■ Drawing upon a significant amount of unpublished data which Margaret Read collected as a part of the 1939 Nyasaland Nutrition Survey and her simultaneous study on the impact of migrant labor on village life in Nyasa land, this article argues that her conclusions that matrilineal practices were being taken over by patrilineal ones is not borne out by the evidence she collected. As peoples in the three nutrition survey villages (Yao, Chewa and Ngoni) and in the surrounding areas experienced much interaction and as, especially, patrilineal Ngoni interacted with matrilineal Chewa, practices usually associated with one or another of these descent systems became mutually modified rather than uni directional. Aspects of patriliny actually failed to give advantage to many people, and aspects of matriliny - such as inheritance, rights to land, and control over labor and children - proved to be more durable than her interpretations implied.