On the Chemistry of Light Production in Luminous Organisms. 105 
erase has sufficient properties in common with the enzymes as a class 
to be considered an enzyme. ‘The peroxidases are well known to be 
used up in the reactions they accelerate. All workers on enzymes 
agree that the more they are purified the less active they become. The 
chemical procedures necessary to remove foreign material bring about 
irreversible changes in the enzyme itself, a characteristic also of many 
protein groups and of the colloidal state in general. This is true in the 
case of luciferase, for the crude luciferase solution is the most active 
preparation that can be obtained. 
I believe that Cypridina luciferase should be placed in a class of 
oxidizing enzymes by itself—a group having the chemical reactions of 
an albumin, possibly in combination with some heavy metal, and 
which, so far as we know, acts specifically on only one substance, 
Cypridina luciferin. It resembles the plant peroxidases in resisting the 
action of chloroform, toluol, etc., but will not oxidize any of the 
hydroxy-phenol or amino-phenol compounds! so readily oxidized by 
the peroxidases, nor will the peroxidases oxidize luciferin with light 
production. Dubois’s researches show that Pholas luciferase differs 
in some properties from Cypridina luciferase and my own work? indi- 
cates that firefly luciferase is more like that of Pholas. A comparative 
study of other species of luminous animals is most needed in order to 
delimit more accurately the class of luciferases as a whole. 
Luciferin presents many characteristics in common with the pro- 
teins, but two, which, to say the least, throw doubt on its protein 
nature. I refer to its peculiar solubility (in alcohols, esters, and glacial 
acetic acid), and its resistance to digestion by proteases, even by tryp- 
sin, which has almost universal digestive action. These two peculiari- 
ties have already been discussed in the body of the paper and there is 
no need of repeating the discussion here. We can only say that if a 
protein, luciferin must belong to a new group differing from known 
natural proteins in these respects. In general characteristics this new 
group would fall somewhere on the border-line between the proteoses 
and peptones. Perhaps it would not be surprising to find in nature 
proteoses or peptones soluble in absolute alcohol. We know also that 
some NH-CO linkages of proteins are broken down with great difficulty 
1 Because of the ease with which many of these hydroxyphenyl compounds undergo auto- 
oxidation one must always compare the color produced by luciferase solution with that produced 
in a control of boiled luciferase solution. I find that a concentrated luciferase solution well 
shaken with chloroform and filtered, which produced a brilliant light with luciferin, had no 
oxidative action on phenol, a-napthol, para-phenylen diamine, ortol, orcin, hydrochinon, resorcin, 
pyramidon, phloroglucin, pyrocatechin, gallic acid, benzidin, pyrogallol, gum guaiac, amidol, 
tannin, or a-napthylamine, either with or without H2O2, Dubois (Ann. Soc. Linn. de Lyon, 1914, 
lxi, 161) reports oxidation of pyrogallol, tannin, hydroquinon, guaiacol, Tromsdorf reagent, 
chlorhydrate of diaminophenol, para-phenylene diamine, napthol, napthol B, and gum guaiac 
plus H2O:2 by a solution of Pholas luciferase. Dubois’ results are of little value, however, as there 
is no evidence that the oxidation is due to luciferase rather than the oxidizing enzymes which one 
finds in cell extracts of all animals, whether luminous or non-luminous. 
2 Harvey, Am. J. Physiol., 1914, xlii, 342; Carnegie Inst. Wash. Pub. No. 251, p. 171, 1917. 
