The aim of this study was to evaluate the possibility to predict outcomes of artificial insemi- nation (AI) in dairy cows based on in-line milk progesterone (P4) concentration. The research was carried out on the herd of loose housing 245 dairy cows of 2-4 lactations, with average milk yielding 11.000 kg per cow. Milk sampling, measuring, and recording of milk P4 concentration was carried out using the Herd Navigator (HN). The grouping was performed according to the following three indices: the first by reproductive condition – pregnant or not pregnant after AI, the second by P4 concentration from day 20 before AI to day 20 after AI, and the third by P4 concentration at AI time. There was a significant difference in P4 concentration in the group of pregnant cows from day 15 to day 9 before AI, and it was by 18.3% higher com- pared to that in the group of non-pregnant cows in the said period (p<0.01). The milk P4 concen- trations began to differ mostly from day 10 after AI. At that time, the average P4 concentration in the group of pregnant dairy cows was by 36.8% higher compared to that in the group of non-pregnant cows (p<0.01). A statistically significant difference between the ratio of the cows with high, medium, and low P4 concentration on days 20-16 before AI (p<0.01) was determined. The highest number of cows with up to 2-3 ng/ml P4 concentration became pregnant at the AI time.
In-line milk P4 records captured on day 10-15 before AI can be used to predict the proper for reproduction period. By P4 concentrations on day10 after AI, the ratio of pregnant cows in herd can be assessed.