1. Thus U.S. cities are less dense than Canadian cities, see Barry Edmonston, Michael A. Goldberg, and John Mercer, "Urban Form in Canada and the United States: An Examination of Urban Density Gradients," Urban Studies 22 (1985), 209-217.

2. Ray Hebert, "Ventura Freeway—It's Now No. 1," Los Angeles Times (10 March 1985), II, 1, 5.

3. Mark S. Foster, From Streetcar to Superhighway: American City Planners and Urban Transportation, 1900-1940 (Philadelphia: Temple University press, 1981), 3-24.

4. David Ward, "A Comparative Historical Geography of Streetcar Suburbs in Boston, Massachusetts and Leeds, England, 1850-1920," Annals of the Association of American Geographers 54 (1964), 477-489, shows that the system in Boston truly was superior to that of Leeds. Scott L. Bottles, "The Making of the Modern City: Los Angeles and the Automobile, 1900-1950" (Ph. D. diss., University of California, Los Angeles, 1984).

5. The auto replaced walking as much as it did public transit in Pittsburgh. Joel A. Tarr, Transportation Innovation and Changing Spatial Patterns in Pittsburgh, 1850-1934 , Essays in Public Works History, no. 6 (Chicago: Public Works Historical Society, 1978), 38. David O. Wise and Marguerite Dupree, "The Choice of the Automobile for Urban Passenger Transportation: Baltimore in the 1920s," South Atlantic Urban Studies II (1978), 153-179, show how the main advantage of the automobile in Baltimore was more housing for the dollar, as the city had a dense streetcar network.

6. Upton Sinclair, The Jungle (New York: Penguin, 1979; 1st ed. 1906).

For a clear demonstration of the relationship between high costs of fixed rail urban transport and suburban homes, see Gary R. Hovinen, "Suburbanization in Greater Philadelphia, 1880-1941," Journal of Historical Geography 11 (April 1985), 174-195. Joel A. Tarr, "From City to Suburb: The 'Moral' Influence of Transportation Technology," in Alexander B. Callow, Jr., ed., American Urban History: An Interpretive Reader with Commentaries 2d ed. (New York: Oxford University Press, 1973), 202-212, and Tarr, Transportation Innovation and Changing Spatial Patterns , 5, 6, 11, 14, 20-21, details the inaccessibility of fixed rail transit for most urban workers, for whom a round trip would cost between 15% and 20% of their daily wage.

7. Gary A. Tobin, "Suburbanization and the Development of Motor Transportation: Transportation Technology and the Suburbanization Process," in Barry Schwartz, ed., The Changing Face of the Suburbs (Chicago: University of Chicago Press, 1976), 101. Tobin's essay provides a subtle analysis of suburban expansion. The socialist government of Milwaukee argued that accessible suburbs would solve the "problem of congestion"; Milwaukee Municipal Campaign Book, 1912 (Milwaukee: Social Democratic Party, 1912), 12. Quotation from Clay McShane, Technology and Reform: Street Railways and the Growth of Milwaukee, 1887-1900 (Madison: State Historical Society of Wisconsin, 1974), 67; on working-class political attitudes, see his chapters 7 and 8.

8. Theodore Hershberg, Harold E. Cox, Dale B. Light, Jr., and Richard R. Greenfield, "The 'Journey-to-Work': An Empirical Investigation of Work, Residence, and Transportation, Philadelphia, 1850 and 1880," in Hershberg, ed., Philadelphia: Work, Space, Family, and Group Experience in the 19th Century (New York: Oxford University Press, 1981), 128-173.

9. John R. Borchert, "American Metropolitan Evolution," The Geographical Review (July 1967), 301-332; for a more precise analysis within one of Borchert's grander periods, see Michael P. Conzen, "A Transport Interpretation of the Growth of Urban Regions: An American Example," Journal of Historical Geography I (October 1975), 361-382.

10. Sam B. Warner, Jr., The Urban Wilderness: A History of the American City (New York: Harper & Row, 1972). Or see James J. Flink, "The Metropolis in the Horseless Age," in Margaret Latimer, Brooke Hindle, and Melvin Kranzberg, eds., "Bridge to the Future: A Centennial Celebration of the Brooklyn Bridge," Annals of the New York Academy of Sciences 424 (May 1984), 289. Flink asserts that "the settlement patterns and lifeways of contemporary metropolitan America, however, have been most of all shaped by the motor vehicle." The construction of this sentence exemplifies technological determinism: the car shapes the city. Such a causal construction is true only in the same trivial sense as, "chunks of hot metal cause death by bleeding," substitutes for, "thousands of Americans deliberately kill one another each year.''

11. Fred Viehe, "Black Gold Suburbs: The Influence of the Extractive Industry on the Suburbanization of Los Angeles, 1890-1930," Journal of Urban History 8 (November 1981), 3-26; Spencer A. Crump, Ride the Big Red Cars: How Trolleys Helped Build Southern California (Los Angeles: Crest Publications, 1962).

12. Kevin Starr, Inventing the Dream: California through the Progressive Era (New York: Oxford University Press, 1985); Richard Henry Dana, Jr., Two Years Before the Mast (New York: Penguin, 1981), 146-147; Bottles, "The Making of the Modern City"; Martin J. Schiesl, "Airplanes to Aerospace: Defense Spending and Economic Growth in the Los Angeles Region, 1945-60," in Roger W. Lotchin, ed., The Martial Metropolis: U.S. Cities in War and Peace (New York: Praeger, 1984), 135-150.

13. Compiled from list of members, National Automobile Chamber of Commerce, Facts and Figures of the Automobile Industry, 1926 (New York: National Automobile Chamber of Commerce, 1926); Bottles, "The Making of the Modern City."

14. For charioteers, see Alan Cameron, Porphyrius—The Charioteer (Oxford: Clarendon Press, 1973), 136-149, and idem, Circus Factions: Blues and Greens at Rome and Byzantium (Oxford: Clarendon Press, 1976). For locomotive names, see Freeman H. Hubbard, Encyclopedia of North American Railroading: 150 Years of Railroading in the United States and Canada (New York: McGraw-Hill, 1981), 306, and Benjamin A. Botkin and Alvin F. Harlow, eds., A Treasury of Railroad Folklore: The Stories, Tall Tales, Traditions, Ballads, and Songs of the American Railroad Man (New York: Crown, 1953), 382-383. For a discussion of Lexington and an exquisitely reproduced picture, see Roy King and Burke Davis, The World of Currier and Ives (New York: Random House, 1968), 78-79. According to King and Davis, Lexington sired 236 champions. They quote the poet Huder All's verse: "Whispers fly about the race-tracks when some mighty deed is done; 'Tis no more than we expected from the blood of Lexington!"

Twain example cited by Frederick Alderson, Bicycling: A History (New York: Praeger, 1972), 104.

15. Gerald Silk, Automobiles and Culture (New York: Museum of Contemporary Art and Harry N. Abrams, 1984); Janicemarie A. Holty, "The 'Low Riders': Portrait of a Youth Subculture," Youth and Society 6 (1975), 495-512; Tom Wolfe, The Kandy-Kolored Tangerine-Flake Streamline Baby (New York: Farrar, 1965).

16. During the last decade of the nineteenth century, the bicycle was the mode of high fashion promenade in Paris: Vanity Fair , 1897, cited by Alderson, Bicycling: A History , 87. The automobile supplanted the bicycle as the mode of promenade during the first decade of the twentieth century, as is perhaps best exemplified in Lartigue's photos of the rich in the Bois de Boulogne. See Fred C. Kelly, "The Great Bicycle Craze," American Heritage 8 (December 1956), 69-73; Sidney H. Aronson, "The Sociology of the Bicycle (1879-1900)," Social Forces 30 (March 1952), 311.

17. This contradicts the analysis of James J. Flink, America Adopts the Automobile, 1895-1910 (Cambridge, Mass.: MIT Press, 1970), 210-213, which merits some detailed attention. Flink first cites contemporary literature, 1899-1909, as evidence "that the development of adequate automobile roads lagged well behind the diffusion of the motor vehicle . . . and that the automobile was widely adopted here [the United States] despite a relative scarcity of suitable roads . . ." (211). As a further empirical test of this plausible argument, he correlates states by their rank ordering on three different variables—automobiles per capita, roads per square mile and percent of all roads that were improved, the latter two for 1904 and 1914, the former for 1910. The relationship with the earlier variables and the number of automobiles was modest, for the later period even slighter. From this he concludes that the number of automobiles had a "short-range deleterious effect . . . on roads" (212).

The two models he tests may be written like this:

I. Number of Automobiles / Population Over 18 = Miles of Roads / Surface Area of State and,

II. Number of Automobiles / Population Over 18 = Miles of Improved Roads / Total Road Miles

(Both equations use rankings of states and both use lagged or led time periods. Flink does not explain why he ranks rather than uses actual values.)

Equation I as tested by Flink makes no sense: presumably the greater the population of a state, the more the miles of roads; yet a greater number of square miles only occasionally meant lesser population, considering the large, sparsely populated western states at the turn of the century. Thus as it stands his results for Equation I test nothing. To capture the relationship Flink intended to, the equation should have been constructed thus:

Ia. Number of Automobiles = Population + Miles of Roads + Miles of Surface Area

Equation II examines the relationship between automobiles per capita and the proportion of all roads which was improved. Presumably, some states could have had very limited yet highly improved mileage, hardly an advantage for automobile owners. Again, to capture the relationship Flink wanted to, the equation might be rephrased as follows:

IIa. Number of Automobiles = Population + Miles of Roads + Miles of Improved Roads

This would capture the sense he intended, that automobiles had some relationship with improved rather than merely all roads.

As Flink's conclusions stand, then, they cannot be justified, for the empirical analysis captures neither the claimed relationship nor any other. Flink does not state his data sources; probably they were from the Department of Commerce, Highway Statistics, Summary to 1955 (Washington, D.C.: Government Printing Office, 1957), cited in his bibliography.

Paul Barrett, The Automobile and Urban Transit: The Formation of Public Policy in Chicago, 1900-1930 (Philadelphia: Temple University Press, 1983), 69-70, 139-153. The only theoretically informed history of roads is Hillaire Belloc's prescient The Road (London: Unwin, 1924). For an engineering history of American nonurban roads, see Department of Transportation, Federal Highway Administration, America's Highways, 1776-1976: A History of the Federal-Aid Program (Washington, D.C.: Government Printing Office, 1977), and Thomas H. MacDonald, "The History and Development of Road Building in the United States," Transactions of the American Society of Civil Engineers 92 (1928), 1181-1206. Edward C. Kirkland, Men, Cities and Transportation: A Study in New England History, 1820-1900 , (Cambridge, Mass.: Harvard University Press, 1948), analyzes transportation only between cities.

18. Calculated from Good Roads I (January-June 1892), "Contract Notes," 171, 233, 291, 351. Philip P. Mason, "The League of American Wheelmen and the Good Roads Movement, 1890-1905" (Ph.D. diss., University of Michigan, 1958). Ballard C. Campbell, "The Good Roads Movement in Wisconsin, 1890-1911," Wisconsin Magazine of History 49 (Summer 1966), 276, 289-290. Quotation: E. G. Harrison from, "Gen. Harrison Tells How the Road was Built," The Road Maker , 1, 3 (Port Huron, Mich., c. 1900), 6, cited in Department of Transportation, Federal Highway Administration, America's Highways , 1776-1976, 47-48.

Interest in high-quality paving came as an early recognition that paying facilitated the movement of commercial freight. In New York in 1887, the issue of annexing surrounding cities such as Brooklyn was partly motivated by the poor quality of its streets, which directly lowered the efficiency of docks and wharfs in the city's harbor. In importance for facilitating commerce, streets—their paving and decongestion—were equal to docks and street railways. See Mayor Abram S. Hewitt, cited in David C. Hammack, Power and Society: Greater New York at the Turn of the Century (New York: Russell Sage Foundation, 1981), 192, 193, 233.

19. Alderson, Bicycling: A History , 85-106; Jerry Sandvick, "Early Airport Development and the Emergence of the Metropolitan Airport Commission," Hennepin County History 43 (Fall 1984), 3-17, describes how a group of Minneapolis businessmen purchased the defunct auto racetrack for airport land by subscription in 1920, not all that different from the funding for mid-nineteenth-century urban projects in England.

20. Glenn Yago, The Decline of Transit: Urban Transportation in German and United States Cities, 1900-1970 (New York: Cambridge University Press, 1984) perpetuates the argument that mass transit did pose a viable alternative to the automobile early in the twentieth century. Few of the historians who study urban technology agree: the main reason for the perpetuation of the myth seems to be ignorance about fixed rail transit as it did exist and experience with successful Technology and Reform , 38-39, for a summary of the reasons the automobile proved to be superior to fixed rail.

21. Of necessity, the uncertain quality of early automobile data render such tests crude and suggestive. On the one hand, if road building represented a passive local response to the new technological imperative introduced by the automobile, then new roads should have followed increases in automobile registrations; on the other hand, should people have purchased automobiles following the expansion of the road system, then we can interpret the automobile transport revolution as technology following essential political decisions. The first hypothesis represents a technological determinist argument, the second a nondeterminist historical argument. Unfortunately, neither hypothesis can be clearly evaluated. If people in fact purchased cars in response to the opportunity to drive them, i.e., in response to more roads, and if they also purchased them on the basis of "neighborhood" or "contagion" effects, having observed the utility of their neighbors' cars, the model for change would look like this: Increase in Cars = Utility + Contagion. This can be operationalized as: Cars Added = Roads Added + Cars Added (For previous year) or, C = R + C (t-1), but, because, by definition, C (t-1) = R (t-2) + C (t-2), etc., the model has circular causation and therefore cannot accurately be tested except in the form, C = R (t-1), or, in the obverse, R = C (t-1).

The following correlation matrix presents a simple test of these two hypotheses. Notice that the variable for cars added is a cumulative one in that it represents change in annual registrations, not simply new cars purchased. The results are ambiguous: both road and automobile expansion seem to account for each other.

Similarly, the hypothesis that the diffusion of the automobile technology responded to increased utility may also be tested. Here a slightly more complex test must be used, for the idea is more subtle. Let us hypothesize that the annual change in the number of automobiles registered results from a combination of two major factors: first, the contagion effect of a spreading consumption item—"everyone else is getting one"; and second, the expanding hard surfaced road network which continuously increased the automobile's utility. Again, the micro-computer is a clarifying analog: purchases result from a combination of the contagion effect and the actual utility, the latter of which is driven by software (like roads for cars). For the automobile, these two factors can be specified as independent variables measured by, respectively, the change in total automobile registration in the previous year, and hard surfaced miles added each year, again including changes in the over-eighteen population in the previous year and again for the period prior to 1930. The second part of the figure reports the estimated regression for this model.

Several suggestive insights emerge from these regression estimates. First, roads did respond to cars and population the previous year, but the previous addition of roads was far more important. Holding all else constant, each added mile of hard surfaced road required only 1.4 miles of road the previous year, while requiring 286 cars or 322 additional people.

For automobiles, the regression must be estimated on annual changes, rather than the actual values of the variables. Contagion best explains the increase in automobiles (this emerged from the correlations, but the regression is much stronger). The number of roads added has virtually no effect, forcing us to strongly question if not simply reject the notion that more places to drive meant more car registrations. On the other hand, it is possible that the all-powerful variable, car registrations in the previous year, cumulatively captures both contagion and utility, for when estimated for hard surfaced roads in the previous year, the results show a positive relationship.

It is likely that these estimated relationships mask more significant urban relationships which conformed more closely to the earlier predictions made in this chapter. The specific urban cases may have been asked by national trends which aggregate city and county, state and local.

Brooklyn after the race, and counted 13,048 persons, of whom fewer than one-fourth were on horses or in wheeled vehicles. Survey citations in Ralph F. Weld, Brooklyn Village, 1816-1834 (New York: AMS Press, 1938), 278-279.

23. For a summary of the urban problems associated with horses, see Flink, "The Metropolis in the Horseless Age," 289-290. Tobin, "Suburbanization and the Development of Motor Transportation," 99, cites the horse-drawn rail statistics. On the horse manure problem, see Joel A. Tarr, "Urban Pollution Many Long Years Ago," American Heritage 22 (October 1971), 65-70. Ironically, in the early nineteenth century, horse manure had been much less of a problem, for the poor kept pigs which roamed the streets, eating manure and other organic waste. Social reformers outlawed pigs in the mid-nineteenth century, creating new waste problems that necessitated the hiring of street cleaners.

24. State of Minnesota, Biennial Report of the Secretary of State, 1907-08 and 1909-10 (Minneapolis, 1908, 1910), 4, 6. Tarr, Transportation Innovation , 26-28, cites the Pittsburgh survey information.

25. National Automobile Chamber of Commerce, Facts and Figures of the Automobile Industry, 1921-29 (New York: National Automobile Chamber of Commerce, 1921-29). In fact, some states did not achieve systematic registration until 1921; Historical Statistics , 730.

26. For a discussion of the Model T Ford and rural road conditions, see Peter J. Hugill, "Good Roads and the Automobile in the United States, 1880-1929," The Geographical Review 72 (July 1982), 327-349, esp. 336-337. This is a useful and comprehensive summary article. J. Interrante, "The Road to Autopia: The Automobile and the Transformation of American Culture," in David L. Lewis, ed., The Automobile in American Culture: Michigan Quarterly Review 19 and 20 (1980/1981), 502-517, argues that the car was more important for rural people—based on proportionally higher ownership by the 1920s.

28. McShane, Technology and Reform , 67, cites a newspaper mention in the 1890s of the "half hour limit." The first commuter train appeared in New York City in 1837, carrying riders up to Harlem; in less than ten years service extended to Westchester County: Kenneth T. Jackson, "Technology and the City: Transportation and Social Form in New York," in Latimer, Hindle, and Kranzberg, eds., "Bridge to the Future," 284. The foremost expert on urban rail transit contrasts rail and auto transit as point-to-point and lateral travel mechanisms: see George W. Hilton, "Rail Transit and the Pattern of Modern Cities. The California Case," Traffic Quarterly 21 (July 1967), 379-393.

29. Compare the major freeways and urban rail lines in Reyner Banham, Los Angeles, Architecture of the Four Ecologies (New York: Harper & Row, 1972), 32, 92.

30. Larry Long and Diana DeAre, "The Slowing of Urbanization in the U.S.," Scientific American 249 (July 1983), 39. Specifically, they report densities of twenty-five "supercities," those over one million, in 1980.

31. Leon Moses and Harold F. Williamson, "The Location of Economic Activity in Cities," American Economic Review LVII (May 1967), 211-222.

32. See Clay McShane, "Transforming the Use of Urban Space: A Look at the Revolution in Street Pavements, 1880-1924," Journal of Urban History 5 (May 1979), 279-307. "These changes in streets literally paved the way for the automobile . . ." (300). McShane develops a convincing argument for the prior causality of the paving revolution, mixing two strands: one, the lobbying efforts of the good roads movement and engineers and two, somewhat less convincing to me, a changed attitude toward streets, which no longer regards them as play areas and undifferentiated public, social space. Pavement innovation came, McShane concludes, for "shifts in the cultural and political climate of nineteenth century cities" (302). Campbell, "The Good Roads Movement in Wisconsin," 290-291, develops convincing evidence based on voting patterns and numbers of automobiles that support for good roads developed independent of and prior to the diffusion of the automobile in Wisconsin.

33. Federal Highway Administration, America's Highways, 1776-1976 , 16. Richard Dennis, English Industrial Cities of the Nineteenth Century: A Social Geography (Cambridge: Cambridge University Press, 1984), 110-140, provides a useful summary and analysis of the role of nineteenth-century urban transport in England; for the generally disputed argument that in Britain demand led transport, see P. W. Daniels and A. M. Warnes, Movement in Cities: Spatial Perspectives on Urban Transport and Travel (London: Methuen, 1980), 2-3.

34. William Fulton, "Those Were the Best Days: The Streetcar and the Development of Hollywood before 1910," Southern California Quarterly 66 (Fall 1984), 235-256: Sherman quoted on 238.

35. Peter Hall, The World Cities , 3d ed. (London: Weidenfeld & Nicolson, 1984), 150, map 6.3b.

36. John C. Schneider, Detroit and the Problem of Order, 1830-1880. A Geography of Crime, Riot, and Policing (Lincoln: University of Nebraska Press, 1980).