STRYCHNINE. 
§§ 408, 409-] 
347 
solution under the skin, for if applied to the surface the same effects 
are produced ; but, if accustomed to manipulation, the operator will find 
the subcutaneous application more certain, especially in dealing with 
minute quantities of the alkaloid. 1 
§ 408. Hypaphorine. —One substance is known which neither physiological test 
nor the colour reactions suffice to distinguish from strychnine, viz. hypaphorine, 2 
the active matter of a papilionaceous tree growing in Java—the Hypaphorus subum- 
brans ; a small quantity of the alkaloid is in the bark, a larger quantity is in the seed. 
Hypaphorine forms colourless crystals which brown, without melting, above 220°, 
and exhale a vapour smelling like naphthylamine. The free alkaloid is soluble in 
water, but has no action on litmus. The salts are less soluble than the free alkaloid, 
so that acids, such as nitric or hydrochloric, produce in a short time precipitates on 
standing. Solutions of the salts are not precipitated by alkalies ; chloroform, ether, 
benzene, all fail to extract it from either alkaline or acid solutions. It gives no 
precipitate with potassic chromate, but most general alkaloidal reagents precipitate. 
It gives a precipitate with iodine trichloride, and has therefore probably a pyri¬ 
dine nucleus, it may be an acid anilide. 3 It gives the same colours as strychnine 
with sulphuric acid and potassic permanganate or potassic chromate ; it causes in 
frogs tetanus, but the dose has to be much larger than that of strychnine. The dura¬ 
tion of life in doses of 15 mgrms. may extend to five days, and frogs may even recover 
after 50 mgrms. 
The distinction between strychnine and hypaphorine is therefore easy; besides, it will 
not occur in a chloroform extract, and it willnot give a precipitate with potassic chromate. 
§ 409. Quantitative Estimation of Strychnine. —The best process of estimating 
the proportion of each alkaloid in a mixture of strychnine and brucine is Keller’s 
method. They may also be precipitated as picrates, and the brucine picrate destroyed 
by nitric acid after obtaining the combined weight of the mixed picrates ; then weigh¬ 
ing the undestroyed strychnine picrate. 
To carry out the latter process, the solution of the mixed alkaloids must be as 
neutral as possible. A saturated solution of picric acid is added drop by drop to com¬ 
plete precipitation. A filter-paper is dried and weighed, and the precipitate collected 
on to this filter-paper ; the precipitate is washed with cold water, dried at 105°, and 
weighed. This weight gives the combined weight of both strychnine and brucine picrates. 
The precipitate is now detached from the filter, washed into a small flask, and 
heated on the water-bath for some time with nitric acid diluted to 1-056 gravity 
(about 11 per cent. HN0 3 ). This process destroys the brucine picrate, but leaves 
the strychnine picrate untouched. The acid liquid is now neutralised with ammonia 
or soda, and a trace of acetic acid added ; the precipitate of strychnine picrate is now 
collected and weighed. The weight of this subtracted from the first weight, of course, 
gives that of the brucine picrate. 
One part of strychnine picrate is equal to 0-5932 strychnine ; and one part of 
brucine picrate is equal to 0-6324 brucine. 
From the strychnine picrate the picric acid may be recovered and weighed by 
dissolving the picrate in a mineral acid and shaking out with ether ; from the acid 
liquid thus deprived of picric acid the alkaloid may be separated by alkalising with 
ammonia and shaking out with chloroform. 
Keller’s method is based on the conversion of the brucine into dinitro-brucine, 
which is insoluble in chloroform. From 0-2 to 0-3 grm. of the crude alkaloids is 
dissolved in 10 c.c. of 10 per cent. H 2 S0 4 ; when cold, 1-5 c.c. of 50 per cent, nitric 
1 Methyl strychnine, as well as methyl brucine, has been shown by Brown and 
Fraser to have an effect exactly the opposite to that of strychnine, paralysing the 
muscles like curare. In the case, therefore, of the methyl compounds, a physiological 
test would be very valuable, since these compounds do not respond to the ordinary tests. 
2 Dr C. Plugge, Arch. f. exp. Path. u. Ph., xxxii. 313. 
3 Julius Tafel ( Ber ., 1890, p. 412) has shown that the colour reactions with H 2 S0 4 
and oxidising agents are the characteristic tests of an acid anilide. 
