Individual body size and composition are important variables for a variety of questions about the behavioral ecology and life histories of non-human primates. Standard methodologies for obtaining body mass involve either capture, which poses risks to the subject, or provisioning, which can disrupt the processes being studied. There are no methods currently available to assess body composition from living animals in the wild. Because of its derivation in muscle, the amount of creatinine that an individual excretes in 24 hours is a reliable and frequently used indicator of relative muscle mass in humans and laboratory animals. Although it is not feasible to collect 24-hour urine samples from wild primates, we apply here a simple method to approximate muscle mass variation from collections of spot urine samples. Specific gravity (SG), an alternative method for assessing urinary water content, is both highly correlated to creatinine and free of mass-dependent effects. Individuals with greater muscle mass should excrete more creatinine for a given SG. We examine this relationship in a dataset of 12,598 urine samples from wild chimpanzees in the Kibale National Park, Uganda. As expected from known differences in body composition, the slope of the relationship between SG and creatinine is significantly greater in adult males than adult females and in adults versus immature individuals. Growth curves generated through this method closely approximate published weight curves for wild chimpanzees. Consistent with the role of testosterone in muscle anabolism, urinary testosterone predicted relative creatinine excretion among adult male chimpanzees. Am J Phys Anthropol 2012. © 2012 Wiley Periodicals, Inc.