The purpose of this study was to investigate intergenerational transmission of reproductive behaviour in Botswana. The major source of data was the 2001 Botswana AIDS Impact Survey where a nationally representative random sample of men and women aged 10–64 years was selected using a stratified two-stage probability sample design. Covariates in the analysis include age, education, marital status, religion, age at first birth, residence, duration at residence and contraceptive use. The main analytical technique is linear regression. The results indicate that the reproductive behaviour of older generations has a significantly positive influence on the reproductive behaviour of the subsequent generation, but does not affect the subsequent generation homogenously. The effect appeared much stronger for women who initiated childbearing at an older age, for women who had never been to school, and for the cohort aged 50–59 years. These findings suggest that number of siblings, as a reproductive behaviour determinant, may very well have confounded previous reproductive behaviour analyses in Botswana. The study draws attention to the importance of the effect of origin family size in determining reproductive behaviour outcomes in Botswana.