496 



SCIENCE. 



[N. S. Vol. XXII. No. 564. 



succeeding in getting them there in the form 

 of untruths, or of partial truths which lead 

 to errors of practice so soon as they are ap- 

 plied. Lamentable examples of these are fur- 

 nished in ' Bulletin 22 ' and in the extended 

 press promulgation regarding what may be 

 expected through bacterial inoculation of the 

 soil. Much expense was incurred in conduct- 

 ing the investigations referred to and in 

 getting them before the public; very much 

 more is being incurred by those who are giving 

 them practical trial; but by far the greatest 

 expense will accrue during the time required 

 to outgrow the disappointment and the smart 

 of defeat, of wasted effort. It is this condi- 

 tion of things, more than ignorance and more 

 than conservatism, which maintains with ter- 

 rible effectiveness, as a brake on agricultural 

 progress, the dogma, ' Book farming don't pay.' 

 In calling attention to the results of the au- 

 thor's investigations it is important to point 

 out that rates of transpiration, as measured in 

 the laboratory trials, are quite inapplicable 

 for use in giving a measure of the rate of flow 

 of air through soils under field conditions. It 

 must be noted that in preparing the soils for 

 the measurements of rates of transpiration 

 they were ' first broken up fine in a mill.' This 

 condition is very wide from what is found in 

 the field and represents more nearly a puddled 

 soil which is always a condition of sterility, 

 and we believe that one of the chief causes of 

 this sterility is the inadequate aeration possible 

 under such conditions. In considering the re- 

 sults of the author therefore it must be borne 

 in mind that he has measured the rates of trans- 

 piration — and of diffusion also— through a thin 

 layer whose field structure had previously been 

 altered by what may be designated dry-pud- 

 dling. In illustration of the effect of dry- 

 puddling we shall cite two series of observa- 

 tions made by Mr. Nelson and Mr. Hogenson, 

 under our direction, while connected with the 

 Bureau of Soils. They are taken from the 

 records in the office of the bureau, which con- 

 tain several hundreds of measurements cover- 

 ing many types of soil which have been ex- 

 amined as to the rates of transpiration in the 

 first, second, third and fourth feet. In the 

 imrticular cases cited we select two of the 



soils which were under investigation in 1903' 

 upon which corn and potatoes were grown, as 

 reported in Bulletin 26 of the Bureau of Soils. 

 The rates of transpiration through these soils 

 were measured under five different conditions, 

 as indicated in the table. By ' field condition ' 

 is to be understood the granulation into which 

 the soil falls naturally when plowed in good 

 condition of moisture, but using only such 

 portions of it as readily pass a one millimeter 

 screen in the air dry condition and without 

 rubbing. This soil was firmly packed in the 

 transpiration tube and the rate of flow of air 

 through it measured, after which it was re- 

 turned to a mortar and pulverized by gently 

 working it under a rubber pestle. When in 

 this condition the transpiration was again 

 measured, after which it was pestled a second 

 time, the process being repeated until the rates 

 of transpiration were obtained for the flve 

 different conditions. 



MEAN RELATIVE BATES OF FLOW OF AIR THROUGH 

 AIR DRY SOILS MORE OR LESS FINELY PULVERIZED. 



Norfolk Sandy Soil. Janesville Loim. 



Pore Seconds. Pore S conds. 



Space. Space. 



Field condition, 37.0 69 51.6 83 



Pestled once, 31.8 1,050 48.5 " 600 



Pestled twice, 29.9 1,724 48.5 800 



Pestled three times, 29.1 2,025 47.9 1,200 



Pestled four times, 28.5 2,550 46.8 1,350 



It is clear from this table that a very pro- 

 found change in the permeability of the two 

 soils has been effected by the dry-puddling; 

 the rate of flow of air through the Norfolk 

 sand being finally reduced to only about one 

 fortieth of what it was at first, and that of 

 the elanesville loam to about one seventeenth. 

 It will be observed also, if computations are 

 made, that the rates have not varied as the 

 sixth or seventh power of the porosity. 



Under undisturbed field conditions the rate 

 of transpiration would in all probability be 

 very different from what is given in the first 

 line of the table and for the surface foot; in 

 air dry condition it is quite certain to be larger- 

 than there found. In actual field conditions 

 the body of the soil is ramified by channels 

 and passageways which are often larger than 

 capillary and through which the a,ir moves. 



