A Model for Estimating Final Clutch Size from Follicle Counts in Plethodontid Salamanders

Abstract. Estimates of clutch sizes are essential for modeling population dynamics, yet for many species of amphibian, clutches can be difficult to observe or methodologically problematic to measure. Clutch sizes for direct-developing Plethodontidae are regularly estimated from counts of ovarian follicles. Because many more follicles begin to develop in an ovary than will ultimately reach full size and be deposited, follicle counts change dramatically over an individual female's follicle development cycle, and a high level of subjectivity is inherent in the process of estimating clutch sizes from follicle counts. Many published studies are not transparent in how they determine clutch sizes from follicle counts. Some investigators address this bias using threshold sizes or other characteristics to separate those follicles that will ultimately mature and be deposited from those that will not, but our experience indicates that such approaches still likely overestimate clutch sizes. To move beyond the subjectivity inherent in estimation of final clutch size from follicle counts, we modeled large Plethodon clutch size as a function of female body size (snout–vent length, SVL) and follicle diameter, then used that model to predict the likely number of mature eggs deposited. We propose that this approach provides reasonable estimates of clutch sizes and variances for use in demographic models.