Understanding past as well as present selection pressures that have influenced the origin, evolution and maintenance of existing communication systems presents a considerable challenge to biologists. Among the multitude of factors known to influence signal form, characteristics of the signalling environment play a crucial role. For example, numerous empirical studies have demonstrated a clear effect of signalling habitat on visual signal evolution in fish (Endler, 1991; Endler, 1992; Boughman, 2001; Maan et al., 2006), birds (Marchetti, 1993; Endler and Thery, 1996; Cynx et al., 1998; Lengagne et al., 1999; Lengagne and Slater, 2002; Heindl and Winkler, 2003a; Heindl and Winkler, 2003b;Uy and Endler, 2004), and lizards (Fleishman and Persons, 2001; Macedonia et al., 2003; Peters and Evans, 2003; Leal and Fleishman, 2004). In addition, evidence that air-borne signal evolution is influenced by habitat characteristics is provided by studies involving singing insects (Michelsen and Larsen, 1983; Romer, 1990;VanStaaden and Romer, 1997;Schul and Patterson, 2003), birds (Richards and Wiley, 1980; Ryan and Brenowitz, 1985;Wiley, 1991) and frogs (Ryan et al., 1990; Ryan and Wilczynski, 1991). Fewer studies have assessed habitat-specific effects on the evolution of substrate-borne (seismic) signals (Michelsen et al., 1982; Magal et al., 2000; Cokl and Dober-
AbstractSignals used in communication are often hypothesized to be optimally designed for their signalling environment. Here, we explore the importance of signalling substrate on seismic signal efficacy and reproductive behavior in the wolf spider, Schizocosa retrorsa: a species found on multiple signalling substrates (pine litter and/or red clay or sand). In this multimodal signalling species, simultaneous with conspicuous visual displays, males produce percussive seismic signals via an impulse mechanism which tends to excite a substrate evenly across a wide band of frequencies. We first quantified the transmission characteristics of this broadband percussive signal by playing recorded signals back across three naturally occurring substrates, two of which represent substrates upon which S. retrorsa is commonly found: leaf litter, pine litter and red clay (the latter two exemplify their natural habitat). The substrates varied in their transmission characteristics with respect to both attenuation (higher on red clay) and filtering. Next, we compared copulation success, courtship behavior and microhabitat choice among these same substrates. Copulation frequency was higher on the natural substrates of pine litter and red clay as compared with leaf litter. Males took longer to initiate courtship on leaf litter, but once initiated, courtship behavior did not vary across substrates and we were not able to discern any choice with respect to the first, or the most common, substrate chosen. Our results show that while S. retrorsa's percussive signals may not be matched to the specific properties of any one substrate, copulation success was substrate dependent and we d...