Identification of phenotypic predictors of milk yield losses in response to once-daily milking : Preliminary results

Extended milking intervals can reduce milk yield but this response is highly variable among cows. Identifying cows with the low milk yield loss associated with milking interval is difficult, restricting the development of milking practices such as once-daily milking. A trial was carried out to identify phenotypic predictors linked to a physical regulation of milk yield due to milk accumulation in the udder. Thirty-two primi- and multiparous Holstein cows producing 34.1 ± 5.4 kg/d of milk at 184 ± 29 d in milk were milked twice a day at 0700 and 1600 h during the 7-d control period (C) and then morning milking was omitted one day (MO, milking omission). Indeed milk loss was shown to occur very rapidly, from the 1st day of implementation. Milk yield and flow were recorded at each milking. Milk fat, protein, lactose, and cell contents were analyzed at each milking for 3 consecutive days of the control period, and on the MO day. Lactose was analyzed in tail blood samples collected 1 h before milking 2 d before MO and on MO day; distances between the 4 teats tips were also measured 1h before milking. Data were analyzed to determine Pearson correlations using SAS. Milk yield decreased by 6.8 kg/d (20%) in response to MO (P = 0.0001). The absolute and relative milk losses showed high variability between cows (−0.1 to 13.4 kg/d and −0.5 to 42%, resp.). Milk loss (absolute or relative) was not correlated to C milk yield. Absolute milk loss was negatively but poorly correlated with C milk fat and protein content (r = −0.34 and −0.40; P = 0.05) and positively but poorly also with C milk lactose content (r = 0.33; P = 0.06). Milk loss was not correlated with C plasma lactose levels and the C total distance between tips. Absolute milk loss was positively correlated to C milk average flow (r = 0.59 and 0.46, resp.; P < 0.01) and negatively to C milking time (r = −0.58; P < 0.001). In conclusion, milk flow could be a promising key phenotypic predictor of individual milk yield losses in response to once-daily milking and in relation to physical regulation of milk secretion