SCOPE 53 - Methods to Assess the Effects of Chemicals On Ecosystems

13.1 INTRODUCTION

This chapter is devoted to birds and mammals, since many species in these classes are capable of moving long distances and thus moving in and out of study areas. Although some species are sedentary, many bird and some mammal species in temperate latitudes are migratory, often moving for hundreds or even thousands of miles from one ecosystem to another. In the tropics, movements may be triggered by dry and wet season cycles rather than by temperature and day length. In some parts of the world, animals are nomadic and their movements seem unpredictable, making an analysis of the problems they face difficult. In all cases, having a detailed knowledge of the movements of the population under study is necessary. The requisite knowledge can vary even within a species: for example, the arctic population of the peregrine falcon (Falco peregrinus) migrates to South America, whereas the race in eastern North America moves only short distances. For studies of pollution, mobility has the advantage that examination of a few specimens gives some information on a wide area; conversely, it provides no detailed information on point sources of pollution. Overall the mobility of birds and mammals makes it challenging to develop methods to examine the effects of chemicals on their populations.

Nonetheless, birds and mammals are important as bioindicators of the effects of chemicals, because they are often at the top of the food web and because of wide interest in, and knowledge of, these higher vertebrates. Chemicals that enter the food chain at low concentrations can become amplified as they move up the food chain. By carefully selecting the indicator, bird or mammal, they can be used to detect traces of chemicals in the environment long before they are apparent in most other organisms, and to detect potential deleterious effects before they become widespread.

In this paper, a framework for evaluating the effects of chemicals on wildlife is presented, the general methods to evaluate the effects of chemicals are reviewed,  the advantages and disadvantages of using birds and mammals to evaluate the effects of chemicals are examined, and several case studies illustrate the methods of assessment. Although the methods to evaluate effects on individuals will be mentioned, the focus is on levels of organizations above the individual, at the population, community, and ecosystem levels. The case studies were chosen to illustrate the methods that have been used with some success to evaluate the effects of chemicals. Most of the case studies are with birds, because these animals are diurnal, large, and obvious members of ecosystems. Thus, any changes in their health or population characteristics are obvious immediately. People notice when birds start dying, are hatched without feathers or with other deformities, or entire colonies or populations may fail to hatch eggs because they all crack when the parents incubate them.

13.2 FRAMEWORK TO EVALUATE EFFECTS OF CHEMICALS

Methods to evaluate the effects of chemicals range from the molecular to the ecosystem level. Evaluating effects becomes more difficult as one moves toward the more complex organizational level (Figure 13.1). Detailed descriptions of the methods used at the level of the individual and lower can be found in Burgess et al. (in press) and at higher organization levels in Burger (in press).

One can start at either end of Figure 13.1. If one starts at the left-hand end, one has to extrapolate laboratory or limited field trials to ecosystems. Starting at the right-hand end, effects at the individual or physiological level are surmised by examining contaminated ecosystems. Ideally, both approaches  should be used to determine how chemicals are affecting wildlife and ecosystems at all levels.

Each approach, either from the molecular level upward or the ecosystem level downward, has advantages and disadvantages. The first approach has the advantage of laboratory testing where a clear causal relationship can be established. However, the first approach suffers from the increasing difficulty of extrapolating from lower to higher organizational levels, whereas the second requires a well-defined, contaminated ecosystem to be studied.

Ecosystems are not simply the sum of the component species, but encompass several ecosystem functions that are products of the interactions of the species. For example, predators and competitors affect the species around them. Particular predators, known as keystone predators, can maintain a high species diversity in a system by selectively preying on some species, which, if not controlled, would eliminate other species from the system by competition or predation (Paine, 1966).

Measuring ecosystem changes directly has the advantage of allowing us to determine the effects of chemicals on these systems; extrapolation is not needed. However, the costs in time and money might be prohibitive to measure the effects of even one chemical on all aspects of the structure and function of a single ecosystem. Thus, even for ecosystems, indicators of structures and functions must be selected, and that selection can prove problematic.

Figure 13.1. Levels of organization to evaluate the effects of chemicals

The effects of chemicals on ecosystem functions can only be determined by studying these functions, and cannot be directly measured or estimated by merely examining individual species. As with measuring effects on species, measuring ecosystem effects requires indicators for endpoints. These indicators should be sufficiently sensitive to provide early warnings, distributed over a broad geographical area, be capable of providing assessment over a wide range of stresses, independent of sample size, and be cost-effective (Sheehan, 1984a; Noss 1990). Ecosystem indicators must be ecologically significant phenomena (Sheehan, 1984b). Indicators for ecosystems might include indices of species diversity, relative species abundances, indices of species richness, or landscape parameters.

In an ideal world, laboratory studies should be predictive of ecosystem effects. Such extrapolation should be possible as one learns how to use microcosm and macrocosm studies to duplicate the structure and function of ecosystems. However, at least for higher vertebrates, this goal is as yet unattainable. In practice, ecologists evaluate the potential effects of chemicals on individual species, or species groups (such as songbirds, seals, ducks, whales). Even attempts to evaluate all resources are usually compilations of species effects (i.e., Bolze and Lee, 1989)  rather than an evaluation of community and ecosystem effects.

13.3 ADVANTAGES AND DISADVANTAGES OF USING BIRDS AND MAMMALS TO EVALUATE THE EFFECTS OF CHEMICALS

Birds are ideal targets to evaluate the effects of chemicals, because they represent several trophic levels, they are visible and conspicuous, their populations and behaviour are observable, reproductive success can be measured relatively inexpensively, comparative data for many species are available, and the young feed on local resources. The main disadvantage of using birds (that they migrate from place to place) can be eliminated by examining the young that have received all of their food from local sources. Population and community characteristics of birds such as population and colony size, reproductive success, behavioural abnormalities, physiological and morphological abnormalities, and their effects on other species in the ecosystem can be monitored. Since birds are diurnal, they are easy to observe; many are large and conspicuous, and they remain in one place during their breeding season. Additional advantages include their high interest to the public, their ability to integrate exposure over time and space, and their ability to serve as early warning sentinels before other components of the ecosystem have been affected. The disadvantages of using birds include: (1) they are mobile, (2) their populations may be endangered, making it difficult to obtain specimens for toxic analysis, and (3) they usually feed over a large area making point-source determinations difficult.

Mammals are useful indicators of the effects of chemicals on ecosystems, because, being mammals, they share some physiological characteristics with humans. Additionally, some mammals, such as rodents, are relatively sedentary, and thus reflect local conditions over a number of years. Since rodents and other small mammals are relatively common and inconspicuous, they can be used to monitor without undue public outcry. This approach will continue to be an important consideration as the public awareness of ecosystem protection continues to increase. Mammals share many of the advantages of birds, including their role at the top of the food web, and their role as important competitors and predators on other organisms (influencing community and ecosystem functions such as species abundance and diversity, and productivity). Furthermore, they are excellent laboratory models, with an extensive literature of experiments on many aspects of their behaviour and physiology.

The disadvantages of using mammals are that they are often nocturnal, their population sizes sometimes vary widely, making interpretations difficult from year to year, and many are difficult to trap. Moreover, reproductive success is more difficult to measure than it is for birds; and few studies of reproductive success have been available over several years. Such baseline data are essential to evaluate the effects of chemicals.

13.4 CASE STUDIES

Four case studies have been selected to illustrate the methods available to evaluate the effects of chemicals on birds and mammals. These cases have been chosen because they evaluated effects at several different levels within ecosystems. These case studies are:

1.  population declines of raptors caused by DDE-induced egg-shell thinning;
2. mortality of songbirds caused by forest spray programs;
3. effects of pollution in the North America Great Lakes on fish-eating birds and mammals; and
4. effects of oil on marine organisms and ecosystems

13.4.1 CASE 1: POPULATION DECLINES OF RAPTORS CAUSED BY DDE-INDUCED EGGSHELL THINNING

This particular investigation started at the population level when declines of populations of the peregrine falcon and several other raptorial birds were noted in the 1960s. Decreases in population were reported again throughout the Holarctic; and indeed in eastern North America the species had disappeared completely (Hickey, 1969). No studies appear to have been made on the effect of the decline or, in some areas, the disappearance of this species on community structure.

Even current preregistration tests are unlikely to have detected eggshell thinning that led to reproductive failure. The reason is the wide difference in the sensitivity of different species; several of the commonest avian test species¾quail, pheasant, chicken¾are almost completely insensitive, and others¾the duck¾only moderately so (Peakall, 1975). However, many species of fish-eating birds and raptors are extremely sensitive.

Only if experimental work had been carried out on a sensitive species such as the American kestrel, Falco sparverius, would reproductive effects have been seen (Lincer, 1975). Then egg breakage would have pointed to eggshell thinning and to the relationship between the degree of eggshell thinning and reproductive failure. Thus, the dose and residue levels of DDE could have been established.

The mechanism is considered to be via the inhibition of Ca-ATPase in the oviduct, reducing the transport of calcium to the site of eggshell formation (Peakall et al., 1973), but details of the pharmacodynamics involved in inter-species variation still remain unclear. Again this effect would not likely have been detected at an early stage. Thus, even today, preregistration studies are unlikely to have found this particular adverse effect of DDT.

13.4.2 CASE 2: MORTALITY OF SONGBIRDS CAUSED BY FOREST SPRAY PROGRAMMES

Attempts to control spruce budworm ( Choristoneura  fumiferana) in the forests of eastern Canada by the use of pesticides is the longest and largest spraying programme undertaken. A detailed review of the amount of pesticide used and the area sprayed from the start of the program in 1952 to 1981 was made by Peakall and Bart (1983). During that time, 18 million kilograms of pesticide were used on some 55 million hectares. The advantage of the programme from the authors' viewpoint is that the areas were so large that problems of immigration of birds occurred after mortality vanished and pesticides were the only chemical added to the system.

Initially DDT was the pesticide used. The most obvious effect of DDT on non-target organisms was the mortality of fish and the aquatic invertebrates on which they depended. The loss of whole year classes of Atlantic salmon, Salmo salar, were reported from rivers in New Brunswick in the mid-1950s (Logie, 1975). Although fish are outside the scope of this paper, laboratory studies that showed that fish died at dilutions of 1:10000000 were published by Ginsburg (1945); and detailed field studies made by the Ontario Department of Lands and Forests (Langford, 1949), showing just the type of effect later found as a side effect of the forest spray programs in eastern Canada, were available well before the operational use of DDT. The second, much more subtle effect on raptorial birds was discussed in the previous section.

Table 13.1. Calculated and observed effects on songbirds of insecticides used in forest spray operations

A similar, although much smaller, programme, was carried on in the United States against gypsy moth, Porthetria dispar. DDT was last used on an operational scale for forest spraying in the US in 1967 and in Canada in 1968. The choice of pesticides in the two countries in the post-DDT era was quite different. In temporal terms it was carbaryl (1959) and malathion (1964) in the US and phosphamidon (1963) and fenitrothion (1967) in Canada. The reasons for the difference in the pesticide usage pattern between the two countries is unknown, but the choices can be examined in toxicological~terms. The mechanism of action of each of these pesticides is the inhibition of the enzyme acetylcholine esterase which disrupts nerve function. The main concern is acute mortality, and this is used as the criterion to rank the pesticides. The risk is dependent on the actual toxicity (the LD₅₀) and the dosage used. The area of spray necessary to kill a small (10 g) songbird such as a kinglet, Regulus spp. can be calculated (as noted in Table 13.1). On this basis, phosphamidon is clearly the most hazardous material, as borne out by several events.

Two methods, one intensive and the other more extensive, were used to assess the impact of pesticides on canopy songbirds in New Brunswick in 1975. The intensive studies were based on counts of singing males along walked transects. The extensive studies were based on motored transects. This technique has been widely used to assess bird populations in North America (Robbins and Van Velzen, 1967). Even extensive techniques only cover a small proportion of the area, as can be appreciated, since the total area sprayed that year in New Brunswick was 2740000 ha. The total avian casualties were estimated at 2 million canopy songbirds and another 0.9 million of wide-ranging species (Pearce et al., 1976). The initial reduction of population in some areas was 80 percent. No long-term effects on population or community structure could be demonstrated. The relationship of field studies to calculate for the forest spraying programme is shown in the last column in Table 13.1. Although the data are incomplete, the general sequence follows the calculated effect.

13.4.3 CASE 3: EFFECTS OF POLLUTANTS IN THE NORTH AMERICAN GREAT LAKES ON FISH-EATING BIRDS AND MAMMALS

The third case study is the reproductive failure of fish-eating birds and mammals in the North American Great Lakes, as an example of two entirely different lines of research¾molecular biology and field investigations¾that eventually blended to provide a comprehensive answer to the problem.

The initial observations in the early 1970s showed severe reproductive impairment of the herring gull, Larus argentatus, and the disappearance of the double-crested cormorant, Phalacrocorax auritus, in the lower Great Lakes (Gilman et al., 1977). Trapping records indicate that population declines of two fish-eating mammals, the mink Mustela vison and the otter Lutra canadensis, also occurred (Environment Canada, 1991). The population increase of the cormorant, from ten to twenty times the pre-pollutant level, after restriction on pollutants had been put in place indicates that effects were exerted on the community structure (Price and Weseloh, 1986).

A correlation was found between polyhalogenated aromatic hydrocarbon (PHAH) levels and reproductive effects. Relating these effects to a specific chemical(s) was much more difficult. Apart from the difficulties inherent in tackling the effects of complex mixtures, two additional problems arose. First, the levels of most PHAHs, cross-correlated strongly to each other, and second the effects seen¾embryotoxicity, edema, structural abnormalities, behavioural changes-were known from laboratory studies to be caused by a wide range of PHAHs. Only in the case of eggshell thinning in cormorants was it possible to assign a specific cause with a high degree of certainty. Detailed field studies, which included egg injection, egg exchange experiments, and nest attentiveness studies, were undertaken (Mineau et al., 1984). These investigations were all the more difficult to conduct because the levels of PHAHs rapidly decreased following bans and restrictions on their usage in the late 1970s.

The identification of the Ah receptor (Poland et al., 1976) by highly specific binding to 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8- TCDD) was a key finding in bringing molecular biology into the realm of toxicology. This receptor controls the induction of the mixed function oxidase (MFO) systems that are of considerable importance in detoxifying a wide variety of pollutants. Studies were broadened to cover several polychlorinated biphenyls (PCBs), polychlorinated dibenzofurans (PCDFs), and polychlorinated dibenzodioxins (PCDDs). A strong correlation between potency to induce MFO activity and their toxicity was found.

The application of this complex biochemistry into field investigations by means of expressing the complex mixtures of PHAHs as "dioxin equivalents" is based on this correlation. The most toxic compound and most potent inducer is 2,3,7,8- TCDD, whose assigned value is 1. All other compounds have lower values; however, when they are multiplied by their concentration, they can give larger values than that obtained for 2,3,7,8- TCDD. This value, potency times concentration, is the "dioxin equivalent." Since the potency is correlated with the ability to induce mixed function oxidases enzymes, the induction of these enzymes can be used to calculate dioxin equivalents (TCDD-EQs).

Significant differences in the ability of egg samples from fish-eating birds from various regions around the Great Lakes to cause induction was found by Tillitt et al. (1991). The relative ranking of colonies correlates well with known areas of contamination. When the overall reproductive success of double-crested cormorant and Caspian tern (Hydroprogne caspia) colonies was plotted against TCDD-EQs of eggs from each colony, a high degree of correlation was found, suggesting that these pollutants, expressed as dioxin equivalents, are the cause of reproductive failure.

The data on mammals are not as clear.  Trapping data, which have severe limitations, indicate decreases of both species, especially along the lake shore where contamination would be expected to be highest (Environment Canada, 1991). Feeding Great Lakes fish to ranch mink can cause reproductive failure, as was demonstrated many years ago. Detailed studies have shown that PCBs caused adverse reproductive effects that are at the environmental levels (Aulerich and Ringer, 1977); however, no such examination has been undertaken in terms of dioxin equivalents.

13.4.4 CASE 4: EFFECTS OF OIL ON AQUATIC AND ADJACENT TERRESTRIAL ECOSYSTEMS

The fourth case study deals with the effects of oil on aquatic ecosystems, including birds and mammals within these systems. This is an example of three levels of research, (1) field observations, (2) field experimentation, and (3) laboratory investigations; these blend to provide a comprehensive understanding of the effects of chemicals. It is also an example where ecosystem functions have been examined extensively both in the field and in the laboratory. In this case, the presence of the oil has stimulated research on lethal and sublethal effects and on immediate and long-term effects. Oil and its breakdown petroleum hydrocarbons are of global concern, because the potential for exposure is cosmopolitan.

Another aspect of oil contamination relevant to methods to assess the effects of chemicals is the potential for acute and chronic exposure. Media and scientific attention is often directed at the massive pollution events that provide acute exposures; yet on a global scale, chronic, low-level exposure may prove to be more challenging to evaluate and regulate the effects of chemicals.

As early as the 1920s the effects of oil were noted on birds. The most obvious effect was mortality due to severe oiling. The increase in oil transport following the Korean War and the use of larger oil tankers led to major oil spills, and these spills still continue today. Although the image of oiled fish, birds, and mammals was very potent, other delayed and sublethal responses were soon suspected. Oil affects all components of the ecosystem from beach grass and Spartina to marine mammals, birds, and terrestrial animals that come to these ocean, estuary, or river waters to feed or drink.

In 1967, the Torrey Canyon went aground in the United Kingdom, and spilled over 80,000 tons of oil into the sea. Since then, over twenty accidents of this magnitude have occurred, including the Amoco Cadiz, the Ixtoc, and finally the Exxon Valdez. Although many oil spills have occurred, the variation in environmental conditions (geography, latitude, time of year) make comparisons challenging. The large number of spills facilitates hypothesis formulation to not only assess adverse effects in the field but also test in the laboratory (NRC, 1981).

Physical characteristics of the oil that determine the effects on marine resources include the volatile fraction, saturated hydrocarbon content, specific gravity, and viscosity. Initially, laboratory studies were designed to examine how variations in these characteristics affect particular organisms. Now, however, laboratory and field experiments and observations are used to understand and evaluate the effect of oil and its constituents. Species and ecosystem characteristics also determine the effects and methods of assessment including: life stage (eggs, young, adult), differential vulnerability by season, mobility, natal habitat (i.e., bottom-dwelling, shoreline), and the presence of refugia (Cairns and Elliot, 1987). The methods to evaluate the effects of oil, from the molecular to ecosystem levels, are numerous and varied (Tables 13.2, 13.3, and 13.4). These tables provide only a partial list of methods at each level of organization, to provide a sense of the type of assessments that have been conducted. Birds and mammals are the focus herein; other examples are only when they indicate the potential of the methodology. For ecosystem effects, examination of both structure and function is essential.

The methods used to assess the effects of oil on ecosystems are further developed than for other contaminants because: (1) many large oil spills have occurred in the past 20 years; (2) many places have produced chronic exposure; (3) the area of exposure is often a well-defined ecosystem such as a river, estuary, or bay; (4) the area of exposure is easy to define visually and with sophisticated oil fingerprinting techniques; and (5) since the cause of the exposure is readily apparent, blame can be established, and funds levied for scientific study, bioremediation, and rehabilitation.

Table 13.2. Illustrations of methods to assess the toxicity of oil on the molecular and physiological levels of organization for wildlife

13.5 PROACTIVE VERSUS REACTIVE APPROACHES

Obviously a need exists to be proactive rather than reactive. The first case study (DDE-induced eggshell thinning) is completely reactive, and is unlikely to have been detected proactively, even with today's methods. The interspecies variation means that, in this case, experimental studies would not have detected eggshell thinning. Historically, although studies have been carried out on the interaction of DDT and its metabolites with ATPases, the extension of the findings to avian eggshell thinning is unlikely even if the chain was not blocked by interspecies variation.

Table 13.3. illustrations of methods to assess the toxicity of oil on the individual and population levels of organization for wildlife

The second case suggests that simple toxicological calculations on chemicals based on toxicity and dose could have prevented serious environmental effects. This case is comparatively straightforward as single chemicals of known mode of action were involved.

The third case study, despite its complexity, gives a hope that eventually a proactive approach based on biochemical mechanisms is possible. In the actual event, an approach combining field observations and molecular biology studies which was successful. However, in theory, the progression from the specific high-affinity binding of TCDD, a known environmental pollutant, to the Ah receptor to toxic effects in the field could have been anticipated. Work on receptors is moving forward rapidly. The receptor approach is being used in the case of the rodenticide flucoumafen. When the hepatic binding sites become saturated the anticoagulant effect becomes lethal (Huckle et al., 1989). This basis is now being used to investigate the possible impact of rodenticides on the barn owl, Tyto alba. A fruitful line of investigation might be to examine the extent that known receptors can be blocked by environmentally important compounds and to examine the implications of the positive findings.

Table 13.4. Illustrations of methods to assess the toxicity of oil on the ecosystem level of organization for wildlife Ecosystem

The fourth case study, although complex, spans the range from molecular to ecosystem effects, and is both reactive and proactive. Studies continue at all levels, from molecular to ecosystem, to evaluate the effects of oil. Since the potential for oil spills continues, oil represents a class of chemicals whose assessment will continue to pose problems for a long time. In theory, the development of methods of assessment, providing data on effects, will ultimately help develop methods of prevention or clean-up.

Overall, these case studies indicate that birds and mammals are important indicator species to evaluate the effects of chemicals. They have served as early warnings in the past, and are expected to continue to do so.

13.6 CONCLUSIONS

Evaluating the effects of chemicals individually or as mixtures on ecosystems is a difficult task, because of the levels of organization and complexity of ecosystems, and the matrix of interactions between all components of the ecosystem. Measurement of effects is not possible on all components or levels, because of constraints on manpower, time, and money. The task of evaluating chemical effects on ecosystems thus converges on selecting a series of indicators, regardless of whether they are at the molecular, individual, population, or ecosystem level.

Birds and mammals are ideal as indicators for several reasons. They are usually high on the food chain, and thus they can bioaccumulate chemicals, making chemicals easier to detect before they can be detected in other organisms, and making the organisms vulnerable to morphological, behavioural, and physiological effects that can be observed easily. These sublethal effects, as well as direct mortality, can result in catastrophic population declines, that is an immediate and clear sign of intolerable levels of contaminants in the environment. Because birds are so visible to the general public, as well as to conservationists and scientists, any changes in their health or population levels are likely to be noted. The public cares about birds and mammals, not only for themselves, but as indicators of environmental health, and ultimately of their own health. The disadvantages of using birds (e.g., many migrate) or mammals (e.g., some migrate, and some are nocturnal) can be overcome by examining the levels of chemicals and their effects in young, using laboratory experiments to ascertain causation and dose-response relationships, and carefully monitoring reproductive success in wild populations.

Overall, the advantages of using birds and mammals, as indicated by these case studies, indicate that each is an important indicator species to evaluate the effects of chemicals. They have provided early warnings of adverse effects of chemicals (e.g., DDT), even before any other organismic or larger ecosystem effects were noted. After birds indicated a problem, similar or other effects were noted in other organisms. Thus, birds and mammals are likely to continue to serve as early warnings of adverse chemical effects, and they should be incorporated into any suite of indicators used to evaluate the effects of chemicals on ecosystems and their component parts.

13.7 REFERENCES

Addison, R.F., Brodie, P.F., Edwards, A., and Sadler, M.C. (1986) Mixed function oxidase activity in the harbor seal (Phoca vitulina). Comp. Biochem. Physiol. 85, 121-124.

Aulerich, R.J., and Ringer, R.K. (1977) Current status of PCB toxicity to mink, and effect on their reproduction. Arch. Environ. Contam. Toxicol. 6, 279-292.

Bolze, D.A., and Lee, M.B. (1989) Offshore oil and gas development. Marine Policy, July, 1-6.

Brzorad, J., and Burger, J. (1993) Fish and shrimp populations in the Arthur Kill. In:

Burger, J. (Ed.) The Arthur Kill: Anatomy of an Oil Spill. Rutgers University Press, New Brunswick, New Jersey.

Burgess P.S., Peakall, D.B., and Landa, V. (in press) Wildlife species as monitors of hazardous waste dumps. In: Moore, J.A., Subramanyam, B.V.R., and Tardiff, R.G. (Eds.) Methods to Assess the Effects of Chemicals on Hazardous Waste Sites. John Wiley & Sons, London.

Burger, J. (in press) Ecological effects of exposures to hazardous waste sites. In: Moore, J.A., Subramanyarn, B.V.R., and Tardiff, R.G. (Eds.) Methods to Assess the Effects of Chemicals on Hazardous Waste Sites. John Wiley & Sons, London.

Burns, K.A. (1976) Microsomal mixed function oxidases in an estuarine fish, Fundulus heteroclitis, and their induction as a result of environmental contamination. Comp. Biochem. Physiol. 53B, 443-446.

Cairns, D.K., and Elliot, R.D. (1987) Oil spill impact assessment for seabirds: the role of refuging and growth centres. Biol. Conserv. 40, 1-9.

Costa, D.P., and Kooyman, G.L. (1982) Oxygen consumption, thermoregulation and the effect of fur oiling and washing on the sea otter, Enhydra lutris. Can. J. Zool. 60, 2761-2767.

Eastin, W.C., and Hoffman, H.P.J. (1979) Biological effects of petroleum on aquatic birds. In: Proceedings of the Conference on Assessment of Ecological Effects of Oil Spills Vol. 8, pp. 561-582.

Engelhardt. F.R. (1982) Hydrocarbon metabolism and cortisol balance in oil exposed ringed seals, Phoca hispida. Comp. Biochem Physiol. 72C, 133-136.

Englehardt, F.R., Wong, M.P., and Duiey, M.E. (1981) Hydromineral balance and gill morphology in rainbow trout Salmo gairdneri, activated to fresh and sea water as affected by petroleum exposure. Aquat. Toxicol. 1, 175-186.

Environment Canada. (1991) Toxic Chemicals in Great Lakes and Associated Effects. Synopsis. Vol. II. Effects, pp. 495-755. Environment Canada, Ottawa.

Frink, L. (1993) Rehabilitation of oiled birds. In: Burger, J. (Ed.) Anatomy of an Oil Spill: The Arthur Kill. Rutgers University Press, New Brunswick, New Jersey.

Fritts, T.H., and McGehee, M.A. (1981) Effects of Petroleum on the Development and Survival of Marine Turtle Embryos. Report to US Department of the Interior, Fish and Wildlife Service, Contract No.121-16-0, Washington, D.C., 41 pp.

Gilman A., Peakall, D.B., Hallett, D., Fox, G.A., and Norstrom, R.J. (1977) Herring Gulls (Larus argentatus) as monitors of contamination in the Great Lakes. In: National Research Council, Animals as Monitors of Environmental Pollutants, pp. 280-289. National Academy of Sciences, Washington, D.C.

Ginsburg, J.M. (1945) Toxicity of DDT to fish. J. Econ. Entomol. 39, 274-275.

Hartsough G.R. (1965) Great Lakes fish now suspect as mink food. Am. Fur Breeder 38(10), 25-27.

Hickey, J.J. (1969) The Peregrine Falcon Populations: Their Biology and Decline. University of Wisconsin Press, Madison, 596 pp.

Huckle, K.R., Hutson, D.H., Logan, C.J., Morrison, B.J., and Warburton, P.A. (1989) The fate of the rodenticide flocoumafen in the rat: retention and elimination of a single oral dose. Pest. Sci. 25, 297-312. 

Hutchinson, J., and Simmonds, M. (1991) A Review of the Effects of Pollution on Marine Turtles. Greenpeace International, Washington, D.C., 20 pp.

Kinako, P.D.S. (1981) Short-term effects of oil pollution on species numbers and productivity of a simple terrestrial system. Environ. Pollut. Ser. A 26, 87-91.

Langford R.R. (1949) The effect of DDT on freshwater fish. In: Ontario Department of Lands and Forests. Forest Spraying and the Effects of DDT, pp. 19-27. Biological Bulletin 2. Ottawa, Ontario.

Leighton F.A. (1991) The toxicity of petroleum oils to birds: an over-view. In: White, J., and Frink, L. (Eds.) The Effects of Oil on Wildlife, pp. 78-94. IWRC, California.

Lincer, J.L. (1975) DDE-induced eggshell thinning in the American kestrel: A comparison of the field situation and laboratory results. J. Appl. Ecol. 12, 781-793.

Lindenof, O., Elmgren, R., and Boehm, P. (1979) The Tsesis oil spill: its impact on the coastal ecosystem of the Baltic Sea. Ambio 8, 214-253.

Logie, R.R. (1975) Effects of aerial spraying of DDT on salmon populations of the Miramichi River. In: Prebble, M.L. (Ed.) Aerial Control of Forest Insects in Canada, pp. 293-300. Environment Canada, Ottawa.

Loya, Y. (1975) Possible effects of water pollution on the community structure of Red Sea corals. Mar. Biol. 29, 177-185.

Mineau, P., Fox, G.A., Norstrom, R.J., Weseloh, D.V., Hallett, D.J., and Ellenton, J.A. (1984) Using the herring gull to monitor levels and effects of organochlorine contamination in the Canadian Great Lakes. Adv. Environ. Sci. Technol. 14, 425-452.

Noss, R. (1990) Indicator for monitoring biodiversity: a hierarchial approach. Con. Biol. 4, 355-364.

NRC (National Research Council). (1981) Testing for Effects of Chemicals on Ecosystems. National Academy Press, Washington, D.C., 128 pp.

Paine, R.T. (1966) Food web complexity and species diversity. Am. Natur. 100, 65-75.

Peakall D.B. (1975) Physiological effects of chlorinated hydrocarbons on avian species. In: Hague, R., and Freed, V.H. (Eds.) Environmental Dynamics of Pesticides, pp. 343-360. Plenum Press, New York.

Peakall, D.B., and Bart, J.R. (1983) Impacts of aerial applications of insecticides on forest birds. CRC Crit. Rev. Environ. Control. 13, 117-165.

Peakall, D.B., Lincer, J.L., Risebrough, R.W., Pritchard, J.B., and Kinter, W.B. (1973) DDE-induced egg-shell thinning: Structural and physiological effects in three species. Comp. Gen. Pharm. 4, 305-313.

Pearce P.A., Peakall, D.B., and Erskine, A.J. (1976) Impact on Forest Birds of the 1975 Spruce Budworm Spray Operations in New Brunswick. CWS Progress Notes No.62, 7 pp.

Perez, P. (1979) State of exploitable algae populations in January 1979. In: Rapport Inst. Peches Marit. Roneo., pp. 1-3.

Poland, A., Glover, E., and Kende, A.S. (1976) Stereospecific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin by hepatic cytosol. Evidence that the binding species is receptor for induction of aryl hydroxylase. J. Biol. Chem. 251, 4936-4946.

Price, I.M., and Weseloh, D.V. (1986) Increased numbers and productivity of double-crested cormorants, Phalacrocorax auritus, on Lake Ontario. Can. Field-Natural. 100, 474-482.

Ridjke, A.M. (1970) Wettability and phylogenetic development of feather structure in waterbirds. J. Exp. BioI. 52, 469-479.

Robbins, C.S., and Van Velzen, W.T. (1967) The Breeding Bird Survey, 1966. Special Sci. Report No.102. US Department of the Interior, Fish and Wildlife Service, Washington, D.C. 43 pp. 

Schwartz, J.P. (1985) Effect of oil contaminated prey on the feeding and growth rate of pink salmon prey (Oncorhyncus gorbuscha). In: Vemberg, F.J., Thurberg, F.P., Calabrese, A., and Vemberg, W. (Eds.) Marine Pollution and Physiology, Recent Advances, pp. 459-476. South Carolina Press, Columbia.

Sheehan, P.J. (1984a) Functional changes in ecosystems. In: Sheehan, P.J., Miller, D.R., Butler, G.C., and Bourdeau, P. (Eds.) Effects of Pollutants at the Ecosystem Level, pp. 101-146. SCOPE 22. John Wiley & Sons, Chichester.

Sheehan, P.J. (1984b) Effects on community and ecosystem structure and dynamics. In: Sheehan, P.J., Miller, D.R., Butler, G.C., and Bourdeau, P. (Eds.) Effects of Pollutants at the Ecosystem Level, pp. 51-99. SCOPE 22. John Wiley & Sons, Chichester.

Teal, J.M., and Hawarth, R.W. (1984) Oil spill studies: A review of ecological effects. Environ. Manag. 8, 27-44.

Tillitt, D.E., Ankley, G.T., Verbrugge, D.A., Giesy, J.P., Ludwig, J.P., and Kubiak, T.J. (1991) H4IIE rat hepatoma cell bioassay-derived 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalent in colonial fish-eating waterbird eggs from the Great Lakes. Arch. Environ. Contam. Toxicol. 21, 91-101.

Weis, J.S., and Khan, A.A. (1991) Reduction in prey capture ability and condition of mummichogs from a polluted habitat. Trans. Am. Fish. Soc. 120, 127-129.