Anthropologists frequently use musculoskeletal stress markers to reconstruct past activity patterns. Yet, researchers have called into question the reliability of muscle marker measurements in part because body size and age affect muscle marker scores. In this study, the author examined an aggregate upper limb muscle marker to determine if after controlling for the effects of body size and age, one could reconstruct activity patterns of a prehistoric Amerind population. Analyses were made of a sample of 102 (43 males, 59 females) prehistoric central California Amerinds. Muscle markers were measured using two-point observer rating scales; body size was measured by humeral articular surfaces; age and sex were determined previously through standard procedures. Using sex separated rankings and partial correlations, disaggregated muscle markers were examined for correlations with age and size to determine if specific muscle markers may be useful in pinpointing to activity patterns. Aggregate upper limb muscle marker correlated with: age, r = 0.44; humeral size, r = 0.44; and sex, r = 0.43; Ps < 0.001. Older individuals had greater muscle markers, as did larger individuals, and males. Rankings seemed to be confounded by the effect size had on the muscle markers. However, based on partial correlations controlling for size and age, the differences that remained between males and females could be used to reconstruct male activities of throwing in hunting and interpersonal aggression acts. Am J Phys Anthropol, 2007. © 2007 Wiley-Liss, Inc.