Comparative effectiveness of Longworth and Sherman live traps

Despite the widespread use of Sherman (H. B. Sherman Inc., Tallahassee, Flor.) and Longworth (Penlon Ltd., Oxford, U.K.) live traps in small-mammal-community assessment, few studies have directly compared the effectiveness of these 2 popular models. This study compared the relative efficacy of both trap types in capturing small mammals in southern Wisconsin grasslands. As trap size may cause capture bias, we compared Longworth traps with equal numbers of small and large, folding Sherman traps. We also deployed a small number of pitfalls. We carried out trapping at 12 sites over a 2-year period (1996-1997). We observed a significant year effect, so we analyzed differences in capture success, species diversity indices, and percent community similarity between trap types separately for each year. Two-way contingency table analyses indicated that all 3 trap types exhibited species-specific differences in capture rates. We assessed standardized deviates for each cell within this two-way design, and we considered departures greater than 2 standard deviations (SE+1.96) from the mean to show an either significantly positive (> + 1.96) or significantly negative (<—1.96) association. In the first year, Longworth traps captured greater numbers of long-tailed shrews {Sorex spp.) whereas small Sherman traps captured more western harvest mice (Reithrodontomys megalotis) and white-footed or prairie deer mice (Peromyscus spp.). In the second year, small Sherman captures were greater for long-tailed shrews and western harvest mice while large Sherman traps captured more meadow voles (Microtus pennsylvanicus) and jumping mice (Zapus hudsonius). Although estimates of community diversity were similar between trap types, percent community similarity estimates were lowest for Longworth-Sherman trap comparisons. Mortality rates were highest for Longworth traps and small Sherman traps and lowest for large Sherman traps. Pitfalls caught proportionally more long-tailed shrews than conventional live traps in the first but not the second year of study. In general, body mass of the animal had little effect on trap capture rates. However, in the first year of this study, small Sherman traps caught lighter (P =0.028) long-tailed shrews than the large Sherman traps. Similarly, Longworth traps caught significantly lighter white-footed/prairie deer mice than either small (P=0.022) or large (P= 0.035) Shermans. When used in combination, both Longworth and Sherman traps can diminish overall sampling error and yield less biased estimates of species composition than either trap type alone. The use of new as opposed to used Sherman traps in the second year of this study might account for the greater capture efficacy of these traps and contribute to differences in relative trap type success between years.