Intergenerational effects on linear growth are well documented. Several generations are necessary in animal models to ‘wash out’ effects of undernutrition, consistent with the unfolding of the secular trend in height in Europe and North America. Birthweight is correlated across generations and short maternal stature, which reflects intrauterine and infant growth failure, is associated with low birthweight, child stunting, delivery complications and increased child mortality, even after adjusting for socio-economic status. A nutrition intervention in Guatemala reduced childhood stunting; it also improved growth of the next generation, but only in the offspring of girls. Possible mechanisms explaining intergenerational effects on linear growth are not mutually exclusive and include, among others, shared genetic characteristics, epigenetic effects, programming of metabolic changes, and the mechanics of a reduced space for the fetus to grow. There are also socio-cultural factors at play that are important such as the intergenerational transmission of poverty and the fear of birthing a large baby, which leads to ‘eating down’ during pregnancy. It is not clear whether there is an upper limit for impact on intrauterine and infant linear growth that programmes in developing countries could achieve that is set by early childhood malnutrition in the mother. Substantial improvements in linear growth can be achieved through adoption and migration, and in a few selected countries, following rapid economic and social development. It would seem, despite clear documentation of intergenerational effects, that nearly normal lengths can be achieved in children born to mothers who were malnourished in childhood when profound improvements in health, nutrition and the environment take place before conception. To achieve similar levels of impact through public health programmes alone in poor countries is highly unlikely. The reality in poor countries limits the scope, quality and coverage of programmes that can be implemented and modest impact should be expected instead. The Lancet series on Maternal and Child Undernutrition estimated that implementation to scale of proven interventions in high burden countries would reduce stunting by one-third; this is perhaps a realistic upper bound for impact for high quality programmes, unless accompanied by sweeping improvements in social services and marked reductions in poverty. Finally, because so much can be achieved in a single generation, intergenerational influences are unlikely to be an important explanation for lack of programme impact aimed at the window of the first 1000 days. Failure to prevent linear growth failure in developing countries has serious consequences for short- and long-term health as well as for the formation of human capital. The nutrition transition has created a double burden by adding obesity and related chronic diseases to the public health agenda of countries still struggling with the ‘old’ problems of maternal and child undernutrition. The challenge ahead is to increase efforts to prevent linear growth failure while keeping child overweight at bay.