Priority Substances List Assessment Report for Non-pesticidal organotin compounds

2.0 Summary of Information Critical to Assessment of "Toxic"

2.1 Identity, Properties, Production, and Uses

There are few data in the literature on the physical and chemical properties of non-pesticidal organotin compounds. However, many organotin compounds dissociate in water into an organotin cation and a companion anion. In water, these species will likely exist as hydrates or complexes depending on the nature and concentration of other solutes (e.g., chloride ion). Consequently, the aquatic persistence, fate, and toxicity of such compounds may be limited to considerations of just six species, those of mono- and di- methyltin, butyltin, and octyltin.

Non-pesticidal applications of organotin compounds are reviewed elsewhere (Gitlitz and Moran, 1983) and summarized here. Mono-organotin compounds are mainly used in the stabilization of poly(vinyl chloride) (PVC) films during manufacture. Smaller quantities are used for glass coating (e.g., butyltin trichloride). Diorganotin compounds are also used mainly as PVC stabilizers, the most important of which are dimethyltin, dibutyltin, and dioctyltin compounds. Dioctyltin compounds are generally used as additives for PVC food packaging products. Dimethyltin compounds are also used for this purpose in some countries, but not in Canada. Other important industrial uses for diorganotin compounds are as catalysts in producing:

• polyurethane foams;
• esters used as plasticizers, lubricants, and heat-transfer fluids; and
• room-temperature-vulcanized silicone elastomers to produce flexible silicone rubbers.

Smaller quantities are used for glass coating (e.g., dimethyltin dichloride), as anthelmintics for poultry, and as stabilizers for lubricating oils, hydrogen peroxide, and polyolefins.

For Canada, the relative importance of various uses of organotin compounds is not definitely known but it is assumed that the use pattern is similar to that reported for the United States (Wilkinson, 1984). In 1982, the most important non-pesticidal applications in the United States were as PVC stabilizers and as catalysts for polyurethane and silicone elastomers, accounting for 67% and 8% of organotin consumption, respectively. Biocidal uses accounted for 20% of the market, and all other uses (including uses as latex paint preservatives, anthelmintics, and coccidiostats) accounted for 5%.

All organotin compounds now used in Canada are imported. In 1984, an estimated 290 tonnes of methyltin compounds were imported to Canada (CIS, 1985a). None was exported. The estimated demand was 170 tonnes for PVC pipe and 120 tonnes for PVC siding and profiles (moulds). In the same year, 1020 tonnes of butyltin compounds were imported to Canada (CIS, 1985b). None was exported. The estimated demand was 405 tonnes for PVC pipe and 615 tonnes for siding and profiles.

There are two major manufacturers of PVC at four locations in Canada - Sarnia, Niagara Falls, Shawinigan, and Fort Saskatchewan (CIS, 1988). The largest use of PVC is in the manufacture of pipe and fittings, and most PVC pipes and fittings used in Canada are made in Canada (CIS, 1988; Statistics Canada, 1989). Organotin compounds are not used only at the four locations previously mentioned. Many smaller companies buy PVC resin and organotin stabilizers, and prepare organotin-stabilized PVC on site (Shermet, 1992). Although the major use of organotin compounds is as PVC stabilizers, only about 30% of all PVC was stabilized with organotin compounds in 1991 (Miller, 1992).

There is no information on amounts of octyltin compounds imported to Canada, but it is assumed that the amounts are far smaller than amounts of methyltin and butyltin compounds because they are used only as stabilizers in some PVC for food packaging (Health and Welfare Canada, 1986).

2.2 Entry into the Environment

Possible routes of entry of non-pesticidal organotin compounds to the environment are:

1. atmospheric transport from other countries;
2. losses during transportation;
3. losses from the manufacture of organotin-containing materials, and from organotin-catalyzed production of polyurethanes and silicones;
4. losses from weathering or leaching of organotin-containing materials;
5. leaching of organotin-stabilized PVC pipe by water;
6. losses from landfill-disposed, organotin-containing materials;
7. losses from incineration of organotin-containing materials; and
8. biological production, in the case of environmental methylation of tin to produce methyltin species.

Although it is reasonable to assume that organotin compounds are entering the Canadian environment from non-pesticidal uses, unequivocal evidence of their presence in effluents or emissions is not available. It is clear that some organotin compounds are entering the environment, but it is not clear if they arise from pesticidal or non-pesticidal uses, or from natural methylation of inorganic tin.

It is likely that the most important route of entry of non-pesticidal organotin compounds to the Canadian environment would be through leaching of PVC pipe by water. This is because the most important non-pesticidal use of organotin compounds is for PVC stabilization, PVC is used principally in the production of pipe, and continuous contact with water is assured. Available information on non-pesticidal sources of methyltin, butyltin, and octyltin species to the Canadian environment is discussed in the following text.

Methyltin species. In Canada, monthly sampling of sewage treatment plant influents, effluents, and sludges in Montreal, Toronto, Hamilton, Sarnia, and Vancouver over a seven-month period (1990 to 1991) revealed no contamination by methyltin species (detection limit 40 ng Sn/L for influents and effluents and 2 ng Sn/g dry weight for sludge) (Chau et al., 1992). No methyltin species were found in leachate samples collected from five landfills in southern Ontario in 1990 (detection limit 40 ng Sn/L) (Chau et al., 1992).

Methyltin species may also be introduced to the environment through microbial methylation of tin (Chau et al., 1980; Weber and Alberts, 1990). In addition to anthropogenic origins for the methyltin species observed in fish, it is possible that the methyltin species were produced naturally, i.e., through microbial methylation of inorganic tin in the fish gut and/or in micro-organisms that are eventually consumed by fish. It should be noted that the monomethyltin and dimethyltin species found in the environment are not degradation products of trimethyltin compounds introduced to the environment, because trimethyltin compounds are not used industrially or as pesticides.

Butyltin species. The monobutyltin and dibutyltin found in a limited study of influents and effluents from sewage treatment plants in five Canadian cities (Chau et al., 1992) may have been present due to their use as PVC stabilizers or because they were degradation products of the pesticide tributyltin, which was also found in sludge samples collected at the same time. The fact that butyltin species were found in sewage treatment plant effluent indicates incomplete removal during sewage treatment. These data are too few to allow meaningful generalizations to be made or to allow the quantitation of releases of butyltin species to the Canadian environment, but they do demonstrate that releases are occurring as a result of pesticidal organotin use or non-pesticidal organotin use, or both.

It should be noted that for both butyltin and methyltin PVC stabilizers, concentrations of organotin stabilizers leached from PVC pipe in the laboratory decline fairly quickly (i.e., over days to weeks) with continuous leaching by water (Boettner et al., 1981; Wu et al., 1989; Quevauviller et al., 1991), and are likely to do so in PVC pipe used in Canada. The United States Environmental Protection Agency has estimated that organotin concentrations in surface waters resulting from the leaching of PVC-stabilized pipes and tubing would be in the pg Sn/L range (U.S. EPA, 1983). The Canadian Standards Association has set a limit for extractable organotin compounds from potable water pipes and fittings (Allidina, 1992).

No butyltin species were found in leachate samples collected from five landfills in southern Ontario in 1990 (detection limit 40 ng Sn/L) (Chau et al., 1992).

Octyltin species. There is no information available about octyltin species entering the environment. Monthly sampling of sewage treatment plant influents, effluents, and sludges in Montreal, Toronto, Hamilton, Sarnia, and Vancouver over a seven-month period (1990 to 1991) revealed no contamination by octyltin species (detection limit 40 ng Sn/L for influents and effluents and 2 ng Sn/g dry weight for sludge) (Chau et al., 1992). No octyltin species were found in leachate samples collected from five landfills in southern Ontario in 1990 (detection limit 40 ng Sn/L) (Chau et al., 1992).

2.3 Exposure-related Information

2.3.1 Fate

Most of the data in the literature on the persistence of organotin compounds refer to the aquatic environment. Data on the persistence of the mono- and di- methyltin, butyltin, and octyltin species in aquatic environments are in some cases fragmentary or nonexistent, but by analogy with the tributyltin species, it is assumed that none of these species would be persistent in aquatic environments. Half-lives are estimated to be less than a few months at 20°C, and longer at lower temperatures. Relevant information for the methyltin, butyltin, and octyltin species follows.

Methyltin species. Inorganic tin and methyltin species undergo biological or abiotic methylation in aquatic environments, although it does not appear to be an important process in short-term studies (Maguire, 1991). Except for the examination of natural methylation mechanisms, little work has been done on the persistence of methyltin species in aquatic environments and distribution among various environmental compartments.

In comparison with butyltin species, for which there is more information, it is likely that sunlight photolysis and microbial degradation of methyltin species will be important, and that in the absence of methylating organisms, methyltin species in aquatic environments would not be persistent, with half-lives of less than a few months at. 20°C.

Butyltin species. Tributyltin does not volatilize significantly from water (Maguire et al., 1983), and it is unlikely that the more hydrophilic dibutyltin and monobutyltin would volatilize. Thus these species, and non-pesticidal methyltin and octyltin species, are not expected to contribute to phenomena such as ozone depletion, global warming, or the formation of ground-level ozone.

In 90% water/10% acetonitrile solutions, monobutyltin and dibutyltin were stable for at least 9 days in the dark, but were degraded by light of wavelength 300 nm (Maguire et al., 1983). The half-life of degradation of monobutyltin was 0.4 days, and after 9 days, the concentration of inorganic tin accounted for about 70% of the initial monobutyltin concentration. The half-life of dibutyltin was > 9 days, and monobutyltin and inorganic tin were the only products observed. The butyltin species are less stable at acidic pH values (Burns et al., 1987).

The logarithms of the n-octanol-water partition coefficients (log Kow) for monobutyltin trichloride, dibutyltin dichloride, and tributyltin chloride are 0.09, 0.05, and 2.2, respectively (Tsuda et al., 1988). This indicates that monobutyltin and dibutyltin would not be bound to the organic portion of sediment to the same extent as the more lipophilic tributyltin species. Hinga et al., (1987) determined that the mean sediment-water partition coefficient for dibutyltin was about 100 times smaller than that for tributyltin (values were 450 and 40 000, respectively). Stang and Seligman (1987) noted that there was a wide variation in adsorption coefficients of butyltin species to sediments, which may reflect lack of equilibrium and differences in sediment composition. They determined partition coefficients for different sediments in the following ranges: monobutyltin, 1 700 to 29 000; dibutyltin, 2 100 to 26 000; and tributyltin, 6 200 to 55 000.

The biological availability of sediment-associated dibutyltin and monobutyltin has received little attention. Oligochaetes can take up tributyltin from sediment and can degrade it to inorganic tin (Maguire and Tkacz, 1985), and it is assumed that dibutyltin and monobutyltin in sediment would also be bioavailable and biodegradable.

There are few data on the uptake of monobutyltin and dibutyltin species by aquatic organisms. Tsuda et al., (1988) determined the logarithms of the bioconcentration factors in carp (Cyprinus carpio) to be 2.1, 1.0, and 3.5, for monobutyltin, dibutyltin, and tributyltin, respectively, in agreement with their octanol-water partition coefficients. Martin et al, (1989) noted a similar pattern for rainbow trout (Oncorhynchus mykiss).

The persistence of dibutyltin and monobutyltin species in aquatic ecosystems is likely to depend strongly on ecosystem-specific characteristics, such as temperature and the kinds and concentrations of butyltin-tolerant and -degrading organisms, as is the case with tributyltin (Maguire, 1987). Most work indicates that these chemicals are not persistent in aquatic ecosystems (i.e., overall half-lives from all degradation processes would be less than a few months at 20°C).

Octyltin species. Only two laboratory-based persistence studies on dioctyltin compounds were identified (Akagi and Sakagami, 1971; Mazayev et al., 1976). Both studies indicated that dioctyltin was not persistent. It is difficult to estimate the persistence of mono-octyltin and dioctyltin in aquatic ecosystems using only these two studies as a basis for prediction. However, compared with the butyltin species, it is expected that the octyltin species would not be persistent in aquatic environments, with half-lives of less than a few months at 20°C.

2.3.2 Concentrations in the Environment

Most of the data on the occurrence of organotin compounds in Canada refer to the aquatic environment. They are the result of one national survey (Maguire et al., 1986) and several regional or local surveys (Maguire et al., 1982; Maguire and Tkacz, 1985; Maguire et al., 1985; Kaye et al., 1986; Harding and Kaye, 1988; Seakem Oceanography Ltd., 1989; Cullen et al., 1990) at a total of about 275 locations. Detection limits vary between the studies depending on the technique and sample size.

The findings for the methyltin, butyltin, and octyltin species in Canada are summarized separately in the following. More detailed discussions of the analytical methods for these species, their environmental occurrence in Canada and other countries, and their persistence and fate are given in the Supporting Document. Analytical methods for organotin compounds do not determine the anionic moiety (Maguire, 1991), and for this reason the species are only referred to as methyltin, butyltin, and octyltin species.

Data on concentrations of non-pesticidal organotin compounds in ambient or indoor air in Canada (or elsewhere) were not identified.

Methyltin species. Monomethyltin and dimethyltin have been found in fresh water, seawater, and sediment in Canada at concentrations similar to those observed in other countries. They were found in about 10% of all water samples analyzed.

For monomethyltin, the highest concentration observed in fresh water in Canada was 1 100 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 32, n.d. values for an additional 242 samples not included in calculations, detection limit 10 ng Sn/L) were 324 ng Sn/L and 200 ng Sn/L, respectively. Monomethyltin was found in only 2 of 70 seawater samples, at concentrations of 80 and 120 ng Sn/L (n.d. values for an additional 68 samples, detection limit 10 ng Sn/L).

For dimethyltin, the highest concentration observed in fresh water in Canada was 320 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 27, n.d. values for an additional 242 samples not included in calculations, detection limit 10 ng Sn/L) were 92 ng Sn/L and 40 ng Sn/L, respectively. Dimethyltin was not found in any of 70 seawater samples (detection limit 10 ng Sn/L). In some locations, environmental methylation of tin seems to be the most likely explanation for the presence of methyltin species. In industrial areas, it is possible that there is input of anthropogenically-derived methyltin species, or that anthropogenically-derived tin is biologically methylated.

There are few data on the occurrence of methyltin species in Canadian biota.

Monomethyltin and dimethyltin species have been found outside Canada in algae, seaweed, eelgrass, mussels, oysters, limpets, and other marine organisms, generally at concentrations less than 120 ng Sn/g wet weight. The exception is eelgrass, in which monomethyltin has been found at concentrations of up to 2 300 ng Sn/g wet weight François and Weber, 1988).

A potential source of exposure to non-pesticidal organotin compounds in drinking water is migration of stabilizers from PVC pipe, which is used fairly extensively in distribution systems in Canada (Lister, 1992). Despite this, few data have been identified on concentrations of these compounds in water supplies. Available information is restricted to one United States report of concentrations of monomethyltin (range: 0.49 to 8.1 ng Sn/L) and dimethyltin (range: 0.40 to 2.2 ng Sn/L) in a limited number of tap water samples collected in Florida in 1977 (Braman and Tompkins, 1979).

Butyltin species. Most of the data on the environmental occurrence of monobutyltin and dibutyltin in Canada were obtained during surveys for tributyltin before its antifouling uses were regulated under the Pest Control Products Act in 1989 (Agriculture Canada, 1989). Although it is reasonable to presume that this regulation has resulted in a substantial decrease in environmental butyltin concentrations, especially in fresh water, no large-scale surveys have been undertaken in Canada since 1989. Concentrations of butyltin species in water and shellfish have decreased after similar regulations were introduced in France (Alzieu, 1991), England (Waite et al., 1991), and the United States (Valkirs et al., 1991; Wade et al., 1991).

Monobutyltin and dibutyltin have been found in fresh water, seawater, and sediment at many locations across Canada, especially in surveys conducted before 1986. Many sampling locations were not randomly selected but were in fact sites where maximum concentrations would be expected. The presence of the monobutyltin and dibutyltin species in, or close to, harbours, marinas, and shipping lanes was attributed to the degradation of the antifouling agent tributyltin. Concentrations were similar to those observed in water in areas of heavy boating or shipping traffic in other parts of the world. The two species were found in about 15% of all water samples analyzed.

For monobutyltin, the highest concentration observed in fresh water in Canada was 5 700 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 65, n.d. values for an additional 209 samples not included in calculations, detection limit 0.1 to 10 ng Sn/L) were 216 ng Sn/L and 27 ng Sn/L, respectively. In seawater, the highest concentration observed was 170 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 25, n.d. values for an additional 104 samples not included in calculations, detection limit 0.1 to 10 ng Sn/L) were 18 ng Sn/L and 3 ng Sn/L, respectively.

For dibutyltin, the highest concentration observed in fresh water in Canada was 3 700 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 65, n.d. values for an additional 209 samples not included in calculations, detection limit 0.1 to 10 ng Sn/L) were 148 ng Sn/L and 25 ng Sn/L, respectively. In seawater, the highest concentration observed was 830 ng Sn/L. The mean and median concentrations derived from original data in all studies (n = 38, n.d. values for an additional 91 samples not included in calculations, detection limit 0.1 to 10 ng Sn/L) were 98 ng Sn/L and 25 ng Sn/L, respectively.

There are few data on the occurrence of butyltin species in Canadian biota. Concentrations of monobutyltin and dibutyltin in fish and molluscs ranged from 7 to 179 ng Sn/g wet weight (Maguire et al., 1986; Cullen et al., 1990; Scott et al., 1991; Wong and Chau, 1992) (detection limits 1 to 10 ng Sn/g wet weight). Such concentrations have been observed in biota in other countries.

A limited number of samples (n = 27) of several marine foodstuffs has been analyzed for tributyltin and its degradation products by Health and Welfare Canada (Forsyth and Cléroux, 1991). Monobutyltin and dibutyltin were not detected [detection limits < 1.0 ng/g (0.7 ng Sn/g*) and < 1.0 ng/g (0.5 ng 5n/g*) wet weight, respectively, for monobutyltin and dibutyltin] in any samples of fish (cod, haddock, perch, trout) but were present in canned and fresh molluscs [concentrations of monobutyltin ranged from 1.2 ng/g (0.8 ng Sn/g*) wet weight in fresh clams to 5.9 ng/g (4.0 ng Sn/g*) in canned mussels and those for dibutyltin ranged from 3.1 ng/g (1.6 ng 5n/g*) in fresh clams and canned cockles to 46.7 ng/g (23.9 ng Sn/g*) in canned mussels]. The monobutyltin and dibutyltin in these samples likely resulted from the use of the antifouling pesticide tributyltin, applied either to boats or to nets in aquaculture; while certain organotin compounds are still registered in Canada for use on boat hulls, no products are currently registered for aquaculture uses. Analyses have also been conducted for butyltin species in fruit drinks (Forsyth, 1992). A limited number of fruit drinks contained monobutyltin [0.1 to 0.2 ng/mL (70 to 140 ng Sn/L*) in 4 of 42 samples, detection limit 0.6 ng/mL (41 ng Sn/L*)].

Based on a U.S. Department of Agriculture survey, it was reported that 8% of 1 031 samples of turkey liver contained dibutyltin (limit of detection 0.04 mg Sn/kg) which resulted from the use of dibutyltin dilaurate as a coccidiostat and anthelmintic in turkeys (Epstein et al., 1991). It was not detected in companion muscle tissue samples. Data are not available to indicate whether use patterns of this compound for this purpose are similar in Canada to those in the United States.

Octyltin species. The mono-octyltin and dioctyltin species have not been found to date in Canada or elsewhere in biota or any environmental medium. These species (if present above their detection limits of 10 ng Sn/L in water) would have been detected by the analytical methods used in the major Canadian surveys for butyltin and methyltin species (Maguire et al., 1982; 1986), and possibly in other Canadian studies as well.

The leaching of octyltin species from PVC plastic packaging, in which they are used as heat and light stabilizers, is a potential source of octyltin compounds in foodstuffs. Analyses have been conducted for octyltin species in selected Canadian foodstuffs (Forsyth, 1992). Out of 15 samples of edible oils, 5 contained both mono-octyltin [5.5 to 26.3 ng/g (2.8 to 13.5 ng Sn/g*); detection limit 1.0 ng/g (0.5 ng Sn/g*)] and dioctyltin [25.2 to 113.3 ng/g (8.7 to 39.1 ng Sn/g*); detection limit 1.0 ng/g (0.3 ng 5n/g*)] A limited number of fruit drinks contained mono-octyltin [4.5 to 16.3 ng/mL (2300 to 8400 ng 5n/L*) in 5 of 42 samples] and dioctyltin [0.9 to 4.3 ng/mL (300 to 1500 ng Sn/L*) in 3 of 42 samples]; however, detection limits were not specified.

2.4 Effects-related Information

2.4.1 Experimental Animals and In Vitro

In a bioassay conducted by the United States National Cancer Institute (1978), dibutyltin diacetate was administered to rats in their diet at dose levels of 0, 3.33, or 6.65 mg Sn/kg (b.w.) for 78 weeks, followed by 26 weeks of observation. There was a significant dose-related increase in mortality in male but not in female rats. An accidental loss of tissue samples in females (high dose only) precluded an evaluation of carcinogenicity. There were no significant increases in the incidence of neoplasms in males, and the United States National Cancer Institute concluded that there was "no conclusive evidence" for carcinogenicity in this sex.

Concurrently, mice were administered dibutyltin diacetate in their diet at dose levels of 0, 9.9, or 19.8 mg Sn/kg (b.w.) · day for 78 weeks, followed by 14 weeks of observation (United States National Cancer Institute, 1978). A significant increase in mortality occurred in females. Though there was a small increase in the incidence of hepatocellular adenomas in female mice, it was not statistically significant. The United States National Cancer Institute concluded that there was "no conclusive evidence" for carcinogenicity in mice of either sex.

Mosinger (1979) reported the results of two-year studies in which mixtures of monomethyltin and dimethyltin compounds were administered to Wistar rats in their diet at dose levels of 10 mg/kg (b.w.) · day. The mixtures consisted of 25:75 dimethyltin bis(iso-octyl thioglycolate) [otherwise known as dimethyltin bis(iso-octyl mercaptoacetate) - see Table 1] and monomethyltin tris(iso-octyl thioglycolate) [otherwise known as monomethyltin tris(iso-octyl mercaptoacetate)]; 80:20 monomethyltin bis(iso-octyl thioglycolate) sulphide [otherwise known as monomethyltin bis(iso-octyl mercaptoacetate) sulphide] and dimethyltin (iso-octyl thioglycolate) sulphide [otherwise known as dimethyltin (iso-octyl mercaptoacetate) sulphide]; and 80:20 monomethyltin (mercaptoethyl oleate) sulphide and dimethyltin (mercaptoethyl oleate) sulphide.

On the basis of the results of these studies, it was concluded that none of these mixtures was carcinogenic. However, because of several limitations, including limited group sizes (n = 20 of each sex), administration of a single dose level, inadequate statistical analyses, and examination of a limited range of non-neoplastic endpoints, these studies are considered to be inadequate and contribute little meaningful information for assessing the evidence of carcinogenicity for these compounds.

Studies on the genotoxicity of monomethyltin have not been identified and only one report appears to be available in which dimethyltin did not bind irreversibly to DNA in vitro (Barbieri and Silvestri, 1991 in: Nieboer and Bryant, 1992). Results have been mixed for the genotoxicity of dibutyltin compounds, which have been more extensively investigated. While not genotoxic in the Salmonella/microsome test or in a dominant lethal assay in Drosophila melanogaster, a mutagenic response in ovary cells of Chinese hamsters has been reported in a limited study in which there was no positive control or replication (Li et al., 1982). Dibutyltin dichloride has also been reported to be positive in a micronucleus test in mice orally administered 50 mg/kg (b.w.) (Life Science Research Limited, 1991).

2.4.2 Humans

No quantitative information was identified on the toxicological effects produced in humans following chronic exposure to non-pesticidal organotin compounds.

2.4.3 Ecotoxicology

The toxicity of tin compounds has been studied extensively (Hall and Pinkney, 1985; Snoeij et al., 1987; Cooney and Wuertz, 1989; and references therein). Organotin compounds are more toxic to aquatic biota than inorganic tin compounds. Progressive introduction of organic groups to the tin atom in any R_nSn^(4-n)+ series produces maximal biological activity against all species when n = 3 (i.e., for the triorganotin compounds). Within the class of triorganotin compounds, however, toxicity varies considerably with the nature of the organic substituents (Davies and Smith, 1980).

The most toxic compounds to insects are the trimethyltin compounds; to mammals, the triethyltin compounds; to Gram-negative bacteria, the tripropyltin compounds; and to Gram-positive bacteria, yeasts, fungi, and fish, the tributyltin compounds. Further increase in the alkyl chain length produces a sharp drop in toxicity. Triphenyltin compounds are particularly toxic to phytoplankton (Wong et al., 1982), while tricyclohexyltin compounds show high acaricidal activity (Davies and Smith, 1980). The variation of the anionic moiety, X, within any particular series of R₃SnX compounds usually has little effect on biological activity (Davies and Smith, 1980; Polster and Halacka, 1971).

In this report, environmental concentrations of the organotin species are compared using the most sensitive aquatic organism, with emphasis being placed on ecologically significant toxicity tests (e.g., LC₅₀ values are chosen in preference to tests such as the Microtox assay).

Only limited information was identified on the toxicity of most non-pesticidal organotin species to fresh water and marine organisms, and no information was identified on the toxicity of the mono-octyltin species to aquatic organisms. No information was identified on the toxicity of non-pesticidal organotin compounds to wild birds or mammals.

Given that tributyltin is considerably more toxic than monobutyltin or dibutyltin, contamination of mono- or di- butyltin with small amounts of tributyltin in testing solutions can lead to apparent toxicity that in fact results from the presence of the tributyltin contaminant (Wester and Canton, 1987). Toxicity thresholds reported for mono-organotin and diorganotin compounds may therefore reflect the toxicity of triorganotin contaminants and should be considered as upper limits.

For monomethyltin species, the water flea, Daphnia magna, is the most sensitive freshwater organism in acute toxicity tests. The 48-h LC₅₀ for this organism was 0.46 mg Sn/L (Steinhauser et al., 1985). The most sensitive marine organism tested is the diatom Skeletonema costatum. The 72-h EC₅₀ for growth for this diatom was 0.04 mg Sn/L (Walsh et al., 1985). No data were identified for the chronic toxicity of monomethyltin to aquatic organisms or for the toxicity of monomethyltin in sediment to benthic organisms.

For dimethyltin species, Daphnia magna is the most sensitive freshwater organism in acute toxicity tests. The 48-h LC₅₀ for this organism was 0.03 mg Sn/L (Steinhauser et al., 1985). For marine organisms, a "worst case" was chosen for the most sensitive organism, the diatom Skeletonema costatum. The 72-h EC₅₀ for growth for this diatom was assumed to be 0.27 mg Sn/L (it was reported as >0.27 mg Sn/L - Walsh et al., 1985). No data were identified for the chronic toxicity of dimethyltin to aquatic organisms or for the toxicity of dimethyltin in sediment to benthic organisms.

For monobutyltin species, the red killifish (Oryzias latipes) is the most sensitive freshwater organism in acute toxicity tests. The 48-h LC₅₀ for this fish was 16 mg Sn/L (Nagase et al., 1991). The only data identified for marine organisms are for yeasts, and the most sensitive marine yeasts are Aureobasidium pullulans, Candida albicans, and Sporobolomyces alborubescens, with 48-h IC₅₀ values for growth of 2.1 mg Sn/L (Cooney et al., 1989). No datawere identified for the chronic toxicity of monobutyltin to aquatic organisms or for the toxicity of monobutyltin in sediment to benthic organisms.

For dibutyltin species, the mosquito larva (Culex pipiens) is the most sensitive freshwater organism in acute toxicity tests. The 24-h LC₅₀ for this larva was 0.1 mg Sn/L (Gras and Rioux, 1965). The most sensitive marine organism tested is the diatom, Skeletonema costatum. The 72-h EC₅₀ for growth for this diatom was 0.01 mg Sn/L (Walsh et al., 1985). There are no data on the toxicity of dibutyltin in sediment to benthic organisms. There are some data on the chronic toxicity of dibutyltin to clams and fish. Exposure of freshwater clams (Anodonta anatina) to 0.015 mg Sn/L dibutyltin dichloride for seven months (weekly static renewal) caused decreases in weight and carbohydrate stores, but no mortality (Holwerda and Herwig, 1986). The no-observed-effect- concentration (NOEC) for histopathological effects in guppies (Poecilia reticulata) was <0.125 mg Sn/L for exposure of three months (Wester and Canton, 1987). De Vries et al. (1991) investigated the comparative toxicity of various organotin compounds in early life stages of rainbow trout (Oncorhynchus mykiss). Beginning with yolk sac fry, trout were continuously exposed for 110 days. Dibutyltin dichloride, with a no-lethal-effect-level of 0.019 mg Sn/L for trout yolk sac fry, was about 1000 times less toxic than tributyltin chloride. The lowest-observed-effect-concentration (LOEC) for mortality was 0.095 mg Sn/L. At the end of the exposure period, resistance to infection was examined by an intraperitoneal challenge with Aeromonas hydrophila, a bacterium pathogenic to fish. Resistance to the bacterial challenge was decreased with dibutyltin dichloride at about 0.1 mg Sn/L. This might be indicative of a suppressed immune function or generally diminished fish health. No data were identified for the toxicity of dibutyltin in sediment to benthic organisms.

No data were identified for the toxicity of mono-octyltin to aquatic or benthic organisms.

For dioctyltin species, Daphnia magna is the most sensitive freshwater organism in acute toxicity tests. The 48-h LC₅₀ for this species was 0.001 mg Sn/L (Steinhauser ^{et al.,} 1985). No data were identified for the toxicity of dioctyltin to marine organisms or for the toxicity of dioctyltin in sediment to benthic organisms.

---

* units converted to remove ambiguity when interpreting analytical data, and to be consistent with units used elsewhere in this report