120 



Gooeh and Gruener — Method for the 



The iodide was introduced into the retort because having* 

 chosen to collect the halogen in alkaline solution it became 

 necessary to take steps to break up, before it should reach the 

 receiver, all nitrosyl chloride, the formation of which our ex- 

 perience in former lines of work not here detailed had led us 

 to expect under the circumstances. In acid solution containing 

 an iodide, nitrosyl chloride liberates iodine and is registered"; 

 in alkaline solution it breaks up with the formation of a 

 chloride and a nitrite, the latter having no immediate action 

 upon the arsenite. The results of experiments made in this 

 manner are recorded in the accompanying table. 







Table 



I. 





KN0 3 



Klin 



MnClo 



KX0 3 





taken. 



retort. 



mixture. 



found. 



Error. 



0-1036 grm. 



0-8 grm. 



10 cm 3 



0-1009 grm. 



0-0027 grm 



0-1083 " 



0-8 " 



10 " 



0-1082 " 



0-0001 " 



0-1064 " 



o-s " 



10 " 



0-1053 " 



o-oon " 



0-1068 " 



0-8 " 



10 " 



0-1033 " 



0-0035 " 



0-0551 " 



0-8 " 



10 " 



0-0531 " 



0-0020 " 



The experiments of Table II were carried out in a manner 

 essentially similar to that of the experiments of Table I, 

 excepting the single point that the iodine evolved in the 

 process of decomposition of the nitrate was received in potas- 

 sium iodide instead of in an alkaline arsenite. The contents 

 of the receivers were united, made alkaline with hydrogen 

 sodium carbonate, treated with an excess of decinormal arse- 

 nious acid, and the unoxidized arsenious acid was determined 

 by decinormal iodine. 











Table 



ir. 









KNO 



3 



KI 



in 



MnCl 2 



KNO. 









taken 





retort. 



mixture. 



found 





Error. 





0-2039 grm. 



l-6g 



rm. 



20 



cm 3 



0-2025 g 



rm. 



0-0014 g 



rm 



0-1060 



cc 



0-8 



cc 



10 



cc 



0-1035 



cc 



0-0025 " 



cc 



0-1036 



c< 



0-8 



cc 



10 



CC 



0-1016 



cc 



0-0020 



cc 



0-1013 



cc 



0-8 



cc 



10 



CC 



0-1002 



cc 



o-oon 



cc 



0-0521 



cc 



0-5 



cc 



10 



cc 



00521 



cc 



o-oooo 



cc 



0-0235 



cc 



0-5 



cc 



5 



cc 



0-0227 



cc 



0-0008 



cc 



0-0273 



cc 



0-5 



cc 



5 



cc 



0-0262 



cc 



o-oon 



cc 



0*0136 



cc 



0-2 



cc 



5 



cc 



0-0132 



CC 



0-0004 



cc 



o-oon 



u 



0-2 



cc 



5 



cc 



0-0009 



cc 



0-0002 



cc 



The errors of both sets of experiments, those of Table I 

 and those of Table II, are considerable, all lie in the same 

 direction, and are indicative of too low registering of the 

 action of the nitrate, since of the complete decomposition of 



