STATE HOETICULTUEAL SOCIETY. 113 



of our figures may not be actual av^erages, and due allowance must 

 be made for this; bat I tliink we can now say that the four hundred 

 barrels per acre each day is not an over statement. There certainly 

 is a much greater quantity lost by evaporation from naked soil. An 

 average of experiments gave Dr. Andre,* of Philadelphia, J;he ratio 

 of six to one in favor of the soil. Mr. T. F. Hunt, of the University 

 of Illinois, found that about oue-tenth of an inch per day escaped 

 from moist soil during clear weather. (See ad-interim report Central 

 Illinois, page 102. ) This last is twice as great as that given above for 

 forest trees, and the comparative effects of drouth in forests and in 

 the open ground shows that some such proportion must prevail. 

 Where trees are set at a considerable distance apart, as in orchards, 

 we may have, however, nearly the normal evaporation from the soil, 

 plus that given off by the trees; for in this case the ground is not 

 much shaded, and there is not likely to be much water-holding 

 mulch accumulated. 



Tarning now to the investigations of Mr. Hunt, as given on 

 page 102 (ad-interim report Central Illinois), we may be able to gain 

 further lessons of much importance. The process consisted as stated, 

 in place cited, of taking specimens of soil two feet deep and finding 

 out the amount of water contained in them by weighing them, then 

 evaporating in a steam chest and weighing again. The per cent, of 

 loss was calculated upon the dry weight. The specimens were col- 

 lected between the 1st and 19th of August, when the pastures were 

 mostly brown and bare, when corn was severely suffering for water, 

 and when many trees, especially those transplanted within two or 

 three years, were dying from drouth, and when fires were catching 

 from locomotive and other sparks and running wild over lots and 

 fields. Still the water in this drouthy soil averaged 13.2 per cent, of 

 the dry weight ! When we assert that this proportion indicates for 

 the water in the two feet of soil, an amount equal to a surface layer 

 four inches deep, or 2,750 barrels of forty gallons each per acre, some 

 notion of the enormous aggregate quantity can be gained. It is cer- 

 tainly a surprising result, and one well worth our careful attention 

 and consideration. The amount of water in this two feet of very 

 dry soil is equal to the average rainfall for a summer month, and, 

 what is more significant, equal to the amount, according to the fig- 

 ures above, given off by rank vegetation during four-fifths of the 

 growing season ! 



If these things are so, or nearly so, surely there is something 

 else for us to think about, in connection with summer drouths, than 

 the simple amount of rainfall. A saturated soil holds varioiis pro- 

 portions of water, according to its physical and chemical constitu- 

 tion. As stated in the tests by Mr. Hunt, four of the dried samples 

 were wetted to saturation, and held, after all drainage ceased, 49.3 



* American Naturalist, Yol. XVI, p. 19. 

 9 



