F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 131 



teria and a simple hydrogenated flavin has been designated bacterial 

 luciferin (Strehler, Harvey, Chang, and Cormier, 1954). McElroy 

 (private communication) has suggested that firefly luciferin is related 

 to the pteridins. 



Space does not permit a review of the complete history or the evi- 

 dence for and against the above hypotheses. The important findings 

 concerning the chemistry of Cijpridina luciferin, facts which appear 

 to have been established without question, may be summarized as 

 follows: Analyses of the purest available luciferin, prepared by An- 

 derson's procedure with two cycles of benzoylation, indicate the 

 presence of nitrogen (Chase and Gregg, 1949), confirmed by the 

 carbylamine test (Mason, 1952a) and by a ninhydrin positive reaction 

 after (but not before) acid hydrolysis of material purified both by 

 paper chromatography (Mason, 1952b) and by paper electrophoresis, 

 as described in a later section. Recent studies, now reported for the 

 first time, indicate no phosphorus in doubly cycled luciferin by the 

 molybdenum blue reaction of Fiske and SubbaRow as modified by 

 Sumner (1944). The reaction is capable of detecting one part in 5 

 million of phosphorus. No carbohydrate has been found by the an- 

 throne test, capable of detecting one part in 900,000 of starch. Control 

 tests with minimal amounts of phosphate and glucose indicate that 

 the two reactions were carried out properly. Luciferin thus appears 

 to contain C, H, O, and N as the primary constituents. 



Anderson's benzoylation method of preparation, apparently yield- 

 ing a compound nonluminescent with luciferase and nonoxidizable 

 with oxygen, but hydrolyzed to active luciferin by HCl, indicates an 

 OH, NH, or NHa group. Acetic anhydride will form a similar non- 

 active acetyl-luciferin (Anderson, 1935). 



The doubly cycled luciferin has been found to combine irreversibly 

 with cyanide, possibly indicating cyanhydrin formation with aldehyde 

 or ketone groups (Giese and Chase, 1940). With azide, luciferin forms 

 a reversible combination, possibly analogous to the reaction of hydra- 

 zoic acid with hydroquinones forming azidohydroquinones and finally 

 aminoquinones, after liberation of nitrogen (Chase, 1942). Chase (un- 

 published ) found no reaction with 2,4-dinitrophenylhydrazine. 



Additional chemical tests have been made by Mason and Davis 

 (1952), which led them to the conclusion that a ketophenol and a 



