The aim was to examidne the nature of the roots for a set of net fertility function expressing the full variety of fertility experience to be found in large human populations. One segment of the project was an attempt to create a family of model fertility schedules encompassing the full range of human experience, an attempt that culminated in the tables presented here. Several sets of model tables have been developed representing in different ways and in different detail of coverage typical age patterns of mortality found in human populations at every recorded level of mortality (United Nations 1955; Coale and Demeny 1966; Ledermann 1969; Brass 1971). A single model schedule of first marriage frequencies (with a corresponding model schedule of the proportion ever married by age, and also a schedule of person years lived in the ever married state) has been found to fit a wide range of age patterns of nuptiality, given the proper choice of parameters specifying the origin and the appropriate horizontal and vertical scales for the standard nuptiality function (Coale 1971).

In Appendix B are printed a set of model age-specific fertility schedules analogous to the earlier model tables of mortality and nuptiality. These schedules represent age patterns of fertility rather than the level of fertility. Since the sum of the tabulated fertility rates, taken over all reproductive ages, is 1.0, age-specific fertility rates can be calculated by multiplying each model rate by an actual population's total fertility rate. In these new tables the fertility in each single year of age is calculated as the product of a number representing the proportion cohabiting at that age and a number representing the age-specific fertility of those who cohabit. By such combinations we have been able to construct schedules that we believe express essentially the full range of age structures of fertility likely to be found in large human populations. The source of this belief is, first of all, the regularity, both in the age pattern of nuptiality, and in the variation of marital fertility with age, noted in an earlier article (Coale 1971). The further and sounder basis for the belief in the validity of the model fertility schedules is their extraordinarily close fit to various accurately recorded fertility schedules of rad-* This project was conducted as part of a graduate course in mathematical demography at Princeton