Two studies were conducted in order to examine three major issues arising from recent studies of mothers' reactions to their crying infants. These issues, which arise in connection with the Aversive Stimulus Model of crying, relate to (a) the relationships between cry characteristics and mothers' affective reactions to crying, (b) the variability of cry characteristics of individual infants, and (c) the influence of context on mothers' reactions to their infants' crying. Resting on the premise that the semantic differential is an appropriate technique for addressing these issues. Study A examined the relationships between three sets of semantic differential scales. These were the scales reported by Brennan and Kirkland (1983), which represent three dimensions labelled Affect, Potency, and Evaluation; the scales reported by Zeskind and Lester (1978); and the scales reported by Mehrabian and Russell (1974a), which represent three dimensions of emotion labelled Pleasantness/Unpleasantness, Degree of Arousal, and Dominance/Submissiveness. A combined factor analysis of cry ratings on these scales uncovered the factor structure of the Brennan and Kirkland scales and of the Mehrabian and Russell scales. The factor representing the Brennan and Kirkland Affect scales also represented both the Zeskind and Lester scales and the Mehrabian and Russell Pleasantness/Unpleasantness scales. The Brennan and Kirkland scales were found to effectively discriminate between perceptually different cry sounds. The Mehrabian and Russell scales, however, were found to be lacking in face validity and therefore unsuitable for use with cry sounds. Study B examined the perceptions and affective reactions of mothers listening to their own infants' cries, in two situations - in the home as the crying occurred, and in an experimental situation involving tape-recorded cry samples. The results indicate that: (a) mothers affective reactions to cries did not simply depend upon the aversiveness of the cry sounds, (b) mothers' affective reactions to cries were strongly associated with their attributions regarding the causes and consequences of the cries, (c) cries from the same infant and cries from different infants varied considerably with respect to their perceived characteristics and the types of affective reactions they evoked, and (d) ratings of the tape-recorded cry samples tended to over-emphasise the relationships between cry characteristics and mothers' affective reactions, and to under-represent the extent to which negative affective reactions were experienced by the mothers in the home situation. Several suggestions were made for future studies. These included the adoption of an individualised approach to study: (a) the cry repertoires of individual infants, (b) the types and patterns of affective reactions experienced by individual mothers, (c) mothers' attributions regarding to their own feelings and their infants' behaviors, and (d) the relationships between mothers perceptions, attributions, affective reactions, and actual caregiving behaviors.