We provide evidence for previously undetected population structure in a wild run of summer steelhead Oncorhynchus mykiss within a river that has considerable recreational importance (Dean River, British Columbia). Data were gathered from an existing catch‐and‐release fishery and examined for phenotypic and genetic variation through the migratory season. Specifically, we compared fish captured in different periods during the migration: early (July 2‐30), middle (July 31‐September 5), and late (September 6‐30). Age (freshwater and saltwater), sex ratio, and body girth did not differ significantly among these groups for females or males. Body length increased through the migratory season for both sexes, perhaps because late‐migrating fish had more time to feed in the ocean. Based on genotypes at 10 microsatellite loci, early and late groups showed highly significant genetic differences (P < 0.001). Assignment tests were able to classify individuals back to early or late groups with 84% accuracy (122 of 145 tests). These results suggest the presence of at least two populations that migrate at different times in the Dean River system. The magnitude of the genetic difference was small (FST = 0.007; Nei's unbiased D = 0.0149, Reynolds coancestry coefficient = 0.007) but comparable to values for other anadromous Pacific salmon species over similar spatial scales. Moreover, the coarse level of our sampling, and possible overlap in migratory timing among populations, suggests that the observed differentiation underestimates the true differentiation. A deficit of heterozygotes in the late group suggests further population substructure within late‐migrating groups of fish. Examining temporal variation through a migratory season proved a useful approach for obtaining preliminary evidence of population structure in migrating salmonids within a small river system.

Organismal traits are presumed to be well suited for performance in the tasks required for survival, growth, and reproduction. Major injuries to such traits should therefore compromise performance and prevent success in the natural world; yet some injured animals can survive for long periods of time and contribute to future generations. We here examine 3 years of camera trap observations along a remote trail through old‐growth forest in northern British Columbia, Canada. The most common observations were of moose (2966), wolves (476), and brown bears (224). The moose overwhelmingly moved in one direction along the trail in the late fall and early winter and in the other direction in the spring. This movement was clustered/contagious, with days on which many moose traveled often being interspersed with days on which few moose traveled. On the video recordings, we identified 12 injured moose, representing 1.4% of all moose observations. Seven injuries were to the carpus, three were to the antebrachium, and two were to the tarsus—and they are hypothesized to reflect damage to ligaments, tendons, and perhaps bones. The injured moose were limping in all cases, sometimes severely; and yet they did not differ noticeably from uninjured moose in the direction, date, contagiousness, or speed of movement along the trail. We discuss the potential relevance of these findings for the action of natural selection in the evolution of organismal traits important for performance.