Intelligence and Miscellaneous Articles. 235 
portions left were therefore mixed, distilled with soda, and the distil- 
late received in hydrochloric acid. The solution was evaporated to 
dryness, traces of chloride of ammonium were removed, and the 
solution treated with avery small quantity of platinic chloride. 
The solution was then allowed to repose until next day ; and the 
platinum salt formed was filtered off and rejected. The filtrate 
was treated with more platinic chloride dissolved in alcohol. A 
copious precipitate was obtained. On analysis it gave 14:16 per 
cent. of carbon. 3°79 of hydrogen, and 37°81 of platinum. ‘Trime- 
thylamine requires 13°57 per cent. of carbon, 3°77 of hydrogen, and 
37°20 of platinum. The following is a summary of the above 
results :— 
Alkaloid 
Platinum obt. cale. Platinum, 
No. of Salt as Trime- NH,Cl per cent. 
experiment. Gas used. obtained.  thylamine. obtained. in Salt. 
Grains. Grains. 
iP — — — — — 
2. — 2°70 6 — — 
3. 21°51 4-70 Lp _- 33°83 
4. 100-00 7:70 de 7 15:0 38°40 
5. 101-00 2°03 0-5 6:2 37°04 
6. == 7°20 16 1:9 38°47 
‘2 42-00 8-00 1°8 271° 38-05 
Mixture .... — — — — 37°81 
The mean percentage of platinum is 38°10, which is 0°90 per 
cent. too high for trimethylamine, and 1-43 per cent. too low for 
dimethylamine. As the errors of experiment tend to give too high 
a number for the platinum, I conclude that the base formed was 
principally trimethylamine. It might also be propylamine, which 
requires the same numbers, and be formed from cyanide of ethyl 
as in Mendius’s well-known process. Such a supposition, however, 
does not account for the ammonia formed at the same time— 
C,H,CN+2H,=C,H,N. But as the ammonia formed simul- 
taneously with the base is in such large excess, further experiments 
are necessary to explain all the phenomena. The base butylene- 
diamine, obtained by Fairley from cyanide of ethylene, requires 
9°59 per cent. of carbon, 2°80 of hydrogen, and 39°45 per cent. of 
platinum * ; and moreover would not have the characteristic smell 
of trimethylamine. 
If we assume that the base is formed from prussic acid, the 
equation becomes 83CHN + 6H,=C,H, + 2NH,. The produc- 
tion of sal ammoniac in the experiments was too irregular to enable 
this equation to be confirmed from its amount. 
According to Lange t a polymeride of prussic acid exists having 
the formula C,H,N,. If formed from this substance, the equation is 
the same as that given above. 
It appeared to me that some light might be thrown upon the 
* Fairley, loc. cit. p. 363. + Deut. chem. Ges. Ber. vi. 99. 
