¹ See Briggs, H. E., Frontiers of the Northwest (New York: Appleton-Century Co., 1940)Google Scholar. and Studness, C. M., ,“Economic Opportunity and the Westward Migration of Canadians During the Late Nineteenth Century”, Canadian Journal of Economics and Political Science (November 1964)CrossRefGoogle Scholar.

² Mackintosh, W. A., Economic Problems of the Prairie Provinces (Toronto: The Macmillan Company of Canada, 1935), p. 282Google Scholar.

³ See Mackintosh, W. A., Prairie Settlement: The Geographical Setting (Toronto: The Macmillan Company of Canada, 1934), pp. 62–68, Figures 43–49Google Scholar. Agricultural expansion after 1911 consisted largely of filling in areas already thinly settled. The only new regions opened up were the northern fringes of the Park Belt and the Peace River Country.

⁴ See Easterbrook, W. T. and Aitken, H. G. J., Canadian Economic History (Toronto: The Macmillan Company of Canada, 1965)Google Scholar; Mackintosh, W. A., The Economic Background of Dominion Provincial Relations (Appendix III of the Royal Commission on Dominion Provincial Relations, Ottowa, 1939)Google Scholar; Hedges, J. B., Building the Canadian West (New York: Macmillan Company, 1939)Google Scholar.

⁵ The price of “Good Red Foreign Wheat” at Liverpool averaged $1.54 in the decade 1871–1880, $1.15 in the 1880's, $0.90 in the nineties and $1.00 in the period 1901–1910. Mackintosh, Economic Problems, p. 283, Table IV.

⁶ Fowke, V. C., Canadian Agricultural Policy (Toronto: University of Toronto Press, 1946)Google Scholar; Dales, J. H., The Protective Tariff in Canada's Development (Toronto: University of Toronto Press, 1965)Google Scholar.

⁷ For capital flows see G. Meier, “Economic Development and the Transfer Mechanism: Canada, 1895–1913,” Canadian Journal of Economics and Political Science (February 1953). For labor migration see D. C. Corbett, “Immigration and Economic Development,” Canadian Journal of Economics and Political Science (February 1953). See also Hartland, P., “Factors in Economic Growth in Canada,” Journal of Economic History, XV (March 1955)Google Scholar.

⁸ Bracken, J., Dry Farming in Western Canada (Winnipeg: Grain Growers' Guide, 1921), pp. 4–8Google Scholar.

⁹ Buckley, K. A. H., Capital Formation in Canada (Toronto: University of Toronto Press, 1955), p. 15Google Scholar.

¹⁰ Historical Statistics of the U.S., Colonial Times to 1957 (Washington: U.S.G.P.O., 1960), p. 237Google Scholar; Hartland, “Factors.”

¹¹ Studness, “Economic Opportunity.” For a critical evaluation of his calculations, see K. H. Norrie, “Economic Opportunity and the Westward Migration of Canadians During the Late Nineteenth Century: A Comment,” Canadian Journal of Economics (February 1974).

¹² The contract with the CPR in 1881 gave the company, among other things, 25 million acres of land. The grant was to be taken in alternate sections on each side of the main line. A novel provision was that the land had to be “fairly fit for settlement.” If the company did not consider land along the trunk line to be suitable it could be replaced by a similar grant elsewhere. By careful selection then the railway early on controlled considerable acreage in southwestern Manitoba and southeastern Saskatchewan which was just at the margin of settlement. Studness is arguing that ownership of the even-numbered sections of every township in this prime area deterred settlement since “… it appears that Canadian settlers occupied the good homestead land that was within a reasonable distance of railroads, and then turned to homestead land in northern North Dakota rather than purchase the more productive Canadian land from the CPR or Hudson's Bay Company.” Studness, “Economic Opportunity,” 582. The latter institution, in surrendering its exclusive claims to the area granted by royal charter in 1670, received land grants totaling one-twentieth of the fertile belt.

¹³ Studness, “Economic Opportunity,” 583, ff. 21.

¹⁴ The introduction to the paper summarizes the results as follows:

(1) The problems of overcoming the physical environment had definitely been solved by the mid 1880's;

(2) … the principal reason for lack of settlement after the mid 1880's was a demand for the region's major potential export which was insufficient to ensure its profitable exploitation; and

(3) that development after 1897 was primarily due to an increase in the profitability of producing that export. …

See Stabler, J. C., “Factors Affecting the Development of A New Region: The Canadian Great Plains, 1870–1897,” Annals of Regional Science (June 1973), 75–76Google Scholar.

¹⁵ Johnston, J., Econometric Methods (New York: McGraw-Hill, 1972), p. 301Google Scholar.

¹⁶ Easterbrook and Aitken, Canadian Economic History, pp. 282–286.

¹⁷ The normal assumption of a first-order auto-regressive scheme was made. Thus u_t = ϱu_t–1 + h_t where h_t is not autocorrelated. See, for example, F. Fisher and P. Temin, “Regional Specialization and the Supply of Wheat in the United States, 1867–1914,” Review of Economics and Statistics (1970), 140–141.

¹⁸ The data are from Mackintosh, Economic Problems, Tables III and IV, pp. 282–283. The series for wheat prices is not consistent since the figures to 1889 are for No. 2 White Wheat at Toronto while those after 1889 are for No. 1 Northern at Winnipeg. The direction of bias is not clear since the Toronto price will exceed the Winnipeg price by an amount equal to the transportation charges between the two points, but No. 1 Northern Wheat will bear a premium over No. 2 White. If a statistically significant break had been found for the period 1879 to 1896 there would have been the problem of deciding whether to ascribe it to this defect in the time series for wheat prices or to more fundamental economic reasons. But since the regression equation over the period did not pick up a structural change, the biases must be largely offsetting.

¹⁹ For twenty-four degrees of freedom the required value of t for significance at the 95% level is 1.711.

²⁰ Hargreaves, M. W. M., Dry Farming in the Northern Great Flains, 1900–1925 (Cambridge: Harvard University Press, 1957), p. 3Google Scholar. Other useful historical references to dry farming are Bracken, J., Dry Farming in Western Canada, and J. A. Widtsoe, Dry Farming (New York: Macirullan, 1913)Google Scholar.

²¹ Employing Widtsoe's classification, areas receiving less than ten inches of precipitation are arid; those receiving between ten and twenty inches are semi-arid; those receiving between twenty and thirty inches are sub-humid; and those with over thirty inches are humid (p. 24). In semi-arid areas dry farming is “indispensable” (p. 1).

²² The expected yield of semi-arid land farmed in the conventional manner can be written as:

where yield, u, is a function of rainfall, X, and f(x) is the probability density function of rainfall. The expected yield of the same piece of semi-arid land under summerfallowing is:

where v(x) is the new functional relationship between yields and rainfall and p(0 ≤ p ≤ 1) is the proportion of the land cropped. The latter term enters since the portion of the land left fallow (1 – p) will nave a zero yield. The contention is that the gains to be made in higher yields from planting on land left fallow the previous crop year more than compensates for the necessity of leaving a portion of the land unproductive. The universal use of summerfafiowing in the semi-arid Canadian Prairies after 1900 would seem to confirm this. Obviously p becomes a choice variable reflecting the trade-off between foregone production and more certain yields. The more arid the region or the more uncertain the rainfall, meaning the greater the probability of a poor crop sown on stubble, the higher will be the value of p for given wheat prices.

²³ The differences are illustrated by the diversities of opinion among area experts reported in Bracken, Dry Farming in Western Canada, Chapter XVI.

²⁴ If the expected yield of sub-humid land is written as:

and that for the same average of semi-arid land under summerfallowing as in footnote 22:

then it is being argued that for contiguous areas such as then under discussion there would be no significant difference in v(x) versus w(x) or, even if there were, it would generally be insufficient to offset the foregone production in the latter case.

²⁵ The terms are from Webb, W. P., The Great Plains (Boston: Ginn and Company, 1931), pp. 3–9Google Scholar.

²⁶ Ibid., p. 3. Webb puts the eastern boundary at the 98th meridian, while G. C. Fite places it at the 100th. See Fite, G. C., The Farmer's Frontier, 1865–1900 (New York: Holt, Rinehart and Winston, 1966), p. 3Google Scholar.

²⁷ See United States Bureau of the Census, Census of the United States, 1890, Vol. I and Census of the United States, 1920, Vol. VI, part I.

²⁸ This paragraph is based on Fite, Frontier, Chapter 7.

²⁹ Mackintosh, Prairie Settlement, pp. 48–49, Figures 32–34. Ibid., pp. 86–89.

³⁰ Calculated from Canada Bureau of Statistics, Census of Canada, 1921, Vol. V.

³¹ Mackintosh, Prairie Settlement, p. 90.

³² Vernon Fowke, Canadian Agricultural Policy, pp. 236–237, has argued that an earlier discovery of Marquis would have hastened western settlement “only slightly.” The analysis here suggests modifying this negative viewpoint somewhat. More adequate rainfall and a slower evaporation rate meant that the more northern Park Belt regions were nearly comparable in moisture conditions to the Red River Valley area. Thus if Marquis had been available in the 1870's it is possible these northern areas would have been settled in the period after 1879 along with southern Manitoba and the eastern Dakotas. The extension onto the semi-arid prairies, and thus the bulk of Canadian settlement, would still have had to await the end of the sub-humid lands, but the relative sizes of settlement would not have been so skewed in favor of the U.S.

³³ In contrast the Canadian semi-arid lands did not receive this stretch of favorable rainfall. In the period from 1883 to 1890 only 1888 was very favorable moisture-wise, with 1887 being “fairly good.” See Morton, A. S. and Martin, C., History of Prairie Settlement and Dominion Lands Policy (Toronto: The Macmillan Company of Canada, 1938), pp. 65–95Google Scholar. Thus the Canadian settlement phase from 1879 to 1886 or 1887 lacked this large scale premature expansion onto semi-arid land, accounting in part for its relatively much smaller scale.

³⁴ Studness, “Economic Opportunity,” 580–581.