346 Scientific Proceedings^ Royal Dublin Society. 



On making up a 1 : 5 mixture with distilled water, and eentrifuging for 

 ten minutes, after standing with rei^eated shaking for three hours, it was 

 found that a sohition resulted which was clear enough to permit of print 

 being read through it, though it still had an opalescence. The liquid had a 

 very faint straw tint. Another portion was passed through a sieve, with 100 

 meshes to the inch, and a 1 : 5 mixture was made up. This portion was, there- 

 fore, richer in the clay fraction. When centrifuged beside the unsieved sample, 

 and for the same time, the supernatant liquid was somewhat less clear 

 and had a marked straw tint. The pH value, determined colorimetrically, 

 was close to 9-4, using thymol phthalein as an indicator, but with thymol 

 blue the untreated sample appeared to be about pH 98 and the sieved pH 9-55. 

 It seems that small amounts of carbon dioxide readily affect the soil extract 

 when no longer buffered by soil particles. It was also suggested by Dr. Joseph 

 that the results of Table V, showing the pH values of filter-papers, might be 

 tested further by adding to a number of papers just enough water to soak 

 them thoroughly and leave a small residue of liquid when they were lightly 

 pressed. This was accordingly done, and it was found that 2 c.c. thus obtained 

 from Whatman No. 1, 11 cm. diameter paper, gave the value pH 11 with care- 

 fully adjusted broni thymol blue, thus agreeing well with the results of the 

 spot test. Munlrtell's No. 0, 9 cm. diameter, gave a press liquid at pH 55, 

 that of Whatman No. 40, 7 cm. diameter, being at pH 5' 55 with methyl red. 

 These results are considerably less acid than the values of Table V. Wlien 

 spot tests were made on these papers, using brom cresol green as indicator, 

 Munktell's No. gave a yellow spot with a green edge and Whatman No. 40 

 a yellow spot with a blue edge, thus denoting a lesser acidity. 



The writer wishes to acknowledge his indebtedness to the Department of 

 Scientific and Industrial Research, London, for a grant coveiing the cost of 

 the hydrogen ion determination outfit ; to the Marine Biological Association, 

 Plymouth, for general laboratory facilities ; and to Dr. A. P. Joseph, Khartoum, 

 for a sample of Sudan soil and his helpful criticism. 



Summary. 



1. As a general rule the effect of increasing or decreasing the soil to water 

 proportion within limits does not alter the p// value appreciably — by as mueii 

 as 2)H 01 — for soils between jjff 6 and 8. With more acid or more alkaline 

 soils an alteration in pH value may be introduced by altering the proportions. 

 For lightly buffered acid soils one part of soil to two of water seems a safe 

 proportion to adopt; for other soils a one to five proportion seems to be 

 convenient and reliable. 



2. The pH value of some soil extracts is markedly modified by filtration, even 

 when a first filtrate is rejected. Both untreated and acid-extracted filter-papers 

 may reduce the acidity. The use of large volumes, about 160 c.c, of filtrate 

 and a filter appropriate to the soil reduces these errors. Where possible clearing 

 by the centrifuge is desirable, and has been the writer's regular practice. 



3. The fibres of acid-extracted filter-j^apers act towards indicators as if as 

 acid as about pH iS, but washing was not found to render them less. acid. 

 Unextracted papers are at about pHl-l-Q and give up traces of alkali to 

 distilled water. 



4. The indicator brom cresol green is to be preferred to methj'l red for 

 the same pll range. 



