Effects of diet and body condition on fecal progestagen excretion in elk.

Recent research demonstrated the utility of fecal progestagens (P4) for detecting pregnancy in elk (Cervus elaphus) during mid- to late gestation. Several factors, however, may influence fecal P4 excretion and limit its use in free-ranging animals. We investigated the effects of nutrition and body condition (percent ingesta-free body fat) on fecal P4 concentrations and incidence of abortion. During mid-gestation (late December 1997 through early March 1998), 40 gravid cow elk varying in body condition were placed on three diets (high, medium, and low) in which the amount of food offered varied. Feces were collected periodically and analyzed for P4 via radioimmunoassay. We found no significant effect of dietary treatment on P4 concentrations, but as body condition declined, P4 concentrations declined significantly. This decline did not impede the ability to detect pregnancy based on previously reported criteria, even for elk in such poor condition that they aborted. However, fecal P4 concentrations in 10% (4/39) of samples collected from 13 non-pregnant animals maintained on a high plane of nutrition were false-positive for pregnancy. We suggest alternate criteria for determining pregnancy in elk using fecal P4 values: > 1.25 micrograms/g feces as pregnant, < 1.0 microgram/g feces as non-pregnant, and 1.0-1.25 micrograms/g feces as inconclusive. Finally, two cows that aborted did not abort until weeks after being classified as emaciated and near death, suggesting that nutrition-associated abortion in elk may not occur during mid-gestation except under extremely harsh conditions.

Toward an Experimental Basis for Protecting Forest Wildlife.

Social and economic debates over allocation of old-growth forests have spawned conservation strategies that are aimed at protecting sensitive wildlife species while allowing limited timber harvesting. We are interested in improving the scientific underpinnings for such conservation strategies, because doing so might both minimize costs of resource development and provide more reliable protection. Here, we discuss potential consequences from inductive inferencing systems used to develop technical support for protecting wildlife in temperate forests. For examples, we refer to recent conservation strategies for Northern Spotted Owls (Strix occidentalis caurina) and Red-cockaded Woodpeckers (Picoides borealis). Soft inferencing systems could result in conservation strategies that fail to meet intended goals, thereby exacerbating forestry-wildlife debates. Greater emphasis should be placed on hypothetico-deductive inferencing processes that vigorously employ adaptive management principles. Such processes simultaneously test alternative landscape patterns and forestry options as rigorous management experiments, and thus could incrementally predicate forest policy upon an experimental basis.