In extensive pastoral dairy farming systems herds graze 12 months of the year with the majority fed a near-100% pasture or conserved pasture diet. The viability of automatic milking in these systems will depend partly upon the amount of supplementary feed necessary to encourage cows to walk from the pasture to the milking unit but also on the efficient use of the automatic milking system (AMS). This paper describes a study to determine the importance of offering concentrate in the milking unit and the effect of minimum milking interval on cow movement and milking performance in a pasture-based AMS. The effects of feeding rate (FR0=0 kg or FR1=1 kg crushed barley/d) and minimum milking interval (MM6=6 h or MM12=12 h) on cow movement and behaviour during milking were studied in a multi-factorial cross-over (feeding level only, 4 weeks per treatment) experiment involving 27 mixed-breed cows milked through a single AMS. Feeding 1 kg barley in the milking unit resulted in a higher visiting frequency to the pre-selection unit (FR0=4.6 visits/d, FR1=5.4 visits/d, sed=0.35, P<0.05) and a higher yield (FR0=22.5 kg/d, FR1=23.6 kg/d, sed=0.385, P<0.01) but had no effect on milking frequency (FR0=1.6 milkings/d, FR1=1.7 milkings/d, sed=0.04, NS). Minimum milking interval was the major factor influencing milking frequency (MM6=1.9, MM12=1.4 milkings/d, sed=0.15, P<0.01). The absence of feeding in the milking unit had no negative effect on behaviour during milking or the number of cows that had to be manually driven from the paddock. The results show that automatic milking can be combined with a near-100% pasture diet and that milking interval is an important determinant for maximizing milk harvested per AMS.

Economic viability of automatic milking systems (AMS) within an Australian pasture-based farming system will be largely determined by the throughput (cows milked/h), which is the result of processes occurring while the cow is in the AMS milking crate. Premilking udder preparation is automated and optional on all AMS. Yet, very few conventional farms in Australia conduct premilking teat preparation regimens, with the majority (78%) strategically washing only visibly dirty teats before milking cup attachment. The objective was to determine the impact of udder preparation in an AMS on the total time spent by cows in the AMS milking unit (crate time). An experiment was conducted with 80 lactating Holstein-Friesian cows in a crossover design over two 5-wk periods to determine the effect of premilking teat preparation (no wash vs. wash) on milk yield, milk harvest rates, and total crate time per milking session in an AMS. Within this study there was no significant effect of treatment on quarter milk conductivity (no wash = 4,858 vs. wash = 4,829 +/- SE = 17 microS/cm), milk blood concentration (no wash = 115.7 vs. wash = 112.3 +/- 7.3 ppm) or test-day somatic cell counts (no wash = 2.044 vs. wash = 2.039 +/- 0.025 log(10) SCC). There was similar total daily milk yield for the 2 treatments (no wash = 20.5 vs. wash = 20.1 +/- 0.2 kg of milk), but a greater mean quarter milk flow rate resulting from the wash treatment (no wash = 0.950 vs. wash = 0.981 +/- 0.013 kg of milk/min). The faster milking was not sufficient to counter the time associated with washing, resulting in longer crate time (no wash = 6.02 vs. wash = 7.12 +/- 0.08 min/milking session) and therefore, lower harvest rate (no wash = 2.08 vs. wash = 1.74 +/- 0.02 kg of milk/min crate time). Not washing teats would allow more efficient AMS utilization by potentially allowing more cows to be milked per machine, which would likely have a positive effect on the economic viability of this technology. The results indicate that a longer term study, investigating the effect of washing teats on udder health and milk quality, is warranted.

Cow throughput in an automatic milking system (AMS) is limited by system parameters such as the time required for pre-milking udder preparation and cup attachment, physiological responses of the cow (such as milk let-down and milking-out rate), milking machine features and cow behaviour. A single-factor cross-over design was used to investigate the effect of pre-milking teat brushing on milk processing time in an AMS operating in an extensive grazing farming system. Teat brushing consisted of two roller brushes tracking up each teat three times (total brushing time of up to 45 s/cow). Cows were allocated to one of two treatment groups with either no brushing (NB) or brushing (B) for a 4-week period before being changed to the other treatment. Teat brushing resulted in shorter average cups-on-time (B = 506.1 s, NB = 541.0 s, P = 0.0001), longer average milk processing time (B = 10.30 min, NB = 9.76 min, P = 0.001) and no difference in daily milk yield (B = 14.67, NB = 14.71 kg/cow, P = 0.826). There was no difference between the two treatments in the success of cup attachment (B = 3.76%, NB = 5.10% unsuccessful milking attempts, P = 0.285). The estimated time cost of pre-milking teat brushing was 53 min for every 100 milkings, equivalent to an additional 5-6 milkings for every 100 milkings by an AMS. The importance of these potential time savings is discussed in relation to automatic milking in farming systems that aim for a lower per cow milking frequency and high ratio of cows to AMS.

Over the past 5 years a herd of up to 180 cows has been milked by two automated milking systems (AMS) at the Greenfield Project research farm near Hamilton. The performance parameters pertaining to pasture management and pasture utilisation during the 2004/05 season are described. The 54 ha farm had an estimated net pasture accumulation of 16.2 t DM/ha/yr. The estimated pasture consumption was determined using back calculations to be 12.3 t DM/ha/yr resulting in an estimated efficiency of pasture utilisation of 76% (based on consumed vs. net accumulation). The data presented suggest that it is possible to achieve satisfactory pasture intakes, pasture utilisation and to be able to maintain pasture quality with automatic milking in a low-input, pasture-based system. Keywords: automatic milking, grazing, pasture management, pasture utilisation