INSTITUTE, PUSA, FOR 1918-19 39 



tion of the P 3 O s between the solvent and the soil. Under 

 these conditions the results obtained with both types of 

 soil were almost identical. This is explained by the fact 

 that the Kalianpur soil contains a small proportion of 

 CaC0 3 and as the proportion of soil taken was large there 

 was sufficient CaC0 3 present to combine with the quanti- 

 ties of P 2 0. taken. It may therefore be accepted that even 

 small quantities of CaC0 3 can, under certain conditions, 

 entirely mask retention through adsorption. 



The conclusion arising out of these experiments is that the 

 distribution, throughout the mass of the soil and immediate 

 subsoil, of the P 2 5 contained in a dressing of superphos- 

 phate applied to the type soils would be of a very different 

 order in each case. To test this conclusion a definite amount 

 of P 2 5 dissolved in a known volume of water was allowed 

 to percolate through a column of soil of 20" depth contained 

 in a glass tube. When percolation had ceased the tube of 

 soil was cut into two inch sections and the amount of P 2 5 

 retained in each section determined. An approximate 

 determination of the P 2 5 held in solution in each section 

 and the P a O s retained by the soil was also made. 



In the case of the Kalianpur soil the percolate contained 

 very appreciable quantities of P 2 5 , and the amount held 

 by the soil varied from 12 per cent, of the original amount 

 taken in the first layer of soil to 2\ per cent, in the lowest. 

 The total amount of P 2 5 (soluble and in solution) retained 

 by any one section was in accordance with conditions 

 demanded by adsorption. 



The distribution through the column of Pusa soil was 

 totally distinct. No P 2 5 , or only traces, were obtained in 

 the percolates, and 69 per cent, of the original amount taken 

 was retained in an insoluble condition in the first, two 

 inches and 76 per cent, in the first four, w T hilst practically 

 no P 2 3 penetrated to a greater depth than 12 inches. It 

 is also noteworthy that the amount of P 2 5 present in solu- 

 tion in the Pusa soil was very much less than the amount 

 present in the corresponding section of Kalianpur soil. 



The addition of 5 per cent. CaC0 3 to the Kalianpur soil 

 brought about a distribution analogous to that obtained in 



