From 1985, when I first met Channi Kumar, he played a seminal role in the research I have conducted with my colleagues on the effects of postnatal depression on child development. At the time I was a very junior researcher, struggling to get my work underway, and his enthusiasm and encouragement were invaluable. Channi played a particularly important role from 1989 when, with the support of the Tedworth Charitable Trust and the Winnicott Trust, the Winnicott Research Unit was established in Cambridge, under the joint Directorship of myself and Peter Cooper, together with Alan Stein. Channi became a key member of the Unit’s advisory committee and, right up until the time of his death, he regularly, and very kindly, provided the wisest counsel and support, and he is still very sorely missed. He was an inspiring friend and colleague, with the deepest compassion and humanity for the plight of the families he helped, and boundless enthusiasm for new perspectives and ideas that could advance understanding and clinical treatment, and I am greatly indebted to him for all his support. This chapter describes research on a prospective longitudinal study of the development of children of depressed and well mothers conducted in the Winnicott Research Unit; much of it took place under Channi’s watch. The work has involved a large number of colleagues, aside from the contributions of Peter Cooper and Alan Stein, and I am particularly grateful to Alison Hipwell, Matt Woolgar, Sheelah Seeley, Janet Edwards, Sarah Halligan, Adriane Arteche, Ian Goodyer, and Joe Herbert for their involvement and support. A diagnosis of ‘postnatal depression’ includes a wide range of possible symptoms, and therefore this unitary term can mask considerable variation in the nature of its presentation. For example, one mother could be slowed down, sleeping excessively, and barely eating, while in another, the episode may manifest itself in restlessness and agitation, with the mother being hardly able to concentrate and feeling constantly irritable. Not surprisingly, then, studies of the effects of postnatal depression on mother-infant interactions have also identified striking variability.