Adult females in a female-bonded, cercopithecine species such as baboons are characterized by hierarchically ranked matrilines, i.e., female offspring assume rankings just beneath those of their mothers. In this system of closely ranked matrilines, a female should engage in significantly more affiliative interactions with those individuals who are closely ranked to herself than with those individuals who are more distantly ranked. We examine the hypothesis that females in this troop of feral yellow baboons (Papio hamadryas cynocephalus) who are closely ranked will also show close social affiliation. We collected focal data on 23 feral, adult female subjects (253 possible dyads) over approximately 1 year at the Tana River National Primate Reserve, Kenya. Following Bramblett's [Behav Brain Sci 4: 435, 1981] method of dominance tabulation and utilizing a modified version of Smuts' [Sex and friendship in baboons, Hawthorne: Aldine Publishing Co., 1985] preferred partner index, we describe and compare the dominance matrix and hierarchy, preferred proximity partner and grooming partner sociograms, and the social networks of these 23 focal females. Over 1,400 interactions were utilized in the dominance tabulations, 41 statistically significant proximity partner preferences were documented, and 100 grooming dyads were recorded. We examine both partners' ranks and the presence of an infant as possible factors influencing proximity and grooming partner preferences. We find that in this population there is no direct correspondence between females' ranks and their affiliation partners. Neither proximity nor grooming preferences are consistently predictable from partners' ranks. While proximity preferences were not significantly influenced by the presence of an infant, grooming partner preferences were. Females with infants had more grooming partners and were more often involved in unidirectional grooming relationships as the recipients than were females without infants. We conclude that females' dominance rankings are not good predictors of either proximity partner or grooming partner preferences and that the presence of an infant does have a significant impact on grooming partner preferences in this population.