30 New Zealand Entomologist Vol 33: 30-37 (February 2010) The efficacy of baited and live capture pitfall traps in collecting large-bodied forest carabids. DAVID S. SELDON, JACQUELINE R. BEGGS School of Biological Sciences, University of Auckland, Private Bag 92019 Auckland. E-mail: d.seldon@auckland.ac.nz ABSTRACT Pitfall trapping is commonly used to sample epigaeic invertebrates throughout the world. However, this technique can be ineffective for capturing some species, whilst capturing high numbers of non- target invertebrates. This study aims to improve the efficacy of pitfall trapping for sampling Mecodema spp. (Coleoptera: Carabidae), but also examines the potential of live pitfall traps to reduce the by-catch of other invertebrates. Four types of pitfall traps were used in two separate sampling regimes: baited kill trap versus nonbaited kill trap, and chambered live trap versus non-chambered live trap, in regenerating kauri/ broadleaf mixed forest (Waitakere Ranges, Auckland, New Zealand). Results show all four types of traps sampled different components of the invertebrate community. Baited kill traps caught extremely high numbers of some groups, particularly Diptera and Coleoptera, whereas non-baited kill traps caught far fewer individuals of these groups. Both types of live capture traps caught few individuals. Three genera of large-bodied Carabidae ( Ctenognathus, Holcaspis and Mecodema) were caught in similarly low numbers in both baited and non-baited traps indicating that baiting with squid is ineffective as an attractant to increase catch rates of the genus Mecodema. Although baiting did not increase carabid catches it appears that live traps worked as well as kill traps, but live traps caught fewer invertebrates and therefore significantly reduce the bycatch. A negative aspect of live pitfall trapping is the potential of captured invertebrates to prey upon each other. However, no predation was observed within the traps during this study. KEYWORDS Diptera, Coleoptera, Mecodema, insect sampling, by- catch. INTRODUCTION Pitfall traps are frequently used as an inexpensive and practical method to sample ground invertebrates, particularly beetle (Coleoptera) and ant (Hymenoptera) assemblages (Southwood 1968; Spence & Niemelä 1994; Hansen & New 2005; Woodcock 2005; Neville & Yen 2007). Pitfall traps are especially useful for sampling ground beetles (Carabidae), which are often nocturnal and found in low densities (Thiele 1977; Spence & Niemelä 1994; Lövei & Sunderland 1996; Hutchison 2001; Raworth & Choi 2001). Several studies in the Northern Hemisphere have shown that densities of large-bodied carabids (e.g., Carabus L. 1758) can be as low as one per square metre in cultivated fields (Obrtel 1971), and are often even lower in grassland and forest habitats (Lövei & Sunderland 1996; Eyre et al. 2004, 2005; Koivula et al. 2004). Most large carabids are active polyphagous predators that can cover a relatively large area during their daily activity period (Lövei & Sunderland 1996; Hutchison 2001; Cartelliere 2003; Eyre et al. 2004). Therefore, if large carabids are the focal organism of a study, it would be beneficial to increase the effectiveness of pitfall traps in order to increase the number of specimens caught (Greenslade 1964; Morrill et al. 1990; Spence & Niemelä 1994; Hansen & New 2005). Increasing the number of individuals of the target species in pitfall traps can be achieved in several ways, such as enlarging the catching area (see Digweed et al. 1995; Brennan et al. 2005; Woodcock 2005). Baiting traps can be used to attract beetles (Walker 1957; Southwood 1968; Spence & Niemelä 1994; Coyle & Larsen 1998; Woodcock 2005) and can be very successful for species from the families Scarabaeidae, Staphylinidae, Silphidae, Ptiliidae, Histeridae, Leiodidae and Hydrophilidae (Southwood 1968; Newton & Peck 1975; Archer & Elgar 2003). For Carabidae, carrion has been the most common bait used (Lövei & Sunderland 1996), but several European studies show that carabid numbers do not increase with the use of baits and that passive pitfall traps are equally successful (Luff 1968; Thiele 1977; Spence & Niemelä 1994; Woodcock 2005). In New Zealand there are very few ecological studies on Carabidae, and no published studies on the use of baited pitfall traps to attract carabid beetle species, including species of the endemic genus Mecodema Blanchard, 1843. There is an urgent need for research on this genus and other large-bodied carabids because of concern about their conservation status (McGuiness 2002). Mecodema species are relatively large (10 - 45 mm); flightless, nocturnal predators and many species have proved difficult to collect by methods such as leaf-litter sifting and hand collecting.