Lygodactylus pakenhami Loveridge 1941

4.1. Lygodactylus pakenhami At present, the taxon ‘ pakenhami’ is considered a subspecies of L. grotei in most publications. In our study, L. grotei and L. g. pakenhami are genetically distinct lineages, confirming previous results (Gippner et al. 2021). Lygodactylus g. pakenhami from Pemba Island differs greatly in 16S and ND2 pairwise distances from mainland individuals considered as L. grotei. While L. grotei and L. g. pakenhami as far as known cannot be distinguished in scalation, they clearly differ in coloration. Furthermore, hatchlings of L. grotei and L. g. pakenhami are easily identifiable and distinguishable by the coloration of their trunks and especially of their tails. The reciprocal monophyly suggested by mitochondrial DNA, conspicuously different coloration—both in adults and hatchlings—and the high genetic distance suggest that these taxa most likely represent two distinct species. This is also in agreement with differences in the RAG1 sequences (although these data are not fully conclusive due to low sample sizes: one sequence available each for grotei and pakenhami). Therefore, we elevate L. g. pakenhami to full species status, as Lygodactylus pakenhami Loveridge, 1941, being aware that this taxonomic hypothesis requires further testing, especially from nuclear-encoded DNA data sets, ideally at the phylogenomic level. Lygodactylus pakenhami is endemic on Pemba Island, the northernmost and second largest island of the Zanzibar Archipelago. Pemba lies off the continental shelf and is surrounded by water 500 to 850 m deep (Moreau & Pakenham 1941). Geological evidence indicates that Pemba Island was separated from the African mainland by faulting that produced the Pemba Channel, possibly during the late Miocene or early Pliocene, 6 million years ago (Stockley 1942; Clarke & Burgess 2000). Corresponding to its long period of isolation, Pemba is characterized by a remarkable number of endemic species, including plants, mammals and reptiles (R̂dder et al. 2010). The simplest explanation for the existence of L. pakenhami on Pemba is vicariance. Its nearest extant relative, L. grotei, is widely distributed in south-eastern Tanzania and northern Mozambique, including the coastal regions. Presumably, a population of an ancestor of L. grotei and L. pakenhami already existed on Pemba Island before its separation from the mainland. Thereafter, there was no further genetic exchange between the continental and the insular population, so that the latter evolved isolated from its continental congener. While a natural dispersal from Tanzania after the separation of Pemba Island cannot be excluded, it is considered as unlikely because the geographic position of Pemba suggests that the East African Coastal Current or a similar paleocurrent conveyed any drifting material into the open ocean very rapidly (Moreau & Pakenham 1940; Hawlitschek et al. 2016a). A recent, human-mediated transportation is unlikely, as after such a short period of isolation we would expect the sharing of mitochondrial haplotypes between mainland and island populations, which we did not detect. Although only two samples yielding DNA sequences of L. grotei have precise georeferenced information, one of these was collected on the mainland directly opposite Pemba Island. However, as is always the case with inferences of allopatric occurrence, we cannot fully exclude that undiscovered populations of L. pakenhami may occur on the African mainland, which would invalidate our biogeographic hypotheses but not our taxonomic conclusions. In contrast to Pemba Island, Zanzibar and Mafia islands lie on the continental shelf and are separated from the mainland only by shallow waters with an average depth of 30 to 35 m, rarely 50 m (Moreau & Pakenham 1941). Geological data of coastal eastern Africa point to a land connection of Zanzibar and Mafia islands with the African mainland up to the end of the Pleistocene, probably only 10,000 –18,000 years ago (Stockley 1942; Clarke & Burgess 2000). Zanzibar and Mafia are inhabited by L. grotei, L. picturatus and L. viscatus, all conspecific with continental populations. Presumably, the distribution of these three species on the islands is probably best explained by recent vicariant events. In contrast to L. pakenhami on Pemba Island, the species on Zanzibar and Mafia have been isolated on the islands for a relatively short time, about 10,000 years. Additionally, both natural dispersal over water after the separation of Zanzibar and Mafia from the mainland and/or a human-mediated transportation due to long-standing trading in the recent past cannot be excluded.Published as part of Röll, Beate, Sanchez, Mickaël, Gippner, Sven, Bauer, Aaron M., Travers, Scott L., Glaw, Frank, Hawlitschek, Oliver & Vences, Miguel, 2023, Phylogeny of dwarf geckos of the genus Lygodactylus (Gekkonidae) in the Western Indian Ocean, pp. 232-250 in Zootaxa 5311 (2) on page 244, DOI: 10.11646/zootaxa.5311.2.4, http://zenodo.org/record/809427