536 POISONS : THEIR EFFECTS AND DETECTION. [§ 7 IO. 
the researches of Vaughan and others^ that diazobenzol is liable to be 
formed in milk and milk products, especially in summer-time. It is 
confidently asserted by many that the summer diarrhoea of infants is 
due to this toxine ; however that may be, it is well established that 
diazobenzol is a violent poison, causing sickness, diarrhoea, and, in large 
doses, an acute malady scarcely distinguishable from cholera, and which 
may end fatally. There will always be difficulty in detecting it, because 
of its instability. The following is the best process of extraction from 
milk. The milk will probably be acid from decomposition ; if so, the 
whey must be separated by dilution and filtration ; without dilution it 
may be found impracticable to get a clear filtrate. In order to keep the 
bulk down, 25 c.c. of the milk may be diluted up to 100 c.c., and, having 
obtained a clear filtrate from this 25 c.c. thus diluted, the filtrate is used 
to dilute another 25 c.c. of milk, and so on. The acid filtrate is neutral¬ 
ised by sodium carbonate, agitated with an equal volume of ether, 
allowed to stand in a stoppered vessel for twenty-four hours, and the 
ether then separated and allowed to evaporate spontaneously. The 
residue is acidified with nitric acid and then treated with a saturated 
solution of potash, which forms a stable compound with diazobenzol, and 
the whole concentrated on the water-bath. On cooling, the tyrotoxicon 
compound forms six-sided plates. Before the whole of this process is 
undertaken, it is well to make a preliminary test of the milk as follows : 
—A little of the ether is allowed to evaporate spontaneously. Place 
on a porcelain slab two or three drops of a mixture of equal parts of 
sulphuric and carbolic acids, and add a few drops of the aqueous solu¬ 
tion ; if tyrotoxicon be present, a yellow to orange-red colour is pro¬ 
duced. A similar colour is also produced by nitrates or nitrites, which 
are not likely to be present under the circumstances, milk having mere 
traces only of nitrates or nitrites ; it may also be due to butyric acid, 
which, in a decomposed milk, may frequently be in solution. Therefore, 
if a colour occurs, this is not absolutely conclusive ; if, however, no colour 
is produced, then it is certain that no diazobenzol has been separated. 
That is all that can be said, for the process itself is faulty, and only 
separates a fractional part of the whole. 
§ 710. Toxines of Hog Cholera. —Toxines have been isolated by F. G. 
Novy 1 from a cultivation of Salmon’s bacillus in pork broth. The 
fluid possessed a strong alkaline reaction. For the isolation Brieger’s 
method was used. The mercury chloride precipitate was amorphous and 
was converted into a chlorine-free platinum compound, to which was 
assigned the composition of C 8 H 14 N 4 Pt0 8 . After separation of this com¬ 
pound, the mother liquor still contained a platinum salt, crystallising in 
needles, and from this was obtained the chlorhydrate of a new base, to 
which was given the name of susotoxine ; it had the composition of 
1 Med. News, September 1890. 
