Barcoding gap, but no support for cryptic speciation in the earthworm Aporrectodea longa (Clitellata: Lumbricidae)

DNA-barcoding, using the mitochondrial marker COI, has been found successful for the identification of specimens in many animal groups, but may not be suited for species discovery and delimitation if used alone. In this study, we investigate whether two observed COI haplogroups in the earthworm Aporrectodea longa correspond to two cryptic species or if the variation is intraspecific. This is done by complementing COI with two nuclear markers, ITS2 and Histone 3. The variation is studied using distance methods, parsimony networks and Bayesian coalescent trees, and the statistical distinctness of the groups is tested on gene trees using the genealogical sorting index, Rosenberg's P_AB and Rodrigo et al.'s P_(RD). We also applied multilocus species delimitation based on the multispecies coalescence model. The two haplogroups were found in COI, and all tests except P_(RD) found them to be significantly distinct. However, in ITS2, the same groups were not recovered in any analyses or tests. H3 was invariable in A. longa, and was, therefore, included only in the multilocus analysis, which preferred a model treating A. longa as one species over a model splitting it into two. We also compared two measurements of size, body length, and no. of segments between the groups. No difference in body length was found, and although a significant difference in no. of segments was noted the haplogroup with the lower mean showed both the highest and the lowest value. When combined, these results led us to the conclusion that there is no support for the separation of A. longa into two cryptic species. This study again highlights the importance of complementing mitochondrial barcodes with more data when establishing species boundaries.