Outline
Graphics
Recently, this phylogenetic perspective has been more often considered in the literature (Brooks and McLennan 1991, 1993, Gorman 1992, Cornell 1993, Losos and Miles 1994). However, few case studies have still been carried out (Losos 1992, 1995, Brooks and McLennan 1993).
I address here two points related to the way some ecological concepts have been under-emphasized in these recent studies. I also assess what bias could be produced by the misleading use of ecological concepts in the framework of phylogenetic analysis.
A community is not necessarily a monophyletic group^
The concept of community has been defined and used by ecologists mostly as "an assemblage of species populations which occur together in space and time" (Begon et al. 1986). This definition does not differ basically between classical textbooks such as Whittaker (1975), Giller (1984) or Ricklefs(1990). It was designed to promote widely the studies of the ecological interactions existing between different species and between the species and their environments:"The principal focus of the community ecologist is the manner in which groupings of species are distributed in nature and the ways in which these groupings can be influenced, or caused, by interactions between species and by the physical forces of their environment" (Begon et al. 1986). As emphasized by Brooks and McLennan (1991), the exact meaning of the community concept has been debated since a long time (e.g. Clements 1905). Discussions mostly deal with the type of ecological relations existing between the species of the community (competition or no competition) and their kind of distribution along the axis of resource use (saturated or unsaturated communities) (Strong et al. 1984).
Owing to these definitions, one should admit that a community is not necessarily a monophyletic group but merely a collection of species occurring together whatever their relationships (Grandcolas 1995). However, using phylogenetic analysis to understand the evolution of community structure makes it necessary to deal with monophyletic groups. This is partly contradictory. Either one should study the phylogeny of all the clades which are pro parte present in a community, or one should select communities which are monophyletic by chance. Brooks and McLennan (1991, 1993) did not especially mention this problem and presented some examples based on the study of monophyletic groups which are called communities. For sake of simplicity, Losos and Miles (1994) recommended favoring the study of selected monophyletic communities(i.e. guilds according to most authors).
The ecological conclusions drawn from the study of some monophyletic communities could be strongly biased. This bias was already denounced by Whittaker (1975) who discarded the definition of communities based on their taxonomic contents. In both cases, i.e. monophyletic communities or taxonomically defined communities, the bias is similar: the ecological interactions between related species will be emphasized and the interactions between unrelated species will not be studied. Such a bias could prevent discovering important relationships existing between unrelated species. For example, the competition between granivorous ants and granivorous rodents in deserts could have been unraveled (Brown and Davidson 1977).
Phylogenetic analysis provides evolutionary inferences only at regional scale^
The seminal paper of Ricklefs(1987) distinguished between local and regional levels in ecological studies: he stressed that community diversity is the outcome of both regional processes such as speciation and local processes such as predation, competition, stochastic variation(which may cause local extinction). This opinion is now widely acknowledged in ecological studies (Barbault et al. 1991, Cornell 1993). Local level can be operationally characterized as a level at which population studies could be carried out. Studies of community structure must clearly be made at this level: population studies allow the determination of abundances, niche widths and niche overlap of the constituent species which are fundamental attributes of community structure. Conversely, the only ecological information pertaining to community at regional level is occurrence or non-occurrence of component species. It is clear that phylogenetic analyses do not match local level but regional level: phylogenetic analyses bear on taxa which are mostly distributed at a level higher than local because there are generally more than one population of a taxon (Grandcolas 1993). The discordance between phylogenetic and ecological analyses was not addressed in previous works: phylogenetic analysis will provide evolutionary inferences at regional level which should be transposed to local level with care (Losos 1992). Different ecological mechanisms and evolutionary processes could account at each level for shaping local community structure.
Possible phylogenetic insights in community structure studies^
The object and the framework of phylogenetic and ecological analyses do not match perfectly (Vrba and Eldredge 1984). Phylogenetic analysis deals with monophyletic groups which are generally distributed at regional level (Cracraft 1985) while the ecological analysis deals with communities (without regard to phylogenetic affinity, i.e. if they are either polyphyletic, paraphyletic or monophyletic entities) and populations which are sampled at local level. These two possible discordances, respectively qualitative and quantitative, could bias or invalidate the results of phylogenetic analyses of communities if they are not taken into account.
Community studies in a phylogenetic perspective should not be limited to monophyletic groups, on pain to minimize or ignore significant ecological processes occurring between unrelated taxa. In this way, these studies would be much more difficult to achieve because many groups which occur in a same community should be phylogenetically analyzed (Losos and Miles 1994).
Also, some procedures must be established which allow the comparison of character changes in the different clades analyzed in a same community. In community studies, most evolutionary assumptions concern several taxa belonging to different clades and their interactions at the population level. Testing these evolutionary assumptions with phylogenetic analyses would thus require to put into relation some character changes which occurred in different clades. These procedures may be unusual in comparative biology because the phylogenetic patterns test for evolutionary processes in a same clade (Coddington 1988, 1990, Carpenter 1989, Grandcolas et al. 1994). Some tests are still possible while some are less powerful, or are prevented.
First example: ecological convergence^
Convergence is an important phenomenon since it provides the opportunity to analyze evolution with respect of the principle of statistical independence (Ridley 1989). The evolutionary relationship between two ecological characteristics observed in several species may be generalized if these species are independent, i.e. unrelated. Convergence may be favored by similar selective pressures onto unrelated organisms. These selective pressures may be exerted through abiotic conditions, thus disconnected from community composition. Or they can be directly related to biotic factors such as other organisms belonging to the same community.
Convergence in a phylogenetic context is a question of polarity and homology. The occurrence of ecological convergence in communities is thus testable using phylogeny. The characteristics under study must be phylogenetically analyzed in each clade constituting the community.
Fig. 1 shows the example of two simple communities comprising some species representative of two clades I and II. The question is: Is the occurrence of convergence related to selective pressures within community? Or, less generally, do species with same diets present convergent life histories, if they belong to the same community?
In the second community, the phylogenetic analysis of two clades does not corroborate the community-based convergence hypothesis. Life history "LH1" occurs in species which do not all belong to the community. Thus, convergence cannot have been favored by biotic interactions within the community.
Second example: competitive displacements^
Comparisons of character changes in two clades along a temporal scale are no more possible, because cladistic analyses have no exact time scales (no assumptions of evolutionary clocks). For example, phylogenetic tests for competitive displacements between two clades present in the same community cannot be very powerful. Phylogenetic analyses could refute a competitive displacement if the polarity of changes is not consistent with the displacement hypothesis. But these analyses could not corroborate strongly the displacement hypothesis even if the polarity of changes is such as expected. The temporal co-occurrence of changes in two clades implied by the hypothesis of competitive displacement cannot be established using cladistics.
Fig. 2 shows communities comprising two clades. Each time, two taxa belonging each to a different clade are believed to have displaced their resource use - presently the prey size - because they experienced symmetrical competition.
In the second case, the displacement is not refuted because the changes in predatory behavior (prey size) occur with the expected polarity. The ancestors of "B" and "T" had more similar prey size. However, this phylogenetic pattern does not corroborate strongly the hypothesis of competitive displacement. As in any other phylogenetic study, one does not study the real function of characters in populations and cannot assess directly the existence of population processes (here, competition). Moreover, the corroboration of the competitive displacement hypothesis would require evidence that changes in resource use have been broadly simultaneous. The cladistic phylogeny cannot provide us with such information because of lack of absolute time scale within cladistics.
According to these examples, phylogeny is good at refutation when used to test evolutionary hypotheses in communities. If the phylogeny does not present the expected pattern, the hypothesis is not plausible. On the contrary, when it presents the expected pattern, it could corroborate very poorly the previous hypothesis because evidence external to the phylogenetic analysis is also needed. It could be, as depicted in the example, that evidence for evolutionary events have occurred at the same time.
In conclusion, because phylogenetic analyses address their results at a regional level, they could not be strictly a test for evolutionary assumptions raised from only one local community study. The ecological interactions observed at a local level between several species are one particular case of the possible interactions in other local communities which also comprise these species. The results of phylogenetic tests achieved at a regional level may be used in two different ways at a local level: either many communities are studied at local level to generalize the results at regional level and to test them in a phylogenetic perspective, or a general model of relationships between local and general levels is devised which allows the application at a local level of the test derived from a regional level. The former possibility requires much more field work and the latter one relies on the validity of the underlying assumptions of the model.
Acknowledgements- I thank P. Deleporte, L. Desutter and J. Losos for their comments.
References^
Barbault, R., Colwell, R. K., Dias, B., Hawksworth, D. L., Huston, M. and Laserre, P. 1991. Conceptual framework and research issues for species diversity at the community level. - In: Solbrig, O. (ed.), From genes to ecosystems: a research agenda for biodiversity. IUBS, Cambridge, MA, pp. 37-71. [Context Link]
Begon, M., Harper, J. L. and Townsend, C. R. 1986. Ecology. Individuals, populations and communities. - Blackwell Scientific, Oxford. [Context Link]
Brooks, D. R. and McLennan, D. A. 1991. Phylogeny, ecology, and behavior. A research program in comparative biology. - Univ. of Chicago Press, Chicago. [Context Link]
Brooks and McLennan, D. A. 1993. Historical ecology: examining phylogenetic components of community evolution. - In: Ricklefs, R. E. and Schluter, D. (eds), Species diversity in ecological communities. Historical and geographical perspectives. Univ. of Chicago Press, Chicago, pp. 267-280. [Context Link]
Brown, J. H. and Davidson, D. W. 1977. Competition between seed-eating rodents and ants in desert ecosystems.- Science 196: 880-882. [Context Link]
Carpenter, J. M. 1989. Testing scenarios: wasp social behavior. - Cladistics 5: 131-144. [BIOSIS Previews Link] [Context Link]
Clements, F. E. 1905. Research methods in ecology. - University Publishing Company, Lincoln, NE. [Context Link]
Coddington, J. A. 1988. Cladistic tests of adaptational hypotheses. - Cladistics 4: 3-22. [BIOSIS Previews Link] [Context Link]
Coddington 1990. Bridges between evolutionary pattern and process. - Cladistics 6: 379-386. [BIOSIS Previews Link] [Context Link]
Cornell, H. V. 1993. Unsaturated patterns in species assemblages: the role of regional processes in setting local species richness. - In: Ricklefs, R. E. and Schluter, D. (eds), Species diversity in ecological communities. Historical and geographical perspectives. Univ. of Chicago Press, Chicago, pp. 243-252. [Context Link]
Cracraft, J. 1985. Biological diversification and its causes. - Ann. Mo. Bot. Gard. 72: 794-822. [Context Link]
Giller, P. S. 1984. Community structure and the niche. - Chapman and Hall, London. [Context Link]
Gorman, O. T. 1992. Evolutionary ecology and historical ecology: assembly, structure, and organization of stream fish communities. - In: Mayden, R. L. (ed.), Systematics, historical ecology, and North American freshwater fishes. Stanford Univ. Press, Stanford, CA, pp. 659-688. [Context Link]
Grandcolas, P. 1993. The origin of biological diversity in a tropical cockroach lineage: a phylogenetic analysis of habitat choice and biome occupancy.- Acta Oecol. 14: 259-270. [BIOSIS Previews Link] [Context Link]
Grandcolas 1995. Phylogénie et origine de la Biodiversité. - In: Bourgoin, T. (ed.), Biosystema, 13: Systématique et Biodiversité. SFS, Paris, pp. 51-58. [Context Link]
Grandcolas(ed.) 1997. The origin of biodiversity in insects: phylogenetic tests of evolutionary scenarios. - Mém. Mus. Nat. Hist. Nat. 173: 1-354. [Context Link]
Grandcolas, Deleporte, P. and Desutter-Grandcolas, L. 1994. Why to use phylogeny in evolutionary ecology? - Acta Oecol. 15: 661-673. [BIOSIS Previews Link] [Context Link]
Losos, J. B. 1992. The evolution of convergent structure in Carribean Anolis communities.- Syst. Biol. 41: 403-420. [Context Link]
Losos 1995. Community evolution in greater antillean Anolis lizards: phylogenetic patterns and experimental tests. - Philos. Trans. R. Soc. Lond. B 349: 69-75. [Context Link]
Losos and Miles, D. B. 1994. Adaptation, constraint, and the comparative method - phylogenetic issues and methods. - In: Wainwright, P. C. and Reilly, S. M. (eds), Ecological morphology: integrative organismal biology. Chicago Univ. Press, Chicago, pp. 60-98. [Context Link]
Ricklefs, R. E. 1987. Community diversity: relative roles of local and regional processes. - Science 235: 167-171. [Context Link]
Ricklefs 1990. Ecology. - Freeman, New York. [Context Link]
Ridley, M. 1989. Why not to use species in comparative tests. - J. Theor. Biol. 136: 361-364. [Context Link]
Ross, H. H. 1972. The origin of species diversity in ecological communities. - Taxon 21: 253-259. [BIOSIS Previews Link] [Context Link]
Strong, D. R., Simberloff, D., Abele, L. G. and Thistle, A. B. (eds) 1984. Ecological communities. Conceptual issues and the evidence. - Princeton Univ. Press, Princeton, NJ. [Context Link]
Vrba, E. S. and Eldredge, N. 1984. Individuals, hierarchies and processes: towards a more complete evolutionary theory. - Paleobiology 10: 146-171. [BIOSIS Previews Link] [Context Link]
Wanntorp, H. E., Brooks, D. R., Nilsson, T. and Nylin, S. 1990. Phylogenetic approches in ecology.- Oikos 57: 119-132. [BIOSIS Previews Link] [Context Link]
Whittaker, R. H. 1975. Communities and ecosystems. - Macmillan, New York. [Context Link]
Section Description^
FORUM is intended for new ideas or new ways of interpreting existing information. It provides a chance for suggesting hypotheses and for challenging current thinking on ecological issues. A lighter prose, designed to attract readers, will be permitted. Formal research reports, albeit short, will not be accepted, and all contributions should be concise with a relatively short list of references. A summary is not required.