422 



SCIENCE PROGRESS 



The curves shown in Fig. 2 * bring out some features of 

 " buffer action " that are shown even more strikingly by soil- 

 water mixtures. The curves represent the changes in Ph when 

 I per cent, and 5 per cent, solutions of Witte's peptone are 



N . . N 



titrated with — lactic acid and — NaOH respectively. An 

 10 10 ^ -^ 



examination of the two curves will show that the amount of 



buffer action is dependent upon a number of factors, among 



which are : (i) The nature of the constituents as indicated by 



the comparison of the slopes of the curves in Fig. 2 with the 



slope of that in Fig. i ; (2) the concentration of the constituents, 



the 5 per cent, solution being more resistant to change in Ph, 



i.e. has greater buffer effect, than the i per cent, solution ; 



(3) the buffer effect is not the &ame at all points of the curve, 



8 ' 



6 • 



1.* 



5 - 



1 



1 



CCS. ^Lactic Acid per Wees, —- -^ccs.-j^ NaOH per lOccs. 

 -^ peptone solution peptone solution 



Fig. 2. — Titration curves of peptone solutions. 



but depends on the region of Ph at which the buffer action is 

 measured. This last point is important, and is brought out 

 better in Fig. 3,* which is the titration curve of 0-phosphoric 

 acid. From this curve it will be noticed that there is practically 

 no buffer action at Ph values of 4-4 and 9-3, but that there is 

 very considerable buffer action in the region of Ph 7*4 to 6-2. 

 This is the region between which the reaction of normally fertile 

 soils varies and the possibility that a part may be taken by acid 

 phosphates as reaction regulators in the lighter soils has been 

 very largely overlooked. Acid phosphates, of course, cannot 

 be the only buffers in a soil — salts of lime may also act as 

 such, although not, perhaps, in the same region of Ph- The 

 soil colloids also are powerful buffers, but nothing is known as 

 to the region of Ph in which they act best. Fig. 2 indicates that 



1 Reproduced from Clark and Lubs' paper, Journ. Bad., 2 (1917), pp. i, 

 109, 191. 



2 Ibid. 



