Abstract
A comparative survey was done of the spider assemblages found in grass tussocks of three grass species (Cymbopogon pospischilii, Eragrostis lehmanniana and Themeda triandra) occurring in three different habitats (hillside, open grassland and around a freshwater dam) in a central Free State grassland. We aimed to determine whether spiders displayed preferences for particular grass species and habitats with varying characteristics. Spiders were caught from the base of grass tussocks by hand using vials. A total of 491 spiders representing 24 families and 56 species were sampled from 270 tussocks, with Philodromidae (16.8%), Gnaphosidae (15.0%), Lycosidae (11.3%) and Trachelidae (10.9%) the dominant families. We found that neither grass species nor habitats significantly affected the abundance, species richness or assemblage structure of spiders, but that sampling month did affect abundance and richness, particularly when considering its interaction with habitat and grass species. Two of the four dominant species showed distinct preferences for particular grass species, Tibellus minor for the thicker-stemmed Cymbopogon and Themeda, and Obatala sp. for the more sparse tussocks of Eragrostis. We suggest that tussock-dwelling spiders are, for the most part, habitat generalists, occupying any available grass species. However, further study is necessary to determine the role of grass structure on particular spider species.
ACKNOWLEDGEMENTS
Funding for this study was provided as part of a National Research Foundation of South Africa (NRF) Thuthuka grant to the second author (TTK2008050500003). The South African National Biodiversity Institute and the staff of the Free State National Botanical Gardens, particularly Elliot Lithudza and Tshilidzi Netshisaulu, are thanked for supporting this project and allowing access to the study site. Jason Botham and Sylvia van der Merwe are thanked for assisting with some of the statistical analyses. The anonymous reviewers provided useful suggestions to improve the manuscript.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.