The effect of contrasting white clover contents on diet selection by grazing sheep provides an insight into grazing behaviour mechanisms. When sheep have a choice between ‘strips’ containing either 70% white clover or 100% grass they consistently choose to graze the former. When clover is intermingled with grass in normal sheep-grazed swards evidence for selective grazing is less clear. The objectives of this experiment were, firstly, to define the pattern of discrimination in response to variation in the contrast in clover content between alternative swards. And, secondly, to test whether the pattern is affected by the mean clover content about which the range is established.

The ‘base’ levels of clover content: 30 and 60% together with 7 ‘contrast’ levels from -30 to +30% of ‘base’ level were planned. Swards of equal content were included as controls. Thus the 30% base was contrasted with 0, 10, 20, 30, 40, 50 and 60%; and the 60% base with 30, 40, 50, 60, 70, 80 and 90% white clover. Each combination of treatments was repeated at two nitrogen levels: C and 30 kg N/ha and the whole replicated twice. The experimental area was at Grasslands Division, Palmerston North, New Zealand. The pasture comprised well-established Ruanui ryegrass - Huia white clover with some weed species. Fifty-six plots (5.6 x 3.5 m), each containing 2 ‘contrast’ areas at 2.8 x 3.5 m were sprayed with different herbicides on 29 July 1983 to achieve the above contrasts.