According to Gause’s principle of competitive exclusion, species using the same resource cannot coexist. If coexisting species exceed a certain limiting similarity, i.e., if they are too similar, selection will lead to the extinction of all except one species, or to ecological character displacement. Although Gause’s principle is unlikely to be valid in a strictly deterministic world, competitive exclusion is to be expected in our stochastic world. There is evidence for competitive exclusion from some experiments, but many cases of character displacement thought to be due to interspecific competition for limiting resources, are better explained by a Wallace effect i.e., reinforcement of reproductive barriers.
According to the principle of competitive exclusion (in its original and widely used form), species using the same resource cannot coexist (e.g., Gause 1935 ; Levin 1970 ; May 1981 ). In other words, coexisting species must differ in certain aspects, enabling them to exploit different resources. – Their similarity cannot exceed a certain “limiting similarity” (e.g. MacArthur and Levins 1967 ). If species are too similar, selection will lead to character displacement and thereby exploitation of different resources.
The principle predicts that in an absolutely homogeneous world with a single resource, only a single species could be stably maintained. Or, according to Levin (1970 ), “No stable equilibrium can be attained in an ecological community in which some r components are limited by less than r limiting factors. In particular, no stable equilibrium is possible if some r species are limited by less than r factors”. Levin concludes that biodiversity is a consequence of the spatial and/or temporal heterogeneity of the world in which species can exploit heterogeneous resources and patterns (Levin 2000 ).
Objections to the principle
Several authors have shown that Gause’s principle is valid only because of environmental stochasticity. According to May and MacArthur (1972 ) the limiting similarity in a deterministic system is zero, i.e., even very similar competitors will not go extinct. Hubbell and Foster (1986 ) have shown that extinction in identical species may take very long, and in populations of a few thousand, extinction time may be as long as speciation time. Hence, Gause’s principle is wrong. However, this is not the case in stochastic systems: extinction still may take a long time, but one of two competing similar species will finally be pushed over the rim by some environmental fluctuation. – Furthermore, the values of limiting similarity change according to environmental conditions and properties of the niche (Abrams 1983 ). – Modelling of plankton communities and experimental studies have shown that even in homogeneous and constant environments plankton may never reach equilibrium, because multi-species competition may lead to oscillations and chaos, contributing to the maintenance of a great biodiversity; Gause’s principle therefore does not apply (see here).
If resources are indeed limiting, competition for similar resources should lead to some displacement between species (i.e., make species less similar), in order to reduce competition and thereby the chance of competitive exclusion of one or more competitors. But how close can two competing species be? Hutchinson (1959 ) believed that there was a body size difference (in length units) of 1:1.3 in coexisting species pairs, indicative of the difference necessary for species to coexist at the same level of the food web. Others (e.g., Schoener 1986 ) found a similar ratio in the size of feeding organs. However, such a ratio is not always found, and differences in body size and feeding organs may be fortuitous, since they sometimes at least also occur among species using similar food resources that are in unlimitedsupply (e.g., blood on the gills of fish used by Monogenea: Rohde 1991 ).
Evidence from experiments and field studies
Is there empirical evidence for competitive exclusion and limiting similarity? Gause in his study on competition between two ciliate protozoans, i.e. Paramecium caudatumandP. aureliaand predator-prey dynamics of Parameciumand another ciliate, Didinium, has shown that coexistence of both species of Parameciumwas made possible by periodic immigrations from adjacent habitats. In simple laboratory experiments which did not replicate the spatial component of natural habitats, coexistence was not possible. Leslie et al. (1968 ) demonstrated competitive exclusion in two Trilobiumspecies. However, they also noted unexplained coexistence between the species. As an explanation of this observation, Edmunds et al. (2003 ) proposed a model in which, as interspecific competition increases, there is a sequence of bifurcations, that is, a scenario with two stable competitive exclusion equilibria is replaced by a scenario with two competitive exclusion equilibria and a stable coexistence cycle. In other words, two species may well coexist on one limiting resource, even if (or because) there is increased competition. Among parasites, an example given for competitive exclusion is the trematode Gorgodera euzeti and the monogenean Polystoma integerrimum, both infecting the urinary bladder of the frog Rana temporaria in the Pyrenees (Combes 2001 , reference therein). The number of frogs examined was 1,941, the number infected with the first species alone was 576, with the second species 280, with both species 39. If double infections had occurred by chance alone, the number should be at least 576×280/1,941 = 83. This example seems convincing, but can the possibility be entirely excluded that slight differences in habitat preferences of the two species are responsible for the smaller than expected number of double infections?
Reinforcement of reproductive barriers (Wallace effect)
Character displacement is not always due to interspecific competition, in many cases it is better explained by selection which has led to the prevention of hybridization between closely related species. (Interspecific hybridization generally produces no offspring at all or offspring that is less fit). In other words, we can distinguish ecological character displacement (due to interspecific competition) and reinforcement of reproductive barriers. A good example was provided by Kawano (2002 ), who examined male morphology of two closely related rhinoceros beetles, Chalcosoma caucasusand C. atlas in Laos, Thailand, Malaysia, Indonesia and Mindanao from 12 allopatric (where species do not co-occur) and seven sympatric (where species co-occur) locations. The qualitative features and the variation in each species is the same in allopatric and sympatric locations, and there is almost complete overlap in dimorphism, body size, horn size, and size of genitalia between the two species in the allopatric locations. In all sympatric locations, however, that is in those locations where species occur together, differences between species in all characters are highly significant. In particular, differences in the size of genitalia are much greater than expected if due to general body size displacement. Kawano suggested that the differences have evolved to avoid interspecific competition and bring about reproductive isolation. However, evidence for interspecific competition is fairly weak if there is evidence at all, but there can be no doubt (in view of the much greater size differences of genitalia than of body length in sympatric locations) that reinforcement of reproductive barriers is an important, probably the most important and perhaps the only factor involved. The same explanation may apply to the observation that one species occurs at higher and the other at lower altitudes in sympatric than in allopatric locations. In ectoparasitic Monogenea (flukes) infecting the gills of marine and freshwater fish, species with identical (or very similar) copulatory organs are always spatially segregated, i.e., live on different parts of the gills, whereas species with different copulatory organs coexist happily together, convincing evidence that segregation is not due to competition, but to reinforcement of reproductive barriers (reviews in Rohde 1991 , 2005 , pp. 121-127) (Figures 1 and 2).
Figure 1. Copulatory organs of three species of Kuhniaand one species of Grubea(left), and of Pseudokuhnia minor(right). Note that four species have identical copulatory organs, whereas the fifth, Pseudokuhnia, has a markedly different one. Modified from Rohde 2002 .
Figure 2. Distribution of five species of Monogenea on the gills of the mackerelScomber australasicus. X = Pseudokuhnia minor, A-C = Kuhniaspp., D = Grubea.Each symbol stands for several to many individuals of the same species.
Note that the species with a different copulatory organ, i.e. Pseudokuhnia minor (Figure 1 right), overlaps substantially with three other species, whereas the species with identical copulatory organs (Figure 1 left) are strictly segregated. Modified from Rohde 2002 .
Although Gause’s principle is unlikely to be valid in a strictly deterministic world, competitive exclusion is to be expected in our stochastic world. Some experiments have provided evidence for it, but many cases of character displacement, thought to be due to interspecific competition for limiting resources, are better explained by a Wallace effect (reinforcement of reproductive barriers).
Gause, G.F. (1935). The struggle for existence. Williams & Wilkins, Baltimore.Levin, S.A. (1970). Community equilibria and stability, and an extension of the competitive exclusion principle. American Naturalist 104, 413-423.May, R.M. (1981). The role of theory in ecology. American Zoologist 21, 903-910.MacArthur, R.H. and Levins, R. (1967). The limiting similarity, convergence and divergence of coexisting species. American Naturalist 101, 377-385.Levin, S.A. (2000). Multiple scales and the maintenance of biodiversity. Ecosystems 3, 498-506.May, R.M. and MacArthur, R.H. (1972). Niche overlap as a function of environmental variability. Proceedings of the National Academy of Sciences USA 69, 1109-1113.Hubbell, S. P. and Foster, R.B. (1986). Biology, chance, and history and the structure of tropical rain forest tree communities. In: Diamond, J. and Case, T.J. eds. Community ecology. Harper and Row, New York, pp. 314-329.Abrams, P.A. (1983). The theory of limiting similarity. Annual Review of Ecology and Systematics 14, 359-376.Hutchinson, G.E. (1959). Homage to Santa Rosalia, or why are there so many kinds of animals? American Naturalist 93, 145-159.Schoener, T.W. (1986). Resource partitioning. In Kikkawa, J. and Anderson, D.J. eds. Community Ecology: Patterns and Process. Blackwell, Oxford, pp. 91-126.Rohde, K. (1991). Intra- and interspecific interactions in low density populations in resource-rich habitats. Oikos 60, 91-104.Leslie, P.H., Park, T. and Mertz, D.B. (1968). The effect of varying the initial numbers on the outcome of competition between two Trilobium species. Journal of Animal Ecology 37, 9-23.Edmunds, J., Cushing, J.M., Constantino, R.F., Henson, S.M., Dennis, B. and Desharnais, R.A. (2003). Park’s Trilobium competition experiments: a non-equilibrium species coexistence hypothesis. Journal of Animal Ecology 72, 703-712.Combes, C. (2001). Parasitism. The ecology and evolution of intimate interactions. University of Chicago Press, Chicago and London.Kawano, K. (2002). Character displacement in giant rhinoceros beetles. American Naturalist 159, 255-271.Rohde, K. (2005). Nonequilibrium Ecology. Cambridge University Press, Cambridge.Rohde, K. Niche restriction and mate finding in vertebrate hosts. In: The Behavioral Ecology of Parasites, pp. 171-197 (eds. Lewis, E.E., Campbell, J.F. and Sukhdeo, M.V.K.). CAB International, Wallingford, Oxon.
The importance of ecological character displacement and reinforcement of reproductive barriers is also discussed in “Niche restriction and segregation”.