Barrier fences have been installed along roadsides in many parts of the Mojave Desert to protect the threatened Desert Tortoise (Gopherus agassizii) and other wildlife species from being killed by vehicles. Many studies have shown that these fences effectively protect not only the tortoises, but many other small vertebrates, as well. Some species, occasionally including the desert tortoise, manage to get across the barrier fences and, if not killed by collision with a vehicle, remain trapped in the small shoulder area next to the roads. We report several observations made in the Red Cliffs Desert Reserve near St. George, Utah, of reptiles that had been trapped by barrier fences. We make suggestions on how to improve the current design of tortoise barrier fences to avoid accidental entrapment and death.

Phylogenetics has been shown to be a useful forensics tool, and this use can be expanded to the conservation arena. Here we employ phylogenetics to test a cryptozoological specimen for species identification. Over the past decade, one of us (JW) has been collecting and recording stories about unusual animals in the Manitoba wilderness of Canada. Recently, a two meter plus shed was found in the wilderness of Manitoba, Canada. We isolated DNA from the shed and amplified and sequenced a portion of the cyt-b gene. This sequence was compared in a phylogenetic context with cyt-b sequences from a number of snake taxa. Based on these analyses, the shed appears to have come from a Boa constrictor from northeastern South America.

We examined 152 Ameiva plei from four sites on Anguilla and from Scrub Island, a nearby satellite, and 12 A. corax from Little Scrub Island, another Anguillian satellite, generated indices of condition by dividing mass (g) by SVL (mm), and quantifi ed degrees of eutrombiculid chigger mite infections by measuring the total areas (mm2) of each lizard covered by one or more clusters of mites. Prevalence in infected A. plei (N = 77) varied significantly by site, but frequencies of infected males and females within sites did not differ significantly. Indices of condition of infected and mite-free lizards did not differ significantly, nor was area covered by mites significantly correlated with condition, suggesting that mite infections are relatively asymptomatic. All Ameiva corax were infected, and area covered by mites was not signifi cantly correlated with condition. Indices of condition for A. corax were signifi cantly lower than for infected A. plei, probably reflecting the poorer condition of lizards occupying a food-deficient habitat.