1'. H. JOHNSON, E. H.-C. SIE, AND Y. HAN EDA 21:5 

 of evidence, the lollowing provisional stiiuturt' has been suggested: 



C— (CH.)^— NH— C— NH., 



II II 



N NH 



With reference to the firefly, the luciferin, in both oxidized and 

 reduced forms, as well as the luciferase have been crystallized and 

 their properties studied (2, 4, 12, 38) . The reduced form of luciferin, 

 active in light emission, is slightly yellow in alkaline solution, and 

 has a greenish yellow fluorescence, ft has the empirical formula 

 CisHjoNoSoOa, and is converted into the oxidized form with the loss 

 of two hydrogens. The latter form is a potent inhibitor of lumines- 

 cence through the formation of a relatively stable, intermediate com- 

 plex with the luciferase, from which it can be removed by the addition 

 of pyrophosphate or coenzyme A. Extensive research by McElroy and 

 his associates has clarified the role of ATP and various factors that in- 

 fluence the amount and intensity of light emitted by the firefly sys- 

 tem. Although the details can scarcely be touched upon in the present 

 paper, it is interesting to note that this system has the extraordinarily 

 high quantum efficiency of one, or close to one (41) . The important 

 point with respect to luciferin-luciferase reactions is the requirement 

 of ATP (35) , a cofactor required tor certain reactions involved in 

 energy transfer within virtually every living cell. 



Finally, the luminescent system of bacteria likewise requires a 

 type of cofactor that has a widespread importance in cellular oxi- 

 dations, namely, a flavin (39, 44) . The precise role of the FMN is 

 uncertain; like Cypridina luciferin, flavin mononucleotide is yellow- 

 colored and fluoresces yellow, whereas the bio- or chemi-luminescence 

 is much more in the blue-. Although an exact correspondence be- 

 tween fluorescent and chemiluminescent spectra of a substance would 

 not necessarily be expected, such correspondence is sometimes quite 

 close, e.g., among the 2,3-dihydrophthalazine-l,4-diones (6) . Marked 

 differences between the emission spectrum of fluorescence and that 

 of chemiliMiiinescence suggest that the same molecule luidergoes con- 

 siderable modification in the chemical reaction before it emits. In 

 bacterial luminescence the aldehyde can scarcely be the emitting 

 molecule inasmuch as it has no absorption in the visible. The 



