Introduction and hypothesis: In women, the risk of pelvic floor prolapse is known to be associated with age and parity. Different studies suggested that it is also related to pelvic dimensions, e.g. biomechanical modelling showed that a larger pelvic canal results in higher values of displacement, stress and strain in the pelvic floor muscles, which can increase the risk of pelvic floor disorders. To better understand the multiple factors contributing to pelvic floor disorders, we assessed how age, body weight, body height, parity (in women), pelvic canal size and overall muscle development affected pelvic floor geometry. Methods: A comprehensive geometric morphometric analysis of variation in pelvic floor muscle shape was conducted based on a dense set of 3D landmarks measured on CT scans in a cohort of 49 deceased men and 52 deceased women. The multivariate association between biological variables (parity, dimensions of the true pelvis, age, body weight, height) and pelvic floor muscle morphology was explored by reduced rank regression in both sexes. Results: In women, advanced age, high body weight relative to body height and a large pelvic canal were associated with a deeper pelvic floor. Surprisingly, parity did not have any strong association with overall pelvic floor shape. In men, high body weight was associated with a deep pelvic floor. Age had little effect on male pelvic floor shape, except for the thickness of the ischiocavernosus muscle, which reduced with age. Conclusion: These results suggest that age, relative body weight and the size of the pelvic canal contribute to the risk of female pelvic floor disorders via their effect on pelvic floor shape, independently of birth-related factors such as injury and avulsion of pelvic floor muscles.