Abstract The paper compares the findings of four surveys, conducted between the years 1964 and 1970, in three different countries (STEINMANN and FOX, USA; ROCHEBLAVE-SPENLE, France; PFEIL, Germany: Hamburg and Büdingen); it attempts to collate these findings with a view to diachronic interpretation and, with respect to the leading question of changing notions of sex roles, it offers a final evaluation. First, there are those results which have been established before, namely that (a) there are divergencies between women’s self-images and their notions of the ideal woman; that (b) men as well as women do but have a very scanty knowledge about the conception of their role as developed for them by the opposite sex, while at the same time (c) new role attributes, obviously of more general relevance, are becoming discernible; and, lastly, that it is possible (d) to observe typical sequences in the processes of sex role change; added to those findings are these more specific ones: that (a) role images vary markedly with respect to internal versus external signs of emancipation (with orientations towards ‘intra-familial values’ remaining rather conservative, whereas those towards values of ‘self-realization’ in the spheres of employment, public life, and leisure activities tend to be more progressive); that (b) the drifting apart of these role dimensions is being reinforced by the two sexes’ differing value attitudes, and is conducive to insecurities for both the male and the female (he experiencing a shrinking of the universe of roles open to him, whilst she has not as yet found a definitive new role ideal of her own); and that (c) such insecurities call forth family crises as well as an atmosphere of general distrust between the sexes,which is due mainly to a lack of openmindedness, frankness, and mutual information. These results are made more specific by means of comparisons along lines of employment groupings, societal strata, and affiliations to different cultures. The picture revealed by such diachronic comparisons shows that, on the one hand, role concepts associated with ‘companionate marriage’ have been gaining additional ground during recent years - and are likely to become the predominating ones in the future -, whilst there still are extant, on the other hand, sex roles that are determined, heavily in some parts, by tradition; all this leads to a constellation which carries with it a double burden especially for the woman. Women’s emancipation, it appears, is not being achieved in one big stride but rather, step by step and under tensions. This follows from data according to which women’s ideal notions about themselves do manifest, paradoxically enough, features of a more conservative character than is true of their realistic self-images. In conclusion, the paper takes a stand for a pragmatic solution (and, incidentally, a liberal one) to present-day role problems. This should be sought, it argues, not in mutual ‘assimilation’ of men and women but in the opening up of a new range of diversified role potentials. The difference between the male and female roles could then be expected to take on a new shape, of hitherto unknown appearance, and to attain to a new equilibrium.