Radziszewski (1877, 1880) gives a long list of substances, chiefly essential oils, which luminesce if slowly oxidized in alcoholic solutions of alkalis. Formaldehyde, dioxymethylen, paraldehyde, metaldehyde, acroleïn, disacryl, aldehydeammonia, acrylammonia, hydrobenzamid, lophin, hydroanisamid, anisidin, hydrocuminamid, hydrocinamid, besides waxes, and such biological substances as glucose, lecithin, cholesterin, cholic, taurocholic, and glycocholic acids, and cerebrin, all luminesce on oxidation. Radziszewski himself and many other authors have compared the light of organisms to this type of luminescence. Indeed the incorrect identification of granules found in the cells of practically all luminous tissues as oil droplets, is largely due to the influence of Radziszewski's work. Dubois (1901 b) has added esculin, and Trautz (1904-5) many aldehydes and phenol derivatives, including vanillin, papaverin, tannic and gallic acids, besides glycerol and mannite to the list of biological substances oxidizing with light production. Guinchant (1905) has described oxyluminescence of uric acid and asparagine, Weitlaner (1911) of substances in humus and McDermott (1913) of substances in urine and the anaerobic alkaline hydrolysis products of glue and Witte's peptone. Pyrogallol is especially prone to luminesce, as was first noticed by Lenard and Wolf (1888) in developing a photographic plate with pyrogallol developer. Later the luminescence was studied in some detail by Trautz and Schloringin (1904-5) who developed the well-known luminescent mix

ture of pyrogallol, formaldehyde, K2CO3 and H2O2. As I have shown, pyrogallol can be oxidized in a great many different ways, and some of these are of great interest, for they very closely imitate the mechanism for the production of light in organisms. These are recorded in [Table 3], which also includes various other types of oxyluminescence of general or biological interest.

TABLE 3
Types of Oxyluminescent Reactions

The spectrum of chemiluminescent reactions has been described in a few instances as continuous but no definite measurements of its extent have been made. Radziszew

ski (1880) found the light of lophin oxidized in alcoholic caustic alkali, examined with a two-prism spectroscope, to give a continuous spectrum, brightest at E, with the red and violet ends lacking. Trautz (1905, p. 101) states that the pyrogallol-formaldehyde-Na2CO3-H2O2 reaction gives a continuous spectrum from the red to the blue green with maximum brightness in the orange. Weiser (1918 a) has studied the spectra of some chemiluminescent reactions by photographing the light behind a series of color screens. He finds also that the spectra are short, with maximum intensity in various regions. Thus, amarin oxidized by chlorine or bromine, extends from the yellow to greenish blue with a maximum in the green while phosphorus, dissolved in glacial acetic acid and oxidized with H2O2, luminesces from yellow green to violet.

The spectra of luminous animals are quite similar to those of chemiluminescent reactions. Moreover, as we have seen, chemiluminescence is essentially an oxyluminescence, since oxygen is necessary for the reaction. All luminous animals also require oxygen for light production. Therefore, bioluminescence and chemiluminescence are similar phenomena and they differ from all the other forms of luminescence which we have considered. The light from luminous animals is due to the oxidation of some substance produced in their cells, and when we can write the structural formula of this photogenic substance and tell how the oxidation proceeds, the problem of light production in animals will be solved.

CHAPTER III
PHYSICAL NATURE OF ANIMAL LIGHT