Long-term studies of individual animals in nature contribute disproportionately to our understanding of the principles of ecology and evolution. Such field studies can benefit greatly from integrating the methods of molecular genetics with traditional approaches. Even though molecular genetic tools are particularly valuable for species that are difficult to observe directly, they have not been widely adopted. Here, we used molecular genetic techniques in a 10-year radio-telemetric investigation of the western diamond-backed rattlesnake (Crotalus atrox) for an analysis of its mating system and to measure sexual selection. Specifically, we used microsatellite markers to genotype 299 individuals, including neonates from litters of focal females to ascertain parentage using full-pedigree likelihood methods. We detected high levels of multiple paternity within litters, yet found little concordance between paternity and observations of courtship and mating behavior. Larger males did not father significantly more offspring, but we found evidence for size-specific male-mating strategies, with larger males guarding females for longer periods in the mating seasons. Moreover, the spatial proximity of males to mothers was significantly associated with reproductive success. Overall, our field observations alone would have been insufficient to quantitatively measure the mating system of this population of C. atrox, and we thus urge more widespread adoption of molecular tools by field researchers studying the mating systems and sexual selection of snakes and other secretive taxa.

In many species of snakes, particularly viperids from temperate regions, production of offspring by individuals occurs on a less-than-annual schedule. Accordingly, acquiring sufficient energy and nutrient reserves for reproduction in females often requires more than a single active season. This is termed capital mode. Yet, in some instances, annual reproduction occurs under conditions where foraging success is high and environmental factors are compliant. This is termed income mode. Here, we addressed the hypothesis of annual versus less-than-annual reproduction from a long-term radio-telemetric study involving female western diamond-backed rattlesnakes Crotalus atrox from a population of the Sonoran Desert in Arizona. From 2001 to 2008, 16 of 20 radio-telemetered females produced 36 litters, which 32 were informative in addressing the hypothesis of reproductive frequency. In 14 females, litters were produced on a biennial or atleast-biennial (Zbiennial) cycle. However, seven females demonstrated annual reproduction, of which several had previously reproduced on a biennial or greater cycle. Because our study was non-experimental, we were unable to unambiguously identify specific proximate factors that contributed to the shift in annual reproduction. Nonetheless, we established that greater annual rainfall was significantly correlated with shifts to annual reproduction. Based on other studies, we hypothesize that increased rainfall was causally linked with increases in rodent densities and the foraging success of female C. atrox, which in turn is linked to reproduction. We describe, moreover, several characteristics of female C. atrox that appear to facilitate the potential for annual reproduction. In long-lived species, such as C. atrox, our research underscores the necessity to follow individuals for extended periods to gain insights on reproductive cycles not captured by point sampling methods, such as short-term field studies or reliance on museum specimens.

No abstract

For various reasons, reduction or cessation of feeding (anorexia) can occur in either sex during periods of reproduction among vertebrates, from cichlids to elephant seals. Anorexia is commonly associated with gestation in snakes. Using radiotelemetry, we investigated the feeding and spatial ecology of a live‐bearing viperid snake, the western diamond‐backed rattlesnakes (Crotalus atrox). Specifically, from 2001 to 2010, we determined the feeding frequency and home range size of adult females (n = 27) during the active season (March–October) in a population from the Sonoran Desert of Arizona. We addressed a central hypothesis: Do hunting and feeding occur throughout pregnancy? Also, we tested a corollary hypothesis: Does pregnancy influence home range size? We documented hunting and feeding from March to October and during pregnancy (June to mid‐September). Feeding frequency was significantly greater in late pregnancy, a result that is in sharp contrast to most other large‐bodied vipers. Furthermore, home range sizes in gestating subjects did not differ from those in nonreproductive years. Births occurred from mid‐August to mid‐September and mean litter size was 3.4. Frequent feeding in C. atrox during gestation unquestionably provides energy and nutrients to the mother, which is likely important for survival, but such food consumption does not imply that nutrients are used by the fetuses. There is, however, recent evidence in other snakes, including a pitviper, that amino acids are transferred to fetuses. Feeding during pregnancy in C. atrox may be important for both income and capital mode reproduction. Hunting and feeding throughout gestation might be accomplished by having relatively small litters not burdened by a body cavity filled with fetuses. Reduction in litter size may thus be a life‐history (fecundity) trade‐off that permits females to survive and maintain pregnancy in regions where drought and high temperatures are often extreme and chronic.

No abstract