The reliability and validity of three measures of infant temperament were compared in this study. The measures included the revised Carey (1978) Infant Temperament Questionnaire, a version of the Bayley (1969) Infant Behavior Record revised for completion by the parent, and a modified version of Buss and Plomin's (1975) EASI, an acronym standing for its four subscales--Emotionality, Activity level, Sociability and Impulsivity. Sixty-six mothers of healthy infants were asked to complete the battery of three infant temperament scales and a random subsample of mothers was asked to take a retest 21 days after the initial testing. Results showed that subscales from all three measures were correlated in meaningful ways, suggesting that there are independent, salient characteristics of the infant that can be measured by quite different rating scales. The strengths and weaknesses of each measure were identified and discussed. (JMB)