The single greatest obstacle to the restoration of large, healthy, populations of bighorn sheep (Ovis canadensis) in the western United States is epizootic outbreaks of bronchopneumonia that may kill 20–100% of the animals in populations. Although the species is capable of rapid initial growth rates following restoration into new habitat (λ = 1.23–1.30 have been observed), these rates of increase are typical only a few years following the release of a population, and then most populations either decline to extirpation or remnant status (<30 animals) or remain at <100 individuals. We studied the fecundity and survivorship of three increasing, and three declining and suspected diseased, populations of bighorn sheep (the latter were subjected to outbreaks of bronchopneumonia) located in or near several large national parks in the western United States from 1991 to 1996. Titers verified both population categories were exposed to the bacteria Pasteurella haemolytica serotypes 3; 4; and 3, 4, 10; Moraxella sp., and parainfluenza-3 and bluetongue (BT) viruses. Pregnancy rates of adult ewes were not different in increasing or decreasing populations (pooled rate = 0.93; p = 0.57), but pregnancy rates of yearlings were lower (0.00 for decreasing vs. 0.33 for increasing populations), initial production of lambs and annual recruitment of lambs was lower (0.14, decreasing vs. 0.66, p < 0.05). Adult survival was lower during the first year of an epizootic, 0.62, in one population, but recovered to 0.85 by the second and subsequent years. Survival of adult rams was variable in diseased populations; in two populations rams appeared to be disproportionately impacted, but in a third population rams survived better during the epizootic. In all the increasing park (unhunted) populations, adult ram survival (0.94 ± 0.01) was higher than adult ewe survival (0.89 ± 0.02) (p = 0.10), in contrast to published information from hunted populations where ram survival was lower. Removal of about 20% of one population for restorations severely impacted one declining population. Removals of 12–20% appeared to be excessive and were not readily compensated for in the Canyonlands National Park desert bighorn population. Disease was a significant limiting factor to restoration of bighorn sheep in the study areas; six of 11 total recovering populations we monitored closely were negatively influenced by apparent disease at some time during our observations.