Ancient oceanic archipelagos of similar geological age are expected to accrue comparable numbers of endemic lineages with identical life history strategies, especially if the islands exhibit analogous habitats. We tested this hypothesis using marine snails of the genus Conus from the Atlantic archipelagos of Cape Verde and Canary Islands. Together with Azores and Madeira, these archipelagos comprise the Macaronesia biogeographic region and differ remarkably in the diversity of this group. More than 50 endemic Conus species have been described from Cape Verde, whereas prior to this study, only two nonendemic species, including a putative species complex, were thought to occur in the Canary Islands. We combined molecular phylogenetic data and geometric morphometrics with bathymetric and paleoclimatic reconstructions to understand the contrasting diversification patterns found in these regions. Our results suggest that species diversity is even lower than previously thought in the Canary Islands, with the putative species complex corresponding to a single species, Conus guanche. One explanation for the enormous disparity in Conus diversity is that the amount of available habitat may differ, or may have differed in the past due to eustatic (global) sea level changes. Historical bathymetric data, however, indicated that sea level fluctuations since the Miocene have had a similar impact on the available habitat area in both Cape Verde and Canary archipelagos and therefore do not explain this disparity. We suggest that recurrent gene flow between the Canary Islands and West Africa, habitat losses due to intense volcanic activity in combination with unsuccessful colonization of new Conus species from more diverse regions, were all determinant in shaping diversity patterns within the Canarian archipelago. Worldwide Conus species diversity follows the well-established pattern of latitudinal increase of species richness from the poles towards the tropics. However, the eastern Atlantic revealed a striking pattern with two main peaks of Conus species richness in the subtropical area and decreasing diversities toward the tropical western African coast. A Random Forests model using 12 oceanographic variables suggested that sea surface temperature is the main determinant of Conus diversity either at continental scales (eastern Atlantic coast) or in a broader context (worldwide). Other factors such as availability of suitable habitat and reduced salinity due to the influx of large rivers in the tropical area also play an important role in shaping Conus diversity patterns in the western coast of Africa.

http://www.ncbi.nlm.nih.gov/pubmed/25121824?dopt=Abstract