The objective of this study was to compare the effect of milking frequency (once vs. twice-daily milking) and breed (Holstein-Friesians vs. Jerseys) on milk and milk solids (MS; milk fat + milk protein), yield per cow, milk composition, somatic cell count and lactation length; cow body weight, body condition score, and reproductive performance over a 4-yr period. Total cow numbers in each herd were 30, 35, 36, and 42 for Holstein-Friesians milked once or twice daily, and Jerseys milked once or twice daily, respectively. Forty hectares of pasture were subdivided into 4 smaller pastures of 10 ha each. Stocking rates for the once-daily herds were 16.7% greater than the twice-daily herd in their respective breed. An increased stocking rate was chosen to achieve equal milk and MS per ha from the 2 milking frequencies. Annual milk, fat, protein, and lactose yields per cow were less for once-daily than for twice-daily milking. Interactions were detected between milking frequency and breed for annual milk, fat, protein, and lactose yields per cow, because Jerseys were relatively less affected by once-daily than by twice-daily milking than Holstein-Friesians. Holstein-Friesian cows milked once daily produced 31.2% less milk and 29.4% less MS per cow than their twice-daily counterparts. In contrast, Jersey cows milked once daily produced 22.1% less milk and 19.9% less MS per cow than their twice-daily counterparts. Milk per ha was 17.7 and 9% less for the once-daily Holstein-Friesians and once-daily Jersey herds, respectively, compared with their twice-daily counterparts, because the greater stocking rate for the once-daily herds did not fully compensate for the milk loss per cow. Milking once daily increased somatic cell count throughout the year in both breeds. Cows milked once daily conceived 3 d earlier, took 5 d less from calving to conception, and needed 11% fewer controlled internal drug release devices than those milked twice daily. Milking once daily is a viable milking option for New Zealand farmers who are prepared to trade-off loss of MS income for increased time to accomplish other non-milking activities.

Once-a-day (OAD) milking offers a major opportunity to improve labour output on New Zealand farms. However few full lactation studies have been undertaken to investigate the effect of this strategy on the productivity of the farm system. A trial was established in 2000 to measure the effect of milking frequency and breed in farm systems where stocking rate was adjusted to equalise per hectare feed demand in different herds. There were four treatments: two herds of 35 Friesian and 42 Jersey cows, at 3.5 and 4.2 cows/hectare respectively, were milked OAD for the entire season. Another two herds of 30 Friesian and 36 Jersey cows, at 3.0 and 3.6 cows/ hectare respectively, were milked twice a day. A higher stocking rate was used for the two OAD herds in an attempt to compensate for the effects of the expected decrease in feed eaten. In the Jersey herd, milking OAD resulted in decreased milksolids (MS) yield per cow and per hectare of 22% and 9% respectively. In the Friesians the corresponding decrease was 31% per cow and 19% per hectare. OAD milking resulted in significantly higher protein and milksolids concentrations, and lower lactose concentration. Mean individual somatic cell count was significantly increased by OAD milking, despite the absence of any differences in levels of bacterial infections between herds. The reduction in MS yield per cow and per hectare was smaller for the Jersey herd than the Friesian herd, indicating that the Jersey cow is perhaps better suited to OAD milking. Keywords: farm system, Friesians, Jerseys, milksolids, milk composition, once-a -day milking, somatic cell count