Abstract. This Subject Report analyzes fertility differentials by socioeconomic and demographic characteristics for all India and for individual states, based on data from India’s 1992–93 National Family Health Survey. The findings indicate a wide diversity among Indian states in the total fertility rate, which ranges from about two children per woman in Goa and Kerala to about five children per woman in Uttar Pradesh. By socioeconomic characteristics, the total fertility rate tends to be higher among rural women than among urban women, higher among women with less education, higher among Muslim women than among Hindu women, and higher among scheduled-caste (SC) women and scheduled-tribe (ST) women than among non-SC/ST women. Estimates of period parity progression ratios (PPPRs) from birth to first marriage imply nearly universal marriage, with 96 percent of all Indian women eventually marrying. However, this overall figure masks some diversity among the states. The PPPRs from birth to marriage range from almost 100 percent in Punjab and Bihar to 89 percent in Orissa, Assam, and Goa, implying that 11 percent will never marry in these latter three states. In the country as a whole, progression from marriage to first birth is also nearly universal, at 97 percent. Again there are exceptions, most notably Andhra Pradesh, where the PPPR from marriage to first birth implies that 7 percent of married women will never have a first birth. In the country as a whole, the progression ratio from first to second birth is also quite high, at 93 percent. At higher parities, progression ratios fall off more rapidly. How rapidly depends to a considerable extent on the general level of fertility in a particular state. The multivariate analysis indicates much higher parity progression ratios among women who have experienced one or more child deaths than among women who have not experienced any child deaths. Parity progression ratios are much lower among women who have one living son than among women who have no living son, and much lower among women who have two or more living sons than among women who have only one living son. Controls for urban/rural residence and education affect these results hardly at all. Parity progression ratios tend to be higher among rural women than among urban women, but this difference virtually disappears when education is controlled. On the other hand, differentials by education persist when residence is controlled, indicating that urban women have lower fertility largely because they are more educated. Differentials in parity progression ratios by husband’s education largely disappear when residence and wife’s education are controlled, indicating that wife’s education is a considerably more important determinant of fertility than husband’s education. Differentials in parity progression by religion tend to be large and mostly unaffected by controls for residence and education, indicating that differences by religion in levels of urbanization and education do not explain fertility differences by religion. On the other hand, parity progression ratio differentials by caste/tribe, which tend to be small to begin with, are reduced further by controls for residence and education.