FORUM is intended for new ideas or new ways of interpreting existing information. It provides a chance for suggesting hypotheses and for challenging current thinking on FORUM ecological issues. A lighter prose, designed to attract readers, will be permitted. Formal research reports, albeit short, will not be accepted, and all contributions should be concise FOR~tUM~l with a relatively short list of references. A summary is not required. T he 'file drawer problem" of non-significant results: does it apply to biological research?

To explore the possibility that facial morphology, echolocation, and foraging behavior are related in some species of bats, intraspecific and interspecific differences in morphology of noseleaves were quantified by univariate and multivariate methods for 248 specimens representing 32 species in three subfamilies of Phyllostomidae (Phyllostominae, Stenodermatinae, and Glossophaginae), The species showed a range of diets from mainly animals (Phyllostorninae) to fruit (Stenodermatinae), and nectar and pollen (Glossophaginae), with the animal-eating species presumed to depend more upon echolocation to detect, locate, and assess prey than frugivorous or nectarivorous species. The canonical-variate analysis revealed significant differences in morphology of noseleaves among j fie three subfamilies, with three features (greatest length of noseleaf, length of spear, and length of horseshoe) showing the best discriminating power. Stenodermatines were chara-.terized by the mosthomogeneous, intraspecific structure of noseleaves glossophagines by the lowest interspecific variability. Phyllostomines showed the highest levels of variance and the most distinctive noseleaves. Euclidean distance values, calculated from measurements of morphology of noseleaf. were similar for most of the 17 species of phyllostomines suggesting that morphology of noseleaf is related to foraging and orientation behavior. The variety of structure of noseleaves in the Phyllostominae coincides with variation in diet, but not with variety in echolocation calls. The connection between structure of noseleaf and echolocation calls remains unclear, like the role of echolocation in the lives of phyllostomine bats.

The purpose of this study was to determine if common poorwills (Phalaenoptilus nuttallii) resist entering torpor during the breeding season. During the summers of 1991 and 1992, we studied poorwills in the Cypress Hills, Saskatchewan, Canada, near the northern limit of their distribution. Since poorwills are monogamous and share incubating and brooding responsibilities, we predicted that the non‐incubating or non‐brooding bird would enter torpor when stressed energetically (e.g. on cold and/or wet nights). Individuals carrying temperature‐sensitive radio transmitters entered torpor significantly less often during the breeding season (two of 195 bird nights) than the non‐breeding season (27 of 44 bird nights). During the breeding season we found no birds involved in an active nesting attempt in torpor. We conclude that reproduction constrains the use of torpor by adult birds, but why non‐incubating and non‐breeding birds did not enter remains unclear.

We determined the diet of common poorwills (Phalaenoptilus nuttallii) by analysing fecal samples collected from roost and nest sites in the Okanagan Valley of British Columbia. The two major insect orders present in the diet in 1989 were Coleoptera (47% by volume) and Lepidoptera (49%), which comprised only 15 and 3%, respectively, of the insects sampled using sticky traps in 1989. Based on the femur length of coleopterans consumed, all coleopterans eaten were longer than 6.7 mm (excluding the head), whereas 76% of the coleopterans caught on sticky traps were less than 6.7 mm long. We found no evidence to support the prediction, based on optimal foraging theory, that poorwills would broaden their diet to include other insect taxa or size classes during 1990, when wet conditions likely depressed prey availability and foraging opportunities. There was no difference in the proportions of various taxa consumed or the size of coleopterans eaten between 1989 and 1990. Acknowledging the bias of our insect sampling regime, we conclude that the apparent prey selection by poorwills results passively from the constraints imposed by detecting prey under conditions of low light.

One of the more frequently used methods of sampling flying insects is the sticky trap. These traps are often used to evaluate the type and size of prey available for aerial insectivores, such as bats and birds (e.g. Barclay 1991; Brigham 1992). Non-attractant sticky traps are favoured because they are considered relatively bias-free, yielding a more representative sample than attractant traps (Service 1976). Sticky traps are versatile in that they are simple to construct and operate, portable, and relatively inexpensive. When using these traps, it is possible to sample insects economically in similar or different habitats and at various heights within habitats (Kunz 1988).