84 Messrs. Matthiessen and Wright on the Action of [Recess, 



codeia mixed with a little of the new base, which is not quite insoluble in 

 bicarbonate of sodium solution ; the precipitate is washed with ammonia- 

 water to remove the trace of apomorphia (which is much more soluble in 

 ammonia than the new base), and is then dissolved in hydrochloric acid 

 and fractionally precipitated by bicarbonate of sodium : the second fraction 

 is pure white, and is free from apomorphia and codeia ; this is then ex- 

 tracted with ether, which dissolves almost the whole. The clear ethereal 

 solution is then shaken up with a few drops of hydrochloric acid, and the 

 solution of hydrochlorate thus obtained, if coloured, must be fractionally 

 precipitated by bicarbonate of sodium, and the ether and hydrochloric-acid 

 process repeated ; the resulting product is a viscid colourless solution of 

 the hydrochlorate of the new base, which refuses to crystallize. When eva- 

 porated on the water-bath, the dry residue yielded the following numbers 

 on analysis : — 



(I.) 0-32/0 gramme, burnt with lead chromate, gave 0*7230 carbonic 

 acid and 0*1820 water. 



(II.) 0*3080 gramme gave 0*6870 carbonic acid and 0*1720 water. 

 (Ill ) 0*3010 gramme, burnt with lime, gave 0*2450 silver chloride. 



Found. 



I. It. III. 



C 18 216 61*01 60 30 60*83 



H 21 21 5-93 6*18 6*20 



CL 71 20*04 2014 

 N 

 (X 



H 20 C1N0 2 ,HC1 



Calculated. 



216 



61*01 



21 



5*93 



71 



20*04 



14 



3*95 



32 



907 



354 



100*00 



The hydrochlorate, when fractionally precipitated by bicarbonate of so- 

 dium, yielded the base as a snow-white mass, scarcely affected by exposure 

 to air, very soluble in alcohol and ether, but not crystallizable from those 

 menstrua owing to decomposition ; when well washed and dried, first over 

 sulphuric acid and then at about 60°, it gave the following results on 

 analysis : — 



(I.) 0*3380 gramme gave 0*8410 carbonic acid and 0*1900 water. 

 (II.) 0*2680 gramme, burnt with lime, gave 0*1160 silver chloride. 



Calculated. Found. 



I. II. 



10*71 





216 



68 03 



67*87 



H 20 



20 



6*29 



6*25 



CI 



35*5 



11*18 





N 



14 



4*41 





o 2 



32 



10-09 





C 13 H 20 C1N0 2 



317*5 



100*00 





Another portion of hydrochlorate converted into platinum-salt gave a 



