1919] 



ALLEN — AZOTCBACTER CHROOCOCCUM 25 



apparatus was used throughout the "Final" experiments re- 

 ported below. 



From this point the matter of a distillation apparatus has 

 been developed as a separate problem and is reported in a 

 following paper (Allen and Davisson, '19). 



Kjeldahl digestions. — As stated above, the foaming accom- 

 panying solutions high in mannite was most troublesome, and 

 it was realized that unless gome improvement could be devised 

 future progress would be almost blocked. Varying conditions 

 with respect to catalysts, i. e., using different amounts of mer- 

 cury, copper sulphate, and metallic copper, did not give any 

 appreciable aid. The f oamiig appeared to be due to separation 

 at the outset of a large amount of carbon or highly carbo- 

 naceous material which is very resistant to decomposition in 

 boiling sulphuric acid. It occurred to us that the high tempera- 

 ture imparted to the digestion mixtures by the sodium (or po- 

 tassium sulphate) caused a heavier deposit of this material 

 than would otherwise be th€ case. Experiment proved the cor- 

 rectness of this suggestion, and it was found that by carrying 

 on the digestion for 20 to 25 minutes with sulphuric acid and 

 copper sulphate alone, thei adding the sodium sulphate, the 

 danger from foaming was slight and the digestion mixture 

 containing 2 gms. mannite cleared in from 60 to 85 minutes. 



The empirical experiments with catalysts were then re- 

 peated, with the result that copper sulphate alone seemed to 



be the most desirable agent. Judging by the time required for 

 the clearing of the mixtures, mercury adds but little to the ef- 

 fect of CuS04 and possesses the disadvantage, of course, that 

 Na2S must be added to the alkali. The above experiments 

 were made with CUSO4, 1 2.0 gms., Hg, 3 drops, and Cu, 0.5 gm. 

 It was noted that the above amount of CuS04 was very effi- 

 cient in inhibiting foaming. Hibbard 2 found that large 

 amounts of CUSO4 were associated with incomplete recovery 

 of NH3 by distillation. To determine the minimum amount of 

 CuS04 required to prevent foaming varying amounts of this 



1 All references to copper sulphate refer to the hydrate CuS0 4 .5 H 2 0. 



* Hibbard, P. L. Notes on the determination of nitrogen by the Kjeldahl 

 method. Jour. Ind. and Eng. Chero. 2 : 4G3-46G. 1910. 



