Factors associated with the milking cluster and the long milk tube which cause a diminished or a fluctuating vacuum within the liner of the teatcup assembly during simulated milking with an artificial udder fell into 3 groups. 1. Lower milk flow rates and admission of air at the clawpiece greatly decreased vacuum fluctuation within a pulsation cycle and increased the mean level of vacuum in the liner during that part of the cycle in which milk was flowing from the teat. 2. Larger bore of the short milk tube connecting the liner to the clawpiece, pulsation of the teatcup liners in pairs instead of all 4 together, and larger volume of the clawpiece bowl in the main decreased fluctuation in vacuum within the liner. 3. Larger bore and shorter length of the long milk tube connected to the clawpiece and decreased height above the outlet of the clawpiece to which the milk was raised mainly increased the mean vacuum in the liner during that part of the cycle in which milk was flowing from the teat.

In a milking experiment with 75 cows mean overall fluctuations in milking vacuum within the liner in a pulsation cycle at peak flow were 11·5 and 4·8 inHg and corresponding mean levels of vacuum during that part of the cycle when milk was flowing from the teat were 11·75 and 13·0 inHg. There were no differences in milking performance of practical consequence.