A mathematical model of the transmission of HIV-1 within heterosexual populations in Sub-Saharan Africa is described and its properties analysed. The model incorporates epidemiological and demographic processes and extends previous work in this area via the inclusion of age and sex dependency in rates of sexual partner change, and sexual partner choice dependent on age. Parameter assignments are made on the basis of current data on the transmission dynamics of HIV-1 and the demography of human populations in Africa. Both age-dependent rates of sexual activity and the sexual contact of males with females younger than themselves act to enhance the predicted demographic impact. With realistic parameter values, the model suggests AIDS is able to reverse the sign of population growth rates from positive to negative values over a timescale of a few decades. The sensitivity of this prediction is examined in relation to changes in the pattern of sexual contact between different age classes of females and males, different patterns of change in the age-dependent rate of sexual partner change, different assumptions concerning the doubling time of the epidemic in its early stages, and the relative efficiencies of viral transmission between men and women, and vice versa. The impact AIDS is predicted to have on the number of young and elderly persons as a fraction of the number of productive adults (the dependancy ratio) is examined under various assumptions concerning the weighting to be applied to mirror the burden imposed by the care of those with AIDS. The paper includes an assessment of the influence of the timing of changes in sexual behaviour, or the promotion of the use of condoms, on the predicted course of the epidemic. The paper concludes with a discussion of data needs and the model refinements required to more accurately mirror current understanding of the epidemiology of HIV-1.