Snakes on an African plain: the radiation of Crotaphopeltis and Philothamnus into open habitat (Serpentes: Colubridae)

Hanlie M Engelbrecht; William R Branch; Krystal A Tolley

doi:10.7717/peerj.11728

Snakes on an African plain: the radiation of Crotaphopeltis and Philothamnus into open habitat (Serpentes: Colubridae)

PeerJ. 2021 Aug 6:9:e11728. doi: 10.7717/peerj.11728. eCollection 2021.

Authors

Hanlie M Engelbrecht^{1

2

3}, William R Branch^{4

5}, Krystal A Tolley^{1

2}

Affiliations

¹ School of Animal, Plant and Environmental Sciences, University of the Witwatersrand, Johannesburg, Gauteng, South Africa.
² Kirstenbosch Research Centre, South African National Biodiversity Institute, Cape Town, Western Cape, South Africa.
³ Department of Botany & Zoology, Stellenbosch University, Cape Town, Western Cape, South Africa.
⁴ Herpetology, Port Elizabeth Museum (Bayworld), Port Elizabeth, Eastern Cape, South Africa.
⁵ Department of Zoology, Nelson Mandela Metropolitan University, Port Elizabeth, Eastern Cape, South Africa.

Abstract

Background: The African continent is comprised of several different biomes, although savanna is the most prevalent. The current heterogeneous landscape was formed through long-term vegetation shifts as a result of the global cooling trend since the Oligocene epoch. The overwhelming trend was a shift from primarily forest, to primarily savanna. As such, faunal groups that emerged during the Paleogene/Neogene period and have species distributed in both forest and savanna habitat should show a genetic signature of the possible evolutionary impact of these biome developments. Crotaphopeltis and Philothamnus (Colubridae) are excellent taxa to investigate the evolutionary impact of these biome developments on widespread African colubrid snakes, and whether timing and patterns of radiation are synchronous with biome reorganisation.

Methods: A phylogenetic framework was used to investigate timing of lineage diversification. Phylogenetic analysis included both genera as well as other Colubridae to construct a temporal framework in order to estimate radiation times for Crotaphopeltis and Philothamnus. Lineage diversification was estimated in Bayesian Evolutionary Analysis Sampling Trees (BEAST), using two mitochondrial markers (cyt-b, ND4), one nuclear marker (c-mos), and incorporating one fossil and two biogeographical calibration points. Vegetation layers were used to classify and confirm species association with broad biome types ('closed' = forest, 'open' = savanna/other), and the ancestral habitat state for each genus was estimated.

Results: Philothamnus showed an ancestral state of closed habitat, but the ancestral habitat type for Crotaphopeltis was equivocal. Both genera showed similar timing of lineage diversification diverging from their sister genera during the Oligocene/Miocene transition (ca. 25 Mya), with subsequent species radiation in the Mid-Miocene. Philothamnus appeared to have undergone allopatric speciation during Mid-Miocene forest fragmentation. Habitat generalist and open habitat specialist species emerged as savanna became more prevalent, while at least two forest associated lineages within Crotaphopeltis moved into Afromontane forest habitat secondarily and independently.

Discussion: With similar diversification times, but contrasting ancestral habitat reconstructions, we show that these genera have responded very differently to the same broad biome shifts. Differences in biogeographical patterns for the two African colubrid genera is likely an effect of distinct life-history traits, such as the arboreous habits of Philothamnus compared to the terrestrial lifestyle of Crotaphopeltis.

Keywords: Ancestral habitat state reconstruction; Forest contraction; Green and Bush Snakes; Herald Snakes; Lineage diversification; Savanna expansion; Sub-Saharan Africa.

Grants and funding

This research was funded by the Andrew W. Mellon Foundation, the South African National Biodiversity Institute, the National Geographic Society Committee for Research and Exploration (Grant 9281-13), and the National Research Foundation of South Africa (Grant 92766). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.