806 EXPERIMENT STATION EECOHD. 



by Warrinpton and Poak" and l».v Cameron and Rreazpale (E. S. R., 15 j). 744) 

 with varying results. As detorniinations of carbon were needed in 1 i)er cent 

 Lydrocbloric acid and 4 per cent ammonia extracts of soils and the employment 

 of neither the Parr calorimeter nor the combustion furnace was feasible, a 

 thorough test of the chromic and sulphuric acid method was made and the 

 results compared with those given by combustion with copper oxid in a furnace. 

 In addition to this, combustions were made with alkaline permanganate solu- 

 tion witli a concentrated chromic acid solution and continued boiling. 



The soils gave figures which compare well with those yielded by ignition 

 methods. Dilute mixtures of chromic acid and alkaline permanganate gave low 

 results. The work seems to give conclusive proof that the concentrated 

 chromic and sulphuric acid treatment completely decomposes organic and in- 

 organic carbon present in soils and overcomes the objection raised by Cameron 

 and Breazeale and Hall and Miller.^ 



" If boiled for 30 minutes, a mixture of 3.3 gm. of chromic acid in 10 cc. 

 of water to 50 cc. of sulphuric acid (sp. gr. 1.84) will oxidize all of the organic 

 carbon and liberate all carbon dioxid chemically or mechanically held in soils, 

 provided the soil is ground to pass GO-mesh sieve and from 1 to 3 gm. of soil 

 used for each GO cc. of mixture. The Brown and Escombe titration method of 

 determining carbon dioxid, and the modified Amos absorption tower for the 

 same, are applicable to either wet or dry combustion forms of apparatus, and 

 can be relied upon to' give rapid and accurate results with considerable economy 

 of time and space. Carbon may be accurately determined in 1 per cent hydro- 

 chloric acid extracts and 4 per cent ammonium hydroxid humus solutions 

 without concentrating below 50 cc. by using the above chromic and sulphuric 

 acid mixture. The apparatus described is applicable to the determination of 

 carbon dioxid in any form, and a number of other gas determinations, depend- 

 ing on absorption in acid or alkalis, oxidation, or reduction processes. By 

 using the apparatus shown, and following the method as outlined, one analyst 

 can run six determinations at one time, and complete a set an hour when doing 

 routine work." 



An improvement in the electrical method of determining salt in soil, W. 

 Beam and G. A. Freak {Cairo Set. Jour., 8 (1914), No. 93. pp. 130-133, pi. 1).— 

 In operating under field conditions with the electrical conductivity method it 

 was found desirable to eliminate as far as possible the influence of calcium 

 sulphate contained in the soil. This may be readily and satisfactorily done 

 by employing, in place of water, diluted alcohol (40 per cent by volume) for 

 the extraction of the salt, and by comparing with a table of resistance in the 

 Fame solvent. 



" Further, it is possible to extend the method to the determination of the 

 proportion of calcium sulphate, since it is only necessary to make another ex- 

 traction, with water, on a fresh sample, and from its conductivity and that of 

 the salt known to be present, to determine the proportion of calcium sulphate 

 by a simple calculation. An attached chart shows the curves of resistances of 

 sodium chlorid in water, calcium sulphate in water, and sodium chlorid in 

 alcohol of 40 per cent. The greater proiwrtion of salt in most Nile soils is 

 made up of sodium chlorid and sodium sulphate, but the resistances of these 

 two are so nearly alike that for all practical purposes the one curve suffices. 



"The method was tested on solutions of known composition, as follows: A 

 water solution containing 0.03 per cent gyp.sum and 0.037 per cent sodium chlorid 

 was found to have a resistance of 210 ohms as against a calculated resistance 

 of 220 ohms. A water solution containing 0.015 per cent of gypsum and 0.018 



"Jour. Chem. Soc. [London], 37 (1880), pp. 617-625. 



"Jour. Chem. Soc. [London], 80 (1906), pt. 1, pp. 595-597, fig. 1. 



