The social behavior of brown bears (Ursus arctos) was studied during the summers of 1972 and 1973 as bears fished for salmon at McNeil River, Alaska. Study objectives were to determine behavioral characteristics of bears in relation to sex and age, changes in social behavior over a 40-day long fishing season, social and environmental parameters correlated with the occurrence of behavior, and to test the hypothesis that brown bears modify social behavior in a feeding aggregation to exploit a resource limited in time and space. Over one-half of the agonistic interactions consisted of passive deferrals. Encounters that included elements of overt threat were jawing, sparring, charges, and fighting. Jawing was the most prevalent agonistic encounter and generally occurred between individuals of the same sex and age class. Sparring, charges and fights were generally initiated by larger bears against smaller individuals. Females with young were most intolerant. Adult males participated in few encounters that involved overt threat since most bears avoided them. Single adult and adolescent females were neither particularly aggressive nor especially tolerant. Adolescent males adjusted quickly to McNeil Falls and as a group were unaggressive. Subadults were wary and frequently were the objects of aggression of older bears. Social dominance relationships between bears of the same class were often ambiguous, the exception being adult males. Relationships between bears of different classes were mostly stable; adult males were dominant, followed in order by females with young, single adult females, adolescents, and subadults. However, apparent reversals also were common between single adult females and adolescent males. Nonagonistic encounters occurred only when salmon were exceptionally abundant and usually involved adolescent and subadult bears. Behavioral changes over time included a decline in the frequency of running deferrals, a decline in deferrals in total, and a decline in the frequency of charges. The occurrence of fighting and sparring encounters did not change, but the frequency of jawing increased within each fishing season. Various factors determined salmon caught by a bear per hour of fishing effort: salmon abundance, water levels, time of day, and fishing location. The time of day a bear could fish and its choice of location depended on its ability to gain and defend a profitable site. Fishing success was directly correlated with social status, but differences in success are probably unimportant in terms of individual fitness except when salmon are relatively scarce. Changes in encounter intensity over time had no detectable effect on fishing success. Salmon abundance, however, resulted in a further reduction of agonistic encounter intensity and an increase in nonagonistic encounters. Bear social relationships were governed largely by variations in resource abundance. Despite energetic and psychological costs imposed by the bear concentration on individual animals, salmon were evidently sufficiently numerous that these costs were outweighted by returns in protein. Dominance relationships at McNeil Falls did not correspond to predictions of classical dominance theory. This may have been partially attributable to the fact that bears in aggregations derive no benefits from tacit acceptance of subordinate roles; a bear's alternatives were to compete and gain access to food or, if unsuccessful, to try elsewhere. To pose the question if normally solitary bears can adapt behaviorally to efficiently exploit a localized source of food may have been inappropriate. Alternatively, bears can be viewed as occupying and defending areas akin to small territories, with their behavior explicable in terms of energetic costs and benefits based on variations in resource abundance.

The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Title from PDF of title page (University of Missouri--Columbia, viewed on January 26, 2010 Thesis advisor: Dr. W. Thomas McKenney. Includes bibliographical references.