From data gathered in the 1965 and 1970 National Fertility Studies, the total intended family size of white, once-married women was analyzed for the effect of age at marriage and age at 1st birth, and an effort was made to determine the extent to which these effects are open to either demographic or sociological interpretation. The age-at-marriage effect is nonlinear, with much greater extremes of fertility in women who marry before age 18 or after age 25. The age-at-marriage differential was unchanged when the data were controlled for premarital pregnancy or for correlated background variables such as education, parental occupation, and religion. However, controlling for unwanted births and subfecundity reduced the difference between the 2 extreme age-at-marriage groups by about 1/3. Differences in fertility when analyzed by mother's age at 1st birth were over twice as large as those differences that appeared in the age-at-marriage analysis. The theoretical aspects of this clear effect on fertility of the age at which marital and parental roles are begun needs consideration.