¹ This is the “revisionist” view of Canadian development. The older view held that 1860–1896 was a period of “secular depression,” but Firestone's GNP estimates [in N.B.E.R., Studies in Income and Wealth, XXIV (Princeton, 1960), p. 230]Google Scholar suggest higher per capita growth rates for 1867–1896 than for the 1867–1967 period as a whole and he concludes that an industrial revolution took place in Canada in the 1860's with the introduction of the factory system. Since agriculture was the largest sector of the economy and one place where the introduction of the factory system was occurring was the agricultural implement industry, this latter view would make the mechanization of agriculture an important aspect of Canada's economic development.

² Jones, R. L., A History of Agriculture in Ontario 1613–1880 (Toronto, 1946), p. 102Google Scholar.

³ For examples of the former, see quotations from the Canada Farmer in Phillips, W. G., The Agricultural Implement Industry in Canada: A Study in Competition(Toronto, 1956), p. 40Google Scholar, and Landon, F., “Some Effects of the American Civil War on Canadian Agriculture,” Agricultural History, 7, p. 168Google Scholar. The latter reports are contained in Ontario Sessional Papers.

⁴ David, P., “The Mechanization of Reaping in the Ante-Bellum Midwest,” in Rosovsky, H. (ed.), Industrialization in Two Systems: Essay's in Honor of Alexander Geschenkron (New York, 1966), pp. 3–28Google Scholar; and David, P., “The Landscape and the Machine: Technical Interrelatedness, Land Tenure and the Mechanization of the Corn Harvest in Victorian Britain,” in McCloskey, D. N. (ed.), Essays on a Mature Economy: Britain after 1840 (London, 1971), pp. 145–205Google Scholar.

⁵ To be precise we should say “threshold small grain acreage.” Throughout this paper farm size is measured by acreage under small grains unless otherwise stated.

⁶ For fuller treatment of this question see R. W. T. Pomfret, “The Introduction of the Mechanical Reaper in Canada 1850–70” (Ph.D. dissertation, Simon Fraser University, 1974). The estimates of premechanized labor requirements are for the northeastern United States [Parker, W. N. and Klein, J. L., “Productivity Growth in Grain Production in the United States 1840–60 and 1900–10,” NBER, Studies in Income and Wealth, 30 (Princeton, 1966), pp. 523–80]Google Scholar. The data on grain prices are from Taylor, K. W. and Mitchell, H., Statistical Contributions to Canadian Economic History, II (Toronto, 1931)Google Scholar.

⁷ The assumption of indivisibility in David's Midwest study has recently come under criticism in Olmstead, A., “The Mechanization of Reaping and Mowing in American Agriculture, 1883–1870,” Journal of Economic History, 35 (June 1975); 327–52CrossRefGoogle Scholar. Neither, the present author's own research, nor discussion with other Canadian historians has brought to light documentary evidence of contracting or sharing of reapers in Canada comparable to the evidence which Olmstead provides for the Midwest. A tentative explanation of the lack of cooperative purchases is given in Pomfret, “The Introduction of the Mechanical Reaper,” ch. iii. Samples of the spatial distribution of farms in three Ontario counties indicated that potential cooperators' farms were seldom adjacent, which, given the cumbersome nature of the early reapers, suggested high costs of moving reapers between cooperators' farms. Olmstead specifically rejects the argument that transport costs were high in the American Midwest (Olmstead, “The Mechanization,” p. 339), but it should be borne in mind that geographical conditions in Ontario were more irregular than in the Midwest and hence less amenable to the movement of machinery (Pomfret, “The Introduction of the Mechanical Reaper,” pp. 33–5).

⁸ The final assumption is heroic, but is not crucial. Changing the parameters of the distribution does not alter our conclusions regarding the pattern or the causes of reaper diffusion (the more reasonable assumption that small grain acreage was more concentrated than total acreage would even make our 1870 estimate closer to the census figure). What is crucial is the assumption of lognormality, or more generally the assumption of skewness, since a symmetrical distribution would have produced different results (Pomfret, “The Introduction of the Mechanical Reaper,” ch. vi).

⁹ Aitchison, J. and Brown, J. A. C., The Lognormal Distribution (Cambridge, Eng., 1963), p. 10Google Scholar.

¹⁰ This is especially true of the rate of interest. The rate used is 6 percent (see Appendix), but there is no Canadian evidence to support this. Even a small adjustment like the use of 5 percent as the relevant rate of interest would raise our estimate above the census value.

¹¹ An exception to the general tendency to ignore improvements is Rosenberg, N., “Factors Affecting the Diffusion of Technology,” Explorations in Economic History, 10 (Jan. 1973), 3–33CrossRefGoogle Scholar. Olmstead has recently ascribed a major role to technological improvements in explaining the diffusion of reapers in the American Midwest (Olmstead, “The Mechanization,” pp. 344–52).

¹² Although this contention is contrary to the usual emphasis in economic theory, it has a long history going back to Adam Smith; see Young, A., “Increasing Returns and Economic Progress,” Economic Journal (1928), pp. 529–42Google Scholar; and Kaldor, N., “The Irrelevance of Equilibrium Economics,” Economic Journal (1972), pp. 1237–55.Google Scholar

¹³ See note 1.

¹⁴ For a more detailed examination of the data and for consideration of some problems omitted here see Pomfret, “The Introduction of the Mechanical Reaper,” ch. v.

¹⁵ Hutchinson, W. T., Cyrus Hall McCormick, 1 (New York, 1930), p. 336Google Scholar.

¹⁶ Rogin, L., The Introduction of Farm Machinery (Berkeley, 1931), pp. 134–5.Google Scholar

¹⁷ Ibid., p. 128.

¹⁸ Denison, M., Harvest Triumphant (Toronto, 1948), p. 32Google Scholar.

¹⁹ Hutchinson, W. T., Cyrus Hall McCormick, 2 (New York, 1935), p. 647Google Scholar.

²⁰ Denison, Harvest, p. 50.

²¹ Ibid., p. 64.

²² Phillips, The Agricultural Implement Industry, pp. 10, 40.

²³ Hutchinson, Cyrus Hall McCormick, 1, p. 311.

²⁴ sIbid., p. 73.

²⁵ Ibid., p. 365. Olmstead, “The Mechanization,” pp. 331–2, argues that “the useful life of a reaper or mower was typically closer to five years or less” before the 1870's. An increased value of d for 1850 and 1860 would increase the threshold farm size and reduce the estimated number of reapers in these years. It would not, however, alter the shape of the projected diffusion path, nor any of the other conclusions based on Table 3.

²⁶ Rogin, The Introduction, p. 95.

²⁷ Hutchinson, Cyrus Hall McCormick, 1, pp. 337, 369.

²⁸ Innis, H. A. and Lower, A. R. M., Select Documents in Canadian Economic History, 1783–1885 (Toronto, 1933), p. 809Google Scholar.

²⁹ Denison, Harvest, p. 59.

³⁰ As quoted in Denison, Harvest, p. 59.

³¹ Cf. Jones, A History, p. 96; and Phillips, The Agricultural Implement Industry, p. 40.

³² Jones, A History, p. 55n.