| ||Format||Pages||Price|| |
|PDF (256K)||17||$25||  ADD TO CART|
|Complete Source PDF (8.1M)||550||$65||  ADD TO CART|
In the past few years, there has been widespread acceptance of the idea that a scientifically justifiable estimate of the hazard of a chemical to aquatic life can be obtained in a cost-effective way from data systematically generated for that purpose. One cannot assume that data generated for research purposes other than hazard evaluation will be adequate to meet this need, since such data have been demonstrably inadequate for producing criterion documents for the 65 chemicals named in the consent decree imposed on the U.S. Environmental Protection Agency (EPA) to stimulate the production of these documents. Since the EPA has had insufficient time to generate a substantial body of its own evidence, data were obtained from the open literature. Although the evidence was, in most cases, exemplary in terms of the original purpose for which it was intended, it was inadequate in the aggregate for the purpose of hazard evaluation. The Toxic Substances Control Act provided the primary impetus for the development of hazard evaluation protocols that require the systematic generation of data. Although these were designed primarily for new chemicals, the need to use them for some existing chemicals has been recognized.
Although the structure and organization of hazard evaluation or toxicity testing protocols are quite varied, the most familiar ones espouse sequential testing. That is, simple inexpensive range-finding tests involving single species are used at the outset, and one proceeds through tiers or phases in which the tests increase in complexity, sophistication, cost, and, frequently, duration. It is generally recognized that the amount of evidence needed to make a sound estimate of hazard will differ from chemical to chemical, and, therefore, provision has been made for terminating testing at multiple points in the sequence. One expects that only a few of the more dangerous or persistent chemicals will require every test in the series, since the large volume of new chemicals being generated and the scarce resources for testing now prohibit a complete series of tests on all chemicals.
Justification for sequential testing is as follows:
1. More expensive, sophisticated tests can be carried out more efficiently if the results from simpler toxicity tests are available when the complex tests are designed.
2. Carrying out tests in sequence is most likely to ensure that an adequate amount of data is made available for the estimate of hazard without markedly overshooting the point at which a sound evaluation of hazard can be made.
3. In situations in which one is examining an array of chemicals that are all intended for the same purpose (to select the one with the least environmental impact), sequential testing is likely to identify the most suitable candidate chemical at the least cost.
A strong case can also be made for simultaneous testing at different levels of biological organization:
1. No compelling evidence exists that single-species tests can be used to predict multi-species, community, or ecosystem responses accurately. Additionally, the ability to predict sublethal effects, such as altered behavior, growth, reproductive success, and the like, from short- or long-term tests on lethality has not been exemplary. If such a prediction were adequate, the need for additional tests beyond the range-finding test would not be so crucial. Since the information from toxicity tests at the beginning of the sequence is not demonstrably correlated with responses in the latter parts of the sequence, the ability to use information from the first part of the sequence in the design of subsequent toxicity tests is questionable.
2. If the time involved in collecting sufficient information to make an adequate estimate of hazard is important (because of the financial outlay in developing a new chemical and the like), delay in reaching a critical mass of information for a sound decision is a major cost factor. Even if money were saved in the sequential testing procedure, this might be overbalanced by the amount lost elsewhere as a consequence of the additional delay. This is particularly true in view of the fact that short-term tests are carried out early in the sequence and longer-term tests are postponed until the short-term tests are completed.
3. The sequential arrangement of tests from the simple to the more complex possibly reflects, in a general way, the historic development of the field. Therefore, toxicity tests—with which there is a long familiarity—are placed early in the sequence and more recent and more sophisticated tests, which are still in the experimental stage of development, are placed last. Because our awareness of the need for additional information has evolved in this fashion (that is, from simple, short-term crude tests to more sophisticated tests), it does not mean that this evolution of thinking has to be repeated in the hazard evaluation process.
The belief that multispecies, community, and ecosystem toxicity tests are second-order tests that can only be carried out after single-species tests is an assumption that deserves serious attention. There is no compelling evidence that single-species tests can be used to predict reliable responses at more complex levels of organization. Development of suitable tests at higher levels of organization than single species may well have been impaired because of the views just discussed, since both funding and research priorities almost certainly have been influenced by them. Although tests at higher levels of organization may be expensive, many are less expensive than or comparable in cost to long-term, continuousflow exposure tests of single species. If toxicity testing at different levels of organization is to be simultaneous instead of sequential, much more attention needs to be given to increasing the array of suitable test methods for multispecies testing than has been the case in the past.
toxicity testing protocols, hazard evaluation, multispecies toxicity tests, ecosystem toxicity tests, aquatic toxicology, hazard assessment
University distinguished professor, Department of Biology, and director, University Center for Environmental Studies, Virginia Polytechnic Institute and State University, Blacksburg, VA