The National Population Policy for the year 2000 had once more set a target for the achievement of the replacement level of the Indian population. The replacement level is defined in terms of the Total Fertility Rate (TFR) of 2.1 births per woman in the course of her life and should be reached by 2010. Demographic projections would prefer to assume 2016 as a more realistic date. The average Indian TFR had come down from 6 in 1951 to 3 in 2001. To the great surprise of planners and demographers, several south Indian states have proved to be way ahead of the National Population Policy. Kerala registered a TFR of 1.71 in 2001, and Tamil Nadu was at almost the same level with 1.76, closely followed by Andhra Pradesh at 1.94. Karnataka was still above the replacement level, at 2.24; it was estimated that it would reach that level within a few years. Andhra Pradesh was the greatest surprise of them all: its TFR had dropped from 2.39 in 1997 to 1.94 in 2001. It has a high rate of female illiteracy and there has been no significant economic progress in this state. The major assumption of demographers that female education and economic progress would lead to a lower TFR was therefore contradicted by the experience of Andhra Pradesh. Moreover, the decline in the TFR usually takes time and does not happen in such a dramatic fashion as it did in Andhra Pradesh. Perhaps it was an awareness of future deprivation rather than of economic progress which prompted even illiterate women to resort to birth control. This goes against all normal demographic assumptions, but there was a striking parallel to this development in Andhra Pradesh in East Germany at the time of German reunification. The number of East German births dropped by 40 per cent at that time, which must have been due to apprehension of an uncertain future on the part of young East German women. This shows that perceptions of the future rather than long-term social and economic trends may influence the decisions of women. This is, of course, only one aspect of the rapid spread of birth control. Knowledge of the methods of contraception and the will to adopt them are also of great importance. Demographers who have studied the spread of adoption of contraceptives have noticed a snowball effect. After an initial phase when only a few women practise birth control, the demonstration effect catches on and others follow their example. In a strange reversal of the assumption that female education leads to birth control, it has been found that birth control may foster female education. Among illiterate women who adopted contraception there were many who would send their girls to school. The correlation seems to be significant, but of course it does not necessarily indicate a causal relation.