400 Transactions.— Chemistry. 
Art. LXV.—On a Search for the Poisonous Principle of Brachyglottis 
repanda and B. rangiora. By Wruuam Sxey, Analyist to the Geo- 
logical Survey Department. 
(Read before the Wellington Philosophical Society, 11th February, 1882.] 
Tue leaves of Brachyglottis repanda, so well known to be poisonous to 
horses, ete., I examined some years ago for the purpose of determining the 
chemical nature of the principle to which this property is referable, but my 
efforts were unsuccessful, in spite of which I was enabled to report as the 
result of this investigation that the ‘‘ poisonous part is not an alkaloid.” 
A few weeks ago, however, I was kindly presented by the Hon. Mr. 
Mantell with a small quantity of a substance which he had observed to 
exude from a freshly-cut shrub of Brachyglottis rangiora, a variety of the 
plant distinguished and named by Mr. Buchanan, and stated by him to 
be a native only of the provincial district of Auckland. This substance 
proved to be essentially a balsam, that is, a mixture of essential oil and 
resin, these together amounting to 99-42 per cent. 
As in my former experiment, I failed to find anything in it having the 
character of an alkaloid, although I extracted, by means of warm water, @ 
small quantity of a bitter substance from it, and found a little nitrogenous 
matter present. 
Experimenting with the balsam in a tentative way, I soon found that it 
gives a very remarkable, indeed I may say an unique, reaction for a balsam 
with acids generally. In warm hydrochloric acid, for instance, it colours to 
a purplish red ; even at common temperatures, if time be allowed, this scape 
about. But the most striking effect is produced by heating the balsam with 
alcohol acidified by any strong acid, a rich deep-blue liquid being thus 
obtained. Subsequently it was ascertained that the same effect is to be 
got by using the leaf or the fresh shoots of this plant in place of the balsam; 
even the old wood can give a feeble reaction of this sort. And it was farther 
ascertained that the variety of this plant which covers many of the hills . 
the vicinity of Wellington (Brachyglottis repanda) also behaves in this 
way with acids. 
It was considered that the substance giving this reaction is likely to be 
the poisonous part of the balsam, chromatic reaction being obtainable with 
a great number of organic poisons. I therefore at once took means to 
isolate both the oil and resin for separately testing. For this the balsam 
was put in a retort along with water, and heat applied thereto. The distil 
it; both the clear solution, as separated from the floating oil, and the pute 
