Like race in the United States, caste in India is perceived by millions of people today as a particularly rigid and oppressive form of inequality. Many practices, described in earlier textbooks as integral to the normal functioning of caste, would now be considered invidious and discriminatory, and might invite legal and political sanctions. Fifty years ago, it might have made sense to say that discrimination based on race was pathological while discrimination based on caste was normal. To insist on the same contrast would be misleading today. When we consider caste and race together, we are struck at once by the remarkable similarity in the contrasting attitudes towards women of lower and higher ranks characteristic of men in privileged positions in both systems. My argument is that inequalities of caste are illuminated in the same way as those of race by a consideration of gender. There are two aspects of the problem. There is, firstly, the sexual use and abuse of women, which is an aspect of the inequality of power,seen in its most extreme form in the treatment of women of the lowest rank by men of the highest; this is the aspect of the problem that has received most attention. There is, in addition, the unremitting concern with the purity of women at the top, associated with ideas regarding bodily substance that have been discussed separately in studies of American kinship and of caste and kinship in India we can deepen our understanding of both caste and race by exploring these ideas more systematically and in comparative terms.If we believe that the position assigned in thought and life to women is of crucial significance to the understanding of both caste and race, we are much better placed today than anthropologists were a generation ago to pursue the comparison between the two in greater depth. The position of women in society,particularly in modem or contemporary society,received very little scholarly attention from sociologists and social anthropologists in the decades when comparisons of race and caste were most extensively made. It is true that Dollard (1957) wrote about the ‘sexual gain of caste’ in the U.S. South and Berreman (1960) later wrote about the sexual exploitation of both black and untouchable women. But these observations were either lost or ignored in the absence of an adequate conceptual framework for the comparative study of gender. It may well be the case that such a framework does not exist in a fully developed form even now. But there is no doubt that the climate has altered vastly so that the plea for a serious consideration of these issues can no longer be as easily ignored as in the past. The advances achieved in women’s studies in the last two decades have implications not only for a fuller understanding of the relations between the sexes, but also for a deeper insight into the general problem of inequality, of which caste and race are two particular forms. I am referring now not only to new facts but also to new ways of looking at facts that have long been taken for granted. The sexual use of women of inferior rank by men of superior rank would not acquire its characteristic forms in societies divided by caste or race if the ordinary relations between men and women were not marked by asymmetry.The asymmetry characteristic of such relations in general is merely reinforced when the man belongs to a superior race (or caste) and the woman to an inferior one. The normal requirement of asymmetry would be seriously upset if the woman belonged to a superior and the man to an inferior rank. The stricter the demand for asymmetry in the ordinary relations between men and women, the more severe will be the sanctions against the reversal of roles. I would surmise that the distances required to be maintained between castes or between races are likely to vary directly with the disparities established between men and women in the society as a whole. We have to be careful, however, to distinguish between relatively stable societies and those undergoing rapid change as a result of changes in the legal and political systems and in the general climate of opinion. Such changes have been marked in the last four or five decades not only in the United States, but also in India. In these changing conditions, small and gradual reductions in disparities are periodically met with sudden and violent reprisals which bring established patterns into sharp relief It is difficult, when this is happening, to demonstrate or even to discern any clear direction of change. The asymmetry inherent in the link between race and gender is nicely brought out in Dollard’s study of Southerntown: In simplest terms, we mean by a ‘sexual gain’ the fact that white men, by virtue of their caste position, have access to two classes of women, those of the white and Negro castes. The same condition is somewhat true of the Negro women, except that they are rather the objects of the gain than the choosers, though it is a fact that they have some degree of access to white men as well as to men of their own caste This asymmetry sustains and is sustained by contrasting images of the sexuality of black and white women of which exact parallels may be found in the contrasting images of lower and upper caste women in India.