India is a country with a pervasive preference for sons and one of the highest levels of excess female child mortality in the world (female child mortality exceeds male child mortality by 43 percent). In this paper, data from the National Family Health Survey (NFHS) are used to test the hypothesis that son preference influences parity progression and ultimately child mortality. In particular, the survival of female children may be adversely affected by the fact that female births are more often followed by the birth of a younger sibling, especially after a short birth interval. The effects of family composition a r e estimated with hazard models. The results show that family composition affects fertility behavior in every state examined. The effects of family composition on excess female child mortality a r e more complex, but female children with older sisters are often subject to the highest risks of mortality.