PIP: Researchers began following 1920 pregnant women (3rd trimester) from 5 primary health centers (PHCs) in Bangalore Division of Karnataka State in India between August 1977-November 1977 to determine the infant mortality rate (IMR), its components, and correlates. Auxiliary nurse midwives followed them once a week during the 1st month after delivery and then once a month until the index child either died or reached 1 year. Male infants were more likely to die than female infants (113 vs. 90; p.05), especially in the neonatal period (76 vs. 48; p.01). Further IMR was highest among 15-19 year old mothers and 35-44 year old mothers (115 and 128 respectively compared to 87-101 for all other mothers). Infants of birth order 2 had the lowest IMR (85 vs. 102-109 for other orders). Parental education was inversely correlated with IMR, especially mother's education. For example, the IMR for mothers with no education was 115 compared to 60 for those with at least 5 years of education (p.01). Father's education was significant at .05. Mothers who lived on at least 20 acres of cultivable land owned by the family (determinant of economic status) exhibited an IMR of 23 while the IMR for those whose family owned no land was 111 (p.1). Mothers who had experienced a previous stillbirth were more apt to experience infant death than those who had not had a previous stillbirth (150 vs. 90; p.05). Untrained traditional birth attendants delivered 69% of all live births. IMR for these infants was considerably higher than that for trained attendants (108 vs. 87; p.1). Most women delivered at home and IMR was much higher for home births than institution births (103 vs. 89), especially after the 1st 28 days (43 vs. 11; p.1). The IMR was highest at Pavagada PHC (130) and lowest at Doddaballapur PHC (83). Policy makers should consider these findings prove when formulating suitable policies to reduce infant mortality.