529 



The crystalline substance which remains after treatment of the 

 tetrethylated chloride with oxide of silver would thus be the mon- 

 atomic base 



[<ftH^g^)N]i . 



H J 



There is, however, some reason to believe that the compound is de- 

 composed in the moment of the formation, and that the crystals con- 

 tain one molecule of water less, being in fact 



H 5 ) 3 ( * H ' *)N] | Q _II j o=(C si H 8 ) s (C a H 2 O a )N=C a n i7 NO a 



The crystalline product has no alkaline reaction whatever ; moreover, 

 we have seen the body 



C 8 II 17 N0 2 

 associated with the iodide in the compound above described. 



I lay some stress upon these facts, since they would lend at once an 

 additional interest to the compound under consideration, which would 

 thus appear in the light of triethylated glycocoll, 



It deserves to be noticed that normal glycocoll exhibits a tendency 

 to form compounds similar in constitution to the iodide above de- 

 scribed, one of the hydrochloric acid compounds being represented 

 by the formula 



[C 2 H 6 N0 2 ]C1 + C 2 H 5 N0 2 . 



The new triethylated compound, whatever its constitution may be, 

 is remarkable for its stability. Ebullition with the strongest potash 

 is without effect upon it. I have boiled it with fuming nitric acid 

 for hours without producing any alteration. A current of nitrous 

 acid passed through the nitric acid solution leaves it unchanged. 

 Evaporated to dryness, the residue gave, with hydrochloric acid and 

 platinum- or gold-solution, the original platinum- and gold-salts. 



When submitted to the action of heat, the triethylated compound 

 is entirely decomposed. A powerful alkaline liquid distils, whilst 

 a charred residue remains behind. The alkaline distillate contains a 

 highly volatile base, forming with hydrochloric acid and dichloridc 

 of platinum a rather soluble salt. I infer from some preliminary 

 platinum-determinations that the base thus obtained is by no means 

 triethylamine. Further experiments are necessary to clear up the 

 VOL. xi. 2 Q 



