34 



KEPOKT OF CHEMICAL LAHOKATOHY 



The Mechanical Analysis of Arid Soils 



Various 



methods of 

 mechanical 

 analysis 



Importance 



ascertaining 

 the total 

 " clay " 



The mechanical analysis of soils, that is to say the division of their particles into 

 groups according to size, is perhaps of greater importance than any other single 

 determination which the soil analyst is called upon to make. These analyses would 

 undoubtedly be made more frequently than is done at present were the methods suggested 

 more satisfactory and, especially, less laborious. That recommended by Hall,' which 

 appears to be the only one practised in Great Britain, is so tedious and often requires 

 so long a time for its completion as to very seriously impair its usefulness. Its 

 application to arid soils offers a still more serious objection in that it entails a treatment 

 with acid, which is not permissible with arid soils, the larger particles of which may 

 be, and usually are, concretions of calcium carbonate. Hilgard's method, depending upon 

 the use of successive currents of water of velocities adjusted to carry particles of the 

 required size, offers several objections, perhaps the least of which is that, requiring special 

 and expensive apparatus, it is, to quote from Hall (loc. cit.), suited only to laboratories 

 devoted entirely to soil analysis. Osborne's method, which, like that of Hall, is carried out 

 by a series of sedimentations in beakers, proves in our hands, working with arid soils, to be 

 capable of yielding results of the most varying character. It leaves very much to be 

 desired on the score of ease and rapidity, especially when a number of samples are to 

 be treated at the same time. Indeed, the multiplication of vessels alone renders it quite 

 inapplicable in such cases. The method advocated by the U.S. Bureau of Soils appeared 

 to us to be the most promising. It consists in breaking up the soil agglomerations — 

 puddling the clay — by agitation in a mechanical shaker with water and a few drops of 

 ammonia. The separation of the clay from the other groups of particles is then made 

 by means of a centrifuge. 



Given the appliances suited to the purpose the above method would seem to leave little 

 to be desired on the score either of rapidity or accuracj*. Our experience with it, as 

 applied to arid soils, has been most disappointing. It, as do most of the other methods, 

 fails to separate a very considerable proportion of the clay, which goes therefore to swell 

 of the silt and other fractions.' That the entire amount of clay should be taken into 

 consideration has been abundantly shown by the results of observation on partially water- 

 logged soils, inefficiently drained, and lacking oxygen. Under these conditions sodium 

 carbonate is formed in notable proportion, one of the results being the deflocculation of the 

 clay with its attendant disastrous effect on production. 



In the analytical work carried out recently it has been noted that even when so 

 little as two per cent, of clay remains associated with the residue of silt and sands, the 

 binding effect on the soil is very marked. 



Clay aggregates in arid soils are known to be held together much more firmly than 

 in those of humid regions. There is evidence that this condition is associated with the 

 intense baking which these soils receive during the season of drought. (See below.) 



In the tests which were made by the U.S. Bureau of Soils^ on the effect of length 

 of time of agitation in the mechanical shaker on the proportion of clay found, no marked 

 additional change appeared to take place when the time was increased beyond six hours, 

 except in the case of samples shaken as long as 77 hours. After such prolonged action 



1 The Soil, p. 51 



- Bure.nu of Soils BiilUHn, No. 24 



