Water management is one of the keys to success in increasing rice production in paddy fields. Paddy rice production will decrease if the rice plants suffer from water stress, and conversely conventional cultivation is very wasteful of water. Local aromatic varieties have deeper rooting character than improved varieties making it more efficient in using water. The purpose of this study was to obtain more efficient and more productive aromatic local varieties in the use of irrigation water in the SRI organic cultivation system. This research was conducted in organic rice fields in Kebonagung village, Imogiri, Bantul with split plot design of 2 factors with 4 replications. Factor I: The mode of irrigation consists of (a) intermittent irrigation, and (b) continuous irrigation and Factor II: varieties consisting of (a) Mentikwangi, (b) Gabusan, (c) Sintanur and (d) IR64. The observation parameters included (1) root characters i.e. surface area of root, total length of root, and dry weight of root, (2) plant growth analysis i.e. specific leaf weight, net assimilation rate, plant growth rate, (3) and yield components i.e. weight 1000 grains, grain weight per hill, and grain weight per plot. The results showed that intermittent irrigation increased the surface area of root and the dry weight of root of Mentikwangi variety, and increased the specific leaf weight and weight of 1000 grains of IR64 variety, and increased the specific leaf weight and the net assimilation rate of Sintanur variety. Grain weight per hill and per plot showed no difference between intermittent irrigation and continuous irrigation. Grain weight per hill and per plot in intermittent irrigation achieved by Gabusan variety were 35.69 g and 14.28 kg respectively, while in continuous irrigation the grain weight achieved by Gabusan variety were 37.69 g and 15.08 kg respectively. In conclusion, there was no difference in yield between intermittent irrigation and continuous irrigation, so that the use of intermittent irrigation is more efficient in water use.