An evaluation of milk temperature measurement for detecting oestrus in dairy cattle. II. Variations in body and milk temperature associated with oestrus.

Vaginal and milk temperatures were measured at each milking in 15 post-partum Friesian cows showing recurrent oestrous cycles for a total of 33 oestruses. A significant increase (P less than 0.001) in both vaginal and milk temperature of 0.29 +/- 0.05 degrees C was shown between the milking prior to oestrus and the milking when oestrus occurred. Cows were divided into 3 groups: showing either a definite temperature rise (27% of cases), a limited temperature rise (60%) or no temperature change (13%) associated with oestrus. Various methods of analysis were investigated to determine the usefulness of the milk temperature data for oestrus prediction. The optimum compromise of high oestrus detection rate (73.3 +/- 16%) and lowest percentage of false positives (10.8 +/- 3.0%) was based on a temperature rise on the day of oestrus of greater than or equal to 0.2 degrees C over the corresponding mean temperature of the preceding three days. These rates suggest the measurement of milk temperature is less useful for oestrus prediction than some earlier reports have indicated.