¹ See Merges, Robert P., “The Trouble with Trolls: Innovation, Rent-Seeking, and Patent Law Reform,” Berkeley Technology Law Journal 24 (Fall 2009): 1583–1614Google Scholar; and Bessen, James, Ford, Jennifer, and Meurer, Michael J., “The Private and Social Costs of Patent Trolls,” Boston University School of Law Working Paper no. 11–45 (Sept. 2011)Google Scholar.

² See, for example, McCurdy, Daniel P., “Patent Trolls Erode the Foundation of the US Patent System,” Science Progress 1, no. 2 (2009): 78–86Google Scholar.

³ US Federal Trade Commission, “The Evolving IP Marketplace: Aligning Patent Notice and Remedies with Competition” (Mar. 2011), http://www.ftc.gov/os/2011/03/110307patentreport.pdf, accessed 12 Oct. 2012Google Scholar.

⁴ For numbers of patents, see Carter, Susan B.et al., eds., Historical Statistics of the United States: Earliest Times to the Present, Millennial Edition, vol. 3 (New York, 2006), 426–29Google Scholar.

⁵ Arrow, Kenneth J., “Economic Welfare and the Allocation of Resources for Invention,” in The Rate and Direction of Inventive Activity, ed. Universities—National Bureau Committee for Economic Research (Princeton, NJ, 1962), 609–25Google Scholar; Teece, David J., “Technological Change and the Nature of the Firm,” in Technical Change and Economic Theory, ed. Dosi, Giovanniet al. (London, 1988), 256–81Google Scholar; Mowery, David C., “The Boundaries of the US Firm in R&D,” in Coordination and Information: Historical Perspectives on the Organization of Enterprise, ed. Lamoreaux, Naomi R. and Raff, Daniel M. G. (Chicago, 1995), 147–76Google Scholar.

⁶ Lamoreaux, Naomi R. and Sokoloff, Kenneth L., “Inventors, Firms, and the Market for Technology in the Late Nineteenth and Early Twentieth Centuries,” in Learning by Doing in Markets, Firms, and Countries, ed. Lamoreaux, Naomi R., Raff, Daniel M. G., and Temin, Peter (Chicago, 1999), 19–57Google Scholar; Gans, Joshua and Stern, Scott, “The Product Market and the Market for ‘Ideas’: Commercialization Strategies for Technology Entrepreneurs,” Research Policy 32 (Feb. 2003): 333–50CrossRefGoogle Scholar; Arora, Ashish, Fosfuri, Andrea, and Gambardella, Alfonso, Markets for Technology: The Economics of Innovation and Corporate Strategy (Cambridge, MA, 2001)Google Scholar; Graham, Margaret B. W., “Entrepreneurship in the United States, 1920–2000,” The Invention of Enterprise: Entrepreneurship from Ancient Mesopotamia to Modern Times, ed. Landes, David S., Mokyr, Joel, and Baumol, William J. (Princeton, NJ, 2010), 401–42Google Scholar. See also the introduction and essays in Clarke, Sally H., Lamoreaux, Naomi R., and Usselman, Steven W., eds., The Challenge of Remaining Innovative: Insights from Twentieth-Century American Business (Stanford, CA, 2009)Google Scholar.

⁷ Khan, B. Zorina and Sokoloff, Kenneth L., “Institutions and Technological Innovation during Early Economic Growth: Evidence from the Great Inventors of the United States, 1790–1930,” NBER Working Paper no. 10966 (Dec. 2004)CrossRefGoogle Scholar.

⁸ US Congress, “An Act to Promote the Progress of Useful Arts,” ch. 357, 5 Stat. 117 (4 July 1836)Google Scholar.

⁹ Both sets of volumes are stored in the National Archives in Washington, DC, in the Records of the Patent and Trademark Office, Record Group 241. For a primer on how to use these records, see Cooper, Carolyn C., “Thomas Blanchard's Woodworking Machines: Tracking Nineteenth-Century Technological Diffusion,” Industrial Archaeology 13, no. 1 (1987): 41–54Google Scholar.

¹⁰ Edison's patents are available through the Web site of the Thomas Edison Papers at Rutgers University, http://edison.rutgers.edu/patents.htm, accessed 17 July 2012.

¹¹ Liber vol. W–3, Records of the Patent and Trademark Office, Record Group 241, National Archives, 295.

¹² Liber vol. W–3, 9, 335–36.

¹³ Cooper, Carolyn C., “William Woodworth's Planer: The Patent and Its Politics, 1828–1856,” Dutchess County Historical Society Yearbook 91 (2012): 101–16Google Scholar.

¹⁴ Cooper, Carolyn C., “Social Construction of Invention through Patent Management: Thomas Blanchard's Woodworking Machinery,” Technology and Culture 32 (Oct. 1991): 960–98CrossRefGoogle Scholar.

¹⁵ Winder, Gordon M., “Before the Corporation and Mass Production: The Licensing Regime in the Manufacture of North American Harvesting Machinery, 1830–1910,” Annals of the Association of American Geographers 85 (Sept. 1995): 520–52CrossRefGoogle Scholar.

¹⁶ See Lipman, Jean, Rufus Porter: Yankee Pioneer (New York, 1968), 27–48.Google Scholar

¹⁷ Advice manuals showered inventors with suggestions about how to market their patents. See, for example, Simonds, William Edgar, Practical Suggestions on the Sale of Patents (Hartford, CT, 1871)Google Scholar.

¹⁸ Liber vol. G–2, 319–20, 420–21.

¹⁹ For a picture of one such wagon from a somewhat later period, see Cooper, , “Social Construction of Invention,” 963Google Scholar.

²⁰ Simonds, , Practical Suggestions, 28–29Google Scholar.

²¹ See “Patent Rights, and the Way Farmers are Humbugged,” Transactions of the New York State Agricultural Society 28 (1868): 501–3Google Scholar.

²² Hardesty v. Smith, 3 Ind. 41 (1851).

²³ Adams v. Johnson, 15 Ill. 345 (1854). Of course, buyers could use claims of fraud as an excuse to renege, but this would itself be possible only in an environment fraught with information asymmetries.

²⁴ “In the Matter of Hon. Platt Potter and Winsor B. French, Esq.,” Appendix: Breach of Cases, Privilege, Reports of Cases in the Supreme Court of the State of New York, vol. 55 (Albany, NY, 1870), 665–66Google Scholar.

²⁵ Brown v. Wright, 17 Ark. 9 (1856).

²⁶ Foss v. Richardson, 81 Mass. 303 (1860).

²⁷ Simonds, , Practical Suggestions, 28Google Scholar. The classic study of this problem is Akerlof, George A., “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism,” Quarterly Journal of Economics 84 (Aug. 1970): 488–500CrossRefGoogle Scholar.

²⁸ Supplement to the Statutes of Indiana in Force 1870, vol. 3, 364–65Google Scholar. For a list of states passing such laws, see Cresee, F. A., Practical Pointers for Patentees: Containing Valuable Information and Advice on the Sale of Patents (New York, 1902), 88–91Google Scholar.

²⁹ Ex parte Robinson, 20 F. Cas. 961 (1870).

³⁰ Cresee, , Practical Pointers, 88–91Google Scholar. The Supreme Court later ruled that a state could impose licensing requirements on sellers of patented articles so long as it did not discriminate in favor of residents. By that time, however, geographic patent assignments were no longer of much importance. See Webber v. Virginia, 103 US 344 (1881).

³¹ Hardesty v. Smith, 3 Ind. 41 (1851) at 43.

³² Rockafellow v. Baker, 31 Pa. 319 (1862) at 320, 321.

³³ Miller v. Young's Administrator, 33 Ill. 355 (1864). See also Gatling v. Newell, 12 Ind. 99 (1859); Bond and Green v. Clark, 35 Vt. 577 (1863). For a summary of the case law, see William C. Robinson, The Law of Patents for Useful Inventions, vol. 3 (Boston, 1890), 6593–470.

³⁴ Gatling v. Newell, 12 Ind. 99 (1859). For other grounds on which assignments claimed to be fraudulent were upheld, see Adams v. Johnson, 15 Ill. 345 (1854); Myers v. Turner, 17 Ill. 179 (1855); Jolliffe & Holland v. Collins, 21 Mo. 338 (1855); and Galpin v. Atwater, 29 Conn. 93 (1860). For examples where courts did not side with sellers see McClure v. Jeffrey, 8 Ind. 73 (1856); Bierce v. Stocking, 77 Mass. 174 (1858); Pierce v. Wilson, 34 Ala. 596 (1859); Lester v. Palmer, 86 Mass. 145 (1862); Clough v. Patrick, 37 Vt. 421 (1865), and David v. Park, 103 Mass. 501 (1870).

³⁵ See, for example, the evaluations of new technologies that the New York Agricultural Society published in its annual Transactions; The Seventh Exhibition of the Massachusetts Charitable Mechanic Association at Faneuil and Quincy Halls, in the City of Boston, September, 1853 (Boston, 1853), viGoogle Scholar; and “Report of the Twentieth Exhibition of American Manufacturers,” published as an appendix to the Journal of the Franklin Institute 20 (1850)Google Scholar. More generally, see Khan, B. Zorina, “Promoting the Useful Arts: Technological Innovation outside the Patent System, 1790–1880,” unpublished paper (Aug. 2009)Google Scholar; and Sinclair, Bruce, Philadelphia's Philosopher Mechanics: A History of the Franklin Institute, 1824–1865 (Baltimore, 1974)Google Scholar.

³⁶ On this point, see especially Khan, “Promoting the Useful Arts.” Khan's evidence shows, moreover, that prizes were not awarded exclusively or perhaps even primarily on grounds of technical merit.

³⁷ “American Patents,” Journal of the Franklin Institute 20 (Dec. 1835): 385–407CrossRefGoogle Scholar.

³⁸ Borut, Michael, “The Scientific American in Nineteenth Century America,” unpublished PhD dissertation, New York University (1977), 42–62Google Scholar.

³⁹ Scientific American 11 (6 Oct. 1855): 26.

⁴⁰ Ibid.

⁴¹ Ibid., 31. Munn greatly exaggerated the journal's circulation. Based on Munn's own diary, Borut concluded that circulation peaked at around 25,000 during the 1850s. See Borut, , “Scientific American,” 65–71Google Scholar.

⁴² A series of articles that Scientific American published during the late 1840s and early 1850s exposing popular inventions as fraudulent or based on bad science may have helped build the journal's reputation. See Borut, , “Scientific American,” esp. 101–2Google Scholar. However, David A. Hounshell found that, as late as the 1880s, even the magazine's feature articles were for sale. See Hounshell, , “Public Relations or Public Understanding? The American Industries Series in Scientific American,” Technology and Culture 21 (Oct. 1980): 589–93CrossRefGoogle Scholar.

⁴³ This figure is based on an examination (for patentees whose surnames began with the letter “B”) of the correspondents recorded as handling patent assignments in the Patent Office Digests.

⁴⁴ Borut concluded from examining advertisements in Scientific American that Munn & Company did not use the journal to promote the sale of patents. See Borut, , “Scientific American,” 125–26Google Scholar.

⁴⁵ As one might expect, Munn & Company's success fostered imitation. In 1864 two of the agency's employees quit to found a competing journal, the American Artisan, in combination with a patent agency, Brown, Coombs & Company. A few years later another agency called the US Patent Right Association began publishing the Patent Right Gazette, and in the 1870s the American Patent Agency began publishing the American Inventor. Unlike Scientific American, however, none of these other publications lasted very long. Judging from the entries in WorldCat, the American Artisan survived from 1864 to 1875; the Patent Right G azette from 1871 to 1877; and the American Inventor from 1878 to 1887.

⁴⁶ Again, this figure is based on an examination (for patentees whose surnames began with the letter “B”) of the correspondents recorded as handling patent assignments in the manuscript Digests.

⁴⁷ Cresee, , Practical Pointers, 41–42Google Scholar; Hutchinson, W. B. and Criswell, J. A. E., Patents and How to Make Money Out of Them (New York, 1899), 162Google Scholar. See also Simonds, , Practical Suggestions, 7–9Google Scholar; and An Experienced and Successful Inventor, Inventor's Manual: How to Work a Patent to Make It Pay, rev. ed. (New York, 1901), 61Google Scholar. It is difficult to assess the validity of these charges, but it does appear that few patents were actually sold by such agencies. We have collected samples of assignment contracts from the Patent Office Digests and have not found any assignments handled by agencies that appear in these advertisements.

⁴⁸ Simonds, , Practical Suggestions, 24–25Google Scholar.

⁴⁹ Ibid., 19–28; Hutchinson and Criswell, Patents; An Experienced and Successful Inventor, Inventor's Manual; Cresee, , Practical Pointers, 46–52Google Scholar.

⁵⁰ “Testimony Taken on Behalf of James Edward Smith,” Hammerstein v. Smith (1890), 68, Case 13618, Paper No. 48, Box 1868, Interference Case Files, 1836–1905, Records of the Patent and Trademark Office, Record Group 241, National Archives.

⁵¹ T. D. Lockwood, Reports of Inventions (Not Approved), 1904–8, Box 1383, courtesy of AT&T Archives and History Center, Warren, NJ.

⁵² Post, Robert C., “‘Liberalizers’ versus ‘Scientific Men’ in the Antebellum Patent Office,” Technology and Culture 17 (Jan. 1976): 24–54CrossRefGoogle Scholar.

⁵³ Standards for entry into the legal profession were quite lax during this period, and some patent attorneys had little formal training in the law. To conduct business with the Patent Office all patent attorneys had to do was register. From time to time, the Patent Office “disbarred” attorneys, probably because of malfeasance. On professional legal standards in the nineteenth century, see Bloomfield, Maxwell, American Lawyers in a Changing Society, 1776–1876 (Cambridge, MA, 1976)CrossRefGoogle Scholar; and Bloomfield, Maxwell, “Law: The Development of a Profession,” in The Professions in American History, ed. Hatch, Nathan O. (Notre Dame, IN, 1988), 33–49Google Scholar.

⁵⁴ The table probably overstates the rate of growth of patent attorneys during the 1880s. The Patent Office had just begun to compile its list of registered agents, and it is likely that much of this growth represents an increase in the proportion of agents who were registered rather than in the number of agents themselves.

⁵⁵ Because our first list of registered agents is for 1883, the table probably understates the importance of patent agents and attorneys relative to other third parties in 1871.

⁵⁶ See the entries in Van Winkle's business diary for 12 Jan.; 2 Feb.; 22, 23 and 29 Mar.; 6, 20, and 28 Apr.; and 16 Aug. 1905, Edward Van Winkle Papers, Ac. 669, Rutgers University Libraries Special Collections, Rutgers, NJ. On December 29, 1905, the same two men brought Van Winkle a soap shaving machine invented by a Mr. Luis to examine and evaluate.

⁵⁷ See the letter of 22 July 1912 from Conover Fitch, vice president of the Waltham Watch Company to Wright, Brown, Quinby & May, and Arthur H. Brown's report of 27 July 1912, Wright, Brown, Quinby & May Correspondence Files, Mss. 598, Case 2, Waltham Watch Company records, Baker Library Historical Collections, Harvard Business School, Boston, MA.

⁵⁸ Letters of 25 Jan. 1915 and 18 Sept. 1917 from the Waltham Watch Company, Wright, Brown, Quinby & May Correspondence Files.

⁵⁹ Letter of 31 Aug. 1914 from Thurston and Kwis to Rollin H. White, Container 1, Folder 4, Rollin H. White and Walter C. White Papers, Ms. 4734, Western Reserve Historical Society, Cleveland, OH.

⁶⁰ See 10 May 1905, Van Winkle Diary.

⁶¹ See, for examples, 30 Jan.; 16 and 17 Mar.; 1 Apr.; 1 and 7 May 1905, Van Winkle Diary.

⁶² See, for examples, 6 and 28 Jan.; and 13 June 1905, Van Winkle Diary.

⁶³ See, for examples, 4, 16, and 23 Feb.; 7 Apr.; 11 and 20 May; and 6 Sept. 1905, Van Winkle Diary.

⁶⁴ See, for examples, 3 Feb., 21 Mar., and 17 July 1905, Van Winkle Diary.

⁶⁵ See 30 Jan., 7 May, and 18 July 1905, Van Winkle Diary.

⁶⁶ See 6 and 28 Jan.; and 21 Oct. 1905, Van Winkle Diary.

⁶⁷ See 24 Aug. 1905, Van Winkle Diary.

⁶⁸ See the diary entry for 20 May 1905. Oliver and Peters subsequently had some disagreement about the terms of the arrangement, and it is not clear whether the deal actually went through. See also 21 and 24 Jan.; 25 and 28 Feb.; 2 Mar.; 13, 22, and 27 May 1905, Van Winkle Diary.

⁶⁹ See 27 Feb.; 7 and 8 Mar.; 24 and 26 June; 26 July; 6 and 8 Sept.; and 17 Nov. 1905, Van Winkle Diary. Disagreements developed in this case too between Oliver and the inventor. See, for example, 21 and 31 July; 8 and 17 Aug.; and 22 Nov. 1905, Van Winkle Diary.

⁷⁰ Usselman, Steven W., Regulating Railroad Innovation: Business, Technology, and Politics in America, 1840–1920 (New York, 2002), 108–117, 169–76CrossRefGoogle Scholar; and Usselman, Steven W., “Patents Purloined: Railroads, Inventors, and the Diffusion of Innovation in Nineteenth-Century America,” Technology and Culture 32 (Oct. 1991): 1047–75CrossRefGoogle Scholar. For the Supreme Court decision, see Railway Co. v. Sayles, 97 US 554 (1878).

⁷¹ Hayter, Earl W., “The Patent System and Agrarian Discontent, 1875–1888,” Mississippi Valley Historical Review 34 (June 1947): 59–82CrossRefGoogle Scholar; Hayter, Earl W., “The Western Farmers and the Drivewell Patent Controversy,” Agricultural History 16 (Jan. 1942): 16–28Google Scholar; Magliocca, Gerard N., “Blackberries and Barnyards: Patent Trolls and the Perils of Innovation,” Notre Dame Law Review 82 (June 2007): 1809–38Google Scholar.

⁷² Hayter, “Patent System.” Another source of the difficulty was a ruling by the Commissioner of Patents in 1870 that for a couple of decades allowed the Patent Office to grant design patents for “useful” changes of form. For a discussion of the confusion created by that decision, see Magliocca, Gerard N., “Ornamental Design and Incremental Innovation,” Marquette Law Review 86 (Summer 2003): 845–94Google Scholar; and Gerard N. Magliocca, “Blackberries and Barnyards.”

⁷³ Hayter, “Patent System”; Hayter, “Western Farmers”; and Hayter, Earl W., “An Iowa Farmers' Protective Association: A Barbed Wire Patent Protest Movement,” Iowa Journal of History and Politics 37 (Oct. 1939): 331–62Google Scholar.

⁷⁴ The additional examples were flour milling (“Imposition on Millers: The Brua Patent Used to Extort Unearned Royalties,” New York Times, 10 Jan. 1896, 14Google Scholar) and brewing (“Brewers' Patent Suits,” Scientific American 43 [16 Oct. 1880]: 246Google Scholar). Besides these cases and a few references to the problems in railroads and agriculture, we turned up only a couple of articles about patent sharks: “A Note of Warning,” Stone 17 (1 Nov. 1898): 436Google Scholar; and “The Patent Bill,” New York Times, 11 Jan. 1879, 4Google Scholar.

⁷⁵ The one individual on the list in 1891 was at the low end of the scale with 14 patents. Three of the 48 assignees with more than 10 patents in 1911 were individuals. They too were at the low end with 11, 14, and 17 patents respectively, compared to 318 for General Electric, 176 for United Shoe Machinery, and 98 for Westinghouse Electric and Manufacturing Company. The samples include about half the population of assignees. We collected them by recording the number of patents received by assignees that appeared on every other page of the lists of assignees in the Annual Report of the Commissioner of Patents for 1891 and 1911. Because we included all patents received by each assignee, we necessarily included patents from pages that were not part of the sample. Hence, our counts overstate the proportion of assignments obtained by assignees acquiring large numbers of patents.

⁷⁶ Because the results for the two samples are so similar, we combine them in the analysis that follows. The main difference between the two is that productive patentees in 1910–11 found their preferred attorneys slightly more quickly on average than in 1890–91.

⁷⁷ See “Testimony on Behalf of Joseph Arbes,” 10, 22–23, 26, Arbes v. Lewis (1900), Case 20,049, Box 2,715, Interference Case Files, 1836–1905.

⁷⁸ 7 May 1905, Van Winkle Diary.

⁷⁹ We focus our discussion on the second two panels because we do not have any lists of registered patent agents before 1883. To the extent that the patent agents listed as correspondents were not functioning as intermediaries, it should be harder to find support in the data for our hypothesis that patent agents improved the functioning of the market.

⁸⁰ They also handled a disproportionate number of contracts for which we lack information on the patent's date of issue. After searching for these patents in the Lexis-Nexis and Google patent databases, we concluded that most of them, especially in 1871 and 1911, were assignments recorded before issue for which the Patent Office had neglected to go back and add the patent number. In other cases, however, the assignment was for an invention that never made it through the patent approval process. These latter cases are useful reminders that assignees who contracted for patents before the date of issue could not be certain that the patent would ever be allowed, which made the soundness of the patent attorney's judgments all the more important.

⁸¹ The results here and in our subsequent analysis were essentially the same when we used three-year intervals.

⁸² The proportion for which the change in the absolute number of patents was greater after meeting the preferred attorney than in either of the other two comparisons was even larger: 57.1 percent for those with 10–19 patents and 61.0 percent for those with 20 or more patents. This analysis still does not eliminate the possibility that inventors tended to find their preferred attorneys early in their careers, just at the point where their rate of patenting was taking off. To test for this possibility, we regressed the change in patenting that occurred in the five years after the inventor met the preferred attorney compared to the five years before on the number of years that elapsed between the inventor's first patent and the first one filed by the preferred attorney. We found no statistically significant association, which suggests that inventors got a boost from hooking up with a preferred attorney regardless of career stage. We controlled for the year in which the inventor filed his or her first patent, whether the inventor was part of the 1910–11 (as opposed to 1890–91) sample, and the region in which the patentee resided. We also checked for an interaction between the year of the inventor's first patent and the number of years that elapsed before hooking up with a preferred attorney. Because none of the key variables were significant, we do not report the results.

⁸³ H. W. Boardman & Co., Hints to Inventors and Others Interested in Patent Matters (Boston, 1869), 13Google Scholar.

⁸⁴ On the weakness of bar associations, see Bloomfield, “Law.” As mentioned above, the Patent Office maintained a list of patent agents certified to practice before it, and small numbers of agents were from time to time disqualified.

⁸⁵ For a more formal analysis, see Wolinsky, Asher, “Competition in a Market for Informed Experts' Services,” Rand Journal of Economics 24 (Autumn 1993): 380–98CrossRefGoogle Scholar. See also Garmaise, Mark J. and Moskowitz, Tobias J., “Informal Financial Networks: Theory and Evidence,” Review of Financial Studies 16 (Winter 2003): 1007–40CrossRefGoogle Scholar.

⁸⁶ 16 May 1905, Van Winkle Diary.

⁸⁷ 16 and 17 May 1905, Van Winkle Diary.

⁸⁸ 31 July 1905, Van Winkle Diary.

⁸⁹ 31 Mar. 1905, Van Winkle Diary.

⁹⁰ 8 Aug. 1905, Van Winkle Diary.

⁹¹ 24 Jan.1905, Van Winkle Diary. See also 5 Mar.; and 7 and 12 June 1905.

⁹² 7 and 8 June 1905, Van Winkle Diary. One might wonder whether inventors' dependence on patent attorneys enabled the latter to earn supra-normal returns at the patentees' expense. However, the number of patent attorneys increased so dramatically in the late nineteenth century, both in absolute terms and relative to the size of the population, that it is likely that any excess returns would have been competed away. Certainly, no attorney ever attained a dominance in the last quarter of the century comparable to that of Munn & Company in the years immediately following the Civil War. The top attorney in our 1871 sample accounted for 5.5 percent of the patent assignments and the top four attorneys 13.5 percent. Those percentages dropped steadily until 1911, when the top attorney handled only 1.3 percent of the assignments and the top four only 4.6 percent. It would also be desirable to know whether patent attorneys benefitted patentees by securing them better prices for their inventions. Unfortunately, the data do not allow us to explore this possibility. By the late nineteenth century contracts typically specified that the assignment was “for one dollar and other consideration.”

⁹³ Nathan, John E., Fish & Neave: Leaders in the Law of Ideas (New York, 1997), 13, 19Google Scholar.

⁹⁴ In Memoriam: Samuel S. Fisher (Cincinnati, OH, 1874), 24Google Scholar.

⁹⁵ See the Wright, Brown, Quinby & May Correspondence Files for letters among patent agents in different cities. Virtually all agents also had regular dealings with at least one attorney in Washington, who could be called upon to conduct searches of patent records and perhaps represent them in preliminary interviews with patent examiners.

⁹⁶ Letter of 30 Apr. 1870 from Aug. H. [last name illegible] to Jenks, Box 3, Folder 59, Mss. 867, Lemuel P. Jenks records, Baker Library Historical Collections, Harvard Business School.

⁹⁷ Letter of 12 July 1877 from Roger Cartwright of London to Blatchford, Seward & Gris-wold, Box 5, Folder 243, Records of Blatchford, Seward & Griswold, 1841–1910, Archives and Special Collections, Massachusetts Institute of Technology.

⁹⁸ 27 Mar. 1905, 28 Apr., 9 May, 1 and 2 June 1905, Van Winkle Diary.

⁹⁹ On the growing importance of industrial R&D in the middle third of the twentieth century, see Lamoreaux, Naomi R., Sokoloff, Kenneth L., and Sutthiphisal, Dhanoos, “The Reorganization of Inventive Activity in the United States in the Early Twentieth Century,” in Understanding Long-Run Economic Growth: Geography, Institutions, and the Knowledge Economy, eds. Costa, Dora L. and Lamoreaux, Naomi R. (Chicago, 2011), 235–74CrossRefGoogle Scholar. This is not to say that independent inventors completely disappeared or that large firms stopped acquiring technology in the market. See Hintz, Eric S., “The Post-Heroic Generation: American Independent Inventors, 1900–1950,” unpublished PhD dissertation, University of Pennsylvania (2010)Google Scholar; Nicholas, Tom, “Spatial Diversity in Invention: Evidence from Early R&D Labs,” Journal of Economic Geography 9 (Jan. 2009): 1–31CrossRefGoogle Scholar; and Nicholas, Tom, “The Role of Independent Invention in US Technological Development, 1880–1930,” Journal of Economic History 70 (Mar. 2010): 57–82CrossRefGoogle Scholar. Over time, however, large firms increasingly relied on internally generated technology. See Mowery, “Boundaries of the US Firm.”

¹⁰⁰ Hicks, Dianaet al., “The Changing Composition of Innovative Activity in the US-A Portrait Based on Patent Analysis,” Research Policy 30 (Apr. 2001): 681–703CrossRefGoogle Scholar. The authors define small patentees as companies that are either outside the Tech-Line database or that had less than 25 patents in the five years ending in 1993. The Tech-Line database includes the companies most active in patenting, about 400 all told.

¹⁰¹ Arora, Ashish and Gambardella, Alfonso, “The Market for Technology,” in Handbook of the Economics of Innovation, eds. Hall, Bronwyn H. and Rosenberg, Nathan, vol. 1 (Amsterdam, 2010), 641–78CrossRefGoogle Scholar. The authors define large companies as those with more than 25,000 employees. See also the essays in Richard S. Rosenbloom and Spencer, William J., eds., Engines of Innovation: US Industrial Research at the End of an Era (Boston, 1996)Google Scholar.

¹⁰² Serrano, Carlos J., “The Dynamics of the Transfer and Renewal of Patents,” RAND Journal of Economics 41 (Winter 2010): 686–708CrossRefGoogle Scholar. The proportions of patents sold are all seven or eight percentage points higher if the patents are weighted by importance according to the number of subsequent patents that cited them.

¹⁰³ Since 1980, patent holders have had to pay renewal fees to keep their patents active for more than four years. Carlos Serrano, J., “Estimating the Gains from Trade in the Market for Innovation: Evidence from the Transfer of Patents,” NBER Working Paper 17304 (Aug. 2011)CrossRefGoogle Scholar.

¹⁰⁴ Zucker, Lynne G., Darby, Michael R., and Armstrong, Jeff S., “Geographically Localized Knowledge: Spillovers or Markets?” Economic Inquiry 36 (Jan. 1998): 65–86CrossRefGoogle Scholar; and Zucker, Lynne J., Darby, Michael R., and Brewer, Marilynn B., “Intellectual Human Capital and the Birth of US Biotechnology Enterprises,” American Economic Review 88 (Mar. 1998): 290–306Google Scholar. See also Galambos, Louis, with Sewell, Jane Eliot, Networks of Innovation: Vaccine Development at Merck, Sharp & Dohme, and Mulford, 1895–1995 (New York, 1995)Google Scholar; and Powell, Walter W., Koput, Kenneth W., and Smith-Doerr, Laurel, “Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology,” Administrative Science Quarterly 41 (Mar. 1996): 116–45CrossRefGoogle Scholar.

¹⁰⁵ For Silicon Valley, see Saxenian, AnnaLee, Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA, 1994)Google Scholar; Castilla, Emilio J., et al., “Social Networks in Silicon Valley,” in The Silicon Valley Edge: A Habitat for Innovation and Entre-preneurship, eds. Lee, Chong-Moon and Miller, William F. (Stanford, CA, 2000), 218–47Google Scholar; Kenney, Martin and Florida, Richard, “Venture Capital in Silicon Valley: Fueling New Firm Formation,” in Understanding Silicon Valley: The Anatomy of an Entrepreneurial Region, ed. Kenney, Martin (Stanford, CA, 2000), 98–123Google Scholar; Suchman, Mark C., “Dealmakers and Counselors: Law Firms as Intermediaries in the Development of Silicon Valley,” in Understanding Silicon Valley, 71–97Google Scholar; and Suchman, Mark C., “On the Advice of Counsel: Law Firms and Venture Capital Funds as Information Intermediaries in the Structuration of Silicon Valley,” unpublished PhD dissertation, Stanford University (1994)Google Scholar.

¹⁰⁶ On software patents, see Bessen, James and Meurer, Michael J., Patent Failure: How Judges, Bureaucrats, and Lawyers Put Innovators at Risk (Princeton, NJ, 2008), esp. ch. 9Google Scholar. Software and business methods have only recently been made patentable, and the kinds of problems they have created are analogous to those resulting from the recognition of design patents for useful changes of form in the late nineteenth century. See Magliocca, “Ornamental Design.”

¹⁰⁷ A search of mainstream publications on the word “troll” easily turns up hundreds, even thousands, of times more articles than comparable searches for the late nineteenth century.

¹⁰⁸ See Allison, John R., Lemley, Mark A., and Walker, Joshua, “Extreme Value or Trolls on Top? The Characteristics of the Most-Litigated Patents,” University of Pennsylvania Law Review 158 (Dec. 2009): 1–37Google Scholar; and Allison, John R., Lemley, Mark A., and Walker, Joshua, “Patent Quality and Settlement among Repeat Patent Litigants,” Georgetown Law Journal 99 (Mar. 2011): 677–712Google Scholar. The category of non-practicing entities includes inventor-owned enterprises, universities, and many other entities besides trolls. Allison and his co-authors compared frequently litigated patents (that is, those that are the subject of at least eight legal filings) to a random sample of patents that show up in the litigation records only once. Nearly three-fourths of the patents involved in frequent litigation were in computer software and business methods. The figure for litigation failures varies depending on whether default judgments are included. Information on the Stanford database is available at http://www.law.stanford.edu/node/149621, accessed 12 Oct. 2012.

¹⁰⁹ For descriptions of some of these entities, see Millien, Raymond and Laurie, Ron, “Meet the Middlemen,” Intellectual Asset Management Magazine 28 (Feb.-Mar. 2008): 53–58, www.iam-magazine.comGoogle Scholar.