56 THE SIMPLER NATURAL BASES 



spinal fluid in certain degenerative nervous diseases, such as general 

 paralysis of the insane, and they regard it as a break-down product of 

 nerve substance. They used platinic chloride for the isolation, but 

 since the amount of choline to be detected is at most very small, and 

 since potassium and ammonium salts are also present, a good deal of 

 controversy has taken place as to the identity of the platinichloride 

 obtained. 



Probably Mott and Halliburton's salt was contaminated with potassium, since even 

 anhydrous alcohol, as employed by Donath [1905-1906], dissolves ammonium chloride. 

 Donath has attempted to utilise the double refraction and chromatic polarisation of choline 

 platinichloride which is not given by the isotropic crystals of the potassium and ammonium 

 salts. The conclusions of Mott and Halliburton and of Donath have been criticised by 

 Vincent and Cramer [1904], by Allen and French [1903] and by Mansfeld [1904] ; Rosenheim 

 [1905-6, 1907] and Allen [1904] have therefore attempted to find a more characteristic test 

 in Florence's periodide reaction (see below) which may be applied to the platinichloride, 

 or directly to the crude choline chloride. 1 According to Rosenheim and to Allen choline 

 is indeed present in the cerebro-spinal fluid in certain diseases, but Donath's suggestion 

 that choline is present in epilepsy and is the cause of the convulsions cannot be upheld 

 (Allen [1904], Kajura [1908], and especially Handelsman [1908]). At most traces are 

 present, wholly inadequate to account for the convulsions. Other authors, however, do 

 not admit that choline has been demonstrated in the cerebro-spinal fluid even in diseases 

 where there is a break-down of nervous tissue. Webster [1909] considers that no choline 

 test hitherto employed is satisfactory. Kauffmann [1908, 1910] thinks that if traces of 

 choline are present they are too small to be recognised with certainty. Kauffmann and 

 Vorlander [1910] consider that the dimorphism of choline platinichloride (and conversion 

 of the regular crystals into those of the monoclinic system, see below) affords a most char- 

 acteristic test, and Kauffmann has concluded that an organic base is present in the cerebro- 

 spinal fluid, which is not identical with choline. Stanford [1913] has recently arrived at the 

 same conclusion, that the base present in disease gives alkaloidal reactions, but no tri- 

 methylamine. Handelsman [1908] has emphasised the fact that on igniting the platini- 

 chloride the odour of trimethylamine is never observed. It would appear that this con- 

 troversy can only be ended by a satisfactory analysis of the platinum salt ; the only 

 published analysis (by Mott and Halliburton) is of little value (Pt found 34-8 per cent. ; 

 calculated 31/6 per cent.). 



According to Mott and Halliburton the choline set free in nervous lesions passes into 

 the blood, a conclusion shared by Allen [1904], criticised by Vincent and Cramer [1904] and 

 particularly by Vincent's pupil Webster [1909], and maintained by Halliburton [1905]. 



Choline has been synthesised by several methods : 



1. By the action of trimethylamine on ethylene oxide in concen- 

 trated aqueous solution (Wurtz [1867]). 



/0\ /OH 



(CH 3 ) 3 : N + / \ + H 2 = (CH 3 ) 3 i N/ 



\CH 2 .CH.OH. 



2. Trimethylamine combines with dry ethylene dibromide at 

 110-112 to yield trimethylamino-bromethylium bromide (Hofmann 

 [1858, under neurine]). 



1 Possibly the very slight solubility of choline nitric acid ester percblorate might be 

 utilised with advantage. 



