24 



Messrs. F. Gr. Hopkins and S. W. Cole. 



Wholly negative results were obtained with various homologous 

 i'atty acids ; with formic, acetic, and propionic aldehydes ; with acetone, 

 and with various ethereal acetates and other esters. 



But, during this preliminary stage of our investigation, the interest- 

 ing observation was made that formic acid, prepared from pure glycerin 

 <md pure oxalic acid, and used instead of acetic acid under the ordinary 

 conditions necessary for the reaction, may yield the colour in a per- 

 fectly typical manner; the spectroscopic absorption of the product 

 obtained being identical with that seen when acetic acid is used. But 

 from the formic no less than from acetic acid, the chromogenic sub- 

 stance may be distilled off, appearing always in the earlier portions of 

 the distillates, and leaving the remainder of the formic acid to yield 

 ■wholly negative results. 



This result — the explanation of which becomes clear in the sequel — 

 appeared to limit somewhat the ground we had to traverse in our 

 search. 



A further and still more definite limitation came to light when we 

 found that the reactive substance in acetic acid is not an impurity of 

 wholly extraneous origin, but is a derivative of acetic acid itself. 

 When a quantity of acetic acid wholly free from the reactive sub- 

 stance has stood for a few weeks, a reaction may always be obtained 

 once more from the earliest portions of a distillate ; and, after stand- 

 ing for a month or two, even the bulk may yield a colour of moderate 

 intensity. (Cf. infra.) 



When, again, a pure acetate, and especially calcium acetate, is 

 distilled with excess of sulphuric acid, the first runnings always give a 

 marked Adamkiewicz reaction, though later portions give none. This 

 is true even when the acetate has been made by neutralising acid which 

 was itself wholly incapable of giving a reaction. 



Lastly, among the products of the dry distillation of most acetates 

 small quantities of a substance are found which react with proteid in a 

 typical manner. In the case of calcium acetate the reaction obtainable 

 is a marked one — though, as stated above, the active substance is 

 certainly not acetone — while with an aqueous extract of the products 

 of the dry decomposition of mercuric (not mercurous) acetate the 

 reaction with proteid is intense. 



With such indications as these facts afforded, we now fortunately 

 •elected to experiment with various two-carbon compounds of typical 

 structure, such as might conceivably arise from acetic acid, by oxidation 

 or otherwise. 



The first positive evidence came to light when we set out to prepare 

 glycollic aldehyde by Fenton's method.* As a mere preliminary 

 observation, we oxidised tartaric acid in solution, by means of peroxide 

 of hydrogen in the presence of a little ferrous sulphate, without taking 

 * 1 Journ. Cbem. Soc.,' 1895, vol. 67, p. 778. 



