Four loealities in the State of Campeche, Mexico were used during the Spring-Summer of 1982 and 1983 to conduct trial in orferto study the stability of rice genotypes. The studied variables were: days to heading (DH), plant height (PH) and grain yield (GY). Two methodologies (Eberhart and Russell, 1966 and Wricke, 1965) were compared to determine which was most effective in detecting the stability ofthe tested genotypes. The results show that these tested genotypes were different or were environmentaly affected at each locality. The PH mea n was slightly shorter for dry farming cultivars. This shortage was largely due to thc effect ofthe Edzna environment which was drier because of the soil type. Among the assayed genotypes, there is a potential for GY in two of the advanced lines (IR2055-466-6-6 and IR2055-481-2-l Cu). One of these could be released as a new variety for rainfed conditions. The results show that either method can be used, giving the same response, although the Wricke's method is more easily interpreted that Eberhart and Russell.