Ecological opportunity – through entry into a new environment, the origin of a key innovation or extinction of antagonists – is widely thought to link ecological population dynamics to evolutionary diversification. The population‐level processes arising from ecological opportunity are well documented under the concept of ecological release. However, there is little consensus as to how these processes promote phenotypic diversification, rapid speciation and adaptive radiation. We propose that ecological opportunity could promote adaptive radiation by generating specific changes to the selective regimes acting on natural populations, both by relaxing effective stabilizing selection and by creating conditions that ultimately generate diversifying selection. We assess theoretical and empirical evidence for these effects of ecological opportunity and review emerging phylogenetic approaches that attempt to detect the signature of ecological opportunity across geological time. Finally, we evaluate the evidence for the evolutionary effects of ecological opportunity in the diversification of Caribbean Anolis lizards. Some of the processes that could link ecological opportunity to adaptive radiation are well documented, but others remain unsupported. We suggest that more study is required to characterize the form of natural selection acting on natural populations and to better describe the relationship between ecological opportunity and speciation rates.

Why some clades are more species-rich than others is a central question in macroevolution. Most hypotheses explaining exceptionally diverse clades involve the emergence of an ecological opportunity caused by a major biogeographic transition or evolution of a key innovation. The radiation of muroid rodents is an ideal model for testing theories of diversification rates in relation to biogeography and ecological opportunity because the group is exceptionally species-rich (comprising nearly one-third of all mammal species), it is ecologically diverse, and it has colonized every major landmass except New Zealand and Antarctica, thus providing multiple replicate radiations. We present an extension of the conventional ecological opportunity model to include a geographic incumbency effect, develop the largest muroid phylogeny to date, and use this phylogeny to test the new model. The nearly 300-species phylogeny based on four nuclear genes is robustly resolved throughout. Consistent with the fossil record, we identified Eurasia as the most likely origin of the group and reconstructed five to seven colonizations of Africa, five of North America, four of Southeast Asia, two of South America, two of Sahul, one of Madagascar, and eight to ten recolonizations of Eurasia. We accounted for incomplete taxon sampling by using multiple statistical methods and identified three corroborated regions of the tree with significant shifts in diversification rates. In several cases, higher rates were associated with the first colonization of a continental area, but most colonizations were not followed by bursts of speciation. We found strong evidence for diversification consistent with the ecological opportunity model (initial burst followed by density-dependent slowdown) in the first colonization of South America and partial support for this model in the first colonization of Sahul. Primary colonizers appear to inhibit the ultimate diversity of secondary colonizers, a pattern of incumbency that is consistent with ecological opportunity, but they did not inhibit initial diversification rates of secondary colonizers. These results indicate that ecological opportunity may be a general but weak process in muroids and one that requires specific circumstances to lead to an adaptive radiation. The total land area, length of time between colonizations, and rank of colonizations did not influence the diversification rates of primary colonizers. Models currently employed to test ecological opportunity do a poor job of explaining muroid diversity. In addition, the various rate-shift metrics identified different clades, suggesting that caution should be used when only one is applied, and we discuss which methods are most appropriate to address different questions of diversification.

We combined new sequence data for more than 300 muroid rodent species with our previously published sequences for up to five nuclear and one mitochondrial genes to generate the most widely and densely sampled hypothesis of evolutionary relationships across Muroidea. An exhaustive screening procedure for publically available sequences was implemented to avoid the propagation of taxonomic errors that are common to supermatrix studies. The combined data set of carefully screened sequences derived from all available sequences on GenBank with our new data resulted in a robust maximum likelihood phylogeny for 900 of the approximately 1,620 muroids. Several regions that were equivocally resolved in previous studies are now more decisively resolved, and we estimated a chronogram using 28 fossil calibrations for the most integrated age and topological estimates to date. The results were used to update muroid classification and highlight questions needing additional data. We also compared the results of multigene supermatrix studies like this one with the principal published supertrees and concluded that the latter are unreliable for any comparative study in muroids. In addition, we explored diversification patterns as an explanation for why muroid rodents represent one of the most species-rich groups of mammals by detecting evidence for increasing net diversification rates through time across the muroid tree. We suggest the observation of increasing rates may be due to a combination of parallel increases in rate across clades and high average extinction rates. Five increased diversification-rate-shifts were inferred, suggesting that multiple, but perhaps not independent, events have led to the remarkable species diversity in the superfamily. Our results provide a phylogenetic framework for comparative studies that is not highly dependent upon the signal from any one gene.