ABSTRACT
Summary The relationship between botanical composition of the herbage on offer and herbage eaten is a key link in the modeling and simulation of grazing systems, yet there appear to be no data describing this relationship for multispecific or nontemperate swards. This paper describes one such relationship for cattle grazing multispecific swards in the subtropics. Native or improved pastures were grazed by Hereford heifers fistulated at the esophagus. The native pasture, which had never received fertilizer, contained mixed grasses and no legumes. One of the improved pastures consisted of native pasture fertilized
679with superphosphate, and the other improved pastures were fertilized and oversown with white clover (Trifolium repens) alone or in conjuction with lotononis (Lotononis bainesii). Samples of pasture and esophageal extrusa were collected on five occasions from June through February from 2 replications of each pasture. Extrusa were separated into green grass, green legume, and dead herbage components. Relationships were established between the total amount of green herbage on offer (GH, metric tons dry matter [DM]/ha) and the proportion of green herbage ingested (PG, g DM/g total extrusa DM). These relationships were similar for the four pastures and were described by a single pooled Mitscherlich equation PG = 0.924 - 7.901e-3.7006GH (r = 0.93) with an approximate residual standard deviation (RSD) of ± 0.119. Relationships were also fitted between the amount of green legume on offer (GL, metric tons DM/ha) in both of the grasslegume pastures and the proportion of green legume in the diet (PL, g DM/g extrusa DM). There were no significant differences between the relationships for the two pastures. The equation from the pooled data was linear, and the intercept did not differ significantly from zero. PL = 0.432 GL (SEb ± 0.070; RSD ± 0.085; P < 0.001). The proportion of green grass ingested (PGG, g DM/g extrusa DM) can be calculated by subtracting PL from PG.
