Effect of lactation stage on the odd- and branched-chain milk fatty acids of dairy cattle under grazing and indoor conditions.

The pattern of odd- and branched-chain fatty acids (OBCFA) in milk fat reflects rumen microbial activity and proportions of different rumen microbial groups. Therefore, these milk fatty acids (FA) are used to predict rumen proportions of volatile fatty acids, duodenal flow of microbial protein, and occurrence of rumen acidosis. However, current models do not correct for the potential effects of lactation stage on the level of OBCFA in milk fat. Hence, the objectives of this study were 1) to describe progressive changes related to lactation stage in concentrations of milk FA, with emphasis on the OBCFA, using the incomplete gamma function of Wood, and 2) to analyze whether lactation curves of milk FA on the one hand and milk production or milk fat content on the other hand coincide through evaluation of the correlation between the parameters of the Wood functions fitted to individual animal data. Data were collected from 2 trials in which milk FA during lactation were monitored. The first experiment was a stable trial with 2 groups of 10 cows receiving 2 dietary treatments from wk 1 to 40 of lactation. The second experiment was a grazing trial with 9 cows that were followed during the first 18 wk of lactation. Lactation curves of milk production, milk fat content, and individual milk FA were developed using the incomplete gamma function of Wood for each of the 3 dietary strategies separately. For almost all of the milk FA, lactation curve shapes were similar for all 3 dietary treatments. The OBCFA with chain lengths of 14 and 15 carbon atoms followed the lactation curves of the short- and medium-chain milk FA, which increased in early lactation. The OBCFA with chain length of 17 carbon atoms decreased during the early lactation period, following the pattern of milk long-chain fatty acids. The short- and medium-chain milk FA and OBCFA in the early lactation period seemed to be negatively correlated with the starting milk production and milk fat content, but correlations were modest. Information of milk FA lactation curves should be incorporated in predictive and classification models based on these milk FA, to improve their performance.