Seed germination is a complex biological process that is influenced by vari
ous environmental and genetic factors. The effects of temperature on plant
development are the basis for models used to predict the timing of germinat
ion. Estimation of the cardinal temperatures, including base, optimum, and
maximum, is essential because rate of development increases between base an
d optimum, decreases between optimum and maximum, and ceases above the maxi
mum and below the base temperatures. Nonlinear growth curves can be specifi
ed to model the time course of germination at various temperatures. Quantil
es of such models are regressed on temperature to estimate cardinal quantit
ies, Bootstrap simulation techniques may then be employed to assure the sta
tistical accuracy of these estimates and to provide approximate nonparametr
ic confidence intervals. A statistical approach to modeling germination and
estimation of cardinal temperatures is presented with reference to replica
ted experiments designed to determine the effect of temperature gradient on
germination of three populations of an introduced weed species, common cru
pina (Crupina vulgaris Pers.).