To understand the role of strain variability, population dynamics of 2 strains of Bacillus licheniformis, ATCC 6634 and ATCC 14580, were modeled as a function of temperature (4.0–12.0°C) and duration (0–72 h) using regression analysis. Based on the initial spiking of vegetative cells (approximately 4.0 log cfu/mL) and spores (approximately 2.0 log cfu/mL), regression equations, elucidating B. licheniformis growth behavior during raw milk holding at low temperature, were obtained. Contour plots were developed to determine the time-temperature combinations, keeping the population changes to less than 1.0 log. In vegetative cell spiking study of B. licheniformis ATCC 6634 (S1), cell population changes remained below 1.0 log up to 72 h at 8°C. For B. licheniformis ATCC 14580 (S2), 1.0 log shift was not observed only after 80 h at 8°C, indicating higher multiplication potential of S1 as compared with S2. As S2 was a readily sporulating strain, the vegetative spiking study showed spore formation at different storage temperatures. Evidence of some parallel germination was observed for this strain at 8°C or higher, when raw milk samples were spiked with spores. The experimental values obtained for sporeformers and spore counts were validated with contour plot-generated values. Overall, for raw milk samples predominated by the low sporulating strain, the contour plots suggested holding at 8°C or below for up to 72 h. In the case of the readily sporulating strain (S2), raw milk could be held at 8°C for 80 h, where little or no sporulation is observed. Sporulation behavior, germination and multiplication ability, strain variability, and temperature and duration of holding raw milk influenced the population dynamics of Bacillus species. However, in the presence of equivalent numbers of both types of sporulating strains in raw milk, despite strain variability, holding milk at 8°C for not more than 72 h would keep any cell population changes below 1.0 log. In addition, under these storage conditions, the population would remain as vegetative cells that are likely to be inactivated by pasteurization. The contour plots, so generated, would help predict the population shifts and define optimum holding conditions for raw milk before further processing.
