Research on crime in the late 20th century has consistently shown, that despite the public rhetoric, immigrants have lower rates of involvement in criminal activity than natives. The earliest studies of immigration and crime conducted at the beginning of the 20th century produced similar conclusions. We show, however, that the empirical findings of these early studies suffer from a form of aggregation bias due to the very different age distributions of the native and immigrant populations. We find that in 1904 prison commitment rates for more serious crimes were quite similar for the two nativity groups for all ages except ages 18 and 19 when the commitment rate for immigrants was higher than for the native born. By 1930, immigrants were less likely than natives to be committed to state and federal prisons at all ages 20 and older. But this advantage disappears when one looks at commitments for violent offenses. Immigrants in their late teens, in fact, were more likely than their native counterparts to be incarcerated for violent offenses.