systematic sampling stratified according to age, and included complete birth histories of 1,460 women 13-54 years old. (See Ferry, 1977, for more information.)

This rural area is monitored through annual rounds. It included a population of 6,435 persons as of October 1990. (See Lagarde et al., forthcoming and Pison et al., 1993, for more information.)

Niakhar-I, 1962-1967: included a population of 35,187 persons as of January 1, 1966. (See Cantrelle, 1969; Cantrelle and Léridon, 1971; and Project Niakhar, 1992, for more information.)

Niakhar-II, 1983-present: included 26,496 persons as of March 1, 1992. (See Project Niakhar, 1992, for more information.)

Ngayokhème Area (included in the Niakhar-I and Niakhar-II areas), 1963-present: included 5,947 persons as of March 1, 1991. (See Waltisperger, 1974; and Pison, 1978, for more information.)

Included a population of 18,988 persons as of January 1, 1966. The Paos Koto survey was monitored from 1963-1965, and Ndemene from 1963-1978. (See Cantrelle, 1969; and Cantrelle et al., 1980, for more information.)

These studies reconstituted families by retranscribing parish certificates and surveys among resident women in 1965-1966. Diahanor included 284 persons as of June 1, 1966; Fakao included approximately 3,000 persons as of January 1, 1966. (See Lacombe, 1973, for more information.)

This survey involved multistage cluster sampling of the rural population of Sine-Saloum (the current regions of Fatick and Kaolack), a retrospective survey among 1,894 women aged 15-44 having been married or had one live birth, with the birth history being cut off at 6 years. (See Goldberg et al., 1984, for more information.)