Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
L egitimate concerns exist about the biological and eco Classification of logical consequences of introducing new or altered organisms into the environment on a large scale. Although these concerns are not restricted to organ Risks Associated with the· isms altered with R-ONA techniques, they have been brought into focus by the possibility that genetically altered organisms will be used more extensively in the future, in both traditional Introduction of and altogether different ways. Some risks are associated with the introduction of certain organisms, regardless of the method by which they were produced. Therefore, society's task must be R-ONA-Engineered Organisms to classify and manage the risks appropriately. Human beings have moved many organisms from the ecosys tems in which they evolved into different ecosystems, for a into the Environment: variety of reasons. Almost all our food crops and animals have been introduced from other ecosystems, as have many of our ornamental plants and our pets. Bacterial and fungal parasites What Factors Need to Be have been introduced to control harmful insects. Microorgan isms have been added to seed and soil to increase crop growth by improving nitrogen fixation. Although we have less informa Considered? tion about nonpathogenic microorganisms than about patho genic microorganisms, plants, animals, and insects, we know that only a very small fraction of all the attempted or accom plished introductions have led to destructive invasions (Sim berloff, pp. 152-161, in Halvorson et al., 1985). Large-scale plantings of genetically modified nonnative crops, such as wheat, soybeans, and com, have generally not resulted in the escape of plants from cultivated fields into unmanaged ecosys tems as weeds. Furthermore, the biological control of certain
19 insects and pest plants by introduced, nonnative parasites and ignorant or whose properties demand greater concern about predators has had negligible negative environmental impact ecological consequences. Nonetheless, a small fraction of introductions of nonnative If we are to proceed prudendy with the use of R-ONA organisms have gone awry, and these have been the subject of engineered organisms, we must create categories that permit us considerable concern (Mooney and Drake, 1986). japanese to classify relative risks associated with environmental intro beedes, gypsy moths, the kudzu vine, and starlings provide ductions, so that levels of containment and environmental familial; frequendy cited examples of uncontrolled, destructive assesment will be appropriate to the intended use. This sec s invasions of nonnative organisms. Introduced nonnative fish tion of the paper identifies and discusses the scientific consid species have driven many indigenous freshwater fish popula erations that must underlie the effon to categorize risk. tions in the western United States and elsewhere to the brink of extinction. And large shifts in species composition have Source and Target Environments occurred throughout subtropical areas of North America, where Although introductions that have caused major ecological introduced fish species have largely displaced native fish spe disturbances can be cited, most (such as the chesmut blight cies. Thus, introductions of nonnative organisms are associated fungus) have involved the movement of an organism from one with risks, and these must be weighed against the benefits. environment into anothet These are inappropriate models for Moreovet; the capacity to alter organisms to carry out specific R-ONA-engineered organisms being reintroduced into the chemical tasks, such as the recovery of metals from ores or the environment from which the organisms were taken. For crop degradation of toxic organic chemicals, will make it possible to species and other organisms being reintroduced into the source use organisms in new ways in environments not previously environment, traditional experience in the breeding and testing subjected to such alterations. of new strains of plants and microbes is the most appropriate Yet even a casual enumeration of the organisms that have model. However, for introductions involving R-ONA been and will be engineered makes it clear that some kinds of engineered organisms taken from quite different environments engineered organisms warrant greater concern than others. For or geographic locations, the accumulated experience with some, sufficient knowledge of their ecological characteristics introduced species is most appropriate for risk assessment. permits us to alter them in various ways and introduce them Many of the currendy proposed introductions are in agricul into the environment with litde or no risk of adverse conse ture, and the organisms are unlikely to become widely estab quences to either the human population or the target ecosys lished outside the field to which they are applied. For most tem. In contrast, there are others about which we are relatively crop species, the chance of proliferation as weeds is remote.
20 That depends, howevet; not only on the recipient environment and these depend on its environment, which in many cases is but on the organism, because a number of species survive as created or influenced by human beings. For example, modem weeds in some, but not all, noncropland habitats. For introduc com is largely a creation of humans, selected over thousands of tion into unmanaged ecosystems, the characteristics of the years for its usefulness as a food plant. Today's high-yielding existing ecological community must be considered along with hybrid com varieties depend completely on people for propa the environment. For the introduction of nonnative organisms, gation and culture and cannot become widespread weeds in it cannot be said that all ecological communities are stable and nonmanaged areas. Hence R-ONA-engineered com plants are resilient to perturbation, in light of considerable evidence to not likely to cause problems. In contrast, plants with broad the contrary. Some communities are more likely to be invaded dispersal capabilities or weedy relatives merit more careful than others, and such differences are critical in determining the attention. success of any introduction. The different meanings of the term "introduction" must be considered for various organisms. Although testing live vac The Biological and Ecological cines in farm animals, planting R-ONA-engineered crops, and Characteristics of the Organism releasing R-ONA-engineered insects all constitute introduc For the determination of ecological risk, the biological prop tions into the environment, the extent to which the various erties of the R-ONA-engineered organism are paramount. For organisms can become established varies widely. The classifica example, if the organism is a pathogen or if the R-ONA modifi tion of organisms on the basis of such characteristics should cation affects pathogenicity or invasiveness, appropriate safe make it possible to proceed with many experiments either guards are essential. Yet strict and rigid controls for all without significant risk or with no greater risk than we already organisms are not justified. It is inappropriate to treat every accept as part of traditional breeding, biological control, and microorganism as though it were a potential pathogen, because vaccine development. the likelihood of converting a nonpathogen into a virulent pathogen by a small genetic change is extremely slight. Scale and Frequency of Introductions The ecologically important characteristics of an organism Growing evidence indicates that the establishment of many include survival, reproductive potential, dispersal characteris species, such as those used for biocontrol, is unpredictable and tics, pathogenicity, competitiveness, and the manner in which depends on the confluence of such factors as favorable weathet; it is involved in essential processes in the ecosystem. Each favorable sites, and suitable vectors or other means of transport. organism has unique patterns of reproduction and survival, Although the data on accidental introductions do not permit
21 the same level of quantitative analyses as for introductions related to biocontrol of insects or weeds, the conclusions are simila.: Some introductions will not succeed no matter how often they are repeated. More generally, success depends to some extent on the scale and frequency with which organisms are introduced. This applies both to the difficulty of establish ing organisms that we want to succeed and the ease of estab lishment by those that may create problems. Experience in biological control has shown that success is enhanced in some instances if the scale or frequency of application is increased, and thus the scale and frequency of a given introduction are of central importance. The implication for R-ONA-engineered organisms is that large-scale or sustained applications might have consequences different from small-scale or single applications. The attractiveness of R-DNA genetic engineering methods lies in the specificity and efficiency with which they allow genetic manipulations. In tum, this may increase the frequency of introductions. Thus, the cumulative probability of undesirable effects resulting from repeated applications or frequent intro ductions must be considered, although if care is exercised in the preliminary analysis of environmental risk, most introduc tions will pose a low risk of environmental damage.
