The Ist‐year correlates of infant–father and infant–mother attachment were investigated in a longitudinal study of early family development. Mothers and fathers were observed interacting with their firstborn, 3‐month‐old infants. Parents also were interviewed individually at 3 months child age concerning their time with the infant and their attitudes and reports about the infant and their parental role. Mothers and fathers were seen in the strange situation with their infant at 12 months. For infants and fathers, security of attachment was predicted from the qualities of interaction at 3 months, the father's attitudes and reports about the infant and the paternal role, and the father's time with the infant. For infants and mothers, security of attachment was predicted from the qualities of interaction at 3 months and the mother's time with the infant. Over the last 2 decades, Bowlby's attachment theory (Bowlby, 1969) and the work of Ainsworth and others (Ainsworth, Blehar, Waters, & Wall, 1978) have influenced the thinking of many developmental psychologists about the infant–mother relationship and the importance of the emotional bond that develops between the infant and mother during the 1st year. Though much research has established the 1st‐year correlates of infant–mother attachment at 12 months (see Belsky & Isabella, 1988), much less is known about the early correlates of the infant–father attachment relationship. Fathers as Attachment Figures Because, in Western cultures, mothers typically become the preferred attachment figure and fathers typically become the preferred playmate (Bretherton, 1985), it is unclear whether the attachment construct is as important in describing infant–father relationships as it seems to be in describing infant– mother relationships. Though one parent can serve in both roles, Bretherton notes that Bowlby (1969) considered the roles to be conceptually distinct, with the child seeking an attachment figure when under stress and a playmate when in a positive mood. In a number of studies (Ainsworth et al., 1978; Belsky, Rovine, & Taylor, 1984; Crockenberg, 1981; Maslin & Bates, 1983; Miyake, Chen, & Campos, 1985), researchers have found that mothers' behavior during the 1st year covaries significantly with assessment of attachment at 1 year using the strange situation procedure developed by Ainsworth and her colleagues (Ainsworth et al., 1978). That is, when mothers have been more sensitive, warm, responsive, and accepting in observations made of mothers with their infant during the 1st year, infants are more likely at 1 year to be judged as having secure attachments with their mother from observations in the strange situation procedure. Although Bowlby (1969) recognized that fathers were usually attachment figures and could be primary attachment figures, comparable findings concerning the relationship between fathers' behaviors with infants during the 1st year and the security of infant–father attachment have not been reported. In one study of 1st‐year origins of infant–father security of attachment, Belsky (1983) concluded that there was no relation between quality of infant–father attachment and father–infant interactional experience. Belsky suggested that the failure to find such a relation may be due, in part, to measurement of inappropriate dimensions; variables other than those that are typically measured (sensitivity, responsivity, warmth) may indeed predict the development of a secure infant–father attachment. Origins of Infant–Father Versus Infant–Mother Attachment The origins of infant–father attachment may reside in different types of interactions than the origins of the infant–mother attachment. There are several differences between mothers and fathers in Western cultures. Fathers spend considerably less time on average with infants than do mothers, and this decreased time includes less one‐on‐one interaction, less accessibility, and less responsibility for the infant's care (Lamb & Oppenheim, 1989). When with their infants, what mothers and fathers do with them differs: Mothers' interactions typically involve caretaking, but fathers' interactions typically involve play (Lamb, 1981; Yogman, 1982).