The evolutionary success of beetles and numerous other terrestrial insects is generally attributed to co-radiation with flowering plants but most studies have focused on herbivorous or pollinating insects. Non-herbivores represent a significant proportion of beetle diversity yet potential factors that influence their diversification have been largely unexamined. In the present study, we examine the factors driving diversification within the Scarabaeidae, a speciose beetle family with a range of both herbivorous and non-herbivorous ecologies. In particular, it has been long debated whether the key event in the evolution of dung beetles (Scarabaeidae: Scarabaeinae) was an adaptation to feeding on dinosaur or mammalian dung. Here we present molecular evidence to show that the origin of dung beetles occurred in the middle of the Cretaceous, likely in association with dinosaur dung, but more surprisingly the timing is consistent with the rise of the angiosperms. We hypothesize that the switch in dinosaur diet to incorporate more nutritious and less fibrous angiosperm foliage provided a palatable dung source that ultimately created a new niche for diversification. Given the well-accepted mass extinction of non-avian dinosaurs at the Cretaceous-Paleogene boundary, we examine a potential co-extinction of dung beetles due to the loss of an important evolutionary resource, i.e., dinosaur dung. The biogeography of dung beetles is also examined to explore the previously proposed “out of Africa” hypothesis. Given the inferred age of Scarabaeinae as originating in the Lower Cretaceous, the major radiation of dung feeders prior to the Cenomanian, and the early divergence of both African and Gondwanan lineages, we hypothesise that that faunal exchange between Africa and Gondwanaland occurred during the earliest evolution of the Scarabaeinae. Therefore we propose that both Gondwanan vicariance and dispersal of African lineages is responsible for present day distribution of scarabaeine dung beetles and provide examples.

With over 60 000 described species in approximately 5800 genera, weevils (Curculionoidea) represent one of the most diverse and species‐rich superfamilies of eukaryotes on the planet. Recent attempts to resolve the phylogeny of family‐group taxa in weevils using morphological, molecular or combined data sets have produced vastly different patterns of relationships, particularly within the largest family, Curculionidae. Here we present an estimation of the phylogeny of Australian weevils and of the divergence dates of the major lineages based on a multi‐gene data set (28S, 16S and COI) spanning ∼3.5 kbp of DNA sequence. We assess its topological similarities to, and differences from, other recently published phylogenetic trees, particularly in relation to taxon sampling and relative diversity of the lineages, and we discuss the implications for weevil systematics. Our results, derived from a different set of taxa that has different combination of loci and different fossil calibration points, recover a number of relationships and age estimates that are congruent with those obtained by other recent weevil phylogeny estimates, indicating that we are beginning to recognise the major monophyletic lineages and reconstruct the major diversification events within Curculionoidea. Resolution of natural groups within families (e.g. subfamilies and tribes) however remains poor, even for studies with the largest volume of sequence data (whole mitochondrial genome analyses), evidently largely due to deficient taxon sampling. Although taxon sampling is known to be one of the most critical determinants of accurate phylogenetic reconstruction, it is rarely addressed in phylogenetic assessments. Our case study of weevils highlights not only its importance, but also the fact that it is very difficult to achieve comprehensive representative sampling in hyperdiverse lineages because of their sheer taxic diversity. Doubts concerning the monophyly of many subordinate lineages exacerbate this problem, restricting the use of exemplar approaches to taxon sampling. The value of comparing different data sets and analyses for assessing systematic relationships is often overlooked, even despite the presence of many conflicting results in previous phylogenetic hypotheses. A summary of relationships recovered relative to taxon sampling and methodological differences between different phylogenetic reconstructions provides an important stepping‐stone in understanding the diversification of weevils.