Page 755 - The Toxicology of Fishes

Aquatic Ecosystems for Ecotoxicological Research 735

environmental factors that influence fish growth, independently of any chemical dosing, must be under-
stood. Chemical dosing can be viewed as a probe with which to perturb the system. Quantifying biotic
responses to such perturbations helps us understand basic fish population dynamics.

Historical Perspective

Forbes (1887), in his work on lake natural history, detailed basic principles of ecological synergism,
variability, and dynamic equilibrium, as well as the complex interactions of predator and prey. Though
speaking of the lake itself, not of the surrogate systems routinely employed in aquatic research today,
Forbes touched on the rationale for using artificial systems in both toxicological and ecological research:
“It forms a little world within itself—a microcosm within which all the elemental forces are at work and
the play of life goes on in full, but on so small a scale as to bring it easily within the mental grasp.” This
postulates the underlying basis for using microcosms (and mesocosms) in ecotoxicological research: the
assertion that they simulate processes that occur in nature enough to be viable surrogates for natural systems.
Initial applications of artificial aquatic systems such as laboratory microcosms, artificial ponds, and
various in situ enclosures were historically used in ecological studies of productivity (Kevern and Ball
1965; McConnell, 1965), community metabolism (Beyers, 1962, 1963; Copeland, 1965), and population
dynamics (Deegan et al., 1997; Stein et al., 1995; Vanni et al., 1997). This research helped lay the
groundwork for understanding how biotic processes function in artificially bounded and maintained
systems. A fundamental knowledge of systems ecology is necessary if there is to be any understanding
of how introduced perturbations (e.g., chemical insult) may be measured over and above natural pertur-
bations of competition, predation, or background chemical and physical milieu. A subject of considerable
concern and debate is whether microcosms simulate natural systems closely enough to be used as
ecosystem surrogates, particularly when fish are included. Microcosms typically do not closely simulate
natural systems at all levels of ecological organization. The small scale of microcosms has not been a
problem for plankton or invertebrates, but their use remains problematic for fish.
The use of surrogate systems in toxicological research, particularly those encompassing any appre-
ciable scale or complexity, is relatively recent (ca. 1960). Concern over the effects of insecticides used
to control mosquito populations in California prompted a series of field studies on the consequences of
chemical control methods on non-target species such as mosquito fish (Gambusia sp.) and waterfowl.
Keith and Mulla (1966) and Mulla et al. (1966) used replicated artificial outdoor ponds to examine the
effects of organophosphate-based larvicides on mallard ducks. Hurlbert et al. (1970) conducted subse-
quent studies in the same systems, examining the impact of this class of chemicals on a greater number
of species within several broad taxa: phytoplankton, zooplankton, aquatic insects, fish, and waterfowl.
Essentially, system-level responses attributable to the pesticide were studied with concomitant changes
in the fish or waterfowl population of interest.
Broad application of microcosms and mesocosms in toxicological studies arose after the realization
that single-species toxicity tests, alone, were inadequate for predicting effects at the population and
ecosystem levels (Cairns, 1983; Kimball and Levin, 1985). It was felt that single-species laboratory
toxicity tests were protective, but not predictive, of ecosystem responses (Fairchild et al., 1992). In
addition, multispecies tests can demonstrate effects not evident in laboratory tests that use a single species
(Cairns, 1986). As environmental protection goals focus on ecosystem-level organization, testing with
more complex systems may involve less extrapolation, presumably with an enhanced capability of
predicting impacts on natural systems. Yet, it remains a fact that the optimal use to which stream
mesocosms have been put is to help understand the complexity of factors influencing growth and
survivorship (Brooks et al., 2004; Stanley et al., 2005). In those studies, fathead minnows (Pimephales
promelas) and amphipods (Hyalella azteca) were used as focal species to learn how aqueous cadmium
moves through the food chain in model ecosystems.
The ecological risk of pesticide application is assessed under the U.S. Federal Insecticide, Fungicide
and Rodenticide Act (FIFRA). The data collection process detailed under FIFRA involves a progression
of increasingly complex toxicity tests, considered together with an estimate of environmental exposure,
to make a determination whether a chemical may pose an unacceptable risk to the aquatic environment.