68 



BULLETIN F. x 



last four days, showing a downward diffusion which exceeded 

 the capillary rise. 



The changes in sulphates, from period to period, are shown 

 graphically in Fig. 2, p. 67, for the several depths. In this case 

 the surface layer gained in SO 4 until the end while the bottom 

 layer had least in it at this time, indicating that the diffusion 

 rate was too slow to counteract the capillary rise. 



In the case of the phosphates, the absorption was evidently 

 so strong from the first that the amounts left in recoverable 

 form were too small to bring out clearly the movements within 

 so short a series of observations. In the next table the percent- 

 age amounts of phosphates found are given, using that recov- 

 ered from the soil at the start as the basis of comparison. 



Amounts of phosphates recovered in different layers, expressed 



in per cents. 



The amount of phosphoric acid added to the soil with the 

 water was only 4.4 per cent, of the amount present at the start, 

 but the smallest amount recovered was 116.7 per cent, of that 

 found in the soil at the start, while the largest amounts found 

 range near 150 per cent. These differences are larger than the 

 error of the method and indicate that more phosphoric acid has 

 come into recoverable condition with water alone, during this 

 capillary treatment, than existed in the soils before being so 

 treated. It has been -demonstrated, through both field and 

 laboratory studies, that one effect of capillary movement is to 

 concentrate nitrates to such an extent that larger amounts of 

 them may be recovered from a soil than it is possible to recover 

 where capillary concentration has not taken place. That this 

 may also be true for other salts isl to be expected unless it b 

 in those cases where large absorption takes place, as so often 

 happens with potash and phosphoric acid. 



