Incongruences between morphology and molecular phylogeny provide an insight into the diversification of the Crocidura poensis species complex

Abstract

Untangling the factors of morphological evolution has long held a central role in the study of evolutionary biology. Extant speciose clades that have only recently diverged are ideal study subjects, as they allow the examination of rapid morphological variation in a phylogenetic context, providing insights into a clade’s evolution. Here, we focus on skull morphological variability in a widely distributed shrew species complex, the Crocidura poensis species complex. The relative effects of taxonomy, size, geography, climate and habitat on skull form were tested, as well as the presence of a phylogenetic signal. Taxonomy was the best predictor of skull size and shape, but surprisingly both size and shape exhibited no significant phylogenetic signal. This paper describes one of the few cases within a mammal clade where morphological evolution does not match the phylogeny. The second strongest predictor for shape variation was size, emphasizing that allometry can represent an easily accessed source of morphological variability within complexes of cryptic species. Taking into account species relatedness, habitat preferences, geographical distribution and differences in skull form, our results lean in favor of a parapatric speciation model within this complex of species, where divergence occurred along an ecological gradient, rather than a geographic barrier.

A Cybertaxonomic Revision of the “Crocidura pergrisea” Species Complex with a Special Focus on Endemic Rocky Shrews: Crocidura armenica and Crocidura arispa (Soricidae)

“…The specific features of white-toothed shrews in comparison to those of soricines are as follows: (i) morphological “plasticity”, which is shown by a high homoplasy of qualitative features [ 20 ] and a low level of congruency between morphological and molecular datasets [ 21 ]; (ii) the presence of “historical mitochondrial introgression” in some species complexes [ 14 , 22 ]; (iii) a usually low level of abundance in natural habitats and consequently a rarity of specimens in zoological collections; and (iv) access to their natural habitats is usually difficult. Altogether, these features make a comprehensive or “integrative” analysis difficult; such an analysis is usually needed during a taxonomic revision and is based on implementation of several modern approaches, such as molecular and karyological analyses and morphological description, including multivariate morphometric analysis (e.g., Jenkins et al 2013 [ 8 , 11 , 23 ].…”

“…Free accessibility of the datasets ensures the reliability and repeatability of results. (c) There is a possibility of “analytical” approaches to the assessment of congruence between molecular and morphological data; this can be helpful in the case of ambiguous results when using some analyses, for instance, for a choice between morphological datasets with different levels of a “phylogenetic signal” [ 21 , 34 , 35 ]. In the current study, we implemented two of these: (a) and (b).…”