Musculoskeletal stress markers (MSM) at entheses and bone biomechanical properties are used in activity reconstructions. The effect of physical activity on bone biomechanical properties is well established but the relative role of physical activity on MSM is less well known. In this article, it is hypothesized that the same causal mechanisms should affect MSM development as those responsible for bone biomechanical properties. Further, there should be a correlation between MSMs and bone cross-sectional properties as both are considered to reflect physical activity. This was tested using three skeletal samples: early 20th century Finnish (Helsinki) and two medieval English (Blackgate and York) populations. Torsional/average bending rigidity (J) for four cross-sectional locations at 80, 65, 50, and 35% of humeral length from the distal end was calculated and pectoralis major, teres major, and deltoid were scored for MSM. Correlations between MSM and size-standardized J were significant for many comparisons, although they were stronger in males than in females, especially on the right side. In ANOVAs, sample was found to be a significant influence on the right side in both sexes. Using an aggregated MSM score, covariance between J and high MSM scores was again stronger in males. Covariance between J and MSM was found both at cross-sectional locations under muscle insertions and at more distant locations, demonstrating both direct and general effects of muscular loadings applied to diaphyses. Thus, the two types of skeletal markers appear to be related to similar underlying mechanical factors, but effects may also be sex- and sample-specific. Am J Phys Anthropol 2012. © 2012 Wiley Periodicals, Inc.