John J. Obrycki
Department of Entomology
Iowa State University Ames, IA. 50011
Maurice J. Tauber and Catherine A. Tauber
Department of Entomology
Cornell University Ithaca, NY. 14853
John R. Ruberson
Department of Entomology
University of Georgia, Tifton, GA 31793
Apunte aquí para versión en Español [X]
Contrary to common assumptions, many predacious insects are not generalist feeders; some are conspicuously selective in their diet. Dietary breadth represents a key component in the ecology, behavior, and evolutionary diversification of predators. It is also a central issue in the manipulation and use of predators in IPM. For example, prey specificity impinges directly on the use of predators in all types of biological control -- importation, as well as augmentation and manipulation (e.g., Hagen et al. 1976, Hagen 1987). Biological control workers generally favor using host-specific natural enemies, and in choosing natural enemies, they often consider a high level of host specificity as a prime attribute. Indeed, the first modern success in biological control involved a highly specific predator, the vedalia beetle, Rodolia cardinalis (see Huffaker and Caltagirone 1986 for a brief history of this success story). However, prey specificity is not always necessary in pest management; many predatory species that are currently mass-produced and released in field crops and orchards, have relatively broad prey ranges; examples include the anthocorids Orius insidiosus and O. tristicolor and the green lacewings Chrysoperla carnea and C. rufilabris. The desire for dietary specificity presents a dual challenge for classical biological control programs involving predators. On the one hand, the interaction between the predator and target pest should be predictable; on the other hand, the nutritional requirements for an introduced predator may be complex and may include factors that are not readily ascertained, e.g. acceptance or preference for novel prey in the new habitat, obtaining nutrients from symbiotic microorganisms, utilization of supplementary food, such as plant fluids, fungi, pollen, or nectar (e.g., Hagen et al. 1976, Ruberson et al. 1986, Obrycki and Orr 1990). A predator's use of these dietary resources has important implications for the outcome and stability of predator-prey dynamics (e.g., Begon et al. 1996). Given the broad range of variability in the diet breadth of predacious insects, biological control efforts would benefit greatly from detailed investigations of predator-prey interactions; such studies will become a necessity in the near future.
The release of insect predators with relatively broad host ranges raises significant questions about negative effects on non-targeted organisms. It is possible that introduced predators may move out of the target crop area and prey on species other than pests (possibly rare or threatened species) (Howarth 1991). To our knowledge, adverse effects from the introduction of insect predators have been examined for only one species. The dispersal of the polyphagous aphid-feeding coccinellid, Coccinella septempunctata, has been associated with reduced abundance of two indigenous coccinellid species in South Dakota and possible adverse effects on the biological control of the alfalfa weevil in Utah (Elliot et al. 1996, Evans and England 1996). Thus, assessment of diet breadth should be an integral part of the pre-release evaluation of entomophagous species, especially when rare or threatened species may be at risk. In this regard, many aspects of host specificity in biological control programs were recently reviewed (BioScience, June 1996, Vol. 46, No. 6).
Typically, insect predators are characterized by a set of attributes that distinguish them from parasitoids, the other major group of entomophagous insects. They are large relative to their prey and require more than one prey individual to complete development; they have free-living predatory immature stages; and many species of insect predators are predacious as both immatures and adults (e.g., Doutt 1964, Hagen et al. 1976). For most predacious insects, prey consists of other insects, but some may consume animals in other classes. Except for predatory Hymenoptera that provision their nests with prey, predacious insects consume their prey immediately after attack. Predators occur in approximately 20 insect orders; lists vary slightly, depending on the definition of scavenging and fortuitous predation. The only orders of insects not known to contain predatory species are the Isoptera, Phasmatodea, Phthiraptera, Strepsiptera, and Siphonaptera (Hagen 1987, New 1991).
Feeding Habits of Insect Predators
Predacious insects exhibit great variation in their dietary range. Some, e.g, the vedalia beetle, Rodolia cardinalis, and the green lacewing Chrysopa slossonae, are highly specific and feed on only one species of prey. Others, e.g. the aphid-feeding coccinellids Hippodamia convergens and Adalia bipunctata, are stenophagous or oligophagous and restrict feeding to a range of related taxa. Finally, other predators, e.g. the bug Podisus maculiventris, and the lady beetle Coleomegilla maculata, are polyphagous (general feeders) and consume a wide variety of prey and non-prey items, e.g. plant fluid, pollen. It is noteworthy that many species with wide prey ranges contain biotypes and populations that differ in their responses to prey; examples include web-building and jumping spiders and the green lacewing Chrysopa quadripuncata (e.g., Hedrick and Riechert 1989, Tauber et al. 1995). In certain cases, symbiotic microorganisms may supply nutrients for development or reproduction. For example, the common green lacewing Chrysoperla carnea harbors yeasts (Torulopsis) in an enlarged esophageal diverticulum; these yeasts can provide essential amino acids that are lacking in the lacewing's diet (Hagen 1987). Predators can be classified according to the life stage of prey they attack (e.g., egg predators), their foraging strategy (e.g., active searchers, ambush or filter feeders, web or bolas builders). Although these types of categories may be generally informative about the type of prey taken, they have no predictive value for the prey specificity of individual species. In this regard, it is of both practical and theoretical significance that predator phylogeny can be an important clue to prey breadth and prey preference. For example, among the Coccinellidae, the Chilocorninae prey on Homopteran scale insects, most species of Coccinellinae are aphid predators, and the Stethorinae has specialized on phytophagous mite species (Gordon 1985). However, it is critical to note that sister-species may differ in the range of prey that they take within their preferred type of prey; e.g., sister-species may include a generalist and a specialist aphid-feeder (e.g., Albuquerque et al. 1997). Therefore, care must be exercised in using phylogenetic relatedness to generalize about the prey breadth of specific taxa.
Methods for Determining Prey Range
A major challenge in IPM is to assess the action of biological control agents. Unlike parasitoids that leave the exoskeleton of their host, predators often leave no clues after feeding. Observation and gut analysis have been used for a long time to determine what predators eat. For chewing-type predators, analysis of feces can also indicate what prey were in the diet. Considerable advancement has taken place recently in the development of serological techniques for identifying which predators feed on target pests in the field; ELISA and immunodot assays are relatively rapid, inexpensive and easy to interpret (Greenstone 1996). Although quantifying the prey consumed presents significant technical problems, the tests are useful for determining key predators within agricultural systems and for evaluating the efficacy of augmentative biological control agents (Hagler and Naranjo 1996).
Predators that are capable of taking a very catholic diet (e.g., in the laboratory) often take a much narrower range of food. As in the case of herbivores and parasitoids, diet-breadth in predators results from the interaction of diverse physiological, behavioral and ecological factors: (a) the relative availability of specific types of prey, (b) the foraging behavior of predators, (c) the suitability of prey, and (d) the risk of predation or other mortality factors associated with obtaining prey (e.g., Price 1984, Endler 1991, Begon et al. 1996). In many cases parental behavior, especially oviposition behavior, can play a crucial role in determining the prey that are available to predacious larvae. For comprehensive reviews of the factors that influence foraging in many groups of invertebrate predators, including predacious terrestrial and aquatic insects, phytoseiid mites, and spiders, see chapters by Hagen (1987), McMurtry & Rodriguez (1987), Riechert and Harp (1987), Sih (1987a, b).
Prey abundance can determine whether a predator enters or remains in a habitat, as well as the type of prey and relative numbers of prey that the predator consumes. A predator is said to show a preference for a specific type of prey if the percentage of that type of prey in its diet is higher than the proportion of the prey available to it. Although some predators' prey preference is fixed (maintained irrespective of relative availability in the environment), other predators may switch to more common prey species (e.g., Begon et al. 1996). Seasonal availability of prey can also be a major determinant of prey specificity (Evans 1982). Predators show several adaptations to low prey densities; these include changing trophic levels (e.g., intra-guild predation, cannibalism), movement to areas with higher prey densities, switching to new prey, and entering prey-mediated dormancy (e.g., Murdoch and Marks 1973, Tauber et al. 1986, Rosenheim et al. 1995).
Predator Foraging Behavior
Habitat & prey location by insect predators
Adult predators may use a variety of cues, including volatile chemicals from both plants and insects, to find prey. The role of chemical cues in locating prey is relatively well understand for the green lacewing Chrysoperla carnea (Hagen 1987). A complex mixture of volatiles (an attractant and an arrestant) induce C. carnea adults to fly upwind, toward a food source, and having reached the food, to cease movement and to oviposit. One of the volatile chemicals is a metabolite of oxidized tryptophan (most likely indole acetaldehyde) which occurs in aphid honeydew. In addition, for the adults to respond to the tryptophan in honeydew, they must perceive a volatile synomone from plants, caryophyllene, which is a relatively common volatile from cotton and alfalfa plants. Knowing the influence of these compounds on predator behavior forms a strong basis for manipulating these natural enemies in the field. Their application attracted large numbers of lacewings to cotton and potato fields. Several species of predacious beetles and bolas spiders also use chemical cues (plant terpenes, prey pheromones, or mixtures of terpenes and pheromones) to find their prey (e.g., Yeargan 1988). Larval predators have evolved various means of locating and recognizing their prey: phototactic and/or geotactic responses, vision, olfaction, sound or vibration detection, contact (see Hodek 1993). For example, predacious mites use chemical cues to find their prey (Sabelis 1992). Searching behavior may be altered by encountering prey; some larval predators switch from linear movement to area-intensive searching after they contact prey. A predator's previous experience (learning) can influence its searching behavior, as well as the type and proportion of prey taken (see Begon et al. 1996). Plant structure may also affect the pattern of predator movement (e.g., Obrycki 1986, Eigenbrode et al. 1996, Heinz and Zalom 1996). Species and biotypes of predators can differ radically in their habitat associations and in the microhabitats they choose. For example, certain ladybird beetles occur in plant canopies, whereas others inhabit areas closer to the soil (Hodek 1993). Chrysoperla rufilabris prefers moist habitats, whereas C. carnea is well adapted to dry situations (Tauber and Tauber 1983, 1993). These differences in habitat association are reflected in the prey available to predators and they are a very important factor in the choice of species or biotypes to be used in specific biological control situations. Prey Acceptance. After the predator discovers prey, it must pursue, subdue, and consume it. These three functions constitute the predator's handling time, which is a very important determinant of diet breadth (e.g., Begon et al. 1996). By definition, generalist predators attack, subdue and consume a wide range of the prey species they encounter. Several morphological and physiological factors can influence prey acceptance (Hagen 1987). One underlying physical factor is the size of prey (e.g., Sih 1987a, Sabelis 1992). After contact with a potential prey, the characteristics of cuticle or the presence of waxes may affect the predator's responses to the prey. For example, prey acceptance by the vedalia beetle, which has a very restricted diet breadth, appears to be influenced by the waxy texture of its prey, the cottony cushion scale (Hagen 1987). Prey defenses (e.g., behavioral, morphological, chemical) may determine whether a predator accepts or rejects prey. Some prey (e.g. aphids) kick, run, drop, or fly away, or exude noxious chemicals when predators approach them. Sometimes, the chemicals induce the predators to vomit. Sih (1987b) and Endler (1991) discuss the defensive interactions between predators and their prey.
If certain types of prey are not suitable (i.e., they have low nutritional quality for the predator), the predator may ultimately reject the prey, or it may continue feeding, but with detrimental effects. The negative effects include reduced rates of development, reproduction or survival. For example, several species of coccinellids and chrysopids exhibit slow development and reduced fecundity when they eat suboptimal aphid prey (Obrycki & Orr 1990, Hodek 1993; Albuquerque et al. 1997). In some cases, predators continue feeding when the prey contain toxins that result in the predator's death. Prey suitability may or may not be the same for immatures and adults. For example, successful development and reproduction in the green lacewing, C. slossonae, requires that both larvae and adults feed on the predator's specific prey (the woolly alder aphid) (Albuquerque et al. 1997). A diverse array of studies on the nutritional (dietary) requirements of predators have been conducted (see review by Hagen 1987). Despite these studies, relatively few insect predators can be mass-reared successfully, and even fewer can be reared on artificial diets. The development of artificial diets for predators is a very significant area for the future development of biological control (Agriculture Research 1997, pp 2, 4-7).
Natural Enemies Associated with Prey
Virtually all animals are vulnerable to natural enemies; active predators themselves often carry a very high risk of detection and attack by parasitoids or other predators. Thus, an important determinant of diet breadth is the range and abundance of natural enemies that are associated with the predator's food source (Endler 1991). Predators' defenses include becoming immobile, living in protected places (ambush predators, such as pit-dwelling antlion larvae), cryptic coloration and polymorphisms (Hawaiian spiders and lacewing larvae), mimicry (mantispid-mimicing wasps), escape and threat behavior, and noxious chemical exudates (see Endler 1991 for an extensive list). The evolution of a predator's defenses against natural enemies may result in trade-offs in terms of reduced time or efficiency in searching for, attacking, consuming, and metabolizing prey.
Stability in predator-prey interactions constitutes a vital issue in IPM. To evaluate the long-term persistence of prey specificity, it is necessary to understand the genetic variation and evolution of predator-prey associations. Such an understanding requires analysis of the behavioral, physiological, phenological and other traits that underlie predators interactions with prey, as well as comparative experimental studies of predators within a phylogenetic context. In contrast to herbivores, very few such studies exist for predacious insects. Such studies would in;hance the further development of IPM.
Genetic Variation in Predator-Prey Interactions
Genetic variability in predator-prey relations have been examined for only a very few arthropod predators -- the spider Agelenopsis aptera and the green lacewing Chrysopa quadripunctata. For the two species, differences in both foraging and defensive behavior are related to variation in prey availability and predation risks among habitats (Hedrick and Riechert 1989, Tauber et al. 1995). In the case of the lacewing, the expression of the genetic variation in foraging and defensive behavior is enhanced by high levels of individual repeatability in the behavior, even after the predacious larvae moult twice (Tauber et al. 1995). Genetic variation in the seasonal and habitat associations among Chrysoperla carnea populations has also been studied; the importance of the genetic variation to biological control and IPM is discussed by Tauber and Tauber (1993) and Luck et al. (1995).
An effective method for analyzing the evolution and stability of food associations in insects is to conduct comparative studies among phylogenetically related taxa. Such an approach has revealed both patterns and processes in the evolution of insect herbivore associations with plants. In marked contrast, the link between phylogeny and prey associations remains relatively unexplored primarily because phylogenetic relationships within most groups of predatory insects are very poorly known. Nevertheless, some relatively recent comparative studies with syrphids and chrysopids have used a phylogenetic basis to investigate the evolution and stability of specific predator-prey associations. Such studies have strong basic and applied implications. Within the hoverflies (Diptera: Syrphidae), the subfamily Syrphinae is believed to be a monophyletic group, and all described species are predatory. Using a phylogenetic framework, long-term population census data, and comparative data on morphological characteristics (e.g., body size, egg size and number), Gilbert and co-workers tested several hypotheses concerning specialization of prey range (Gilbert et al. 1994). For aphid predators in the Syrphinae, the data do not support predictions that specialized predators lay few but large eggs, or that adults of specialized predators tend to be large. The results indicate that specialization is a derived trait in the Syrphinae and that long-term stability of populations of specialist Syrphinae is similar to that of the generalists. Phylogenetically based investigations examined the evolution and stability of prey specialization in the Chrysopidae. The family is considered monophyletic, but relationships within the tribe Chrysopini, which contains most species used in biological control, need further study (Tauber and Adams 1990). Comparative experimental data from a pair of sister-species (C. quadripunctata, a generalist, and C. slossonae, a derived specialist) demonstrate that a three-step process underlies the evolution of prey specialization (Tauber et al. 1993). First (step 1), establishment of a predator population on a new food involves genetic variability or phenotypic plasticity in prey usage. Second (step 2), adaptation to the specific prey may involve behavioral, physiological, morphological and/or phenological traits. Unlike in the Syrphinae, the evolution of prey specialization in Chrysopa is associated with increased egg size, reduced fecundity and increased body size; trade-offs among the traits are also important (Albuquerque et al. 1997). Finally (step 3), the maintenance of prey specialization requires the evolution of reproductive isolation between the ancestral and derived populations. In the case of C. quadripunctata and C. slossonae, this step included the evolution of seasonal asynchrony, perhaps specific prey requirements for mating, and a postmating, prezygotic barrier to fertilization (Albuquerque et al. 1996). Given the large differences and the high degree of reproductive isolation between the generalist and specialist sister-species, C. slossonae's prey specialization appears very stable.
It is appropriate to remember Prof. Kenneth S. Hagen, University of California, Berkeley, who contributed greatly to the study of the nutrition and ecology of predacious insects.
Albuquerque, G. S., C. A. Tauber and M. J. Tauber. 1996. Postmating reproductive isolation between Chrysopa quadripunctata and Chrysopa slossonae: mechanisms and geographic variation. Evolution 50: 1598-1606.
Albuquerque, G. S., M. J. Tauber and C. A. Tauber. 1997. Life-history adaptations and reproductive costs associated with specialization in predacious insects. J. Animal Ecology 66: 307-317.
Doutt, R. L. 1964. Biological characteristics of entomophagous adults, pp. 145-167. In P. DeBach [ed.], Biological control of insect pests and weeds. Reinhold Publ. Co., New York.
Eigenbrode, S. D., T. Castagnola, M. B. Roux and L. Steljes. 1996. Mobility of three generalist predators is greater on cabbage with glossy leaf wax than on cabbage with a wax bloom. Entomol. exp. appl. 81: 335-343.
Elliott, N. C., R.W. Kieckhefer and W.C. Kauffman. 1996. Effects of an invading coccinellid on native coccinellids in an agricultural landscape. Oecologia 105: 537-44
Endler, J. A. 1991. Interactions between predators and prey, pp. 169-196. In J. R. Krebs and N. B. Davies [eds], Behavioral ecology, 3rd ed. Blackwell Scientific Pubs, Oxford.
Evans E. W. 1982. Timing of reproduction by predatory stinkbugs (Hemiptera: Pentatomidae): patterns and consequences for a generalist and specialist. Ecology 63: 147-158.
Evans, E.W. and S. England 1996. Indirect interactions in biological control of insects: pests and natural enemies in alfalfa. Ecol. Appl. 6: 920-930.
Gilbert, F., G. Rotheray, P. Emerson and R. Zafar. 1994. The evolution of feeding strategies, pp. 323-343. In P. Eggleton and R. Vane-Wright [eds.], Phylogenetics and Ecology. Linnean Society of London, Symposium Series No. 17.
Gordon, R. D. 1985. The Coccinellidae (Coleoptera) of America north of Mexico. J. New York Entomol. Soc. 93: 1-912
Greenstone, M. H. 1996. Serological analysis of arthropod predation: past, present and future, pp. 267-321. In W. O. C. Symondson and J. E. Liddell [eds], The ecology of agricultural pests. Chapman & Hall, London.
Hagen, K. S., S. Bombosch and J. A. McMurtry. 1976. The biology and impact of predators, pp. 93-142. In C. B. Huffaker and P. S. Messenger [eds.], Theory and practice of biological control. Academic Press, New York.
Hagler, J. R. and S. E. Naranjo. 1996. Using gut content immunoassays to evaluate predaceous biological control agents: a case study, pp. 383-399. In W. O. C. Symondson and J. E. Liddell [eds], The ecology of agricultural pests. Chapman & Hall, London.
Hedrick, A. V. and S. E. Riechert. 1989. Genetically-based variation between two spider populations in foraging behavior. Oecologica 80: 532-539.
Heinz, K. M. and F. G. Zalom. 1996. Performance of the predator Delphastus pusillus on Bemesia resistant and susceptible tomato lines. Entomol. exp. appl. 81: 345-352.
Hodek, I. 1993. Habitat and food specificity in aphidophagous predators. Biocontrol Sci. & Technol. 3: 91-100.
Howarth, F. G. 1991. Environmental impacts of classical biological control. Ann. Rev. Entomol. 36: 485-509.
Huffaker, C. B. and L. E. Caltagirone. 1986. The impact of biological control on the development of the Pacific. Agric. Ecosystems Environ. 15: 95-107.
Luck, R. F., M. J. Tauber and C. A. Tauber. 1995. The contributions of biological control to population and evolutionary ecology, pp. 25-45. In J. R. Nechols [ed.], Bioloigcal control in the western U.S.: accomplishments and benefits of regional research project, W84. ANR Publications, Oakland, CA.
McMurtry, J. A. and J. G. Rodriguez. 1987. Nutritional ecology of phytoseiid mites, pp. 609-644. In F. Slansky Jr., F. and J. G. Rodriquez [eds], Nutritional ecology of insects, mites, spiders, and related invertebrates. John Wiley & Sons, New York
Murdoch, W. W. and J. R. Marks. 1973. Predation by coccinellid beetles: experiments on switching. Ecology 54: 160-167. New, T. R. 1991. Insects as predators. Kensington, Australia: New So. Wales Univ. Press. 178 pp.
Obrycki, J. J. 1986. The influence of foliar pubescence on entomophagous species. Chap. 3 (pp. 61-83), In D. J. Boethel and R. D. Eikenbary [eds.], Interaction of host plant resistance and parasites and predators of insects. Ellis Horwood Publ., West Sussex.
Obrycki J. J. and C. J. Orr. 1990. Suitability of three prey species for Nearctic populations of Coccinella septempunctata, Hippodamia variegata, and Propylea quatuordecimpunctata (Coleoptera: Coccinellidae). J. Econ. Entomol. 83:1292-97
Price, P. W. 1984. Insect ecology, 2nd ed. John Wiley & Sons, New York. Riechert, S. E. and J. M. Harp. 1987. Nutritional ecology of spiders, pp. 645-672. In F. Slansky Jr., F. and J. G. Rodriquez [eds.], Nutritional ecology of insects, mites, spiders, and related invertebrates. John Wiley & Sons, New York Rosenheim, J. A., H. K.
Kaya, L. E. Ehler, J. J. Marois and B. A. Jaffee. 1995. Intraguild predation among biological control agents: theory and evidence. Biol. Control 5: 303-35
Ruberson, J. R., M. J. Tauber and C.A. Tauber. 1986. Plant feeding by Podisus maculiventris (Heteroptera: Pentatomidae): Effect on survival, development, and preoviposition period. Environ. Entomol. 15: 894-897.
Sabelis, M. W. 1992. Predatory arthropods, pp. 225-264. In M. J. Crawley [ed.], Natural enemies. Blackwell Scientific Pubs., Oxford.
Sih, A. 1987a. Predators and prey lifestyles: an evolutionary and ecological overview, pp. 203-224. In W. C. Kerfoot and A. Sih [eds.], Predation. University Press of New England, Hanover.
Sih, A. 1987b. Nutritional ecology of aquatic insect predators, pp. 579-607. In F. Slansky Jr., F. and J. G. Rodriquez [eds.], Nutritional ecology of insects, mites, spiders, and related invertebrates. John Wiley & Sons, New York.
Tauber, C. A. and P. A. Adams. 1990. Systematics of the Neuropteroidea: present status and future needs, pp. 151-164. In M. Koszterab and C. W. Schaefer [eds.], Systematics of the North American insects and arachnids: status and future needs. Virginia Agric. Exp. Sta. Information Series 90-1, Blacksburg, Virginia Polytechnic Institute and State University.
Tauber, C. A., M. J. Tauber and L. R. Milbrath. 1995. Individual repeatability and geographical variation in the larval behaviour of the generalist predator, Chrysopa quadripunctata. Anim. Behav. 50: 1391-1403.
Tauber, M. J. and C. A. Tauber. 1983. Life history traits of Chrysopa carnea and Chrysopa rufilabris (Neuroptera: Chrysopidae): influence of humidity. Ann. Entomol. Ent. Soc. Amer. 76: 282-285.
Tauber, M. J. and C. A. Tauber. 1993. Adaptations to temporal variation in habitats: categorizing, predicting, and influencing their evolution in agroecosystems, pp. 103-127. In K. C. Kim and B. A. McPheron [eds.], Evolution of insect pests. John Wiley & Sons, Inc., New York.
Tauber, M. J., C. A. Tauber and S. Masaki. 1986. Seasonal adaptations of insects. Oxford University Press, New York. Tauber, M. J., C. A. Tauber, J. R.
Ruberson, L. R. Milbrath and G. S. Albuquerque. 1993. Evolution of prey specificity via three steps. Experientia 49: 1113-1117.
Yeargan, K. V. 1988. Ecology of a bolas spider, Mastophora hutchinsoni: phenology, hunting tactics, and evidence for aggressive chemical mimicry. Oecologia 74: 524-530.
Hagen, K. S. 1987. Nutritional ecology of terrestrial insect predators, pp. 533-577. In F. Slansky Jr., F. and J. G. Rodriquez [eds.], Nutritional ecology of insects, mites, spiders, and related invertebrates. John Wiley & Sons, New York.
Begon, M., J. L. Haper and C. R. Townsend. 1996. Ecology, 3rd ed. Blackwell
Science Ltd. Oxford. Bioscience June 1996 Vol. 46 No. 6 pp. 401-453.