Progesterone concentrations in milk were not significantly different when quantitated by different methods (RIA vs. GLC). There was a significant day effect (P less than 0.05) on milk progesterone level due apparently to gradually decreasing values as pregnancy advanced over days 30, 120 and 210. The means for the percent fat content were different (P less than 0.05) for all comparisons among four times of collection (immediately premilking, milking pool, immediately postmilking, and 3 hr postmilking). For progesterone concentration, the main effect of time and the three-way interaction (time times antiserum times purification method) were significant (P less than 0.005); all other main effects and interactions were not significant. Within each of 4 assay groups (2 antisera times 2 purifications), the concentration of progesterone was greater (P less than 0.05) for the samples which were collected immediately postmilking than for any of the other times of collection. The three-way interaction seemed due primarily to difference in progesterone determinations among the four assay groups in the samples which were taken immediately postmilking. Over all milk samples within each assay group, the percent fat content and the concentration of progesterone were positively correlated (r = 0.71, P less than 0.01).