We are pleased with Tella et al.'s comment that `commodity selection' offers a promising new approach for understanding some unresolved aspects of coloniality. Our review implies that animals may not be selected to breed colonially, but rather to choose commodities like breeding habitats and mates, which secondarily may have produced breeding aggregations. Tella et al.'s main argument is that commodity selection is a complement to the orthodox economic approach of measuring costs and benefits of high density breeding, rather than a substitute for it.Our general response is to reiterate that the economic approach has failed to provide a general explanation of colony evolution, largely because the wrong questions have been asked. Indeed, when investigating the evolution of coloniality, we do not want to know whether it is better for an individual to breed in a bigger or smaller colony, but we want to understand the processes (rather than the benefits) that lead to breeding aggregations.The estimation of the lifetime reproductive success (LRS) of individuals in a variety of colony sizes may produce information on colonial breeding[1xSee all References[1], but this method cannot reveal the mechanisms of colony formation. If, for example, certain individuals choose small colonies that subsequently expand around them (both in the current and next seasons), then knowledge of their LRS in relation to their `choice' colony size will not elucidate the processes of colony formation. Furthermore, LRS is unlikely to be the right measure of fitness, particularly in this context[2xDanchin, E, Gonzales-Davila, G, and Lebreton, J.D. J. Av. Biol. 1995; 26: 67–75CrossrefSee all References[2].Our main point about reproductive success (RS) is that it is a cue that conspecifics can appraise to aid them in selecting breeding habitats and mates. Their ultimate and proximate processes meet: patterns of spatiotemporal variability of the environment (including the distribution of mates of heterogeneous quality) may have led to the evolution of habitat selection strategies that produce dynamic aggregations of breeding individuals[3xBoulinier, T and Danchin, E. Evol. Ecol. 1997; 11: 505–517CrossrefSee all References[3]. Individuals that prospect for a breeding site for the next season may not need to know the actual causes of local reproductive failure or success (predation, etc.) this season. They only require information on the breeding performance of local breeders in a given year.The discussion on predation underlines a classic problem when studying adaptation: current impact may be misleading, and a comparative study of coloniality showed that coloniality is positively associated with exposure of nests to predation[4xSee all References[4]. But, the high current levels of predation cannot be used to argue that predation played a role in the evolution of coloniality. It may well be that some advantages of coloniality secondarily allowed species to nest in exposed situations[4xSee all References[4]. Similarly, low frequencies of extra-pair paternity in colonial species cannot be used to argue that sexual selection plays a secondary role in the evolution of coloniality. The sexual selection (or `hidden lek') hypothesis[5xSee all References[5]predicts that the frequency of extra-pair copulation is the more direct variable explaining nesting aggregation. In support of this prediction, frequency of extra-pair copulations[6xMoller, A.P and Birkhead, T.R. Am. Nat. 1993; 142: 118–140Crossref | PubMed | Scopus (119)See all References[6]was found to be higher in colonial species.In this context, it is clear that long-term studies are important in allowing to relate the fate of individuals and their breeding distribution to the patterns of spatiotemporal variation of habitat quality, and of factors affecting it (e.g. Ref. [7xSee all References[7]). Experimental approaches are nevertheless necessary to demonstrate causality. The commodity selection framework prescribes the design of much-needed experiments to test predictions of habitat and mate selection in relation to aggregation (e.g. Refs [3xBoulinier, T and Danchin, E. Evol. Ecol. 1997; 11: 505–517CrossrefSee all References, 7xSee all References], Table 1 in Ref. [8xDanchin, E and Wagner, R.H. Trends Ecol. Evol. 1997; 12: 342–347Abstract | Full Text PDF | PubMed | Scopus (241)See all References[8]).
Read full abstract