This paper examines trends and estimated sex ratio at birth (SRB) for India and child mortality (q5) risk, 1981 and 1991, and discusses how the 2001 Census results are foreshadowed. Earlier state-level (rural vs urban) analyses are extended to present these data at the district level in the form of maps. Multivariate statistical analyses exploring social and economic covariates that affect the likelihood of gender bias in death (female disadvantage in child mortality risk) have been conducted. The female disadvantage is evident in birth and death spread over India in the decade 1981-1991. Though infant and child mortality levels fell for girls and boys, gender differences persisted, and penetrated the hitherto egalitarian south. Masculine SRB that were seen only in urban areas of the north-west in 1981 have spread to urban areas of many northern states in 1991. Multivariate analyses suggest that between 1981- 1991, women's status variables become less associated with reduced gender bias. Modernisation variables suggest less association with gender bias, or that a substitution of pre-natal for post-natal elimination of unwanted daughters may be occurring. These findings are situated in the literature on gender, education, paid work participation, and marriage system changes in India.