Studies on diaspore germination in native species with low economic relevance but great ecological significance have been based on a wide range of sample sizes. However, can the sample size change the physiological inferences made from germination measurements? To answer this question, diaspores of six Cerrado species were evaluated for germinability, germination time (initial, mean, and final), germination velocity (mean germination rate and Maguire’s rate), coefficient of variation of the germination time, and synchronization index of the germination process. Germinability, final time, mean time, and synchronization index were robust with respect to sample size fluctuation. Maguire’s rate, initial time, coefficient of variation of the germination time, and mean germination rate, in contrast, were affected by sample size fluctuation, at least in one of the species tested. The robustness of the time measurements and the synchronization index also demonstrates that the germination process occurs in a cadenced rhythm, much like a biological clock. Among the measurements evaluated, Maguire’s rate is the only one that must be avoided, since it is strongly influenced by sample size and by the balance between germinability and mean germination rate. These results demonstrate that sample size can affect inferences about the germination process and can compromise restoration and (or) conservation efforts.