The interaction of temperature (10, 14, 25, 31, and 37&C), pH (pH 5, 5.6, 6.5, 7.4, and 8), and NaCl (0, 2, 5, 8, and 10%) in a laboratory medium affects the specific growth rate of Staphylococcus aureus. From growth curves obtained by the turbidimetric technique, a nonlinear model in which the specific growth rate (m) is fitted directly, without data transformation and with the residual error variations taken into account, is proposed. This model correctly fits experimental data and gives more biological information than the quadratic polynomial model. Moreover, the comparison of five strains of S. aureus was performed by a principal-component analysis in which the specific growth rate was the identifying characteristic for S. aureus strains. The results obtained from model coefficient comparison among the five strains and from multivariate data analysis allow the same classification of strains to be performed. Two of them have similar behaviors during food spoilage, two others could be distinguished by their capacity to grow at a low temperature, whereas the last one was markedly different from the others.