994). Gammarus locusta (amphipoda) como indicador de meios litorais contaminados. Dados preliminares. In Actas da 4a ConfereL ncia Nacional de Qualidade do Ambiente, Lisbon, pp. 116—125. Sexton, E. W. (1928). On the rearing and breeding of Gammarus in laboratory conditions. J. Mar. Biol. Assoc. ºK NS 1 20, 33—54. Society of Environmental Toxicology and Chemistry (SETAC)—Europe (1993). Guidance document on sediment toxicity assessment for freshwater and marine environments. In ¼orkshop on Sediment ¹oxicity Assessment, Renesse, ¹he Netherlands, 8—10 November 1993 (I. R. Hill, P. Matthiessen and F. Heimbach, Eds.). SETAC—Europe. Stock, J. H. (1967). A revision of the European species of the Gammarus locusta-group (Crustacea, Amphipoda). Zool. »erh. ¸eiden 90, 1—56. Swartz, D. C., Ditsworth, G. R., Schults, D. W., and Lamberson, J. O. (1985). Phoxocephalid amphipod bioassay for marine sediment toxicity. In Aquatic ¹oxicology and Hazard Assessment: Seventh Symposium (R. D. Caldwell, R. Purdy, and R. C. Bahner, Eds.), ASTM STP 854, pp. 284—307, ASTM, Philadelphia. Thomas, J. D. (1993). Biological monitoring and tropical diversity in marine environments: A critique with recommendations, and comments on the use of amphipods as bioindicators. J. Natural History 27, 795—806.