¹ For a complete elaboration of the traditional view, see Bhatia, B. M., Famines in India (2nd ed New York, 1967)Google Scholar or Dutt, R. C., The Economic History of India, vol. 2 (1903Google Scholar; rpt., Delhi, 1963).

² I am uncomfortable with the view that the late nineteenth century was the worst period of famines in the history of South Asia. Our information for all earlier periods is much less complete and, in any case, the British viewed food crises much differently than did their native predecessors. Scattered bits of evidence suggest that some other periods may have had comparable frequency of famines. Moreland lists nine famines between 1614 and 1660 but notes that his list is incomplete both because some of the records from the period had been destroyed and because the Englishmen who kept the records were not in touch with all parts of India. Moreland, W. H., From Akbar to Aurengzeb (Delhi, 1962), pp. 205–19Google Scholar.

³ Bombay Presidency, Statistical Abstract of Bombay Presidency (3rd ed., Bombay, 1922), App. 33-35Google Scholar. Throughout this paper information on only 15 of Bombay Presidency's 19 districts is used. The remaining four districts have been excluded because their agricultural statistics are almost unusable and because, being on the coast, they receive in excess of 90 inches of rain a year.

⁴ For the most thorough study of rainfall and “goodness” of season, see Mann, H. H., Rainfall and Famine: A Study of Rainfall in the Bombay Deccan (Bombay, 1955)Google Scholar. Mann found correlations between effective rainfall and yields ranging from 0.262 to 0.482 for four districts. His work suggests there may be some more complex relationship between the distribution of rainfall through the season and crop yields!

⁵ Heston, Alan W., “Official Yields Per Acre in India, 1886-1947: Some Questions of Interpretation,” Indian Economic and Social History Review, 10 (12 1973), 322CrossRefGoogle Scholar. Heston's index, averaged over 15 districts is:

The number of years in which famines were declared in more than 3 districts of the Presidency in the first four of these periods are, respectively, 1, 4, 2, 2 (through 1921 only).

⁶ Three slight famines, 1891-92, 1913-14, and 1915-16, have been omitted.

⁷ McAlpin, Subject to Famine (in progress), Table 6.4.

⁸ It is unclear precisely which of several possible decision-making units (families, lineages, castes, villages) in fact made or affected the decision I wish to discuss. The individual farm family seems a reasonable decision-making unit with which to begin. Until more research on local power structures and cooperative work groups has been done, we lack the information to consider the other units that may have been important actors or constraints in the choice of insurance mechanisms. Even when we know more about the locus of decision making, we will still have to figure out whose survival was of paramount importance to any decision makers. I have assumed that any unit is concerned with its own survival only.

⁹ Marvin Harris has argued that Hindu prohibition on cattle slaughter is basically a way of imposing some order between goals one and two. Since cattle can be food, but since they are also essential for agricultural operations, the culture forbids most of the people to eat cattle under any circumstances. See Marvin Harris, Cows, Pigs, Wars, and Witches (Glasgow, 1977), pp. 14-30. Reports from the nineteenth and early twentieth centuries suggest that preservation of cattle ranked for agriculturists almost as high as their own survival-and perhaps above that of some members of their families. See Bombay Presidency, Proceedings, Famine Department, vol. 5989. no. 3747 of 1900, p. 6901Google Scholar.

¹⁰ See Elphinstone's, Lieut. P. A. report on agriculture in part of Khandesh in Bombay Presidency, “Papers Relative to the Introduction of Revised Rates of Assessment into Eight Talookas and Two Pettas of Khandesh Collectorate,” Bombay Selections, no. 93 (Bombay, 1865), p. 131Google Scholar, for the propor tions of some mixtures common in the late 1850s. See Mann, Harold H. and Kanitkar, N. V., Land and Labour in a Deccan Village (Bombay, 1921), p. 71Google Scholar, for mixtures current in Poona district in the twentieth century and for their analyses of the reasons for planting mixtures of crops. See Bombay Presidency, Annual Report of the Director of Agriculture, 1884-85 (Bombay, 1886), p. 13Google Scholar, for tions on how fields planted with mixed crops were to be counted for the preparation of agricultural statistics.

¹¹ Bombay Presidency, Statistical Report on the Pergunnahs of Padshapoor, Belgam, and Chandgurh, Khanapur, Bagulkot and Badamy, and Hoondgood in the Southern Mahratta by Marshall, Thomas (Bombay, 1822), p. 123Google Scholar.

¹² McCloskey, Donald N., “English Open Fields as Behavior Towards Risk,” in Uselding, Paul, ed., Research in Economic History: An Annual Compilation ofResearch, vol. 1 (Greenwich, Conn., 1976)Google Scholar.

¹³ India, Famine Commission 1898, Report (and Appendicies) of the Indian Famine Commission, 1898 (Calcutta, 1898)Google Scholar, “Evidence of Witnesses from the Bombay Presidency Taken before the Indian Famine Commission, 1898,” p. 6. See McAlpin, Michelle Burge, “Railroads, Prices, and Peasant Rationality: India, 1860-1900,” this JOURNAL, 34 (09 1974), 662–84Google Scholar, for evidence on the prevalence of grain storage.

¹⁴ For some of the gestalt of these administrators, see Stokes, Eric, The English Utilitarians and India (Oxford, 1959), ch.2Google Scholar. While all intervention in the economy was regrettable, types that occurred only once and that might yield revenues adequate to pay for themselves (or might reduce the costs of other operations) were viewed as preferable to those that were likely to have to be repeated often and that could not be administratively defined as “productive.”

¹⁵ In western India, however, irrigation possibilities appear to have been very limited with nineteenth-century technology. Rivers in the SMC and the Deccan were few, and many of them fell so slowly as to make canal irrigation impractical. That is, the only lands that could be commanded by such canals were those already in the river bottoms and not in need of irrigation. In most parts of the SMC and the Deccan ground water was too far down for successful construction and operation of wells on a regular basis. See Bombay Presidency, “A Paper on Irrigation in the Deccan and Southern Maratha Country,” by Colonel, Playfair R. E., Superintending Engineer for Irrigation, in Bombay Selections, Irrigation Series No. 5 (Bombay, 1866)Google Scholar; and Bombay Presidency, A Statement and Remarks Relating to the Expenses of Irrigation from Wells in the Deccan, Khandesh (Bombay, 1856)Google Scholar.

¹⁶ McAlpin, Subject to Famine, Table 5.4.

¹⁷ The discussion of famine relief policy is drawn from Bombay Presidency, Famine Relief Code (Bombay, 1885)Google Scholar, and from discussions in Proceedings and Famine Reports during and after successive famines.

¹⁸ “Circular from Collector of Ahmednagar to All Patil Kulkarnis,” in Bombay Presidency, Proceedings, Famine Department, vol. 5985, p. 331Google Scholar.

¹⁹ McAlpin, Michelle Burge, “Price Movements and Fluctuations in Economic Activity, 1860-1947,” in Kumar, Dharma, ed., Cambridge Economic History of India, vol. II (forthcoming)Google Scholar.

²⁰ Bombay Presidency, Report on the Famine in Bombay Presidency in 1912-13 (Bombay, 1914), p. 6Google Scholar.

²¹ Government of Bombay, Labour Office, Report on an Enquiry into Agricultural Wages in the Bombay Presidency, by Findlay-Shirras, G., Director of the Labour Office (Bombay, 1924), pp. 64–66Google Scholar.

²² Bombay Presidency, Report on the Famine in Bombay Presidency in 1905-06 (Bombay, 1907), p. 11Google Scholar.