APPENDIX TO CHAPTER I AMINES 125 



crystals of the periodide of trimethylamine are sucked off on to glass 

 wool, washed with 3-4 c.c. of a mixture of one part of the above potas- 

 sium tri-iodide solution with three parts of water. The crystals are then 

 dissolved in sodium thiosulphate solution, and after adding excess of 

 sodium hydroxide, the trimethylamine is distilled ; the distillate is 

 titrated with acid. The mother liquor of the crystals of trimethyl- 

 amine periodide yields by a similar treatment the dimethylamine on 

 distillation. 



The separation of ammonia and monomethylamine, which are 

 left behind as hydrochlorides mixed with the sand, is effected by 

 Frangois's process, of which the following is an example : 70 grm. of 

 methylamine + 7 grm. of ammonia (both in the free state) in 2000 c.c. 

 of water are shaken for one hour with 200 grm. of yellow mercuric 

 oxide. The solution is decanted and the precipitate is washed. The 

 filtrate and washings contain all the methylamine, but almost the 

 whole of the ammonia is in the mercury precipitate. To remove the 

 remainder, 40 c.c. of caustic soda and 40 c.c. of saturated potassium 

 carbonate solution are added, together with 100 grm. of mercuric 

 oxide. The solution now only contains monomethylamine. 



Methylamine can be distinguished from ammonia by means of 

 Nessler's reagent ; the amine gives a cream-coloured precipitate, 

 ammonia a brown one. 



The estimation of small quantities of amines in the presence of 

 much ammonia has been described by Bertheaume [1910, 2]. 



Fleck [1896] recommends the separation of trimethylamine from 

 ammonia by means of the sulphates, rather than the chlorides. 

 Ammonium sulphate is insoluble in absolute alcohol, in which am- 

 monium chloride is distinctly soluble ; trimethylamine salts dissolve 

 readily in alcohol. 



de Filippi [1906] has estimated trimethylamine in urine by destroy- 

 ing ammonia, primary and secondary amines by means of sodium hypo- 

 bromite ; this reagent leaves tertiary amines intact. Doree and Golla 

 [1910] by a slightly modified method found 0*014 P er cent, trimethy- 

 lamine in urine. They state that this amine cannot bedistinguished from 

 choline by the alloxan test, nor by the bismuth iodide or periodide test. 



Melting points and solubility of trimethylamine salts : 



Hydrochloride 271-275 soluble in boiling chloroform. 



Picrate . . 216 soluble in 77 parts of cold water. 



Picrolonate . 250-252 in 1121 parts of cold and 166 parts of boiling water, 



794 of cold and 233 of boiling alcohol. 

 Aurichloride . 228 yellow monoclinic crystals, readily soluble in hot alcohol, 



slightly in water. 

 Platinichloride 240-245 regular orange crystals, little soluble in boiling alcohol. 



