SOILS — FERTILIZERS. 815 



" In one case, and possibly in others also, injury to many trees followed the 

 application of water, probably due to the washing of the nitrates down into 

 the soil. . . . 



"The ordinary white alkali that commonly occurs in these sections is, for 

 the most part, free from nitrates, but the brown and white areas frequently 

 overlap. 



" The ground waters, unless derived from niter areas, are free from nitrates 

 and these nitrates cannot be accounted for by the evaporation of such water 

 from the surface of these areas. 



" In some sections pi-actically the whole of the irrigated land has been 

 affected, thus eliminating the question of transportation except as to the river 

 water used for irrigation. Our river waters carry from 0.00 to 0.4 part per 

 million of nitrogen as nitrates. 



"The localization of these areas and the rate of increase also preclude the 

 theory of transportation and concentration. 



"Excessive irrigation for the purpose of washing out the 'black alkali,' 

 nitrates, has not been successful, but this failure has probably been due to the 

 method used. 



" The liberal application of manure has not given permanent, if indeed, any 

 relief. 



"Thorough and frequent cultivation has not been followed by the beneficial 

 results expected. 



"The soils of Colorado are, generally speaking, poor in nitrogen, but the 

 ratio of nitric nitrogen to the total nitrogen is fi-equently very high, 17 to 50 

 per cent being not uncommon. 



" The nitrogen in these soils is fixed by Azotobacter which use the nitrogen 

 in the air to build up their tissues. 



" The fixation of nitrogen, in a sample of ordinary soil from the college farm, 

 collected December 12, 1910, on incubation for 27 days was found to have taken 

 place at the rate of 5.222 lbs. of nitrogen, equal to 17.5 tons of proteids per 

 acre-foot of soil per annum. 



" The nitric nitx'O^en present in this soil at the beginning of the experiment 

 corresponded to 840 lbs. of sodic nitrate per acre-foot of soil ; this had increased 

 in 48 days to 1,999 lbs. as a maximum, a gain of 1,159 lbs. which would give, 

 if this rate were continued for a year of 360 days, a gain of 8,676 lbs., or 4^ 

 tons of nitrates per acre-foot of soil. 



" The incubated samples, with but one exception, showed a darkening of their 

 surfaces. 



" No addition of anything except boiled, distilled water was made to these 

 samples before or during incubation. A large bottle was partially filled with 

 some, approximately 8 lbs., of the original sample just as it was collected. 

 This bottle was loosely stoppered, inverted and kept in a room where the tem- 

 perature was fairly high and even. This soil was analyzed just as the incu- 

 bated samples were and showed a decided increase in both the total and nitric 

 nitrogen, 7.45 and 19.15 per cent respectively. 



" Fixation takes place rapidly in this soil in the presence of from 13.5 to 20 

 per cent of water. The rate of fixation of nitrogen obtained is sufiicient to 

 account for the nitrates found in the soil provided that it is nitrified. 



" The rate of nitrification obtained is sufficient to account for the formation 

 of the nitrates found, in most cases if not all of them." 



Bacteriological studies of the fixation of nitrogen in certain Colorado soils, 

 W. G. Sackett (Colorado 8ta. Bui. 119, pp. 3-42, pis. 2, figs. 5).— With a view 

 to determine the source of the excessive amounts of nitrates in certain Colo- 



