1919] 



ALLEN — AZOTOBACTER CHROOCOCCUM 23 



We obtained most satisfactory results by aliquoting, so that 

 the final amount reduced in the centrifuge tube corresponded 

 to 2 cc. of 2 per cent dextrose nutrient solution. 



Nitrogen is determined on the remainder of the culture 

 solution and computed to the total amount of the original cul- 

 ture. 



NITROGEN METHODS 



Distillation.— The distillation apparatus used above, which 

 had proved very satisfactory in the Wooster laboratory for 

 distillations from weakly alkaline solutions, did not prove 

 entirely satisfactory for distillations from the strongly al- 

 kaline solutions used in the Kjeldahl method. A very slight 

 escape of ammonia was detected from the receiver flask by 

 means of a second receiver flask. The introduction of a cooled 

 condenser was therefore necessary. At the same time we 

 made some experiments on an apparatus without rubber con- 

 nections. This device is shown in fig. 1. The mouth of the 

 Pyrex flask A was flared slightly and ground to fit the head 

 B which was sealed on to the condenser tube C, the latter be- 

 ing provided with the water jacket D 13 inches in length and 

 terminating in the perforated bulb E. 



The difficulty attending the use of a cooled condenser is that 

 complete transfer of the ammonia requires sufficient distilla- 

 tion to increase the volume in the receiver to a point where ac- 

 curate titrations with N/50 solutions are interferred with. 

 This difficulty is overcome to a considerable extent by the use 

 of the principle employed by Benedict 1 , i. e., distilling into 

 the cooled condenser for 15 or 20 minutes, then draining the 

 condenser and completing the distillation. Using the above 

 apparatus and distilling slowly for 20 minutes through a 

 cooled condenser, then draining and continuing the distilla- 

 tion for 20 minutes longer, quantitative transfer of the am- 

 monia was effected and the volume of the receiver kept fairly 

 low. As zinc was used to prevent bumping and as no provi- 

 sion for scrubbing was in3luded in the apparatus, a second 

 distillation over N/10 NaCH was necessary. The magnitude 



1 Benedict, F. G. The distillaton of ammonia in the determination of nitro- 



gen. Am. Chem. Soc., Jour. 22 : 2,'i9-2G3. f. 1. 1900. 



