130 Perspectives in American History urban growth: the nature of the commodities produced and ex- changed, the marketing problems they create, the institutional and legal framework within which economic activity takes place. JFunctionally. we may divide the roles of preindustrial towtjs (i.e., towns of the seventeenth and eighteenth centuries) as follows: (^) civil and ecclesiasriqiljitlrninistratinri with their attendant "court life" and demimondes; (2) maritime transport and external com- mercial exchange; (3) industrial production; and (4) internal ser- jvices. The first three can usually be described as independent vari- ables, while the fourth is essentially a dependent variable. That is, the number of persons employed in a town in service functions (broadly conceived to include not just innkeepers, servants, tailors, dressmakers, barbers, and the like, but also petty retail shopkeepers, building trade workers, most teachers, ministers of religion, and other professionals) will vary roughly with the number of persons attracted to the town as residents or visitors by the other three functions. It is not easy to find acceptable data that might enabje^5ne to construct functional profiles of colonial towns, me^sttfing exactly the number of persons employed in each of dierlour sections just described. However, reasonable approximations can be made. Allan KulikofF18 has reported the njamoer of adult males in each occupation recorded in the Boston tax records for 1790.1 have re- analyzed his figures, classifying them according to the four "sec- tors" enumerated in tbeprevious paragraph (Appendix C). For Philadelphia and ks'suburbs of Southwark and Northern Liberties, there are published and unpublished tax lists giving occupations. Sam B. Warner has supplied me with data for Philadelphia in 1774 (frornjtdx and other official records) giving 3,793 occupations out oŁabout 6,000 heads of household and single men listed.19 I had 18. Allan KulikofE "The Progress of Inequality in Revolutionary Boston," William and Mary Qiiarrer/y, 3d ser., 28 (1971), 375-412, esp. 411-4.12. This is a valuable article, though marred by some errors in arithmetic and eighteenth-century commercial termi- nology. 19. For an explanation of the sources of these data (now stored in a machine archive at the Inter-University Consortium for Political Research, Ann Arbor), see Sam Bass War- ner, Jr., Tlie Private City: Philadelphia in Tliree Periods of Its Growth (Philadelphia, 1968), pp. 226-227.