Five studies were performed to determine factors affecting progesterone concentration in skim milk. Results of the first study indicated that progesterone concentration was higher in skim milk of samples kept 16 hours in an ice bath (0 C) than of those left at room temperature (21 C). In the second study, this temperature effect was found to be reversible, with skim milk progesterone concentration increasing when whole milk samples were cooled prior to centrifugation. In the third study, [3H]-labeled progesterone was used to determine the relationship between fat content of foremilk (the first milk obtained from the teats), midmilk (milk obtained through milking), and strippings (milk obtained immediately after milking machines have been removed) samples and temperature (4 C and 21 C) on the percentage of progesterone in the skim milk fraction. The relationship between percentage of butterfat and percentage of progesterone in skim milk was linear when the log of these variables was used for calculations. In the fourth study, assayable progesterone in the skim milk fraction of foremilk, midmilk, and strippings was affected by temperature. In the fifth study, a multiple-regression procedure was used to determine the amount of variation in percentage of radioactive progesterone in the skim milk fraction. Independent variables (whole milk butterfat and temperature of incubation [1, 3, 13, 22, 37, and 50 C]) and the natural log of each variable, were entered into a stepwise multiple-regression analysis. The log of the temperature and percentage of butterfat of whole milk at the time of centrifugation accounted for 89.2% (r2 = 0.892) of the variation in the log of the progesterone concentration in the skim milk fractions. The equation describing this relationship was: log percentage of progesterone in the skim milk fraction = 4.046 - 0.144 X (log of temperature of whole milk sample) - 0.688 X (log percentage of butterfat in whole milk sample). The loss of progesterone from skim milk fractions of warm whole milk samples is possibly a physical phenomenon dependent on the temperature of the sample and its percentage of butterfat. A nomograph was created to allow others to use these variables in making adjustments in progesterone concentrations.