220 POISONS : THEIR EFFECTS AND DETECTION. [§ 272 . 
becomes of a bright pure red and froths strongly ; if, on the other hand, 
a trace of prussic acid be present, the liquid becomes brown and does not 
froth, or only slightly does so. 
8. Robert’s Test.— A 1-4 per cent, solution of blood, to which 
a trace of ferricyanide of potassium is added, is prepared, and the 
neutralised distillate added to this solution. If hydric cyanide be 
present, then the liquid becomes of a bright red colour, and, examined 
spectroscopically, instead of the spectrum of methsemoglobin, will be 
seen the spectrum of cyanmethaemoglobin. Robert proposes to examine 
the blood of the poisoned, for the purpose of diagnosis, during life. A 
drop of blood from a healthy person, and a drop of blood from the 
patient, are examined side by side, according to the process just given. 
9, An extremely delicate test has been suggested by F. Weehuizen 
( Clnem . Cenir., 1905, i. 1191). To the solution supposed to contain HCN 
is added an alkaline solution of phenolphthalin and weak copper sulphate 
solution (1 : 2000) ; if HCN is present, phenolphthalin is oxidised in the 
cold to phenolphthalein and the liquid turns red ; it is said to be sensitive 
even if 1 part of HCN is diluted up to 500,000 parts. 
§ 272. Separation of Hydric Cyanide or Potassic Cyanide from 
Organic Matters, such as the Contents of the Stomach, etc. —It is 
very necessary, before specially searching for hydric cyanide in the 
contents of the stomach, to be able to say, by careful and methodical 
examination, whether there are or are not any fragments of bitter 
almonds, of apples, peaches, or other substance likely to produce hydric 
cyanide. If potassic cyanide has been taken, simple distillation will 
always reveal its presence, because it is found partly decomposed into 
hydric cyanide by the action of the gastric acids. Nevertheless, an acid 
should always be added, and if, as in the routine process given at p. 51, 
there is reason for suspecting that there will be no cyanide 
present, it will be best to add tartaric acid (for this organic acid will in 
no way interfere with subsequent operations), and distil, as recommended, 
in a vacuum. If, however, from the odour and from the history of the 
case, it is pretty sure to be a case of poisoning by hydric or potassic 
cyanide, then the substances, if fluid, are at once placed in a retort or 
flask, and acidified with a suitable quantity of sulphuric acid ; or if the 
tissues or other solid matters are under examination, they are finely 
divided, or pulped, and distilled, after acidifying with sulphuric acid as 
before. 1 It may be well here, as a caution, to remark that the analyst 
must not commit the unpardonable error of first producing a cyanide 
by reagents acting on animal matters, and then detecting as a poison 
the cyanide thus manufactured. If, for example, a healthy liver is 
carbonised by nitric acid, saturated with potash, and then burnt up, 
1 Domenico Ganassini (Bull, de Soc. med. chir. de Pavia , 29) prefers in all cases 
tartaric as the acidifying acid. 
