Archive for December 11th, 2013


PrintAbstract

Tarantulas in the North American genus Aphonopelma are poorly known due to their challenging patterns of morphological variation and questionable taxonomy; few specimens can be confidently identified using existing keys or comparisons to original descriptions. In an effort to identify new strategies for resolving what has been characterized as a “taxonomic and nomenclatural nightmare”, we employed five different approaches for delimiting species in a group of closely related tarantulas from the Mojave Desert in the southwestern United States. These methods included the application of single techniques (morphology, DNA barcoding, shared genealogical exclusivity among independent loci, and generalized mixed Yule coalescent) and an integrative approach that incorporates genealogical and ecological information. Results demonstrate that the taxonomy of these spiders as presently defined underestimates actual species-level diversity and the group is in need of revision. The number of species delimited by each approach, however, was variable and we argue that it is this discordance that emphasizes the importance of incorporating multiple lines of evidence into an integrative taxonomic framework that can be used for constructing robust taxonomic hypotheses for Aphonopelma species

URL:  http://www.sciencedirect.com/science/article/pii/S1055790312004101

 

1-s2.0-S1055790313004223-fx1DNA

Abstract

The North American tarantula genus Aphonopelma provides one of the greatest challenges to species delimitation and downstream identification in spiders because traditional morphological characters appear ineffective for evaluating limits of intra- and interspecific variation in the group. We evaluated the efficacy of numerous molecular-based approaches to species delimitation within Aphonopelma based upon the most extensive sampling of theraphosids to date, while also investigating the sensitivity of randomized taxon sampling on the reproducibility of species boundaries. Mitochondrial DNA (cytochrome c oxidase subunit I) sequences were sampled from 682 specimens spanning the genetic, taxonomic, and geographic breadth of the genus within the United States. The effects of random taxon sampling compared traditional Neighbor-Joining with three modern quantitative species delimitation approaches (ABGD, P ID(Liberal), and GMYC). Our findings reveal remarkable consistency and congruence across various approaches and sampling regimes, while highlighting highly divergent outcomes in GMYC. Our investigation allowed us to integrate methodologies into an efficient, consistent, and more effective general methodological workflow for estimating species boundaries within the mygalomorph spider genus Aphonopelma. Taken alone, these approaches are not particularly useful – especially in the absence of prior knowledge of the focal taxa. Only through the incorporation of multiple lines of evidence, employed in a hypothesis-testing framework, can the identification and delimitation of confident species boundaries be determined. A key point in studying closely related species, and perhaps one of the most important aspects of DNA barcoding, is to combine a sampling strategy that broadly identifies the extent of genetic diversity across the distributions of the species of interest and incorporates previous knowledge into the “species equation” (morphology, molecules, and natural history).

URL:  http://www.sciencedirect.com/science/article/pii/S1055790313004223