Field evidence from Namibia does not support the designation of Angolan and Namibian subspecies of Welwitschia mirabilis Hook

Abstract

Welwitschia mirabilis is an ancient, endemic gymnosperm found in numerous disjunct populations in western Namibia north of the Namib Sand Sea to the Bentiaba River in southwestern Angola. Based on six plants grown in the Berlin-Dahlem Botanical Garden, Leuenberger [2001, Welwitschia mirabilis (Welwitschiaceae), male cone characters and a new subspecies. Willdenowia 31: 357–381] differentiated two groups of plants using eight male inflorescence characters. He concluded that the two groups represented geographical subspecies: W. mirabilis subsp. mirabilis in Angola and W. mirabilis subsp. namibiana in Namibia, based on putative location data of the seed sources for the plants. Leuenberger called for field data collection, especially in northern Namibia, to verify his results. However despite this lack of data, numerous authors have cited his work as definitive in scientific papers and online plant descriptions and keys. We analysed five distinguishing male cone characters from 13 Namibian plant populations and found significant variation that encompasses both descriptions of the Namibian and Angolan subspecies. Data collected from plants grown from Angolan and Namibian seed in the Kirstenbosch Botanical Garden also fell within the range of the Namibian field populations. We found significant evidence of latitudinal gradients in cone and peduncle length, but cone colour, secondary branch length and bract shape were highly variable within and between the Namibian populations sampled. Our data do not support the designation of an Angolan and Namibian subspecies. W. mirabilis has probably been evolving in this landscape for more than 100 million years with unknown metapopulation dynamics predetermining current geographical populations. Our study calls for molecular characterisation in conjunction with a thorough investigation of both male and female plant characteristics throughout the current range of the plant to further clarify the phylogenetic relevance of the variation observed in male cone characters.

ACKNOWLEDGEMENTS

Thanks to Mary Jacobson, Mary Abrams and Mary Seely for their help with data collection; Grinnell College for funding provided to KJ and PJ; to MET and NBRI for research permits to work on this protected plant in the parks, conservancies and concessions; and to staff at Gobabeb, Save the Rhino Trust, Wilderness Safaris and Palmwag Lodge for facilitating our visits and work, especially Kapoi Kassona and Jeff Muntifering. Finally, we express our immense gratitude to Mary Seely for so generously sharing her passion for Namibia with everyone over the years. We are all deeply enriched by our experiences in the Namib Desert because of her.