Summary.Three box experiments examining the effects of shoot removal on rhizome carbohydrate reserves are reported.During summer in England the optimum frequency of hand defoliation for maximum loss of carbohydrate was 10–14 days. As in previous experiments, high levels of soil nitrogen sometimes increased the rate of carbohydrate loss from 9‐in. rhizome pieces. This result was obtained when nitrogen was supplied as a nitrate, an ammonium salt, or urea.Repeated applications of 2 or 4 oz/ac paraquat over a period of about 60 days reduced carbohydrate reserves to about 50% of the level obtained with the same number of hand clippings. It is suggested that paraquat may have been translocated to the rhizomes, the carbohydrate content of which may have been reduced by an effect of the paraquat on the respiration rate.When plants which had been subjected to different methods and frequencies of defoliation were uprooted and buried beneath 2 in. of soil, there was a connection between the rhizome carbohydrate content at the time of burial, the ability of the plants to regenerate, and the rate at which new carbohydrate reserves built up during the following two mon