432 EXPERIMENT STATION RECORD. 



the magnesium .salts had little effect, except the nitrate, which increased the per- 

 colation; the ammonium salts used decreased the rate of percolation much more 

 than the potassium salts; the sodium carbonate and chlorid, and especially the. 

 nitrate, had a remarkable effect in reducing the rate of percolation. One-tenth per 

 cent of soda as sodium chlorid reduced the rate of percolation- to one-tenth, and 0.1 

 per cent of nitrogen as sodium nitrate reduced it to one-twentieth of the rate through 

 the soil alone. . . . 



"It is believed that the quantities of these salts which it is practical to add to 

 the soil would be sufficient to make a great difference in the escape of the water ol 

 the soil by percolation, and therefore in the amount of water which a soil so treated 

 can maintain for the growth of a crop. 



" The soil in which there was a great reduction in the rate of percolation, due to 

 the presence of certain salts, in each case had the appearance of being puddled, 

 which it undoubtedly was. Slaked lime had the greatest effect in increasing the 

 rate of percolation, which agrees w T ith the common experience of the farmer that 

 lime makes the soil more open and porous. Molecular quantities of lime added to 

 the soil to which sodium nitrate (0.1 per cent nitrogen) had been added did not 

 restore its permeability, nor did I! times the molecular quantity do so. Nine times 

 the quantity increased the permeability to about one-third more than that of the 

 soil alone. . . 



"After determining the rate of percolation in a Bample it was allowed to freeze in 

 the soil tube, and after thawing the vacant space all around between the soil and 

 the soil tube was tilled in with melted paraffin. The rate of percolation was then 

 determined and was found to be a little more than double that of the soil below 

 freezing. After this experiment enough sodium carbonate was added to the water 

 remaining on top of the soil to give 0.5 per cent sodium oxid, and the next day the 

 rate of percolation was determined and was found to lie reduced to the extremely 

 low ligure of 0.2 cc. per hour.'' 



These results are briefly discussed in their bearing upon alkali soils, 

 and the following - quick method of determining the amount of gypsum 

 necessary to correct the puddled condition of alkali soils is suggested: 



" Determine the rate of capillary rise, rate of saturation, rate of percolation, rate 

 of evaporation, etc., in those semialkali soils which are yielding the best crops, 

 then by adding constantly increasing quantities of the gypsum to samples of soil 

 from the alkali spots to lie treated the quantity required to bring the soil back to 

 the standard may bo determined to a close approximation. In this w r ay the needs 

 of the soil would be indicated without waiting for a year of experimentation with 

 a crop.'' 



"The rate of evaporation was determined by placing 25 grams of the 

 sample of soil to which the various salts had been added in flat dishes 

 of equal size, to each of which enough water was added to saturate the 

 soil, when they were weighed, set aside for evaporation to take place, 

 and weighed each day. The dishes were shallow and the soil was 

 about three-fourths of an inch deep/' In these experiments untreated 

 soil and soil treated with 0.5 per cent of K 2 and MgO and 0.75 of 

 CaO in different forms were used. 



"Potassium nitrate and potassium chlorid exerted the greatest influence in 

 retarding both the rate of evaporation and the rate of capillary rise in the soil, and 

 magnesium oxid and magnesium carbonate increased these two factors most. Asa 

 rule those compounds which affect evaporation most affect the rate of capillary rise, 

 and in the same direction. The calcium superphosphates and the potassium carbon- 

 ates are exceptions to this general rule. The hygroscopic nature of some salts 



i 



