Donald 0. Parsons' recent article in this Journal raises several important questions.' Using a statistical model that has been employed by numerous sociologists for over ten years, Parsons illustrates the effects of various measures of family background on male youths' schooling and earnings.2 Parsons' original contribution to this literature is to employ a human-investment model in order to separate the total schooling effect of a given family characteristic into productivity, cost, and consumption components. For example, one of Parsons' most intriguing results is that a mother's educational attainment has its main influence on her son's labor market achievement through its effects on the structure of the child's tastes for schooling and through the productivity effect of schooling on earnings. Father's schooling, on the other hand, has its effect primarily through reducing the cost of schooling arising apparently from the greater subsidies more educated fathers bestow on their children.3 If this is true, exactly what is it that well-educated mothers do in the home that gives their children an earnings advantage in the labor market? Further, if this advantage reflects primarily environmental rather than genetic factors, can some type of social policy be used to replicate in families with low parental levels of education the process (whatever it is) of intergenerational wealth transmission enjoyed by high-status families? Parents can transmit income and wealth to their offspring in a variety of ways. At one extreme, transmittal occurs directly through bequests and gifts of financial assets. On the other hand, biological parents supply the genetic endowments that may affect skills useful in school and the labor market. Finally, parental time devoted to the rearing of children can also augment a child's skills and traits.4 It is these parental investments of time and goods in the children-particularly in the preschool years-which we wish to discuss here. If it could be shown, for example, that there is a consistent relationship between the well-known family background variables (e.g., parental education, occupational status, family size) used by Parsons, investments of parental time in preschool human capital, and the outcomes of the investment (the child's schooling and earnings), we could an-