This paper uses a new linked sample constructed from full-count census data of 1851–1911 to revise estimates of intergenerational occupational mobility in England. I find that conventional estimates of intergenerational elasticities are attenuated by classical measurement error and severely underestimate the extent of father-son association in socioeconomic status. Instrumenting one measure of the father’s outcome with a second measure of the father’s outcome raises the intergenerational elasticities ( β ) of occupational status from 0.4 to 0.6–0.7. Victorian England was therefore a society of limited social mobility. The long-run evolution and international comparisons of social mobility in England are discussed.