WATER SOILS. 431 



insignificant except in the case of sodium nitrate which, applied at the 

 rate of 0.001 per cent, decreased the rate of absorption 30 per cent. 

 Moreover, in fertilizers as ordinarily applied it is probable that the salts 

 which increased the rate of absorption would neutralize the effect of 

 those which decrease it. The author is of the opinion that variations 

 in the surface tension of the soil water due to different salts " are too 

 small to affect very appreciably the soil's supply of moisture." The 

 arrangement of soil particles under different conditions and the factors 

 which affect it are discussed at some length. 



The water-holding capacity of the soil under the different conditions 

 was tested, but the differences observed were so small and irregular 

 that the results are not reported. 



The percolation of water through the soil was determined on untreated 

 soil and that which had received 0.5 per cent of muriate and nitrate of 

 potash, 0.75 per cent of lime (Ca< >) and 0.1 per cent of sodium carbon- 

 ate. The method used was as follows: 



"Fifty grams of soil was placed in tho soil tube (of 1% inches in diameter) satu- 

 rated with water, and a head of 1 in. of water added on top of tho soil, so geutly that 

 the water was not muddied, and the time required for 5 cc. to pass through was 

 noted. The next day 1 in. head of water was added again and the rate of percola- 

 tion noted, and so on until the rate became nearly constant." 



The results show that in untreated soil the rate of percolation 

 decreased gradually until it reached 9.0 cc. per hour on the fifth day. 

 In the soil to which 0.5 per cent of potassium chlorid and potassium 

 nitrate were added the rate of percolation was' reduced at first to one 

 twenty-fifth and one-thirtieth, finally to one twenty-fifth and one ninety- 

 sixth, respectively, of that through the soil alone. In the soil to which 

 lime was added the rate was more than doubled. "In the case of the 

 potassium salts the percolations slightly increased for a time and then 

 decreased until it was practically nil, when the observations ceased." 



Percolation through the soil to which lime had been added increased 

 for a time and then decreased until it became constant at 20.4 cc. per 

 hour. The indications were that these changes in the rate of percola- 

 tion were due to a change in the size of the soil aggregates. Salts 

 which showed the most marked effect in reducing the rate of percola- 

 tion appeared to puddle the soil. 



The effect of allowing water to remain longer than one hour in con- 

 tact with soil before the rate of percolation was determined was tested 

 on untreated soil and on soil to which had been added in addition to 

 the kinds and amounts of the salts noted above 0.3 per cent of nitrogen 

 as ammonium nitrate; lime (CaO) plus 1 per cent of nitrogen as nitrate 

 of soda; 0.1 per cent of Na..O as chlorid plus 0.1 per cent of acid car- 

 bonate of sodium; sodium carbonate plus calcium sulphate; and 1 

 molecule of sodium carbonate plus 2 molecules of calcium sulphate. 



"All the potassium salts used reduced the rate of percolation, the nitrate and 

 chlorid having the greatest effects; the nitrate, chlorid, sulphate, and calcium 

 hydroxid increased the rate of percolation, the other lime salts having little effect; 



