This study aimed to analyze the population dynamics of the mite Varroa destructor in honey bee (Apis mellifera) colonies in a temperate semi-arid climate in Mexico. Ten colonies homogeneous in population, food stores, and levels of mite infestation, were used. The mite infestation rate in brood and adult bees, total number of mites, daily mite fall, brood and adult bee population, and food stores, were determined periodically for 10 months. There was a significant effect (P <0.05) of sampling period on the population of V. destructor in adult bees, brood, total mite population, and daily fallen mites. The total mite population increased 26% on average per colony. The increase in brood amount reduced the mite infestation rate in adult bees, and the opposite occurred when the brood decreased. Monitoring V. destructor populations by recording fallen mites is more reliable than determining mite infestation rates in bees, as mite fall has a dynamic pattern similar to that of the total mite population. The best period to apply an acaricide treatment in the region of study is between November and December because most mites were in the phoretic phase, since there was less brood in the colonies compared to other times.