1. A group of four Ayrshire and four Friesian cows was strip grazed on a cocksfoot-dominant sward while a second, similar group was zero grazed on herbage cut from the same field. The experiment consisted of five 10-day periods, two in the spring and three in the summer.
2. The intakes of all cows were calculated from values for faeces output, estimated by using chromic oxide, and for herbage digestibility, estimated from faeces nitrogen. The regressions used for predicting digestibility were determined with the zero-grazed cows, digestibility coefficients being calculated from measured intake and estimated faeces output values.
3. Over the whole experiment there was no difference between treatments in milk yield, herbage digestibility or intake. The solids-not-fat content of the milk of strip-grazed cows was significantly higher than for zero-grazed cows.
4. In both spring and summer the increasing maturity of the herbage caused declines from one period to the next in herbage digestibility and intake, and there were declines also in milk yield and solids-not-fat content. The declines were greater for the zero-grazed cows, apparently because they, unlike the strip-grazed animals, were unable to select the more digestible and palatable components of mature herbage. The effects of selective grazing on digestibility, however, were evidently small, for the difference in between the treatment groups was never greater than one unit.
5. The estimated energy intakes of both treatment groups corresponded quite well with their theoretical requirements of energy for maintenance, milk production and live-weight gain, and there was no evidence of the energy cost of free grazing being appreciably greater than that of zero grazing.