The study of nineteenth‐century infant mortality in Britain has neglected the rural dimension to a surprising degree. This article maps the change in infant mortality rate (IMR) between the 1850s and the 1900s at registration district (RD) level. Latent trajectory analysis, a longitudinal model‐based clustering method, is used to identify the clusters into which rural RDs fell, based on their IMR trajectories. Relationships between IMR and population density, fertility, female tuberculosis mortality, female illiteracy, male agricultural wages, and distance from London are examined in a longitudinal study. The tuberculosis (maternal health), illiteracy (education), and distance variables had the most effect. IMR responded most strongly to improving health and education in the east, less in the central area, and least in the north and west. The eastern zone's higher‐than‐average mid‐century infant mortality therefore declined faster than the national average. A central and southern zone had slightly lower IMR in mid‐century but did not keep up with the rate of decline in the east. The peripheral north and west had the lowest mid‐century rates but their decline was overtaken by the other zones. The interpretation of these findings and their relevance to the wider study of infant mortality are discussed.