Estimation of herbage consumption by grazing cattle using measurements of eating behaviour

Feed consumption (I). of a group of eight non-lactating Jersey cows strip-grazing oats at three stages of growth was estimated from number of eating bites (N) and bite size (5) measurements using the expression, I= N × S. N was measured with four non-fistulated animals while S was determined independently with four oesophageal fistulated animals. These estimates were compared with herhage consumption estimates using a cutting technique (fourteen samples of both herbage offered and refused). Eating behaviour estimates of intake based on N and S were 29% higher than cutting estimates due to fistulated animals having a high number of mastication bites with the head up during sampling and a large diurnal variation in bite size. After allowing for these sources of error eating behaviour estimates were only 2% higher than cutting estimates. Sampling variability was similar for both cutting and eating behaviour techniques; on an individual plot basis the percentage coefficients of variation were 11.2 and 10.1 for cutting and eating behaviour techniques, respectively. Methods of reducing sample variability are discussed and tables of numbers of animals required to detect true differences in intake at various levels of precision are presented. It is concluded that herbage consumption by grazing ariimals can be obtained from measurements of eating behaviour.