Seasonal excavations of ramet systems in overlapping natural populations of Typha latifolia and T. angustifolia revealed intrapopulation variation in ramet size and reproduction. Reproductive "states" were recognized based on whether or not a ramet had flowered and the number of offspring ramets it possessed. As the growing season progressed, the populations became increasingly heterogeneous in their reproductive-state composition with the largest size classes of ramets containing the greatest number of reproductive states. Therefore, despite a strong correlation between size and reproduction, average ramet size was considered to be an inadequate predictor of population growth in ramet numbers. Results for vegetative reproduction and ramet mortality indicate that the T. latifolia population was growing while the T. angustifolia population was approximately stable. In contrast to most results reported for other species, the frequency distributions of ramet size classes showed T. angustifolia to have a normally distributed population. A slight degree of positive skewing occurred in T. latifolia with the greatest skewing at the end of the growing season. Despite the lack of strong skewing of ramet weights, evidence from other sources have demonstrated that competition was intense and we hypothesize that competition in clonal populations may not not always act to cause positive skew in weight distributions.