SummaryThe milk-production potential of different sward types was measured in each of the four seasons of the year in a replicated experiment in south-west Victoria, Australia. Dairy cows were offered ad libitum allowances of a 'short-term winter' sward, based on Italian ryegrass (treatment STW), a 'long-term winter' sward, based on a winter-active tall fescue (treatment LTW), a 'longterm summer' sward, based on a summer-active tall fescue (treatment LTS) and a Control sward (perennial ryegrass) in four seasons (days in milk in parentheses): spring (November dairy cows, 124), summer (February, 227), autumn (May, 234) and winter (August, sixtyfour). A 'short-term summer' sward, based on chicory and white clover (treatment STS), was also included in summer. There was a significant season · treatment interaction caused by a more gradual decline in milk yield from peak for cows grazing treatment LTS compared to the Control treatment in the transition period from spring to summer. In summer, cows grazing treatment STS produced more milk (1AE41 kg fat plus protein per cow d ) than cows grazing all other treatments (0AE92 kg per cow d )1 ) because of the superior nutritive value of herbage of pre-grazing pasture and higher apparent dry-matter intakes. Swards based on alternative species to perennial ryegrass are capable of supporting milk production that is at least comparable over an annual cycle.

An experiment was conducted on 2 contrasting soil types for 4 years (1998–2001) to determine the effects of plant species mixture, management inputs, and environment on sown species herbage accumulation (SSHA) and seasonal growth pattern of pastures for dairy production. Five pasture types, combined with 3 management treatments, were established in south-west Victoria. Three of the pasture types were based on perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.). One pasture type included short-term, winter- or summer-active species in the perennial ryegrass–white clover mixture. The final pasture type was based on the perennial grasses cocksfoot (Dactylis glomerata L.), tall fescue (Festuca arundinacea Schreb.), and phalaris (Phalaris aquatica L.). The 3 management treatments involved different levels of fertiliser input and weed/pest control. Pasture type had a significant impact on SSHA in 3 of 4 years. In the first year, the mixture based on cocksfoot, tall fescue, and phalaris had the lowest SSHA, but this pasture matched other types from 1999 onwards and yielded the highest in 2000, the year with the driest summer during the experiment. Ryegrass–white clover mixture based on old cultivars had generally lower SSHA than the other types except in the first year. Higher fertiliser inputs increased SSHA by 16–28% in 1998, 1999, and 2001. There was a significant site × pasture type interaction on SSHA in 2000. The mixture based on cocksfoot, tall fescue, and phalaris produced up to 1–2 t DM/ha.year more than the other types in summer and autumn in dry–normal years. The inclusion of short-term species, or more stoloniferous white clover cultivars, in the ryegrass–white clover mixture, had little effect on SSHA, or on the seasonal distribution of pasture growth. Pastures based on perennial grasses other than perennial ryegrass appear to have potential for altering the seasonality of pasture growth in south-west Victoria, although the benefits resulting from changing pasture type will depend on environment. Overall, increasing management inputs usually had a greater effect on SSHA than changing pasture type, but management responses were also affected by environment, particularly through the effects of a dry season on a sandy soil type.

The effects of sward surface height (SSH) and daily herbage allowance (HA) on the defoliation pattern and grazing mechanics of early lactation dairy cows grazing on irrigated perennial ryegrass–white clover pasture were studied. The hypothesis tested was that SSH and HA affect intake and diet selection through their effects on the pattern of defoliation which is influenced by the resistance to prehension bites. Factorial combinations of two initial SSH (14 and 28 cm) and two daily HA (35 and 70 kg DM cow−1 d−1) were examined in a replicated experiment. The peak longitudinal tensile force required to break the sward portion encompassed in a 100 cm2 area [bite fracture force (BFF100)] was measured as an index of the resistance to prehension. The volume of herbage defoliated and herbage intake increased with SSH (P < 0·05) and HA (P < 0·01). Corresponding to an increase in HA from 35 to 70 kg DM cow−1 d−1, there was a proportional increase in the total defoliation area (TDA) and intake by 0·24 and 0·55 in the short sward compared with 0·16 and 0·32 in the tall sward respectively. The results of this experiment suggest that a consistent spatial pattern of reduction of the canopy exists during defoliation by cows and that the volume of sward canopy defoliated is the major variable affecting herbage intake. The BFF increased down the sward profile at a rate that was higher (P < 0·05) for the taller sward than for the shorter sward. It is proposed that a relatively lower resistance to prehension in the short sward compared with the tall sward explains the greater proportionate increase in TDA and intake corresponding to an increase in HA. The rate at which BFF100 increases down the sward profile is suggested as a sward physical variable that can influence the defoliation process. The estimated time and energy costs of prehension bites are discussed in the context that larger bites are handled more efficiently than smaller bites.