Skip to main content

Exploring Morphospaces in Adaptive Radiations to unravel Ecological Speciation

Final Report Summary - EMARES (Exploring Morphospaces in Adaptive Radiations to unravel Ecological Speciation)

The EMARES project aimed to extend work on the evolution of adaptive traits using a single model species of African butterfly, Bicyclus anynana, that was carried out at Leiden University (NL) from 1987-2010 to the 300 or so related species in the subtribe, the Mycalesina, that includes the model species. These species of small brown butterflies are distributed throughout the Old World tropics in habitats ranging from primary rain forest to more open savannah grasslands. The adult butterflies feed on rotting fruit and can be readily trapped in the field whilst their larvae feed on grasses. The overall goal of the project was to describe patterns of evolution of species diversity over time and to understand more about the processes underlying the evolutionary diversification of these species including patterns of adaptive radiation in different geographical regions. This would in time provide a new case study in the analysis of ecological speciation, the process by which divergent patterns of natural selection in differing environments can lead to local adaptation to those environments and new species that are reproductively isolated from one another.

Over twenty new species have been described in our work and overall we have a much improved understanding of the species relationships, as well as those of the major clades within the whole subtribe. The phylogenetic reconstruction of the whole group has been much improved by employing sequencing of a battery of mitochondrial and nuclear genes. This is now providing an excellent foundation for all ongoing research in this group. We also have a much improved description of the dynamics over time of the various major radiations with the Mycalsina, as well as more rigorous hypotheses about the biogeography of the expansions of the mycalesines in the Old World tropics since the mid-Miocene. The work has established the presence of parallel radiations involving expansions that were adaptive in nature and which began some 20 MYA. The whole genome sequence for the model species, Bicyclus anynana, that is now fully available on LEPBASE provides an important resource for the community of researchers on the Lepidoptera and more widely.

The reconstruction of the phylogeny suggested that a key ecological driver for each of the adaptive radiations could be the expansion of ancestral species from belts of primary, wet aseasonal rain forest in to progressively more seasonal, wet-dry environments and more open, savannah-grassland habitats. Our work has shown how the evolution of phenotypic plasticity and seasonal forms in these butterflies with alternative forms flying in the generations of the wet and dry seasons has been crucial in the expansions of each of the parallel radiations. These new more seasonal environments were also associated with the expansion of grasses that use the C4 photosynthetic pathway and not the C3 pathway that is typical of ancestral species of grass found in wet and shaded forests; the C4 pathway provides advantages to those species growing in higher levels of solar radiation. We have therefore tested the hypothesis that the expansions in to these new more open and seasonal habitats was not only associated with the elaborations of patterns of phenotypic plasticity and seasonal polyphenism (i.e. alternative wet and dry seasonal forms of generations of the adult butterflies flying in the alternating seasons as we hade studied previously) but also with the evolution of the ability of the larvae to feed and develop on C4 grasses. By collaborating with Dr Colin Osborne in Sheffield and Prof. Howard Griffiths in Cambridge and making rich use of our field and lab samples with carbon and oxygen isotope analyses, we have analysed this hypothesis in detail. Several different types of study including in the field in South Africa with Prof. Brad Ripley and of laboratory experiments in Sheffield have led to convincing support that the evolution of the ability to use C4 grass species as larval host plants has facilitated each of the adaptive radiations, and that C3/C4 host plants has therefore acted as a major ecological driver of diversification.

Work has screened the male sex pheromones, androconia and genitalia of many species. We found a much higher than expected diversity of the pheromones, as well as higher amounts, than in other groups of Lepidoptera. Evidence also shows a high rate of evolutionary diversification with extraordinary diversity among the species of a single genus, Bicyclus, in Africa. We also showed that many of the androconial wing structures linked to pheromone production and dissemination in courtship actually produce no pheromones and so presumably have additional functions. The male genitalia appear in general to evolve much more slowly yielding generally subtle differences among species but we have detected groups with more rapid rates of evolutionary diversification. Further work will study patterns of evolution of these sexual traits over the whole phylogeny.

Analysis of the rich image library of the wing patterns of many species of mycalesine that we have obtained in EMARES is now providing support for how developmental bias in certain wing eyespot traits, as revealed in previous experimental work with our model species, can indeed influence patterns of evolutionary diversification over time. Furthermore, a lineage of mycalesine butterflies on Madagascar (using experiments on one of our new species stocks established in EMARES as well as the image library) shows strong evidence of a key innovation that breaks the developmental bias thus facilitating expansion in to a novel area of 'morphospace' combined with a different ecology.