Mass transiency remains the most striking and consistent finding to emerge from quantitative studies of Victorian North America. In almost every place where historians have looked at least half, often two thirds, of the adults present at one end of a decade had left ten years later, and rates based on shorter periods reveal a stream of people constantly flowing through nineteenth-century cities. Although 363,000 people lived in Boston in 1880 and 448,000 in 1890, during the decade about one and one-half million people actually had dwelled within the city. When Victorians sought a symbol of progress, they often chose the steam engine; had they wanted a metaphor for their cities, they could have found none more apt than the railroad station. In this paper we confront the question of transiency. Using the New York State Census of 1855 for the entire city of Buffalo and a 10 percent sample of household heads in rural Erie County, we attempt a method of estimating persistence (the proportion of the population remaining in a given place) that is different from that used by most historians. Given the richness of the census, we are able to inquire with great detail into the factors that determined length of residence in a nineteenth-century city and its surrounding countryside.