The margay (Leopardus wiedii) is a small felid currently facing critical population declines in many regions throughout its Central-South American range. This species is thought to be the most arboreal of the New World felids, which has led to concern that it may be especially sensitive to deforestation and habitat destruction. Due to the margay's elusive nature, little is known about its ecology and natural history. In the present study, a camera trap survey was used to investigate whether margay abundance and activity patterns are correlated to vegetational structure in a mid-elevation forest in the eastern Andes of Ecuador. Structural habitat variables were measured at each camera station to assess whether these factors were predictive of margay camera "trap" success. The data show that canopy cover, average distance to nearest trees, and distance to forest edge were significant predictors of margay trap success, although compositional analysis indicated that the highest margay presence was in areas of 51-75% canopy cover rather than the very densest locations. These results highlight the importance of curbing habitat destruction and deforestation as part of efforts to reverse the margay's declining population trend and protect its habitat.

No abstract

In 1966, island biogeographer Sherwin Carlquist published a list of 24 principles governing long-distance dispersal and evolution on islands. The 24 principles describe many aspects of island biology, from long-distance dispersal and establishment to community change and assemblage. Although this was an active period for island biogeography, other models and research garnered much more attention than did Carlquist's. In this review, over 40 years of support for or against Carlquist's principles is presented. Recent work has supported most of the 24 principles, and improved methodologies have generally substantiated his initial claims. However, Carlquist's original work and ideas remain relatively underrepresented in the biogeographic literature. Use of philosophical model domains provides one explanation as to why Carlquist's work has received little attention. Carlquist's principles are largely natural history tests, and don't translate well into the theoretical, design of preserves, or the experimental domains-whereas other competing models do well in such domains.

Ecuador harbours a diverse assemblage of tropical mammals, yet the natural history and local-scale distributions of many species remain poorly understood. We conducted the first systematic camera-trap survey of terrestrial mammalian carnivores at Wildsumaco Wildlife Sanctuary, a mid-elevation (,-, m), montane rainforest site on the slopes of Sumaco Volcano, in the heart of the Tropical Andes biodiversity hotspot. We quantified trap success, latency to detection and temporal activity patterns for each species detected. We recorded nine carnivore species (four felids, two procyonids and three mustelids), including the first verified record of the jaguarundi Puma yagouaroundi in the region. These species comprise one-third of all terrestrial carnivore species known to occur in Ecuador and % of those thought to occur at mid-elevation. All except one of the carnivores we detected have reported elevational ranges # , m; the one exception, the puma Puma concolor, occurs throughout mainland Ecuador at -, m. No cloud forest or highland species (i.e. those with a reported lower elevational limit of $ , m) were detected. Trap success was highest, and latency to detection smallest, for the margay Leopardis wiedii, and temporal activity patterns for all species were consistent with those reported previously in the literature. Our results demonstrate that the midelevation montane rainforests of Sumaco Volcano support an exceptionally high diversity of co-existing mammalian carnivores, many of which appear to be near their upper elevational limits, and emphasize the conservation value of this area.

Some animals have the capacity to produce different alarm calls for terrestrial and aerial predators. However, it is not clear what cognitive processes are involved in generating these calls. One possibility is the position of the predator: Anything on the ground receives a terrestrial predator call, and anything in the air receives an aerial predator call. Another possibility is that animals are able to recognize the physical features of predators and incorporate those into their calls. As a way of elucidating which of these mechanisms plays a primary role in generating the structure of different calls, we performed two field experiments with Gunnison’s prairie dogs. First, we presented the prairie dogs with a circle, a triangle, and a square, each moving across the colony at the same height and speed. Second, we presented the prairie dogs with two squares of differing sizes. DFA statistics showed that 82.6 percent of calls for the circle and 79.2 percent of the calls for the triangle were correctly classified, and 73.3 percent of the calls for the square were classified as either square or circle. Also, 100 percent of the calls for the larger square and 90 percent of the calls for the smaller square were correctly classified. Because both squares and circles are features of terrestrial predators and triangles are features of aerial predators, our results suggest that prairie dogs might have a cognitive mechanism that labels the abstract shape and size of different predators, rather than the position of the predator.