Mar Biol (2011) 158:287–296 DOI 10.1007/s00227-010-1558-6 123 ORIGINAL PAPER EVects of brooding and broadcasting reproductive modes on the population genetic structure of two Antarctic gastropod molluscs J. I. HoVman · A. Clarke · K. Linse · L. S. Peck Received: 12 July 2010 / Accepted: 29 September 2010 / Published online: 19 October 2010  Springer-Verlag 2010 Abstract Life-history characteristics exert a profound inXuence upon the population structure of many marine organisms. However, relatively few genetic studies have compared direct with indirect-developing species in the same ecosystem or geographical region, and none to our knowledge within an Antarctic setting. To address this issue, we have collected novel ampliWed fragment length polymorphism (AFLP) data from the direct-developing top shell Margarella antarctica to form a comparison with pre- viously published data for the broadcast-spawning Antarc- tic limpet Nacella concinna. We scored 270 loci in 240 M. antarctica individuals sampled from Wve populations spanning the full length of the Antarctic Peninsula. Pro- found diVerences were identiWed in the strength and pattern of population structure between the two species, consistent with gene Xow being highly restricted in M. antarctica rela- tive to N. concinna. Introduction It is widely believed that direct developing species exhibit restricted dispersal relative to indirect developers and should therefore tend to show lower levels of population genetic connectivity (e.g. Hellberg et al. 2002; Kinlan and Gaines 2003; Palumbi 2003). However, surprisingly few studies have tested this hypothesis directly by generating comparable empirical datasets, across similar geographi- cal ranges, for species with contrasting life histories that are otherwise ecologically equivalent. Moreover, although some of these studies have found greater levels of population structure in direct developers in support of theoretical expectations (e.g. Nishikawa et al. 2003; Watts and Thorpe 2006; Sherman et al. 2008; Underwood et al. 2009), others have shown that it is not unusual for species with direct development to disperse long distances (e.g. Ayre and Hughes 2000; Richards et al. 2007). To explain this, authors have invoked the passive transport of individuals via fronds of seaweed or fragments of sponge dislodged during violent storms, or upon driftwood (e.g. Thiel and Gutow 2005a, b; Richards et al. 2007; Waters et al. 2008). Another feature of previous comparative studies is that the majority have focussed on tropical species, with the pri- mary emphasis being on taxa inhabiting coral reefs (e.g. Ayre and Hughes 2000; Bay et al. 2006; Nishikawa et al. 2003; Underwood et al. 2009). A few studies have also examined temperate marine species (e.g. Galarza et al. 2009), but to our knowledge none have yet explicitly com- pared the population structure of direct with indirect devel- opers within a polar setting. This is important because, with polar regions currently warming faster than anywhere else on the planet (Meredith and King 2005), there is now a pressing need to evaluate how factors such as life history, geographical isolation and oceanographic barriers to gene Xow interact to shape the population genetic structure of polar marine organisms. The Antarctic marine environment has been relatively isolated since the formation of the Antarctic Circumpolar Communicated by C. Riginos. J. I. HoVman (&) Department of Animal Behaviour, University of Bielefeld, Postfach 100131, 33501 Bielefeld, Germany e-mail: j_i_hoffman@hotmail.com A. Clarke · K. Linse · L. S. Peck British Antarctic Survey, Natural Environment Research Council, High Cross, Madingley Road, Cambridge CB3 0ET, UK