Optimal collection and preservation protocols for fecal DNA genotyping are not firmly established. We evaluated 3 factors that influence microsatellite genotyping success of fecal DNA extracted from coyote (Canis latrans) scats: 1) age of scat, 2) preservative, and 3) diet content. We quantified genotyping success by comparing rates of allelic dropout, false alleles, and failed amplifications among consensus genotypes. We used a panel of 6 microsatellite loci to genotype 20 scat samples, each of which was subjected to 3 age (1 day, 5 days, and 10 days post‐deposition) and 3 preservation (DET buffer, 95% ethanol [EtOH], and lysis buffer) treatments. Both sample age and storage buffer had a significant effect on success and reliability. Ethanol and DET buffer preserved fecal samples with similar efficiency, and both were superior to lysis buffer. Our analysis of DNA degradation rates revealed that samples collected as early as 5 days of age yielded DNA that was highly degraded relative to samples collected on day 1. We tested the influence of dietary remains on microsatellite genotyping by using scat samples consisting predominantly of insect prey (n = 5), mammalian prey (n = 9), or the remains of juniper (Juniperus spp.) berries (n = 6) and compared EtOH and DET buffer preservation efficacy. We observed a significant interaction effect between storage buffer and diet for the probability of a false allele in a polymerase chain reaction (PCR), suggesting that the optimal preservation technique depended on the food remains comprising the scat. Scats comprised of juniper berry remains were more reliably genotyped when preserved in DET than EtOH. Mammalian prey‐based scats were more reliable when stored in EtOH than DET buffer. Insect‐predominant scats were preserved in EtOH and DET buffer with similar efficiency. Although accurate and reliable results can be obtained from scats collected at ≥5 days of age, we suggest sampling design to include collection of scats <5 days of age to minimize field and laboratory expenses. We suggest EtOH preservation for scats of obligate carnivores and of facultative carnivores with a diet consisting primarily of mammals. We suggest DET buffer preservation for animals with a diet consisting of plant‐derived foods. Lysis buffer protocols that we employed should not be used for fecal DNA preservation. © 2011 The Wildlife Society.

Conserving large carnivores throughout the world will often require that they share the landscape with livestock. Minimizing depredations and increasing tolerance by livestock producers will be critical for conservation efforts. To investigate factors influencing calf mortality and producer detection rates (i.e., number of livestock killed by predators, found by producers, and correctly classified as to cause of death), we monitored radio-tagged domestic calves at two sites in the Mexican wolf recovery area (East Eagle [EE] and Adobe Ranch [AR]). Study areas differed in grazing practices, density of predators (mountain lions, black bears, coyotes, and Mexican wolves), and amount of effort spent monitoring cattle. We radiotagged 618 calves over 3.5 years, and 312 calves over 2 years on the EE and AR, respectively. The overall proportion of radioed calves that died was higher on the EE (6.5%) than on the AR (1.9%). Predators (especially mountain lions) accounted for 85% of mortality on the EE and 0% on the AR. Calves selected by predators were on average 25 days younger than the surviving cohort. Our results indicate that year-round calving, especially in areas with high predator densities, are subject to higher losses primarily because calves are exposed to mortality agents for longer periods of time rather than having higher natural rates of mortality. We found a significant difference in producer detection rates between study sites, likely due to differences in the intensity of monitoring cattle between sites. On the EE, the producer detected 77.5% of mortalities and on the AR, the producer detected 33% of mortalities. Our results support changing husbandry practices to limit calving to a seasonal endeavor and that performance payment may be a better compensation strategy than ex post compensation schemes.

In landscape genetics, model selection procedures based on Information Theoretic and Bayesian principles have been used with multiple regression on distance matrices (MRM) to test the relationship between multiple vectors of pairwise genetic, geographic, and environmental distance. Using Monte Carlo simulations, we examined the ability of model selection criteria based on Akaike’s information criterion (AIC), its small-sample correction (AICc), and the Bayesian information criterion (BIC) to reliably rank candidate models when applied with MRM while varying the sample size. The results showed a serious problem: all three criteria exhibit a systematic bias toward selecting unnecessarily complex models containing spurious random variables and erroneously suggest a high level of support for the incorrectly ranked best model. These problems effectively increased with increasing sample size. The failure of AIC, AICc, and BIC was likely driven by the inflated sample size and different sum-of-squares partitioned by MRM, and the resulting effect on delta values. Based on these findings, we strongly discourage the continued application of AIC, AICc, and BIC for model selection with MRM.