October 6, ]91l] 



SCIENCE 



have been made, chiefly on the basis of extra- 

 polation from figures of past growth in com- 

 parison with past and present population of 

 other countries, and generally on the assump- 

 tion that the sources of life and habitability 

 are either unlimited or limited only by land 

 area. One of the latest and most compre- 

 hensive estimates is that of Henry Gannett, 

 geographer of the tenth, eleventh and twelfth 

 censuses;^ it was made without reference to 

 limitation of sources of life, but in the light 

 of the decreasing percentage increment of 

 population shown by records of this and other 

 countries. His figures for prospective in- 

 crease are essentially arbitrary, decreasing 

 from 21 per cent, during the decade 1900- 

 1910' to 5 per cent, for the decade 2090-2100, 

 giving populations of 90,000,000 in 1910, about 

 250,000,000 in 2000 and 500,000,000 a century 

 later. Thus far no comprehensive extrapola- 

 tions based on the thirteenth census appear to 

 have been made. 



Recent researches tend to indicate that the 

 assumption of unlimited resources, or of re- 

 sources limited only by land area, is unwar- 

 ranted; for while the mineral resources of the 

 United States are vast, while the forests are 

 renewable and the farms susceptible of large 

 increase in productivity, while the atmosphere 

 gives little threat of exhaustion (despite the 

 gloomy anticipations of Sir William Crookes 

 and others concerning the stock of nitrogen), 

 and while the available sun-power is thus far 

 used to but a small fraction of its capacity, a 

 practical limit to the productivity and habita- 

 bility of the country is fixed by limitation in 

 the water supply — and it is worth while to 

 consider prospective population in the light 

 of this limitation. 



Standards for the use of water in relation 



' Eepoi't of the National Conservation Commis- 

 sion (Sixtieth Congress, Second Session, Senate 

 Document 676), 1909, Vol. 2, pp. 7-9. 



- Perhaps through misprint, Gannett 's increment 

 for this decade does not correspond with the popu- 

 lation figures; it is put at 21 per cent. — the rate 

 subsequently determined by the Thirteenth Census 

 — though his estimate of 90,000,000 is only 19.2 

 per cent, above the 75,569,000 (or 18.4 per cent, 

 above the 76,000,000) appearing in his tables. 



to crop production and the maintenance of 

 human existence arise under irrigation in arid 

 regions, where water is measured more care- 

 fully and balanced more exactly against plant 

 and animal life than in humid lands. Here 

 25 acre-feet of water properly used in agricul- 

 ture or horticulture will sustain a family of 

 five for a year, with the requisite surplus pro- 

 duction for exchange; the best results follow 

 application of the water on five acres of land 

 to an aggregate of five feet in depth as needed 

 during the season. Using water in this way, 

 the rural population is one per acre, or 640 

 per square mile, stated in terms of land; but 

 it is justly measured only as one for each 

 5 acre-feet (6,800 tons) of that menstruum 

 which alone renders land productive. 



The standards fijsed in arid regions are not 

 greatly difl^erent from those arising of late in 

 humid lands. Hellriegel in Germany and 

 King in this country have shown that crop 

 plants require for their growth a quantity of 

 water, measured by transpiration, averaging 

 from 300 to 600 (with a mean of about 450) 

 times the weight of the plants after drying; 

 and common field experience indicates that in 

 addition to the moisture passing through the 

 plants the soil requires an even larger quantity 

 to maintain a texture suitable for crop growth 

 — ^much. of which passes away through evapo- 

 ration and seepage. On this basis " the agri- 

 cultural duty of water " in this country has 

 been formulated as the production of one- 

 thousandth part of its weight in average plant 

 crop." Reckoning human food and drink on 

 this basis, and assuming that meats require 

 (chiefly in the growth of plants used as feed 

 for the animals) ten times the quantity of 

 water represented in vegetal food, it appears 

 that the adult who eats 200 pounds each of 

 bread and beef in a year consumes something 

 like a ton of water in drink and the equiva- 

 lents of 400 tons in bread and 4,000 tons in 

 meat, or 4,401 tons in all — figures, correspond- 

 ing fairly with the results of intensive agri- 

 culture in arid districts. Accordingly, the 



" ' ' Yearbook of the Department of Agricul- 

 ture," 1910, pp. 169-176; Bureau of Soils Bulle- 

 tin 71, 1911, pp. 7-14. 



