6 Morris J. Blish 



method also gives an exceedingly low ash content, practically the 

 same as that of the Rather method. It is a combination of the 

 Rather and Hilgard methods, the extraction of humus being 

 carried on in the same way as in the Hilgard method, with the ex- 

 ception that the ammonum carbonate has already been added to 

 the 4 per cent, ammonium hydroxid. This causes the very fine 

 clay particles to flocculate on the filter, and consequently prevents 

 their being carried through. The method has thus far been 

 tried but very briefly, only three such determinations having 

 been made. The results of these, however, indicate that the 

 method should be as reliable as the Hilgard, with the added ad- 

 vantage of the low ash-content, caused by the absence of any 

 clay particles. It is, however, just as tedious as the Hilgard 

 method, and therefore does not seem as practical as the Rather 

 method for most gravimetric work. 



THE COLORIMETRIC METHOD 



For soils which are very low in humus, however, the grav- 

 imetric determination is not entirely reliable and satisfactory. 

 This is shown by the results obtained from gravimetric determina- 

 tions on very light clay subsoil which should contain very little or 

 no humus. Their extracts with 4 per cent, ammonia may have 

 so little color that the eye can scarcely detect it, but a gravimetric 

 determination will show .15 per cent, to .20 per cent, humus, while 

 the humus ash will be practically the same as in soils of high 

 humus content. It would seem from this that the 4 per cent, 

 ammonia must dissolve a certain amount of material other than 

 humus, from a soil. In surface soils, and soils of comparatively 

 high humus content, the error introduced in this way is of slight 

 consequence, but it is readily seen that in subsoils, and soils of 

 low humus content, the percentage of error is considerably in- 

 creased. Assuming that humus is the dark-colored organic ma- 

 terial of the soil which is soluble in dilute alkalies, it was decided 

 that for soils low in humus, a colorimetric determination would be 

 of considerably more value than a gravimetric on the same soil, to 

 say nothing of the great economy of time in the use of the color- 

 imetric method. Colorimetric determinations were run on all of 



116 



