Effects of grazing and haycutting on the yield and persistence of dryland aphid-resistant lucerne cultivars at Tamworth, New South Wales

Abstract

The yield and persistence of 28 lucerne cultivars were assessed in two experiments at Tamworth on the Northern Slopes of New South Wales. Plots were either cut for hay or grazed when all cultivars were at the l/l0th bloom stage or every 6 weeks, whichever occurred first. Rainfall was below average during the experimental period. For purposes of data comparisons, cultivars were grouped into winter-dormant (WD), semiwinter- dormant (SWD), winter-active (WA), highly winter-active (HWA), dormant (D = WD + SWD), and active (A = WA + HWA) categories. In the first experiment (22 cultivars), total yields over 4 years were highest in haycut plots for the WD cultivar WL3 18 (35 690 kg/ha) and lowest for the aphid-susceptible SWD cultivar 77-B-1545 (25 255 kg/ha). In grazed plots, the HWA cultivar CUF101 had the highest yield (20 091 kg/ha) and 77-B-1545 the lowest (12 127 kg/ha). The frequency of all cultivars in haycut plots declined from 71 to 49% during the experiment. Mean frequency decreased from 68 to 11% in grazed plots and was not significantly different between cultivars after 4 years. In the second experiment (seven cultivars), total yields over 3.5 years were highest for the SWD cultivar Pioneer Brand 581 in the haycut plots (21 213 kg/ha) and the HWA cultivar WL514 in grazed plots (11611 kg/ha). The frequency of all cultivars in haycut plots decreased from 89 to 61% during the experiment; those in grazed plots decreased from 90 to 12%. There were no significant differences between the cultivars in either treatment at the end of the experiment. Trends in time between the contrasts WDSWD and WA-HWA (experiment 1) and D-A (experiments 1 and 2) were analysed in a repeated measures analysis. For dry matter, significant trends over time (P<0.05) occurred in the haycut WA-HWA contrast (linear decrease) and for the contrasts WD-SWD (linear decrease) and WA-HWA (concave quadratic) in the grazed plots of experiment 1. Under both grazing and haycutting, the D-A contrasts in experiment 1 were linear and increased with time; in experiment 2, the active lucernes constantly outyielded the dormants. For plant frequency, the haycut WA-HWA contrast increased linearly with time and, with grazing, the WD cultivars had a significantly higher frequency (P<0.05) than the SWD class. For both grazed and haycut plots, the D-A contrasts in experiment 1 were significant and cubic; in experiment 2, they were concave quadratics. These data indicate that, irrespective of frequency, the grazed yield of the active lucernes was always greater than that of the dormants. Despite differences in the plant frequency of the different dormancy classes during the experiment, their persistences after 4 years of grazing were similar.