RECENT ADVANCES IN SCIENCE 547 



solution in water containing a trace of sulphuric acid and 

 reprecipitation with alcohol. The resulting product was 

 found to be a mixture of the enzyme with a nitrogenous 

 glucoside ; the latter forms an insoluble compound with 

 mercuric chloride, and was precipitated out of solution by this 

 means and the filtrate was once more precipitated with 

 alcohol ; after several more solutions and reprecipitations, 

 the enzyme was obtained free from adhering glucoside. 



In the best preparation the yield of enzyme obtained 

 from 5 kilos of horse-radish was 0-45 gram, which was about 

 60 per cent, of the total enzyme present. From the mercuric 

 chloride compound, about 3-4 grams of the glucoside were 

 obtained by liberation with 2N hydrochloric acid. A new 

 method of estimating peroxidase has been worked out which 

 is free from the errors inherent in Bach and Chodat's method; 

 it depends on the production of purpurogallin from pyrogallol 

 and hydrogen peroxide in the presence of the peroxidase. As 

 a standard of measurement the authors have fixed upon the 

 " purpurogallin number " which represents the number of 

 milligrams of purpurogallin which would be produced by 

 1 mgm. of the vacuum-dried preparation. Ihis number is 

 about 0-25 for well-pounded horse-radish, 360 for the crude 

 enzyme, before purification by means of mercuric chloride, 

 and about 670 for the purest sample of the enzyme so far 

 obtained. The enzyme itself also appears to be a nitrogenous 

 glucoside containing over 30 per cent, of a pentose and an 

 equimolecular proportion of a hexose ; if it contains only two 

 sugar molecules its molecular weight would appear to be 

 about 500 with three atoms of nitrogen ; it contains 5-5 per 

 cent, of ash, and about 0-46 per cent, of iron. The companion 

 glucoside has a higher molecular weight, and gives Millon's 

 and the xanthoproteic reactions ; it contains about 50 per 

 cent, of pentose and a hexose residue as well, and three atoms 

 of nitrogen to every two molecules of pentose. It has been 

 stated that oxyhemoglobin does not differ from peroxidase 

 in activity, but this error is no doubt due to the fact that 

 peroxidase has never before been obtained in such a high 

 degree of purity ; as a matter of fact, oxyha.moglobin is only 

 about one-thousandth part as active as a quantity of peroxidase 

 containing the same amount of iron. 



The mechanism of alcoholic fermentation has been 



