This book is about path-dependence in the standardization of railway gauges and spans a period of almost two hundred years from c. 1820 onwards. Puffert claims to have written it with two audiences in mind-those interested in the history of railways and technical standards and those interested in the concept of path dependence (p. xi). Nevertheless, it is abundantly clear that this work is primarily about path-dependence, and uses railway gauges as a case study to illustrate the explanatory power of this concept in economics (p. 7). The model described in part 3 of the book is an explicit attempt to model the economics of path-dependence incorporating a spatial modelling approach. The claim that initial preference and path-dependency dominate gauge choice, resulting in diversity, seeks to inform the debate between the proponents of path-dependency (Brian Arthur and Paul David) and the challengers (Stan Liebowitz and Stephen Margolis). Historians will find this claim neither surprising nor path-breaking.

Douglas J. Puffert has shown how particular preferences, both in initial gauge selection as well as subsequent conversion processes, result in the spatial diversity of gauges. This diversity, he argues, can persist even after concerted efforts to capture network externalities through track conversions. He considers track selection to be a product of both systematic as well as non-systematic elements. These arguments are buttressed by an impressive historical narrative of the gauge selection process, spanning all continents and time periods (part 2) and by the simulation results of the model he constructs (part 3).

While reading the book, one cannot but help draw a comparison of spatial networks that Puffert describes to those studied earlier by Thomas Hughes (Networks of Power: Electrification in Western Society, 1880-1930. Baltimore, MD: John Hopkins University Press, 1983). Puffert omits to make a reference to this substantial literature from the History of Technology, an omission made glaringly obvious by a lack of reference to Steven Usselman's Regulating Railroad Innovation: Business, Technology, and Politics in America, 1840-1920 (Cambridge University Press, 2002). Historians may also wonder about the sources that Puffert has relied upon to construct his historical narrative. There are references to reports and proceedings of various regulatory commissions, railway associations, etc., but most of the narrative appears to be based upon secondary or published material.

These are hardly serious concerns for those interested in this book for its discussion and analysis of the economics of path-dependence. To this audience, concerns may lie in other directions. Puffert models the preference of agents at the initial stages of the track selection process for the early networks and the preferences of subsequent networks as well as early networks in the later stages of the selection process (pp. 257-61). However, the model is unable to distinguish between the preference of the railway companies and those of its engineers, a historically significant distinction that Puffert himself highlights in his narrative (pp. 52-65). The reader is unable to determine if such distinctions are indeed important to the path-dependent process. The narrative in the book suggests that the distinction is important, however, the model cannot account for it. A more serious issue though is an absence of discussion of if and how competing transportation networks, particularly road and/or air transport networks, affect preferences for subsequent track conversions, both in the historical context of the twentieth century as well as in the simulation model. The reader is also left wondering whether productivity differences and operating efficiencies contribute to conversion costs in the model and in gauge preferences.

Such questions, nevertheless, show how engaging this part of the book is and how effectively Puffert is able to get the reader to consider path-dependency as a spatial process, in addition to being a temporal one. This is the most important contribution of Puffert's work to the literature on path-dependency. He provides a substantial refinement to the work of Arthur and David, upon which Puffert clearly bases his motivation and inspiration (pp. 248-52). Another important contribution that the book makes, which Puffert fails to acknowledge, is to the growing literature on voluntary consensus standards (eg., Craig N. Murphy and JoAnne Yates. The International Organization for Standardization (ISO): Global Governance Through Voluntary Consensus. New York: Routledge, 2009). While the literature in economics has traditionally distinguished between de facto and de jure standards, there are few detailed empirical studies of standards that may be legal but are built by consensus by firms, particularly at the international level (see for instance, Marc T. Austin and Helen V. Milner, “Strategies of European Standardization.” Journal of European Public Policy 8, no. 3 (2001): 411-31.) Puffert's work fits into this gap by showing how consensus is built through changing preferences based upon both systematic as well as nonsystematic factors. Thus, while the book is clearly about the economics of path-dependency, it does inform the economics of standardization in an original way.

On the whole, the book is easy to read as it is intelligently organized. Part I raises the historical, technological, and economic issues, while part II provides the rich historical narrative of the gauge selection and conversion process. Part III provides the economic reasoning and analysis behind the path-dependency and the resultant diversity and part IV brings the various issues together in a conclusion. Readers could easily read part IV in advance of the historical narrative. Those interested only in the historical events can skip part IV altogether, although Puffert has made this section accessible to noneconomists.

Puffert's work is firmly rooted in the controversy over path-dependence in economics literature. It does not explicitly relate to the broader literature on standards and standardization. Nevertheless, the book opens up the issue of path-dependence further in its implications not only for economic history, but for economic geography as well. Also, Puffert tells a compelling story about choices in railway gauges.