1 i 0- rrrj \ i 3 of Biological Systems Proceedings of the Conference on Luminescence March 28-April 2, 1954, spon- sored bij the Committee on Photobiologij of the National Academy of Sciences- National Research Council and supported by the National Science Foundation S. €>curcU-i Edited by FRANK H. JOHNSON Princeton University A Publication of the AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE Washington, D.C. 1955 Copyright 1955 by The American Association for the Advancement of Science Library of Congress Catalog Card Number 54-12547 Preface Luminescence — the emission of "cold light" in the form of bioliimi- nescence, chemiluminescence, fluorescence, phosphorescence, et cetera — has long been a fascinating subject of inqpiry. Modern chem- ical and physical research has greatly advanced the understanding of fluorescence and phosphorescence, particularly of inorganic and of many organic substances; within the past generation, practical appli- cations have become a part of everyday experience, as witnessed by such familiar examples as fluorescent lighting, phosphorescent paints or tapes, and television screens. Luminescence in relation to biological processes is a more difficult and complicated problem, the progress of research has been seemingly slow, and the importance of studies on this aspect has been less obvious. Thus, it has not yet proved possible to obtain luminescent extracts from more than about half a dozen of the myriads of visibly luminescent organisms that are scattered over the phylogenetic tree of animals from protozoa to fishes, as well as over the primitive plant world of bacteria and fungi. Furthermore, basic relationships between photosynthesis, fluorescence, and chemi- luminescence of chlorophyll in green plants remain to be clarified, and the full significance of the property of fluorescence in many biological molecules such as riboflavin is yet to be understood. Within the past several years, however, remarkable progress has been achieved, and broad implications of luminescence in fundamental biological research have become convincingly evident. The way has been paved by an increasing number of investigators, as well as by the ever expanding knowledge in various fields and the introduction of new methods of analysis and measurements. Among our contem- poraries, E. Newton Harvey is outstanding for his own extensive and penetrating studies, over a period of more than four decades, as well as for the work he has inspired in his students and colleagues. iii iv PREFACE Prominent also are the studies led by A. J. Kluyver in the Netherlands, and by Yata Haneda and others in Japan. The luminescence of bio- logical systems has now ripened into a fruitful area of general sig- nificance. Against this background, in brief, the Committee on Photobiology of the National Academy of Sciences-National Research Council, under the chairmanship of Sterling Hendricks and with the support of the National Science Foundation, recognized the mutual advantages of bringing together a group of leading investigators for a critical appraisal of present knowledge, for the first-hand interchange of ex- periences and ideas, and for the projection of likely approaches to unsolved problems. The conference that resulted was planned to include all aspects of bioluminescence, together with fundamental aspects of chemiluminescence and fluorescence. It was organized by a sub-committee composed of L. R. Blinks, E. N. Harvey, F. H. Johnson (Chairman), W. D. McElroy, and C. E. ZoBell, and it was held at Asilomar, near Pacific Grove, California, March 28-April 2, 1954. The number of participants was restricted, partly because of limitations of available funds, and partly because it was considered that, with the purposes in view, a small group would be most favorable to success. Although the conference was thus restricted, the papers and chief discussions are made available to all who may be interested through the publication of this book. The first paper, appropriately presented by Harvey, summarizes some important aspects of the present status of our knowledge of bioluminescence and the outlook for further advances. It includes an invaluable tabulation of all the large groups of animals and plants in which there are luminescent representatives, together with typical genera of luminous organisms, their habitat, the availability of histo- logical information about the luminous organ, and the demonstrated biochemical properties of the system, such as the necessity of oxygen or of ATP, the separability of a luciferin and luciferase, the color of fluorescence when present, the color of the bioluminescence, and the susceptibility to inhibition by light. After the introductory survey, various avenues of approach, from the purely physical to the purely biological, are dealt with compre- hensively and in pertinent detail. PREFACE V In their paper on luminescence spectroscopy, Becker and Kasha deal with the primary step for the utilization of energy by the chloro- phyll system, and in particular, the question of which electronic state of chlorophyll is involved in the primary step. Omitting unessential details of tlie physical theory, they consider chiefly the role of n-n-- transitions in the chlorophylls, the importance of intercombinations in these molecules, and the possible interaction of the ethylenic potential function with the electronic transitions in the chlorophylls. In the first of the two papers that follow, Arnold briefly presents some new data on delayed light production in green plants, together with a hypothesis to account for the observation that light saturation of the delayed light emission occurs at lower intensities than light saturation of photosynthesis. French then gives a summary discussion of fluorescence spectrophotometry of major photosynthetic pigments, including phycoerythrin, bacteriochlorophyll, chlorophylls a, b, and c, protochlorophyll, phycobilins and chlorophylls in the red algae, and a new leaf pigment that was discovered with a fluorescence micro- scope. An apparatus for automatically plotting the spectral energy distribution of weakly emitting light sources, and its application to the study of pigments in living cells as well as in solution, is described and illustrated in adequate detail. The kinetic approach to the mechanism of chemiluminescence of the 2,3-dihydrophthalazine-l,4-diones (DPD's or phthalic hydrazides, of which luminol is probably the best known derivative ) is stressed in the paper by Wilhelmsen, Lumry, and Eyring. The theory of absolute reaction rates is applied in interpreting the rates of excitation, radia- tion, and quenching of the molecules involved in light emission. New data are included in regard to the relationships of the reaction to oxygen, and the nature of intermediary compounds and end products is discussed. Chemiluminescent, fluorescent, and absorption spectra in relation to molecular structure, with particular reference to DPD and naphtho- quinone derivatives, as well as dimethylbisacridinium nitrate, are dis- cussed in the joint paper by Spruit and Spruit-van der Burg. The in- fluence of factors such as temperature and pH is included, and the significance of the photochemical inactivation of bioluminescence in yi PREFACE seeking to identify the nature of the Hght-emitting molecule in living cells is as critically evaluated as present information permits. The biochemical approach is emphasized in the next four papers, which deal with the luminescent systems of Cypridina, the firefly, and luminous bacteria. Tsuji, Chase, and Harvey discuss the most recent chromatographic and electrophoretic experiments which have resulted in highly purified Cypridina luciferin and have made it possible to show that such preparations exhibit not only fluorescence but also, in certain solvents, phosphorescence. Changes in the absorption spec- trum resulting from oxidation, or from changes in pH, are described. Data on the infrared absorption spectrum are contributed in the dis- cussion by Mason. Products of hydrolysis furnish a partial clue as to the nature of the luciferin, although the structure of the molecule remains to be established. The biochemistry of firefly luminescence, including the most recent advances, is summarized in the paper by McElroy and Hastings. The role of activators, including ATP, pyrophosphate, and Mn or Mg ions, as well as the action of inhibitors and the influence of variations in concentration of reactants, variations in temperature, pH, oxygen ten- sion, and other factors, is considered in detail. On the basis of the available evidence, a reaction scheme is proposed to account for the control of either steady state or flashing luminescence in both the intact organ and in extracts of the firefly. Additional data on the chemical and physical properties of firefly luciferin are contributed in a brief paper by Strehler, including evidence, from mass spectroscopy of chromatographically purified luciferin, suggestive of a dipyrimidopy- razine nucleus. The biochemistry and mechanisms controlling the emission of light in extracts of luminous bacteria is treated at some length by Strehler, who recently succeeded in obtaining easily visible luminescence in cell-free preparations of these organisms. The roles of flavine mono- nucleotide, coenzyme I, and a long chain aliphatic aldehyde, all of which appear to be necessary for a long-lasting, bright luminescence of the enzyme preparation, are discussed. A detailed reaction scheme postulating terminal reactions with a peroxide is presented to account for the available data, including the effects of temperature and hydro- static pressure. Considerations bearing on this scheme, in particular. PREFACE vii and on the relationships between free energy, activation energy, and emission of visible light in chemiluminescent reactions in general are set forth in discussions by Eyring, Mayer, and Kauzmann, respec- tively. Further data on the bacterial system, based on considerably purified bacterial luciferase, are contributed in the paper that follows by Hastings and McElroy. The physical chemistry of activation and inhibition of intracellular luminescence, with special reference to the fundamental action of temperature, hydrostatic pressure, and chemical agents on the reac- tion rates and equilibria involved in the process of light emission and other biological processes is discussed, apart from the mathematical details of the quantitative theory, in the paper by Johnson. The em- phasis is on points of general interest, with a somewhat more detailed discussion of the significance and interp)retation of recent data on the kinetics of luminescence in cell-free extracts. Physiological control of luminescence in animals of varying degrees of complexity, from protozoa and coelenterates to animals with a central nervous system, including worms and fishes, is dealt with by Nicol. Quantitative data on the characteristics of the luminescent response to electrical stimulation in representative types illustrate facilitation, fatigue, and other phenomena familiar in neuromuscular physiology, as well as some effects that are not so generally familiar. Problems in the analysis and interpretation of the physiological control of luminescence in various types of animals are critically examined in the following paper by Buck. A great diversity of luminescent organisms, terrestrial as well as marine, found in Japan and the Far East is described by Haneda in a comprehensive account that includes first hand observations on distribution, ecological relationships, morphology, histology, physi- ology, natural history, and other aspects of distinct biological interest. An extensive list of references gives access to the important Japanese literature. The ecology of marine dinoflagellates, with particular reference to the biological and environmental factors associated with the occurrence of "red water" conditions, is presented in the paper by Ryther. The discussion includes the influence of nutrient requirements, salinity, temperature, and air and water currents. Instances of red water con- viii PREFACE ditions that have been recorded in various parts of the world from 1891 to 1945 are summarized in a table, along u'ith specific organisms involved, the presence or absence of luminescence, the occurrence of mass mortality, and brief notes on the ecological conditions. The final paper by Haxo and Sweeney deals with the cultivation and some of the physiological characteristics of the photosynthetic, marine dinoflagellate, Gonyaulax polyedra, which occasionally gives rise to remarkably vivid displays of luminescence in the coastal waters of southern California. It is both pleasant and appropriate to record grateful acknowledg- ment of the assistance rendered in many ways toward the success of the Conference: by the National Science Foundation, the National Academy of Sciences-National Research Council, the members of the sub-committee on arrangements, the management of Asilomar, the authors of the papers and discussion, and the publishers. The editorial task has been lightened in various ways, especially by the prompt and efficient cooperation of the authors and of all others concerned with the publication. Editorial changes in manuscripts were rarely made and they were chiefly for correcting a few obvious errors. The editor must assume full responsibility, however, for the author, genera and species, and subject indexes, as well as for whatever errors, shortcom- ings, or usefulness may be found in them. F. H. J. Princeton, N. /. January 1, 1955 Members of the Conference on The Luminescence of Biological Systems Asilomar, Pacific Grove, California, 1954 Dr. Rubert Anderson, Medical Laboratory, Army Chemical Center, Maryland Dr. William Arnold, Oak Ridge National Laboratory, Oak Ridge, Tennessee Dr. E. R. Baylor, Zoology Department, University of Michigan, Ann Arbor, Michigan Mr. Ralph S. Becker, Department of Chemistry, Florida State Uni- versity, Tallahassee, Florida Dr. Lawrence Blinks, Hopkins Marine Station, Pacific Grove, Cali- fornia Dr. John Buck, National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Maryland Dr. Havre Carlson, Office of Naval Research, Washington, D.C. Dr. Aurin M. Chase, Department of Biology, Princeton Universit)% Princeton, New Jersey Mr. William V. Consolazio, Program Director for Molecular Biol- ogy, National Science Foundation, Washington, D.C. Dr. Demarest Davenport, University of California, Santa Barbara, California Dr. L. M. V. DuYSENs, University of Utrecht, Utrecht, Holland Dr. Henry Eyring, Department of Chemistry and Graduate School, University of Utah, Salt Lake City, Utah Dr. C. Stacy French, Department of Plant Biology, Carnegie Institu- tion of Washington, Stanford University, Stanford, California IX X CONFERENCE MEMBERS Dr. Arthur Giese, Biology Department, Stanford University, Stan- ford, California Dr. Yata Haneda, Tokyo Jikeikai Medical College and Yokosuka Museum, Yokosuka City, Japan Dr. E. Newton Harvey, Department of Biology, Princeton University, Princeton, New Jersey Dr. J. Woodland Hastings, Department of Biological Sciences, North- western University, Evanston, Illinois Dr. Francis T. Haxo, Scripps Institution of Oceanography, La Jolla, California Dr. Frank H. Johnson, Department of Biology, Princeton University, Princeton, New Jersey Dr. Michael Kasha, Department of Chemistry, Florida State Uni- versity, Tallahassee, Florida Dr. Walter J. Kauzmann, Chemistry Department, Princeton Uni- versity, Princeton, New Jersey Dr. Howard S. Mason, Biochemistry Department, University of Oregon Medical School, Portland, Oregon Dr. William D. McElroy, McCollum-Pratt Institute, Johns Hopkins University, Baltimore, Maryland Dr. J. A. C. NicoL, Marine Biological Laboratory, Plymouth, England Dr. C. B. van Niel, Hopkins Marine Station, Pacific Grove, California Dr. John H. Ryther, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts Dr. C. J. p. Spruit, Landbouwhogeschool, Wageningen, Holland Dr. H. Burr Steinbach, Assistant Director, Division of Biological and Medical Sciences, National Science Foundation, Washington, D.C. Dr. Bernard L. Strehler, Department of Biochemistry and Institute for Research in Biophysics, University of Chicago, Illinois Dr. Beatrice M. Sweeney, Scripps Institution of Oceanography, La Jolla, California Dr. Frederick I. Tsuji, Department of Biology, Princeton University, Princeton, New Jersey Mr. Paul C. Wilhelmsen, Chemistry Department, University of Utah, Salt Lake City, Utah CONFERENCE MEMBERS xi Dr. E. J. Ferguson-Wood, Marine Biology Laboratory, Cronulla, New South Wales, Australia Dr. Claude E. ZoBell, Scripps Institution of Oceanography, La Jolla, California Contents H Survey of Luminous Organisms: Problems and Prospects E. Newton Harnty 1 Luminescence Spectroscopy of Molecules and the Photosynthetic System R. S. Becker and M. Kasha 25 Light Saturation of Delayed Light Production in Green Plants William Arnold 47 Fluorescence Spectrophotometry of Photosynthetic Pigments C. Stacy French 51 Kinetics of Chemiluminescence of the 2,3-Dihydrophthalazine-l, 4-diones Paul C. Wilhelmsen, Rufus Lumry, and Henry Eyring 75 Spectroscopic Investigations of Luminescent Systems C. J. P. Spruit and A. Spruit-van der Burg 99 Recent Studies on the Chemistry of Cypridina Luciferin Frederick I. Tsuji, Aurin M. Chase, and E. Newton Harvey 127 Biochemistry of Firefly Luminescence W. D. McElroy and J. W. Hastings 161 Firefly Luciferin Bernard L. Strehler 199 Factors and Biochemistry of Bacterial Luminescence Bernard L. Strehler 209 Purification and Properties of Bacterial Luciferase J. W. Hastings and W. D. McElroy 257 Inhibition and Activation of Intracellular Luminescence Frank H. Johnson 265 '^ Physiological Control of Luminescence in Animals J. A. C. Nicol 299 \ xin >A J xiv CONTENTS Some Reflections on the Control of Bioluminescence John Buck 323 Luminous Organisms of Japan and the Far East Y. Haneda 335 Ecology of Autrophic Marine Dinoflagellates with Reference to Red Water Conditions John H. Ryther 387 Bioluminescence in Gonyaulax polyedra F. T. Haxo and Beatrice M. Sweeney 415 Author Index 421 Genera and Species Index 427 Subject Index 435 Survey of Luminous Organisms: Problems and Prospects E. Newton Harvey Department of Biology, Princeton University, Princeton, New Jersey One of the first questions to be settled on discovery of a living lumi- nous organism is whether the species is truly self-luminous, with light from a chemiluminescent system of its own, or whether the light comes from luminous bacteria. The answer is important because light emission from luminous cells is usually associated with coccus-like granules, which look like bacteria, and the existence of symbiotic luminous bacteria in a number of species has been definitely estab- lished. There are in fact three categories of luminous bacteria associated with organisms. In addition to the saprophytic varieties, growing on dead fish or flesh, parasitic bacteria occasionally attack living sand- fleas, midges, and caterpillars, giving rise to a luminous disease which is usually fatal to the infected individual. Nevertheless, the animals are active while living, and have been reported as luminous species. A much more important and widespread type of luminescence re- sults from bacterial symbiosis, common among the squid and the fish. Every individual of a species whose light is associated with symbiotic luminous bacteria must be luminous. In addition, the host often pos- sesses complicated luminous organs in which the symbionts grow. Although the luminous bacteria emit a continuous light, the fish or squid may develop special devices, movable screens, or migrating chromatophores by which the bacterial light can be obscured. The most striking instances of bacterial symbiosis occur in the Indonesian fish, Photobleplioron and Anomalops, two genera of the 2 SURVEY OF LUMINOUS ORGANISMS family Anomalopidae, from the Banda Islands. These fish are charac- terized by a white spot under the eye, made up of a large group of elongated cells, richly supplied with blood vessels, forming a distinct oval organ in which the symbiotic luminous bacteria grow. The light is continuous but can be cut off — in Photoblepharon by a black fold of skin which is drawn over the organ like an eyelid, in Anomalops by rotating the organ so that the light shines against the body, rather than out toward the surroundings. Nothing is known of the embry- ology of these fish, but all individuals are luminous and the bacteria could have entered the light organ from the sea water. There are open- ings between the region of luminous cells and the environment. Usu- ally fish or squid with symbiotic luminous bacteria possess such open luminous organs, often with a long duct through which the bacteria can be squeezed, as in the fish Malacocephalus. Symbiotic luminous bacteria often require a special culture medium for luminescence, whereas saprophytic forms will live on any dead animal. Table I (column 2, type of light) presents the distribution of all groups (arranged phylogenetically ) in which species have been re- ported as luminous while living. The luminous forms in which bacterial symbiosis (Bs) or bacterial parasitism (Bp) have been observed, are contrasted with self-luminous forms, marked E or I. In the first column of the table, the habitat of the luminous groups is indicated, whether marine, M, fresh water, F, or terrestrial, T. It will be observed that bacterial symbiosis is most common among the fish. Sometimes it is difficult to distinguish between harmless parasitic luminous bacteria and the symbiotic variety. Certain squid usually contain luminous bacteria in a special gland, but in some individuals no light is visible on dissecting and opening the gland. In such cases how can one decide between symbiosis and harmless penetration by luminous bacteria? It was once supposed that all bioluminescences were of bacterial origin, but it is now quite certain that symbiosis among luminous animals is not nearly as widespread as had been supposed. The reader is referred to the work of Pierantoni ( 1936, 1939 ) ' and of Buchner ( 1953 ) for a discussion of this field. " Literature prior to 1951 will be found in the bibliography of Harvey in Biohwnnescence, Academic Press, New York, 1952. Only later references are given in this paper. E. NEWTON HARVEY 3 In addition to the genera listed in the table, occasional reports of luminous species are to be found in the literature for the following groups: Ascomycetes (Xijlaria); Phallales (Ileodictyon, Kalchbren- nera); Turbellaria {Monocelis); Polychaeta {Nereis, Pohjopthahnus); Polyzoa {Memhranipora, Flustra, Electra, Acanthodesia*' ) ; Chaetog- natha (Sogitta); Pteropoda (Creseis, Styiola, Clio, Cavolinia); Proso- branchiata (Pterotrochea, Tonna); Octopoda {Cirrothauma, Ele- doneUa); Pycnogonida (Colossendeis); Araneae (Spiders); Isoptera (termites); Lepidoptera (Arctia); Asteroidea (Brisingia); Ascidiacea (Ciona); Chimaerae (Chimaera); Teleostomi {Parehippus, Riwettus, Exocetus, Xenodermichthys, Halosauriis, Bassozetus, Leucicorus, Ma- criirus, Lampr 0 gram mil s, Mixonus, Malthopsis, Ipnops). Nothing is known of the nature of the hght emission and confirmation of lumi- nosity is to be desired. The title of this conference, "The luminescence of biological sys- tems," suggests that self-luminous organisms are the principal ones to be considered and that the chemistry of the various luminescent systems should receive first attention. How may the more than 70 quite different groups of luminous organisms listed in Table I be arranged on such a basis? It is always difficult to classify a subject, because those familiar with the field may disagree regarding the proper basis for classification, and because of the usual exceptions to every rule. One basis for characterizing luminescent organisms depends on whether the light emission takes place within the cell, intracellular, I, or the luminous substances are secreted to the outside, extracellular, E. The groups belonging in these categories are indicated in Table I, column 2. Perhaps a more fundamental grouping, for which there can be httle criticism on biological grounds, is (1) luminous plants and (2) lumi- nous animals. This classification not only reflects the fundamental difference in chemistry in the animal and vegetable worlds, but is also particularly cogent because in the plant kingdom, whose only luminous representatives are bacteria and fungi, f light production is ' See K. Kato, Do/;t/f5ugafcu Za5si, 59, 9-10, 1950. f The chemiluminescence following photosynthetic processes, invisible to the dark adapted eye, which has been described by Strehler (1951) and Strehler and Arnold ( 1951) is in a different category from bioluminescences. 4 SURVEY OF LUMINOUS ORGANISMS continuous and independent of stimulation. The luminescence intensity varies only with environmental changes, such as temperature, pH, and salt content of the medium. Among animals, no light appears until the luminous region is excited in some way — by nerves or directly by mechanical, electrical, and chemical stimulation. The resulting lumi- nescence is momentary or of short duration. In a few fish (Poriclithys, Echiostoma, Maiirolictis, Argyropelecus) light emission of photophores is hormone (epinephrin) controlled and the light lasts a longer time than in the case of direct nerve stimulation. No bacteria or fungi are known, which can be stimulated to lumi- nescence, but a few animals, whose light is not due to luminous bacteria, can luminesce continuously. These exceptions to the general rule are found in the diplopod millipede Ltiminodesmtis (Davenport, Wootton, and Gushing, 1952), and in various stages of firefly develop- ment. The firefly egg and the pupa are continuously luminous, as are adult cream-colored wingless females of the related beetle, Phengodes. The luminous organs of the females may remain brightly luminous for days, with no change in intensity. Lwninodesmus is continuously luminous, with no voluntary control, and luminous bacteria appear to be absent, at least they have not been demonstrated. Light emission of both larval and adult luminous organs of the firefly is controlled by nerves, but during the pupal stage, the larval organ ceases to function and the adult lantern of the firefly develops as a wholly new structure, in a new position on the abdominal seg- ments. It is completely reconstituted by luminous wandering cells ( not subject to nerve stimulation ) whose light shines through the chitinous integument. Protozoa respond directly to stimulation, usually to mechanical disturbance resulting in the "phosphorescence of the sea," but meta- zoan* luminous cells are supplied by nerves, and reflex luminescence All other reported cases of luminescence of plants are due to reflection or in the case of marine algae, due to luminous organisms growing on the sea- weed. Luminous dinoflagellates or peridineae are regarded as animals for this discussion because they luminesce only on stimulation. " Among sponges no electrical excitation of luminescence could be elicited, but response occurred on rubbing the sponge (Harvey, 1921). Sponges do not possess nerves. E. NEWTON HARVEY 5 occurs, either as a result of secretory nerve stimulation in forms with extracellular luminescence, or from an effect similar to the action of a motor nerve on a muscle, from the impulse of what might be called a "luminor nerve," rather than a "luminous nerve." The recording of light from stimulation of nerves or luminous tissue is a field of special interest, first studied with luminous beetles (Snell, 1931, 1932; Brown and King, 1931; Harvey, 1931), and recently greatly extended by Nicol (1952, 1953), and by Harvey and Chang (1954). The results will be considered in detail in this volume. Another field of considerable interest which should be investigated with modern equipment and microelectrode technique is the electrical change accompanying luminescence. A beginning has been made by Hasama ( 1939-1944 ) , who has published "electroluminograms" repre- senting the electrical potential change during luminescence of the glowworm, the firefly, the worm Chaetopteriis, the mollusc Plocamo- phorus, and the pennatulid Cavernularia. No physiological or biochemical investigation can be considered adequate without a knowledge of the histology, particularly the fine structure of luminous cells or tissue. The extent to which this is known among various luminous groups and the regions for further research are indicated in Table I, column 3. That the light generally arises from cellular granules is a most significant and important observation deserving further study. Bacterial and fungal luminescence, in which steady state conditions determine the amount of photogenic material undergoing change, are probably the most difficult to analyze from the chemical standpoint. Precursors of the photogen, as well as reaction products and various enzyme systems, may be expected to influence the continuous lumines- cence. A continuation of the kinetic studies on bacterial luminescence initiated by Johnson and collaborators (see 1954) is much to be de- sired. It is also to be expected that extracellular luminescent systems should be different from intracellular ones as in bacteria or those of the firefly. In the firefly, the cell reactions are designed for building up a supply of photogen which suddenly reacts with a flash of light fol- lowing stimulation. Like bacteria, the firefly luminous cells must con- tain the precursors of the photogen and also the reaction products, as well as various enzyme systems. 6 SURVEY OF LUMINOUS ORGANISMS In an organism with extracellular luminescence such as the ostracod crustacean Cypridina, gland cells are filled with the finished secretion to be extruded to the sea water by the contraction of muscles. In such a case we may expect the principal luminous substances directly concerned with light production to be present, with a minimal amount of precursor. The reaction products form in the sea water. Cypridina does possess the simplest chemical system thus far investigated, one in which the only recognized substances are oxygen, a heat stable, dia- lyzable, oxidizable substance, luciferin and a thermolabile enzyme, luciferase, specific for the light-emitting oxidation of luciferin in aqueous solutions. In the Cypridina gland, the luciferin and luciferase are manufactured in separate gland cells. Since the chemistry of lumi- nous bacteria, of the firefly, and of Cypridina is individually repre- sented in this conference, a discussion of the accessory factors involved in bacterial and firefly luminescence will be found in the appropriate chapter. Studies on luciferase kinetics, similar to those made by Chase ( 1946-52 ) with Cypridina luciferase, should be extended to other organisms. The ability to demonstrate the luciferiu-luciferase reaction according to the method described in the explanation of Table I is presented in column 5. In addition to Cypridina, fireflies, and bacteria, only the marine fireworm of Bermuda {Odontosyllis enopla), the deep sea shrimp (Systellaspis and Heterocarpus), and the fresh water limpet* {Latia neritoides) , of New Zealand, have been reported as positive for a luciferin-luciferase reaction. Although firefly biochemistry will be considered in another chapter, it must be pointed out that in the firefly luminescent system, adenosine triphosphate (ATP) is an important accessory substance. This can be demonstrated by the fact that a cold water extract of the luminous tissue of the firefly allowed to stand until all light disappears, will again luminesce when ATP is added (McElroy, 1947, 1951). A sys- tematic test of similar ATP action in other luminous organisms has been made by Harvey and Haneda (1952) and Haneda and Harvey ( 1954) and is included, together with later tests, in Table I, column 6. The importance of ATP in luminescent reactions appears to have • According to Bowden (1950). E. NEWTON HARVEY 7 been definitely established only among the elaterid and lampyrid beetles. The absence of an ATP reaction does not necessarily mean that ATP plays no part in light production (McElroy and Harvey, 1951). It ma\' indicate that other components of the luminescent system are lacking and that further analysis may be necessaiy to designate the complete system. Similarly, the absence of a luciferin-luciferase reac- tion may indicate that accessory substances are lacking or that lucif- erin and luciferase are particularly unstable substances in the group of luminous organisms tested. In view of the recent work (Strehler, 1953; Strehler and Cormier, 1953; McElroy, Hastings, Sonnenfeld, and Coulombre, 1953; Strehler, Harvey, Chang, and Cormier, 1954) on bacterial luminescence, where the luminous system is complicated by accessory substances such as long chain aldehydes, as well as in firefly luminescence requiring high energy phosphates, a redefinition of lucif erin becomes necessary ( Har- vey and Tsuji, 1954). It is no longer sufficient to claim that luciferin is present in a boiled extract of luminous tissue, whereas the dark cold water extract contains luciferase. McElroy has demonstrated that dark cold water extracts of firefly (Photinus pyralis) lanterns emit no light when purified pyralis luciferin is added, but do luminesce with ATP. Therefore the cold water extract (luciferase) lacks ATP in- stead of luciferin. ATP is the limiting factor under these conditions. Rather than placing the emphasis on the limiting factor, or on heat stability or dialyzability, as has been done previously, light emission should be the criterion for luciferin. In the case of luminous organisms requiring dissolved molecular oxygen for luminescence, luciferin may properly be defined as the oxidizable substance supplying molecules capable of absorbing enough excess energy from a chemical reaction to emit in the visible region. Such a definition implies that some form of luciferin molecule — either free base or acid, either dissociated anion or cation, in reduced or oxidized form, either free or combined with protein, like a prosthetic enzyme group — can pick up the energy of the oxidative reaction in which it is involved. Such a definition does not mean that luciferin is the same substance in different luminous animals, nor does it necessarily designate luciferin molecules them- selves as the ones which emit, but it does imply that a related mole- 8 SURVEY OF LUMINOUS ORGANISMS cule, such as a luciferin-luciferase combination, or an oxidized lucif- erin molecule, or a molecule of an intermediate step, is the emitter. The actual molecule emitting might be designated the "photogen." It has long been recognized that a substance whose molecules are readily excited to fluoresce by the energy of radiation is most likely to be chemiluminescent from the energy of a chemical reaction. Bac- terial luciferin, firefly luciferin and Cypridina luciferin are all fluores- cent, and it was early observed (Harvey, 1925) that fluorescence of the light organs of luminous animals is a widespread phenomenon. The fluorescence of ctenophore luminous organs is particularly notice- able immediately after the bioluminescence has ceased (Harvey, 1925, 1926). The distribution of marked fluorescence in the luminous organs of various groups is given in column 8 of Table I. In addition to bacteria, Cypridina, and fireflies, the only other lumi- nous system which has received chemical attention in recent time is that of earthworms, studied by a group of Czech investigators, Komarek, Backovsky, and Wenig (see Wenig, 1946). They have demonstrated the presence of riboflavin, not flavin phosphate or flavin adenine dinucleotide (Wenig and Kubista, 1949), in the yellow lym- phocytes of the luminous earthworm Eisenia suhmontana, as well as in those of a nonluminous form, E. foetida. The luminous lymph of E. submontarm fluoresces yellow- greenlike riboflavin until the bio- luminescence has disappeared, at which time the fluorescence color changes to blue, that of lumichrome. A corresponding change in the yellow-green fluorescence of the nonluminous lymph of E. foetida does not take place. Consequently the Czech workers first postulated that the bioluminescence of the earthworm is connected with a change from riboflavin to lumichrome, a reaction which does not occur in the nonluminous species. Later they state that the molecules of ribo- flavin are believed to be "absorbed in an oriented layer on the surface of granula of lipoid character . . . [and] the activation energy which brings them into an excited state is probably derived from an oxidative reaction in which molecular oxygen takes part." It is interesting to note that the luminous granules of the earthworm are yellow. The luciferin of Cypridina is also yellow (although its luminescence is blue ) , and a yellow color has been observed associated with luminous cells in at least seven additional groups — Hydrome- E. NEWTON HARVEY 9 dusae, Polychaeta ( Tompteridae and Terebellidae ) , Nudibranchia, Copepoda, Chilopoda, Lampyrid fireflies, and Macruroid fish. In many other kiminous organisms a yellow color cannot be estab- lished with certainty although a faint yellow would be difficult to detect. Of particular interest is the cephalic luminous organ of the railroad worm (PhrixotJirix), whose bioluminescence is a bright red, quite similar to the fluorescence of hematoporphyrin and to the chemi- luminescence of metal porphyrin compounds. One might expect to find a faint red tinge in the organ from a porphyrin, or a red fluores- cence, but the tissue appears quite colorless to the eye and is non- fluorescent in ultraviolet or in yellow to violet light (Harvey, 1944, 1945). It is certain that no red color screen is involved in the red luminescence of PhrixotJirix. Another method of grouping luminous organisms involves the neces- sity or non-necessity of oxygen for luminescence. The author (1926), in a systematic study of oxygen requirements for luminescence of various groups, was amazed to find that the ctenophores Beroe and Eiicharis, the scyphomedusan Pelagia, and the radiolarians, Thalassi- cola and Colozoum, require no dissolved oxygen for light production. The result has been confirmed for another ctenophore, Mnemiopsis (Harvey and Korr, 1938), and for the hydromedusan Aeqtiorea (An- derson, private communication, 1939). It should be emphasized that it is the luminescent system of extracts of various ctenophores which emits light without dissolved oxygen, whereas the ability to stimulate a ctenophore to luminescence through nerves may be lost in absence of oxygen (Chase, 1941). The relation between oxygen pressure and luminescence intensity should be studied for all organisms (see Has- tings, 1952, 1953). Another basis for classification of luminous groups depends on their relation to light, whether light inhibits the luminescence or not. Such a division might involve inhibition of a nerve-stimulating mechanism, or a photochemical action on the chemiluminescent system. Such an efiFect of ultraviolet light and photosensitized visible light on Cypridina luciferin will be referred to in the chapter on Cypridina chemistry. The most important and best known case of light inhibition is to be found among ctenophores. They do not luminesce in sunlight or strong electric light but regain the ability after some twenty minutes in the 10 SURVEY OF LUMINOUS ORGANISMS dark. The effect of strong light is on the luminescent system. Weak light will inhibit the excitation mechanism (Moore, 1924). No day- night rhythm of luminescence has been established in the ctenophore, Mnemiopsis, but among dinoflagellates, a day-night rhythm has been described. The complicated details of light inhibition have been de- scribed in Bioluminescence (Harvey, 1952), and the action on various groups is summarized in Table I, column 7. This survey of luminous organisms has indicated certain well-defined biochemical groups, but there are many others about which too little is known to attempt a logical separation. Luminous animals are scat- tered over the evolutionary tree from Protozoa to fish without any indication of direct evolutionary descent. One species of a genus may be luminous and another not. One variety of a fungus may be lumi- nous and another not. The existence of nonluminous mutants of lumi- nous bacteria has been known since the work of Beijerinck (1912). The author ( 1932, 1953 ) has taken the viewpoint that luminescence has arisen independently in various phyla of the animal kingdom, probably from some slight change in chemical systems common to cells in general, most likely the cell-respiratory systems. There is every evidence that the luciferins of fireflies, bacteria, and Cypridinae are chemically quite different and must be prefixed by the word firefly luciferin, bacterial luciferin, etc., in order to designate them. The bioluminescence emission spectra of various organisms range in the wavelength regions of red to violet, just as do the chemiluminescence spectra of quite unrelated organic compounds — metal prophyrins, red; pyrogallol, yellow; dimethyldiacridinium nitrate, green; 3-aminophthalic hydrazide, blue luminescence. Substituted groups on these compounds will change the chemiluminescent inten- sity markedly and the maximum wavelength of emission slightly, but thus far spectral energy curves have given little clue to chemical structure. It must be emphasized that small shifts in maximum emission of bioluminescences have little meaning, since the presence of absorbing pigments in cells and the phenomenon of differential absorption may change the emission spectrum of any luminous animal. The latter effect is particularly well seen in Spruit- van der Burg's ( 1950 ) study of the relation of density of suspension to bacterial luminescence emission. E. NEWTON HARVEY 11 The maximum wavelength changes from 500 to 470 milhmicrons as the suspension density changes from 40 to 8 arbitrary units. The various species of luminous bacteria do possess different maxima in the blue region when measured in dilute suspension (Spruit- van der Burg, 1950). The light of luminous bacteria looks green to the eye when the intensity is sufficiently high to involve color vision, no doubt because the spectral energy curve is skewed, with greatest energy on the long-wavelengths side of the maximum around 480 millimicrons. In this respect bacterial luminescence differs from that of the blue Ci/pridina luminescence, whose maximum emission is about 480 millimicrons, but the spectral energy curve is more narrow and symmetrical. Because of the above considerations, bioluminescent emission spectra are not too significant. Nevertheless, as a guide to previous investigations, the measured maximum wavelengths and the reported color of the light of various groups have been collected in Table I, column 9. It would be ideal if luminous groups could be separated on the basis of the chemiluminescent systems involved, primarily depending on the structure of luciferin. At the present time that is not pos- sible, but certain organisms can be classed together with reasonable certainty on the basis of similar chemical behavior. There is little doubt but that the fungi and the bacteria contain similar luminescent systems, although the role of flavins in fungal luminescence is as yet not demonstrated. So little is known of the biochemistry of luminescence of the Pro- tozoa that it is very difficult to predict a similarity with any of the known chemiluminescent systems. Noctiluca and smaller forms can often be obtained in enormous numbers and should offer good material for chemical research, except that the percentage of luminous sub- stance is undoubtedly small. In Noctiluco, light is always associated with minute granules, which flash on stimulation and emit a steady glow under conditions of injury to the cell. No form is better adapted for cell physiological studies, particularly of the excitation process. Noctiluca is an ideal organism for the investigation of the all-or-none law, time relations of the flash, repetitive stimulation, fatigue, conduc- tion of local excitation, etc., in a single cell. Among other advantages, Noctiluca is large and nearly spherical, 0.5-1.0 mm in diameter, pos- 12 SURVEY OF LUMINOUS ORGANISMS sessing a single nucleus and a sap vacuole into which substances may be injected or sap withdrawn. It also possesses a flagellum and a ten- tacle, whose movement can be studied. In addition there are visible protoplasmic changes on electrical stimulation. There are also interesting osmotic relations in Noctiluca, connected with a specific gravity less than that of sea water. The low density is due to a definite salt content, not to oil droplets. Permeability, movement, and light emission can all be studied together in this single cell — a most unusual situation. The relation between oxygen pressure, temperature, hydrostatic pressure, pH, salts, drugs, etc., and light emission and movement should be worked out in detail. The smaller dinoflagellates are not as favorable material as Nocti- luca for the cell physiologist. However, the relation between oxygen pressure and light emission should be studied carefully. The Radio- laria do not require oxygen for luminescence, and this peculiarity may be more widespread among Protozoa than is now realized. Day- night rhythms of luminescence might be carefully tested in all species. Should it prove feasible to culture luminous Protozoa in large numbers under laboratory conditions, they should be as valuable for biochemi- cal work as the luminous bacteria. The great groups of luminous coelenterates and ctenophores behave alike in many ways. The luminous tissue can be ground in sea water to a dark extract that will give a brilliant light when added to fresh water. Luminescence comes from granules which dissolve with light emission as a result of treatment with many agents, such as saponin, which cause the cytolysis of cells. The chemistry of these forms is completely unknown and offers a virgin field for investigation, since material can no doubt be prepared by modern methods of freeze- drying. As we have seen, in some species dissolved oxygen is not necessary for luminescence and in other species photochemical changes in the luminescent system prevent luminescence after the animal is exposed to daylight. In addition, the physiology of luminescence in medusae, pennatulids, and ctenophores presents many problems connected with excitation of light by luminor nerves and reflex transmission of impulses. A begin- ning has been made by the work of Parker (1920), Buck (1953), and Harvey and Chang (1954), but much more needs to be done. The E. NEWTON HARVEY 13 wave of light which spreads over a pennatuHd colony stimulated at one point was observed by L. Spallanzani in 1783 and has excited the interest of investigators ever since. Marine annelids are much more diverse than the coelenterates in methods of light production. Some secrete from epidermal glands, some exhibit intracellular luminescence. The color of the luminescence varies from yellow to blue. A luciferin-luciferase reaction has been demonstrated in Odontosyllis, and this worm offers special advantages for chemical study, as it swarms in large numbers at Bermuda and other places in relation to phases of the moon. In annelids, the nerv- ous control of luminescence is particularly favorable for study and has been thoroughly investigated by Nicol in Chaetopterus (1952) and in polynoid worms ( 1953 ) . Earthworms are a favorable group for chemical studies. Reference has already been made to the Czech investigations. In addition to the part played by flavins in luminescence, attention should be par- ticularly directed to the fact that one species of the genus Eisenia may be luminous and another species not, even though closely related. What is lacking in the nonluminous species? Cross breeding should be attempted. Study of such intercrosses between varieties of the fungus Paniis stipticus has already given interesting results (Macrae, 1942). Among molluscs, very little has been added to the chemistry of light production since the work of Dubois, which is summed up in his 1928 article in Richet's Dictionnaire de Physiologie, Vol. X. Pholas is extraordinarily favorable for chemical work, and should be attacked by modern methods. Self-luminous squid offer many species with complicated luminous organs for physiological work, but they are mostly deep sea forms, rare and unfavorable for chemical studies, with the exception of Watasenia scintillans, which breeds in enormous numbers during April-June in Toyama Bay on the western coast of Japan. Nerve and possible hormone control of the lighting mechanism presents an untouched field for investigation. Among Crustacea, ostracods, copepods, mysids, and decapod shrimp all produce an external secretion, often in great abundance. It might be expected that luciferin from one species would react with luciferase 14 SURVEY OF LUMINOUS ORGANISMS from another species of all the groups, but such is not the case. No luminescence appears when luciferin of Cijpridina is mixed with solu- tions which should contain luciferase from copepods, or from decapods, and vice versa. Only if the luciferin is from another genus of ostracods (Pyrocypris) will it react with Cypridina luciferase, or will the re- verse "cross" be positive. This specificity of the luciferin-luciferase reaction is widespread in the animal kingdom. The chemistry of all the Crustacea with external luminous secretions should be investigated in detail. Crustacea with scattered photophores present problems of physi- ological interest, but do not appear to be favorable for chemical work, because of the small volume of luminous material. It is possible that the photophores are hormone controlled. This relation and nerve control of lighting should be carefully investigated. Among myriapods, the facts concerning light production of the diplopod Luminodesmus has been presented by Davenport, Wootton, and Gushing (1952). Chilopods produce an external secretion in great abundance, which should be intensively studied. The chemistry may be similar to that in earthworms, but not enough is known to make a comparison. If luminous centipedes could be bred in cap- tivity they should be highly favorable forms for biochemical work. Among insects, apart from the Coleoptera, whose luminescence has been much investigated, the light of Collembola, Diptera (see Kato, 1953), and the controversial genus Fulgora should be studied in great detail — morphology, histology, physiology, and chemistry. Very little is known of light production in these rather uncommon luminous insects. That fireflies have been of great value for biochemical work is apparent from the report presented in another section. The rapid flash of a firefly still remains an unsolved problem in insect physiology. Behavior studies in different species of fireflies in regard to the use of the light (see Buck, 1937, 1948), present a promising field for the ecologist. The meaning and mechanism of synchronous flashing of tropical fireflies is still a mystery. Finally the spectral energy distri- bution of the light of various species should be explained. Since Coblentz' monograph of 1912, httle has been done to determine whether the range in color of the light from orange to yellow green E. NEWTON HARVEY . 15 is due to a characteristic of the chemiluminescent reaction or to ab- sorption by pigments. The extreme range of color appears in the railroad worm Phrixothrix with both yellow lights and also a bright red light in the head, which is not connected with any red pigment detectable with the eye. The ophiuroids, despite considerable histological study, present many unsolved problems of physiological and biochemical nature, and the same may be said for the balanoglossids and the tunicates. Luminous species are abundant among surface marine organisms and are easily obtainable. The considerable number of fish whose light is not due to luminous bacteria are in the same category as the squid. They are mostly deep sea forms with photophores, not favorable for chemical work but presenting many physiological problems in connection with nerve and hormone excitation. They deserve intensive investigation, whenever the occasion allows. In fact it seems certain that luminous organisms as a whole can supply much information important for an under- standing of fundamental processes in even.' field of biology. PQ < 03 § in -r " § en o O O C O ^ ^ ;feO (^ GO 05 c _o CJ 03 »^ C i '^ O) OJ *^ >-, o r^ Oh:1<: o ■^ lo CO >> O 73 O o o OJ 02 Q, >i rC H t f light ?dge 0 -a c c3 03 0 — 3 .t^ ^ ^ ?, ^ ao 03 >, G 0 WHW ^ iM CO 01 00 o iC '^ ro (M c3 05 c o s o c 'a 3 0 r^ Tt< 3 -a LO r^ 0 03 C M< + I I + + +1 fa On O c3 m o O -C pQ bO Tt* to 3 0 3 tH ■J3 S3 0 •? ^ -ki ■*i 0 00 a • ^H EC a s ^ ^ 3 a fa I o '«3 fl- ee JZ CO CO T- ■3 I I I + + I + +1 +1 +1 s g o o "5 e s o 6^ ^ a s 2 7= ^ 03 O ^ a, '3b 2" c s 3 •" fa 2 >> -a > '^ G CO o «J Si = -g e? c3 o o S in o3 T3 ^ C3I S " Ji t- O 5r h3 .a^ s o -a 53 o o3 in 3 -o a '^ o CO o S2 -SS ■£ Oh S C • T2 O e s ci e s • >; ***» St. S « E c ;3 bO fa pq "S o + + I I I I + I +1 +1 I +1 O- § §§ §§ 8 O eo" Oh :9i as cq 03 CD 3 -a 6 h OJ 03 ^ a o a; M fa g o3 o3 ■« ^ 0) tH «? S 03 X 2 16 J3 ^ ^ o ^ -3^33 = fQ >H pq >i pq pq cQ O ^ I ID ■3 3 c c CD 0) 0 -^ £ 2 1 1 ^ is J2 ■£ _c 0 7; 0 -^ tc -X f— 1 "3 '3 s 1 Yellow-green Yel (55( 3 4= ^ ^ "C £ 2 2 S 3 3 3 a; S S S o I I o -S 3 3 is -^ o .S :=! 3 ^.3 O -* I o 3 5 + + 1 I I + I + + + + 4- + + + + + + + + I +1 +1 I •+ + + + + + I ++ + + + + ^ a HH w w K w S§S^§^ §§H HP^SSg S S e o [^ _ ■+= :^ 01 O e Si. ^ o e Si a, e c ^ 01 . & < 35 o o -a a S o e o CO 55. o -0 -a o o 03 a; O s o |^_ a. 8 I i 2 .y 03 o o I O -S3 S5 a. o • '--a a; ^^ e c -si c3 o ^- o Si u -^ o3 .a K^ .s 4) § 2 .2 ••:3 Pi •«<1 _ -^ 0 chia, oiola -6 . 1 s s^ -^o =0- g ^ ••s* *^ - -^ << cc 3 CO "^ 1-s ■ 50 -S _^ :§ § -c: Lycot roteuth Rondel s,Sto Pyroc s~ 55i ^ '"^ -N •ri 03 e Pi, .C ^< cc -c 3 03 ^ > Q 0 ^ 5 i -^ 0 17 05 00 JS 4= r^ _x cc s IZ "S S2 J1 ■3 X ?^ ■3 a p &^ ■3 ^ ^ t~> 'U E-^ o S ra 0 >> 0 s IN. ^ -C CO 05 ■3 c 3^ 8 cc 1 >> ■3 -fi -= '5 o m cQ o I 3 C a; a; 3 S J3 t- o .ta o = ^ c ^ ^ 3 I S^ii OJ mO Clh ^ S c _0Q l» '3 (-• 0 m o o CQ CQ a I Q lO + I Tf + + + + + + o O CO +1 +1 +1 + + + I + I 1 I +1 (M w a o CQ 3 O c 3 a H -o c a 3 o O W HH a a §S S S § fe e- H H H HHHH HE- 00 o w 5i. e s s- »j 0 B 5ft "5 "S^ Pi" S 0 fc- fe ^^ o3 --^ i-i 0 HJ a 0 C 1^ -<-= .ST 4J a; 3 U CO I K P o v a o pi. o 18 s - "aS "5 r? C M ^ « O o . a a bJDS O 3 C GO 03 ■5 "o a o3 . 4) "O —- c I CO t. t, « -33 >2 O O Pi >>o o 10 d ■^ -5 !C "§ on ^ O Xi o 5 a> P3 bb> O ^ 3 3 -fi _Q -2 X2 tK ^ IB a aj is o) Qi Qj »- r^ 'i- >^ >- O C 4) bC M X a W IE t/3 3 _3 ^ P3 CQ S S a 3 3 2 & .2 "c o , & O! s + o a £ bC I '§' + + I + I + + + + + + + + + + + + + + + + + + + + + + + + + + P3 y-t H P3 fc. I— ( o m papqcQ pQfflS-;pq &- H H H H SS S SSS §§§S s ? g § 1 g 3 '!_ ^ s _ ^ s s> 5 s Si I 0 H s 00 00 2^ 0 0 J^ fc sT N •>si 1 0 0 ? 0 s 1 - 00 2 ^__ 1 0 s a. cc~ 0 0 0 e R "ii 0 ? s l1 Si ■«o ^ ^ •S &0^ ■^ ^. a £ 0 K^J s c 1^ 0 a. 1 s a 0 o3 a >> c3 ;-! a. 8 -3 s 1 s ."2 'E p s c3 0 a; 1 C PL, Rhagophthalmidae (D Drilidae (Diplocladon) Drilidae (Phrixothrix) i 1 Hi a i t K 0) -0 ■§ 3 a, 1 1 tn 3 a a 31 unicata (Pyrosoma, Ap lasmobranchii (Etmopt 0 0 r2 3 2 1 1 "a g: 0 0 i a 12 'E c q; c; 0 fi 0 0 J a 0 c ^ IE o3 tc C te OJ G c ^ QJ.S fl xi y. o O tc"^ Ci O) 03 .3 ?:! S -*^ S 03 03 Sh ^ o3 a2 & 3 S = 2 0--.3 ■s s a O 3 I 8 ? oi . C 5 a; oj o ^ > I O •?- o3 ■^ :^ 3 c T3 ~ O o3 C B -C r/} o3=J S 0^ _ > G -S: ~-!3 o-si QJ X^ '^ cc t. O ^3 !L ^ -O r^ <3 "c E^ Oj G c O o3 a a CiC o o o W Ji r^ ffi "oS -a Oj G 1 o3 0) S-, r\ B -*^ '%-, C O c3 d >i 03 ^ -3 ■^^ a *= 03^"^ -^ -Q -S C 33 ^ O o3 CO S "=: 0) tiC o G '-^ -<- s o.s- a c-2 - 'o3^ — 05 • o-. ^ G^O O „ro o3 O '— I > G'* O s: Qj o a fe a M G O 3 C o3 la w i a-c-^ ^ — lC G +^ 1/^ a; .3 o "^ Sti — -a QJ 03 Q, o G O) o ■a. a G ^ ^1^ - MO 2 o-Q rti ^ 03 ^■0-3 o3 i.a::g bC , " o ^ CO o3 bC CO S a, o bcc H-i 05 C"^ 03 > Csi g G 03"-'*^ S § fi >. I OJ ■3 o3 -O 3 nqG o3 « CO G O G o ag 3 ^ X 3 „ !N O ■— O .G d (K G a ^ 3 3 3 a3 "S O i- o " 2 <^ -* -G !- •« ati ■ o.S r--^ c s " ?■ 5 .2f o > c ^ O G 3 o 5I C: .— ^ =^ -G cj ^ M o :S S"-^ G O C -^--.Ji p.G G O 3 S i ^ 00 5 5j cc O 03 <5 O-G ^■5.3 2 5 ?-o g=~' 20 E. NEWTON HARVEY 21 Explanation of Table I The groups are arranged phylogenetically in order of increasing com- plexity. References to authors and original papers are not given, since the complete literature and additional details regarding light production will be found in the book, Bioluminescence, by E. N. Harvey (Academic Press, N. Y., 1952), containing a bibhography of 1800 papers. Papers not in- cluded in the book are listed at the end of this article. Column 1 records the habitat, whether marine (M), fresh water (F), or terrestrial (T). Column 2 indicates the type of light, whether due to self-luminosity with intracellular (I), or extracellular (E) light emission from secreted material, or to symbiotic luminous bacteria (Bs) always found in the luminous organ, or to parasitic luminous bacteria (Bp) which have infected the living ani- mal. Saprophytic luminous bacteria will grow on any dead organisms, especially fish and squid, and some false reports of luminescence stem from this source (the fish, Harpodon). Luminous fungi will grow on wood and many other vegetable materials. Column 3. Knowledge of the cytology of the photogenic cell or the his- tology of the light organ is indicated by a + if the structure is well known, by a — if unknown and by ± if further study is desirable. Mutation has been investigated only in bacteria, genetics only among the fungi. Knowl- edge of chemistr>^ and physics of light emission is best known among bac- teria, ostracods, and fireflies (columns 4-9). Column 4. The necessity of dissolved molecular oxygen for light produc- tion is indicated by -}-. A — means that hght will appear in complete absence of oxygen (Pt and Ho, or in excess hydrosulfite). A blank in this column and in columns 4-9 indicates knowledge is lacking. Column 5. The luciferin-luciferase reaction is positive ( + ) if light appears when a hot water extract of the luminous cells allowed to cool (luciferin) is mixed with a cold water extract of the luminous cells allowed to stand until the light disappears ( lucif erase ) . The luciferin of one species will give no light with the luciferase of another luminous organism unless the two are closely related, such as different species of fireflies or different species of ostracods. No light appears on mixing luciferin (or luciferase) with lu- ciferase (or luciferin) of a firefly and an ostracod, or Odontosyllis and an ostracod, or Sijstellaspis and an ostracod. Column 6. If a nonluminous water extract of the luminous tissue con- taining magnesium emits light when the sodium salt of adenosine triphos- phate (ATP) is added, the ATP reaction is positive; if no light the reaction is negative. A + and a — sign indicate positive in some species, negative in others. 22 SURVEY OF LUMINOUS ORGANISMS Column 7. Marked inhibition of luminescence of single-celled organisms or of luminescent extracts by sunlight are indicated by + and no marked inhibition by — . A + and a — indicate that some species show light inhibi- tion, others not. DN indicates that there is a day-night rhythm of lumi- nescence such that the intact living organism does not luminesce in the daytime even if kept in the dark. Column 8. Many luminous organs are brightly fluorescent in near ultra- violet light or become so after functional activity. The color of the fluo- rescent Hght (usually near that of the bioluminescence) is indicated. A — sign means no marked fluorescence. Column 9. The color of the bioluminescent light is recorded, but too much reliance must not be placed on color estimation by eye observation. The wavelength of maximum emission is also given where spectrophotometric curves are known. E. NEWTON HARVEY 23 References * Bronk, J. R., E. N. Harvey, and F. H. Johnson. 1952. The effects of hydro- static pressure on luminescent extracts of the ostracod crustacean, Cypri- dina. J. Cellular and Comp. Physiol., 40, 347-65. Buchner, P. 1953. Endosymbiose der Tiere mit pflanzhchen Microorga- nismen. Symbiosen bei leuchtenden Tieren, 61-66; 470-528. Basel and Stuttgart. Buck, J. B. 1953. Bioluminescence in the study of invertebrate nervous sys- tems. Anat. Record, 117, 594. Chase, A. M. 1950. The kinetics of heat inactivation of Cypridlna lucif erase. /. Gen. Physiol, 33, 535-46. Chase, A. M. 1952. Studies on cell enzyme systems. VII. Lucif erase inactiva- tion by alcohols. /. Pharm. Exptl. Therap., 105, 371-79. Chase, A. M., and E. H. Brigham. 1952. Studies on cell enzyme systems. VI. Competitive inhibition of Cypridina luciferase by butyl alcohol. /. Cellu- lar and Comp. Physiol, 38, 269-80. Davenport, D., D. M. Wootton, and J. E. Cushing. 1952. The Biology of the Luminous Millipede, Luminodesmus sequoiae, Loomis and Davenport. Biol Bull, 102, 100-10. Haneda, Y. 1951. The luminescence of the deep-sea fishes, families Gadidae and Macrouridae. Pacific Science, 5, 372-78. Haneda, Y. 1952. Some luminous fishes of the genera Yarella and Polyipnus. Pacific Science, 6, 13-15. Haneda, Y., and E. N. Harvey. 1954. Additional data on the adenosine tri- phosphate and the luciferin-luciferase reactions of various luminous ani- mals. Arch. Biochem. and Biophys., 48, 237-38. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Harvey, E. N. 1953. Bioluminescence: Evolution and comparative biochem- istry. Federation Proc, 12, 597-606. Harvey, E. N., and J. J. Chang. 1954. Analysis of the luminescent response of the ctenophore, Mnemiopsis, to stimulation. Science, 119, 581. Harvey, E. N., and Y. Haneda. 1952. Adenosine triphosphate and bio- luminescence of various organisms. Arch. Biochem. and Biophys., 35, 470-71. Harvey, E. N., and F. I. Tsuji. 1954. Luminescence of Cypridina luciferin without luciferase, together with an appraisal of the term, luciferin. /. Cel- lular and Comp. Physiol, 44, 63-76. * Only recent pubUcations, not included in the bibliography in E. N. Harvey, 1952, Bioluminescence, Academic Press, New York, are listed. 24 SURVEY OF LUMINOUS ORGANISMS Hastings, J. W. 1952. Oxygen concentration and bioluminescence intensity. I. Bacteria and Fungi. /. Cellular and Camp. Physiol, 39, 1-30. Hastings, J. W. 1952. Oxygen concentration and bioluminescence intensity. II. Cypridina hilgendorfii. }. Cellular and Conip. Physiol., 40, 1-9. Hastings, J. W., W. D. McElroy, and J. Coulombre. 1953. The effect of oxygen upon the immobiUzation reaction in fire-fly luminescence. /. Cel- lular and Comp. Physiol., 42, 137-50. Johnson, F. H., H. Eyring, and M. J. Polissar. 1954. The Kinetic Basis of Molecular Biology. John Wiley and Sons, New York. Kato, Kojiro. 1953. On the luminous fungus gnats in Japan. Science Repts. Saitama University, Series B, 1, 59-63. McElroy, W. D. 1951. Properties of the reaction utilizing adenosine triphos- phate for bioluminescence. /. Biol. Chem., 191, 547-57. McElroy, W. D. 1951. Phosphate bond energy and luminescence. In Phos- phorus metabolism. Vol. I. W. D. McElroy and Bentley Glass, editors. Vol. I, pp. 585-600, 730-32. Johns Hopkins University Press, Baltimore, Md. McElroy, W. D., and J. Coulombre. 1952. The immobilization of adenosine triphosphate in the bioluminescent reaction. /. Cellular and Comp. Phys- iol, 39, 475-85. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1953. The requirement of riboflavin phosphate for bacterial luminescence. Science, 118, 385-6. Nicol, J. A. C. 1952. Studies on Chaetopterus variopedatus. I. The light- producing glands. II. Nervous control of light production. /. Marine Biol. Assoc. U. K., 30, 417-52. Nicol, J. A. C. 1952. Studies on Chaetopterus variopedatus (Renier). III. Factors affecting the light response. /. Marine Biol Assoc. U. K., 31, 113-44. Nicol, J. A. C. 1953. Luminescence in Polynoid Worms. /. Marine Biol. Assoc. U. K., 32, 65-84. Strehler, B. L. 1951. The luminescence of isolated chloroplasts. Arch. Bio- chem. and Biophys. 34, 239-48. Strehler, B. L. 1953. Luminescence in cell-free extracts of luminous bacteria and its activation by DPN. /. Am. Chem. Soc, 75, 1264. Strehler, B. L., and W. Arnold. 1951. Light production by green plants. ;. Gen. Plnjsiol, 34, 809-820. Strehler, B. L., and M. J. Cormier. 1953. Factors affecting the luminescence of the cell-free extracts of the luminous bacterium, Achromobacter fischeri. Arch. Biochem. and Biophys., 47, 16-33. Strehler, B. L., E. N. Harvey, J. J. Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Proc. Natl Acad. Sci., 40, 10-12. Luminescence Spectroscopy of Molecules and the Photosynthetic System* R. S. Becker and M. Kasha Department of Chemistn', Florida State University, Tallahassee, Florida I. Scope of Present Discussion In recent years numerous advances in the interpretation of electronic states and transitions in complicated molecules have been made. We shall discuss the application of those results which we believe to be of importance for the chlorophylls in their role of energy transfer agents in photosynthesis. Our discussion will be qualitative and with- out the complications of spectroscopic nomenclature and symbolism. Moreover, we shall omit all discussion of such fine points concerned in the electronic transitions as symmetries, polarizations, vibrational fine structure, and assignments. We believe that these points, however intrinsically interesting they may be to the professional spectroscopist, will not be involved directly in the problem of photosynthesis. On the other hand, we shall discuss all those spectroscopic aspects which we believe will be important in the capabilities for utilization and transfer of electronic excitation energy by the chlorophylls. As will be shown, this discussion will be limited perforce to the lowest electronic states of the molecules, namely, the ground (singlet) elec- tronic state, the lowest triplet excited state, and the singlet excited states (our discussion requires the plural) involved in the red ab- sorption band of the chlorophylls. Specifically, the Soret band and higher energy transitions we shall disregard, in accordance with the role of Internal Conversion discussed in Section II. " Work done under Contract NR-015-318 between the Office of Naval Research, Department of the Nav>', and the Florida State University. 25 26 LUMINESCENCE SPECTROSCOPY OF MOLECULES The main points of our discussion will involve (1) the role of n,7r-transitions in the chlorophylls, (2) the importance of intercom- binations in these molecules, and (3) the possible interaction of the ethylenic potential function with the electronic transitions in the chlorophylls. Our interpretations will be applicable rigorously to the properties of the chlorophylls as physically isolated molecules in vitro. However, by discussing expected changes in properties of the mole- cules upon change in environment, we shall be able to extrapolate occasionally to the behavior of the molecule in vivo as a photosyn- thetic agent. Omitted from our discussion will be any discussion of chemilumi- nescences, of the mode of energy transfer involving chlorophylls, and of other problems which are involved mainly in the biological system, such as accounting for the difference in the chlorophyll ab- sorption bands in the living plant and in vitro. II. Importance of Lowest Excited Electronic States in Utilizability of Excitation Energy ( 1 ) Internal Conversion Extensive studies in the field of molecular luminescence have re- vealed that only the lowest electronic state of a given series of elec- tronic states in a molecule is capable of re-emitting its excitation energy. Upper electronic states of the series, upon excitation, lose energy thermally (by collisions) and radiationlessly go over into the lowest electronic excited state of the series. By series of states is meant states of one electronic multiplicity, or mode of electronic pairing (see Section V). This phenomenon of radiationless combina- tion of excited states is known generally in the field of spectroscopy as internal conversion (for additional discussion, see Kasha, 1950). Its simplest consequence is that only the lowest excited singlet state is important in any energy transfer or energy utilization process: the higher excited states decay far more rapidly and generally cannot be expected to participate in any photochemical process. (2) Observation of Fluorescence In unsaturated molecules, including the chlorophylls, the w-electrons give rise to most of the low-lying states corresponding to absorption R. S. BECKER AND M. KASHA 27 bands in the visible and ultraviolet regions (cf. Coulson, 1947, and Piatt, 1951, for general discussions). A general study of luminescence properties of molecules (Kasha, 1950) shows that if the 7r-electronic states are the lowest singlet states of an excited series, fluorescence generally will be observed. If, however, the lowest excited singlet level is of an n,Tr type (cf. Section IV), the molecule will be generally a nonfluorescent one. In Section IV we shall interpret the behavior of chlorophyll fluorescence activation on this basis. The intrinsic lifetime of the lowest singlet excited state will limit the probability of utilization of the excitation energy of that state. If the state is very short-lived, it may be that no energy transfer or photochemical process will be rapid enough to compete with spon- taneous fluorescence emission. If a molecule has a high intrinsic quantum yield of fluorescence (or phosphorescence, see below) in an undisturbed (isolated) system, this property indicates excitation energy availability. However, if strong fluorescence actually is ob- served for the molecule in any reacting system, the lifetimes of fluorescence must be shorter than those for any other process and the observed quantum yield of fluorescence then indicates energy wasted. In other words, the spontaneous emission rate in such a case was comparable to the rates for some other energy utilization process. A few quantitative relations governing lifetimes and quantum yields will be given in the next section. (3) Lowest Triplet State of the Molecule All normal molecules studied thus far, barring understandable exceptions, have exhibited lowest triplet excited states (Lewis and Kasha, 1944; Kasha, 1947). By normal molecules is meant here that an even number of electrons is present in the molecule, with electron pairing, so that diamagnetic, singlet ground electronic states result. The common electronic excitation is that in which the electron pairing is maintained: this gives rise to all the commonly observed absorption bands and the fluorescence phenomenon. Such transitions are labeled singlet — > singlet. On the other hand, if electron pair uncoupling takes place, states of higher multiplicity are observed, the common state of this type being a triplet state in normal molecules, the singlet-triplet transition being designated an intercombination. Of course, many such triplet 28 LUMINESCENCE SPECTROSCOPY OF MOLECULES states are possible for the molecule. However, in accordance with the above discussion of internal conversion, only the lowest triplet state will be of any importance in photochemical and other energy utiliza- tion processes. We shall be interested in what role the lowest triplet state may play in the photosynthetic reaction. Outstanding in its importance in photochemical and energy transfer mechanisms is the long lifetime of the triplet state. In general, a triplet state will be longer lived by a factor of one million over the lifetime of a corresponding singlet state. This places the range of intrinsic triplet state lifetimes between 10-^ second and 1 second for "normal" cases. There are several mech- anisms by which intrinsic triplet state lifetimes may be varied con- siderably without cjuenching of excitation energy. These phenomena will be discussed mainly in Sections V and VI. It is likely, however, that if a triplet state of a chlorophyll may be excited easily, it may play an important part in any energy utilization process, on the basis of lifetime alone. It is now generally recognized among spectroscopists that the low- est triplet state can be detected readily in most molecules by the study of low-temperature phosphorescence spectrum (Lewis and Kasha, 1944; Kasha, 1947; Kasha, 1950). The intrinsically long Hfe- time of phosphorescence generally makes its observation in fluid systems impossible, since in such cases the quantum yield and life- time are reduced to immeasurable limits by the competitive collisional deactivations. Consequently, the phosphorescence emission of a mole- cule is always sought in rigid glass solvents. The high viscosity in such systems allows the long-lived phosphorescence to be observed. At lower viscosities, quenching will occur, and any phosphorescence emission observed will have a shorter lifetime and a diminished quan- tum yield compared with the intrinsic valves. In Section V we shall discuss the expected wavelength range and other characteristics of the lowest triplet states of the chlorophylls. III. Basic Quantitative Relationships (1) Intrinsic Lifetime of an Excited State In the previous section we have argued that an important criterion of utilizability of excitation energy is the intrinsic lifetime of the excited R. S. BECKER AND M. KASHA 29 state. In lieu of direct measurement of decay time constant ( which in the absence of any form of quenching would of course be the intrinsic lifetime), we may make recourse to the classical relation (for discus- sion, see Lewis and Kasha, 1945; Kasha, 1950 ) : '°={sdi^)^0''<"^'^^'r'* 0 This expression allows the intrinsic lifetime t° of a luminescence to be calculated from the absorption band integral, J'edf, evaluated graphically. The average frequency (cm~^) of the transition va, the re- fractive index of the medium n at the same frequency, and the multi- plicity ratio gu/gi for the upper and lower states are the other variables. The constants are ir, and c, the velocity of light. (2) Summation of Intrinsic Quantum Yields for a Molecule We define quantum yield or quantum efficiency of a luminescence by Number of quanta emitted $ = —^ , Number of quanta absorbed The total intrinsic quantum yield for a molecule (in the absence of external quenching ) may then be described by J] $, = 4>f° + V + <|.i„t = 1 where ^f° is the intrinsic quantum yield of fluorescence, ^p° is the in- trinsic quantum yield of phosphorescence, and *int is the intrinsic quan- tum yield for internal degradation by thermal steps. In the older literature the incorrect expression ^f° + *int = 1 is assumed, leading to the conclusion that if no fluorescence is observed, internal degradation must predominate. Actually, in general the sum of ^f° + *p° may ap- proach unity (Kasha, 1950), so that $int may be negligible in many molecules, at least in rigid solvents. It is of the utmost importance to understand that ^f° and ^p° are complementary in magnitude. Thus, if $int = 0, as seems to be the case in general for rigid molecules, then if (^f° = 0.2, $p° must equal 0.8. In other words, in a fluid solution, even if phosphorescence is not ob- served, due to collisional deactivation, the corresponding triplet state is excited with a probability of 0.8 for each absorbed quantum (and this is followed by deactivation, if fluid). Then, if this triplet state is in- 30 LUMINESCENCE SPECTROSCOPY OF MOLECULES volved in a photochemical process, the limiting quantum yield of the photochemical reaction would be 0.8. Conversely, if, say, an energy transfer or photochemical reaction proceeds uniquely via the lowest singlet excited state, and if ^f° = 0.2, then the limiting value of ^photochem will bc 0.2. (3) Quenching: Lifetime and Quantum Yield If we define observed lifetime of an excited state by t, and intrinsic lifetime by t°, then for collisional bimolecular quenching where $ is the observed quantum efficiency of a luminescence in the presence of partial quenching, and $° is the intrinsic quantum yield of fluorescence {or phosphorescence) as before. Consequently, for collisional bimolecular quenching For example, if ° = 0.2, and $ = 0.02, then the observed lifetime will be 0.1 of the intrinsic lifetime of an excited state. In the older literature, the incorrect expression t = *t° was used; for the example given, the observed lifetime would appear to be 0.02 of the intrinsic lifetime. This error can be traced to the neglect of the complementary luminescence to the one observed, as the reader can confirm by the study of the expression given in Section ( 2 ) . Obviously, the incorrect expression sets 4)° = 1 for any luminescence. IV. Role of n,7r-Transitions in Spectra of Chlorophyll a and Chlorophyll b First we shall give a rough qualitative description of n,7r-transitions. Such transitions have been described by McMurry and Mulliken (1940), and more generally by Kasha (1950) and Piatt (1953). The designation n means nonbonding electron orbital. In addition to pure TT-electron excitations, it is found that nonbonding electrons (re- ferred to by chemists as "lone pairs") may be excited to unfilled TT-electron orbitals in the molecule. Examples of n-» tt transitions are: (a) the excitation of a nonbond- ing 2p orbital electron of the oxygen in formaldehyde H2C = O: to R. S. BECKER AND M. KASHA 31 an excited ( antibonding; cf. Coulson, 1947) 7r-orbital; (b) the excita- tion of a nonbonding electron of the N-atom in pyridine to an excited (antibonding) 7r-orbital. For our purposes the characteristics of n -^ tt- transitions which are of interest are: (1) n ^ TT-absorptions (even if allowed by spectroscopic selection rules ) are weaker tlian corresponding tt -> 7r-absorptions ( Kasha, 1950). This means, in view of the relation discussed in Section III ( 1 ) , that n,7r-excited states will be much longer Hved than analogous 7r,7r-excited states, with the consequences discussed previously. ( 2 ) n -^ TT-absorptions show a blue shift upon change of solvent from hydrocarbon type to hydroxylic type (Kasha, 1950; McConnell, 1952). Blue shifts of as much as a few hundred to a few thousand wave numbers (cm"^) have been observed. Recently it has been estabhshed (Brealey and Kasha, 1954) that these blue shifts may be ascribed largely to hydrogen bonding of the solvent to the n-electrons of the solute molecule. (3) If n -^ TT-absorptions fall at lower energy or longer wavelength than any tt -^ 7r-absorption, then the molecule will be nonfluorescent (Kasha, 1950). Examples are many-fold, in which the nonfluorescence may be attributed to this juxtaposition of energy levels. Thus, all ahphatic ketones and many aromatic ketones and aldehydes are non- fluorescent and have n,7r levels lower than 7r,7r. In these, $f° = 0, and p° ^ 1, so that (only) very strong phosphorescences are generally observed in rigid glass solutions. Other examples are most nitro compounds, quinones, azo compounds, and simple N-heterocychcs. The application of these interpretations to the chlorophylls may be made as follows. Both chlorophyll a and chlorophyll b contain carbonyl groups, C = O. Isolated carbonyl groups will have ultraviolet n -^ -n-ab- sorptions, similar to that in acetone; these will not interest us. How- ever, C = O groups conjugated with other parts of the molecule may have n -^ 7r-absorption at quite long wavelengths. Moreover, the more highly conjugated the carbonyl group, the stronger will the n -^ 7r-ab- sorption be, although in all cases somewhat less strong than pure TT-electron absorptions. Moreover, such n -> 7r-absorptions should blue shift in hydroxylic solvents, or any solvents capable of forming molecular complexes specifically involving the n-electrons of the chlo- rophyll carbonyl groups. 32 LUMINESCENCE SPECTROSCOPY OF MOLECULES The study of fluorescence activation of the chlorophylls by Living- ston, Watson, and McArdle (1949) can be interpreted spectroscopi- cally in terms of the above discussion. These workers found that chlorophyll a and chlorophyll b in hydrocarbon solvents showed very little fluorescence. However, especially in hydroxylic solvents such as alcohols and water, strong fluorescence was observed. ^(iT;ir/ t UJ (triplet) HVPROCARBON HYDROGEN BONDING Fig. 1. Effect of hydrocarbon and hydrogen-bonding solvents on the frequencies of electronic transitions of a molecule with tt.tt and n,7r levels close together. Ground states are equalized arbitrarily, and excited vibrational and triplet electronic levels are omitted. Moreover, the lowest frequency absorption band of these two chlo- rophylls was more complex in hydrocarbon solvents (cf. their Figs. 4 and 5), since it is lower in apparent intensity in the latter (though not necessarily in integrated absorption) and, notably, shows evi- dence of a shoulder on the long-wavelength side of the main peak. These results may be interpreted according to the diagram shown in Fig. 1. On the left side of the diagram we indicate the energy relationships which we picture for chlorophyll a. We assume that the shght shoulder shown in Livingston et al. (1949) is an indication of an n,7r-transition just slightly lower in energy than the stronger, 7r,7r-transition. Upon changing to a hydrogen-bonding solvent, the latter transition would be expected to undergo a normal red shift. R. S. BECKER AND M. KASHA 33 while the former undergoes a strong bhie shift. Thus, we accept the discussion by Livingston et al. (1949) concerning the stabihzation of a keto form by hydrogen-bonding solvents. However, we add to this the spectroscopic point that only the hydrogen-bonded keto form would be capable of fluorescence, if our energy scheme is correct. For chlorophyll h, we observe that an additional conjugated car- bonyl group is present (the fonnyl group). With two ;i,7r-transitions possible for the b species, we anticipate a larger net 7r,7r-/i,7r-separation. In Fig. 4 of the paper by Livingston et al. (1949), a considerable shoulder on the long-wavelength side of the lowest absorption band for hydrocarbon solutions of chlorophyll b is found, which is absent for chlorophyll b in hydrogen-bonding solvents. We attribute the disap- pearance of this shoulder to a blue shift of the additional n,7r-transition which we ascribed to the formyl group, as well as the blue shift of the n,7r-transition, which is also present in chlorophyll a. The above interpretation thus is capable of accounting for the differences between chlorophyll a and chlorophyll b spectra observed by Livingston et al. (1949), as well as for the appearance of fluores- cence in the hydrogen-bonded molecules. Nevertheless, we note that the results of the Russian workers (Evstigneev, Gavrilova, and Kras- novski, 1950) emphasize the role of magnesium in the activation phenomena. The reconciliation of their results with our understand- ing of the spectroscopic aspects of this problem we have not yet achieved. However, we can state that we can picture no mechanism based on the hydration of the magnesium which would account for both spectroscopic observations, i.e., the activation of the fluores- cence as well as the observed changes in the absorption spectrum. Piatt (1951) noted the possibility which we have discussed above, but he did not carry out the interpretation properly; in particular his idea of a negligible blue shift is not supported by subsequent work. V. Lowest Triplet State of Chlorophyll We have in progress a comprehensive study of the luminescence emission properties of porphyrin-like molecules including chlorophyll a and chlorophyll b. In the following discussion we shall outline the theory and approach which we are using as a guide to this problem and shall indicate the positive results obtained thus far. At the outset 34 LUMINESCENCE SPECTROSCOPY OF MOLECULES we can state that we have obtained a triplet —> singlet emission for chlorophyll b, a phosphorescence observed previously by other work- ers under uncertain conditions. We shall give a thorough analysis of the significance of this finding. (1) Singlet-Triplet Split: Expected Lowest Triplet Energy The energy difference between the lowest excited singlet state (S') and the lowest triplet state (T) (cf. Fig. 2) is called the singlet-triplet split'* in a molecule. This split is very variable (Kasha, 1947) from t >- O i UJ Fig. 2. Electronic levels important in energy utilization processes in a molecule (schematic). Only ir,ir levels are shown, and excited vibrational levels are omitted. molecule to molecule. For example, in molecules like naphthalene and anthracene the split is very large, of the order of 10,000 to 12,000 cm~^ On the other hand, in dyelike molecules, such as the cyanines and many others, the singlet-triplet spht is of the order of 2000 cm~^ If we classify a linearly or cyclically conjugated molecule like porphy- rin as dyelike, then we might expect the singlet-triplet split to be of the order of 2000 cm^^ ± 1000 cm^^ In zinc tetraphenylporphyrin Dorough et al. (1951) have observed a singlet-triplet split of approxi- mately 2500 cm-^ (cf. their Fig. 11). In recent experiments in our ' This definition assumes that the singlet and triplet states correspond in orbital configuration. If they do not, the energy difference given is called the singlet-triplet separation. R. S. BECKER AND M. KASHA 35 laboratory we have found an S'-T split of this same order of magni- tude in metal phthallocyanines. We would expect the chlorophyll a and chlorophyll b molecules also to exhibit a similar split. Since the lowest frequency absorption of alcoholic solutions of these chloro- phylls is at approximately 6500 to 6600 A, and the fluorescences at approximately 6700 to 6800 A, the phosphorescences or triplet-singlet emissions should occur at approximately 2500 cm"' lower in fre- quency, which would be at about 8050 to 8200 A. Calvin and Dorough (1947) originally reported a phosphorescence in a pure mixture of chlorophylls commencing at about 8000 A and extending farther into the infrared. Later work by Calvin and Dorough (1948) on chromatographically separated chlorophyll a and chloro- phyll h showed no phosphorescence in chlorophyll a and only a weak phosphorescence in chlorophyll h, commencing at 8600 A and pro- ceeding farther into the infrared. Further investigation by Livingston ( 1949 ) failed to reveal any emission from either chlorophyll a or chlorophyll h. Livingston believes that if a phosphorescence of chlo- rophyll exists, it is at longer wavelengths than 9000 A or has a life- time shorter than 0.01 second. It is obviously important that further work clarify these contradictions of experiment. In our laboratory we have made preliminary observations on the phosphorescence of chlorophyll a and chlorophyll h. In agreement with Calvin and Dorough (1948), we found no phosphorescence for chlorophyll a. Furthermore, in agreement with Calvin and Dorough ( 1948 ) , we did get a very good photograph of the phosphorescence of chlorophyll h. A small amount of chlorophyll a was present with the fo-isomer; however, it is known that energy transfer will not take place in dilute solutions as rigid glasses at low temperature. Our exposure was sufficient to reveal one medium strong narrow band at 8650 A. We believe there is no reasonable doubt as to the reality of this emis- sion as a chlorophyll h phosphorescence. However, as discussion be- low will reveal, we do not simply relate this state with the photosyn- thetic step, contrary to the Calvin and Dorough (1947) hypothesis. The experimental conditions under which our observations were made were as follows. We used a high-speed phosphoroscope having a resolving time of 3 X 10 ~* second. The chlorophylls were dissolved in a rigid glass solvent formed by supercooling the dilute solution 36 LUMINESCENCE SPECTROSCOPY OF MOLECULES (approximately 10~'^ M) of the chlorophylls in standard EPA solvent mixture (of. Kasha, 1947) (2 parts ethyl alcohol, 5 parts isopentane, 5 parts ethyl ether, measured by volume), at a temperature of 77° K. The chlorophylls were chromatogrammed on a column of packed sugar containing 3% starch, the bands developed by a mixture of 15% (vokime) ethyl ether in petroleum ether. The absorption spectrum of chlorophyll a was spectroscopically free of pheophytin, chlorophyll h, and other plant pigments. The absorption spectrum of the chlorophyll h used indicated approximately 10-15% of chlorophyll a but no other plant pigments. We used ammonia-hypersensitized I-N Eastman Kodak spectroscopic plates. The spectrograph used was a Hilger me- dium glass. Type E 495. We are extending our observations to obtain complete vibrational structure of the phosphorescence emission and to obtain a measure of the quantum efficiency and lifetime (decay constant) of the emis- sion. Our exposure time of several hours at a wide slit ( 1 mm ) would seem to indicate that p° is rather small. However, we must point out that until the mean lifetime has been measured, we would have no comparison with the phosphoroscope resolution time quoted above. Only when the mean lifetime is known will it be possible to state with certainty whether our observation was made near the beginning of the phosphorescent decay, or near the tail end of the decay for each excitation cycle. (2) EfFect of Electric and Magnetic Fields: Heavy Atom and Paramagnetic Atom Effects in Phosphorescence Spectroscopically the mechanism of singlet-triplet transitions is un- derstood in terms of a phenomenon known as spin-orbit interaction. For our purposes we can state that in all molecules there will be a spin-orbit interaction due to the electric fields of the nuclei of the atoms present in the molecule. This spin-orbit interaction may be strongly enhanced by the introduction of high atomic number or para- magnetic atoms into the molecule (cf. Kasha, 1950, and references therein). The effect of such introduction results in the following changes ( 1 ) the fluorescence quantum yield ^f° for the substituted molecule is lowered, (2) conversely, *p° is increased greatly, and (3) the lifetime of the phosphorescence is decreased greatly. Spin-orbit R. S. BECKER AND M. KASHA 37 interaction is very sensitive to such effects, because of the high-power dependence of the probabiHty of singlet-triplet transition on atomic number (Z"*). Applied to the porphyrin-like molecules, this means that replace- ment of magnesium ( Z = 12 ) by zinc ( Z = 30 ) should result in a pronounced enhancement of the phosphorescence and a considerable shortening of its mean lifetime. Replacement of magnesium by copper (Z = 29) should have an additional strong effect due to the magnetic field produced by the unpaired copper ti-electrons. Thus, Calvin and Dorough (1947, 1948) found such effects in zinc and copper chlorin and porphyrin derivatives. It is in fact an old observation in the por- phyrin fluorescence field that the porphyrin derivatives of paramag- netic metals such as iron and copper, are nonfluorescent. We expect to find these strongly phosphorescent. In our laboratory we have made some preliminar)' investigations on magnesium and copper phthallocyanines. The magnesium phthallo- cyanine is strongly red fluorescent, while the copper phthallocyanine shows no fluorescence. However, copper phthallocyanine shows a strong infrared phosphorescence at 77° K, in conformity with expec- tation. A different sort of magnetic or electrical effect is also possible for heavy or paramagnetic atoms in the neighborhood of the excited molecule. It has been demonstrated (Kasha, 1952) that heavy-atom fluorescence quenchers induce the conversion of excited singlet energy to triplet energy. It is undoubtedly true that similar quenching effects by paramagnetic molecules such as O2 would be due to the same mechanism. It may turn out that these field effects will have some influence on the energy processes in chlorophylls in the photosynthetic system. (3) Nature of Lowest Triplet State of Chlorophyll According to our discussion in Section IV, both 7r,7r and n,7r excited singlet states may be present in the region of the red absorption band of chlorophyll a and chlorophyll b. If this is so, and that can be established rigorously by further experiment, we would be compelled to consider that below the lowest excited singlet state there will be at least two triplet states in both chlorophyll a and chlorophyll b, one 38 LUMINESCENCE SPECTROSCOPY OF MOLECULES triplet of an n,7r type, and one triplet of a 7r,7r type. In addition, in chlorophyll b, an additional triplet of an n,7r type may result from an n-electron of the formyl group in its triplet excitation to a Tr-molecular orbital. This discussion may appear rather complicated, but fortunately the questions raised are accessible to experimental investigation. (4) Role of Lowest Triplet States of Chlorophylls in Energy Transfer In all the published discussions of excitation energy transfer in- volving chlorophylls, the main emphasis has been on the study of fluorescence. It is obvious that if ^f° is of the order of magnitude of 0.1, the use of fluorescence as a criterion for energy transfer focuses attention on the minor part of the possible available energy. That the remaining 90% of the absorbed energy must be accounted for to ex- plain the efficiency of observed photosynthesis in plants has been indicated previously, e.g., by Livingston (1949). When the quantum efficiency of chlorophyll phosphorescence has been determined, we shall be in a position to evaluate whether the lowest triplet state receives the missing 90% of the energy. If any considerable amount of the excitation energy actually reaches the lowest triplet state, then it may be entirely possible for this state to be important in the energy transfer processes (preceding the photo- synthetic primary step ) . In fact, it is entirely possible that chlorophyll a may transfer its excitation energy to chlorophyll b as far as triplet state energy is concerned. Of course, this would be just the reverse of the commonly accepted course of energy transfer between chloro- phylls, deduced from fluorescence studies. We emphasize that we merely wish to point out the possibility of such a transfer, and not to predict its necessity. We would be inclined to favor the idea of net energy transfer from chlorophyll a to chlorophyll b on two points: first, chlorophyll b is the minor component of chlorophylls in many living plants. It thus would play a minor part in light absorption by the plant, and energy transfer from chlorophyll b to chlorophyll a would have a minor role in photosynthesis. On the other hand, the spectroscopic and structural dis- tinctions between chlorophyll a and chlorophyll b (especially if the R. S. BECKER AND M. KASHA 39 Z7-isomer has a readily available triplet state) may make chlorophyll b the important intermediate acceptor of the excitation energy. Thus, a net transfer of energy from chlorophyll a to chlorophyll b may be important. Moreover, the fact that chlorophyll a is a major compo- nent of plant chlorophylls would make its dominant light absorption the chief role of this isomer in photosynthesis. In the next section we shall discuss one additional spectroscopic phenomenon which may dominate the properties of the chlorophylls in energy utilization. This involves the electronic interaction of the vinyl group present in chlorophylls and the possibility that it behaves as an energy dissipator in isolated chlorophylls in vitro. This may account for possible low quantum yield of luminescence of chloro- phyll as isolated molecules. VI. Interaction of Low Excited States of Chlorophyll with Torsional Potential The theory of the electronic-vibrational interactions in the ethylene molecule has been developed by Mulliken (1933) and Mulliken and Roothaan ( 1947 ) , with particular reference to the effect of twisting or torsion about the double bond. This t>^pe of distortion is unique, among possible intramolecular distortions, in lowering vastly the en- ergy of the excited singlet and triplet states ( cf . Fig. 3 ) of ethylene. The effects of this torsional potential as it is called are profound in determining the spectroscopic behavior of the molecule. For example, the absorption spectra change shape upon change of temperature and viscosity, and no fluorescence is observed (Potts, 1954). The reasons for this are clear from the diagram. Absorption is mainly to a molecule distorted far from its most stable configuration, quite the opposite of normal one-electron excitation in molecules. Upon cascading through torsional levels under the influence of collisions, the molecule reaches a minimum from which no emission could be observed at accessible wavelengths. Even in methylethylenes in rigid glasses (Potts, 1954) no fluorescence can be observed. However, in ^rans-stilbene (trans- diphenylethylene), fluorescence can be observed in rigid glasses or crystals, though not in fluid solutions (Kasha, 1950). Such behavior is general in complex ethylenic molecules. Apparently, the phenyl groups are sufiBciently large to inhibit the 40 LUMINESCENCE SPECTROSCOPY OF MOLECULES torsional mode of vibration, probably by introduction of a bamer (dotted curve, Fig. 3) on the potential curve (we disregard the simi- lar barrier in the ground state; it introduces no new feature). Conse- quently, absorption is to a nearly equilibrium configuration, with re- emission of fluorescence directly observable. sT 1 \l / \ / t il \ liJ Z A y / 1 r ^ - 5r\^ IL- / 90 AN&LE 180 Fig. 3. The variation of potential energy with twist of the double bond in ethyl- enic molecules. The dotted curve represents a barrier to torsion introduced by inhibition of twisting motion in rigid solvents by substituted groups on the ethylene. A, fluid; A', rigid. The vinyl group — CH = CH2 in the chlorophylls is conjugated with the main 7r-electron system of the molecule. If this vinyl group is free to undergo torsion upon excitation, it is possible that in the isolated molecule this could serve as the chief dissipator of electronic excitation energy. Its efficiency as a dissipator would depend on the magnitude of the interaction of the main 7r-electron system with the torsional potential. The low quantum yield of fluorescence in alcohol solution in both chlorophyll a and chlorophyll h may be due to such an interaction. The absence of phosphorescence in chlorophyll a and R. S. BECKER AND M. KASHA 41 the possible low quantum yield of phosphorescence (not yet deter- mined) of chlorophyll b would be similarly accounted for. In the living cell, it is possible that these dissipative effects, if ac- tual, would be absent. This could be the secret of chlorophyll in the living cell. Certainly, if electronic energy is dissipated as described above, and if attachment of chlorophyll to some other molecule is through the vinyl group in vivo, then most of the absorbed light energy could become photochemically available. VII. Conclusion From our analysis of the spectroscopic interpretation of the chlo- rophyll problem, it is apparent how little experimental investigation has been made on the nature of the lower electronic states, which would be involved in energy utilization processes. As a result, much of our discussion has necessarily been speculative. However, we hope that the viewpoint presented will stimulate further examination of these points. In our developing research program we shall have the opportunity to investigate experimentally many of the interpretations made. Since our discussion is a preliminary one, we have refrained from a discussion of the follow-steps in photosynthesis that have been proposed recently. Once the primary step of energy utilization for chemical reaction initiation has been established for the chlorophyll system, the relation to the follow-steps should become apparent. Our basic aim is to establish which electronic state of chlorophyll is in- volved in this primary energy step, as contrasted with simple light absorption. References Brealey, G. J., and M. Kasha. 1954. The role of hydrogen-bonding in the n ^ TT blue-shift phenomenon. (Presented at the symposium on Molecular Structure and Spectroscopy, Ohio State University, Columbus, June 14- 18). Calvin, M., and G. D. Dorough. 1947. The phosphorescence of chlorophyll and some chlorin derivatives. Science, 105, 433-34. Calvin, M., and G. D. Dorough. 1948. The possibility of a triplet state inter- 42 LUMINESCENCE SPECTROSCOPY OF MOLECULES mediate in the photo-oxidation of a chlorin. /. Am. Chem. Soc, 70, 699-706. Coulson, C. A. 1947. Representation of simple molecules by molecular orbitals, Quart. Rev. Chem. Soc. (London), 1, 144-78. Dorough, G. D., J. R. Miller, and F. Huennekens. 1951. Spectra of metallo- derivatives of alpha, beta, gamma, delta-tetraphenylporphine. /. Am. Chem. Soc, 73, 4315-20. Evstigneev, V. B., V. Gavrilova, and A. A. Krasnovski. 1950. Doklady Akad. Nauk SSSR., 70, 261. Kasha, M. 1947. Phosphorescence and the role of the triplet state in the electronic excitation of complex molecules. Chem. Revs., 41, 401-19. Kasha, M. 1950. Characterization of electronic transitions in complex mole- cules. Discussions Faraday Soc, No. 9, 14-19. (This paper summarizes results of a general study now in preparation for publication.) Kasha, M. 1952. Collisional perturbation of spin-orbital coupling and the mechanism of fluorescence quenching. /. Chem. Phys., 20, 71-74. Lewis, G. N., and M. Kasha. 1944. Phosphorescence and the triplet state, /. Am. Chem. Soc, 66, 2100-16. Lewis, G. N., and M. Kasha. 1945. Phosphorescence in fluid media and the reverse process of singlet-triplet absorption, /. Am. Chem. Soc, 67, 994- 1003. Livingston, R. 1949. The photochemistry of chlorophyll, pp. 179-96, in Photosynthesis in Plants, J. Franck and W. E. Loomis, eds. Iowa State College Press, Ames, Iowa. Livingston, R., W. F. Watson, and J. McArdle. 1949. Activation of the fluorescence of chlorophyll solutions. /. Am. Chem. Soc, 71, 1542-50. McConnell, H. 1952. Effect of polar solvents on the absorption frequency of n ^ TT electronic transitions. /. Chem. Phys., 20, 700-04. McMurry, H. L., and R. S. Mulliken. 1940. Mechanism of the long wave- length absorption of the carbonyl group, Proc Natl. Acad. Sci. U. S., 26, 312-17. Mulliken, R. S. 1933. Electronic states, quantum theory of the double bond. Phys. Rev., 43, 279-302. Mulliken, R. S., and C. C. J. Roothaan. 1947. The twisting frequency and barrier height for free rotation in ethylene. Chem. Revs., 41, 219-31. Piatt, J. R. 1951. Electronic structure and excitation of polyenes and por- phyrins, in Radiation Riology, Vol. Ill, Biological Effects of Visible Radia- tion, Sterling Hendricks, ed. McGraw-Hill Book Company, New York. Piatt, J. R. 1953. Classification and assignments of ultraviolet spectra of conjugated organic molecules. /. Opt. Soc. Amer., 43, 252-57. Potts, W. J., Jr. 1954. Low-temperature absorption spectra of selected olefins in the farther ultraviolet region. Technical Report, Part 2, for 1952-3, Laboratory of Molecular Structure and Spectra, Dept. of Physics, Uni- versity of Chicago (submitted for journal publication). D ISCUSSION Dr. Duysens: I should like to point out two things which seem to argue against the suggestion made by Mr. Becker that the light energy is passed on to the photosynthetic reaction via a triplet state of chlorophyll b. The striking similarity between the action spectrum of the fluores- cence of chlorophyll a and the action spectrum of photosynthesis of the many photosynthetic organisms investigated indicates that the energy absorbed by the various pigments is passed on to the fluores- cent (lowest singlet) state of chlorophyll a and in this way becomes active in photosynthesis. It might now be argued that the energy is passed from chlorophyll a to the triplet state of chlorophyll b. As far as I know, no mechanism is known by which such a transfer will be accomplished with high efficiency. There is further the fact that many algae do not contain chloro- phyll b. The pigment common to all of them is chlorophyll a. Appar- ently, an energy transfer from a back to a triplet state of b is, in general, not necessary for photosynthesis. It seems therefore not neces- sary to postulate that such a transfer back to b occurs. Dr. Kasha and Mr. Becker: The philosophy of our proposal concern- ing energy transfer between chlorophyll a and chlorophyll b has apparently not been made clear. Our aim was merely to point out a possibility that had been overlooked previously. Most of the conferees seem to have taken it for granted that we implied that energy transfer from chlorophyll a to chlorophyll b (via triplet states) was a neces- sary phenomenon, which was not our intention. Thus, in answer to Dr. Duysens' second point, it is obvious that if algae do not contain chlorophyll b, energy transfer between chloro- phyll a and b cannot be considered, in either direction. His statement is not a valid criticism of our suggestion on energy transfer. In answer to Dr. Duysens' first point, it is true that there is no highly developed mathematical theory to account for a transfer of energy from the triplet state of one molecule to another. However, such a possibihty is qualitatively very plausible and probably is quite a common photochemical mechanism. Highly efficient excitation of triplet states is common; the long lifetime of the latter makes a highly efficient transfer of energy plausible. 43 /K^^^^^L y 44 LUMINESCENCE SPECTROSCOPY OF MOLECULES The main quantitative theory on energy transfer, due to Forster, predicts that efficiency of energy transfer is proportional to intensity of absorption of the acceptor molecule. This would argue against our suggestion. However, it is doubtful whether Forster would insist that his mechanism would apply to all types of energy transfer. We should like to point out that Dr. Duysens' statement (that the similarity of the "action spectrum of fluorescence" of chlorophyll a and the "action spectrum of photosynthesis" requires that the "fluo- rescent" level be involved in photosynthesis ) is incorrect. In the normal excitation of phosphorescence the "action spectrum" of fluorescence and phosphorescence are necessarily identical, since both first involve lowest singlet-singlet absorption. Dr. Mason: Would you describe the operations which characterize fluorescence and phosphorescence? Have phenomena such as delayed fluorescence or premature phosphorescence been observed? What op- erations must be performed in order to decide in which category the emission falls? Dr. Kasha and Mr. Becker: Dr. Mason's first question is answered in the published literature (Kasha, 1947; Lewis and Kasha, 1944). In answer to the second question, there is no sharp distinction be- tween fluorescence and phosphorescence of molecules on the basis of lifetime. Thus, fluorescence is usuaUij very short lived, in the range 10~^ to 10" ^'^ second (mean life) because the emission corresponds to intense absorption bands (cf. Section III, 1, our paper). However, singlet-singlet luminescence, or fluorescence may be long lived, as long as 10"'^ second ("delayed" has unhappy connotations) if the singlet-singlet absorption is weak. On the other hand, phosphores- cence lifetimes are usiioUy long, in the range 10 to 10 "^ second. Nevertheless, under the action of certain perturbations, such as sub- stitutions in the molecule involving atoms of high atomic number (e.g., Br I), the inherent phosphorescent lifetime may be shortened as much as by a factor of 1000. Thus, a 10 '-second phosphorescence could be observed. Obviously, there is overlap between lifetimes of fluorescence and phosphorescence. Distinctions which are qualitative may be made, however, between fluorescence and phosphorescence as follows : ( 1 ) The excited state giving rise to phosphorescence is paramagnetic. This has been proved by direct susceptibility measurements in one case of phosphorescence. It is a difficult measurement and cannot be applied easily enough to R. S. BECKER AND M. KASHA 45 make it worth while. There is a possibihty that the new paramag- netic resonance absorption methods can be used to detect triplets generally. (2) The phosphorescence lifetime is sensitive to high- atomic-number-atom substitution in a unique way, whereas fluores- cence lifetime is generally insensitive to such substitutions. However, in some simple molecules (e.g., acetone) this distinction is not very clear, since substitution has several effects. It is possible that the new method of environmental perturbations (Kasha, 1952) may resolve such difficulties. (3) The clearest distinction between fluorescence and phosphorescence is the wavelength of the emission, if both emis- sions exist. In a few molecules, only one or the other emissions exists; this is a rare circumstance, but where it occurs, recourse can be made to steps ( 1 ) or ( 2 ) above. However, when both emissions can be observed (for which, of course, rigid glass solutions at low tempera- tures are used; see text), it is always true that phosphorescence is the longer-wavelength emission, whereas fluorescence is the shorter- wavelength emission (see text on "singlet-triplet splits"). Moreover, it can also be said that, if two emissions are found for a molecule, the shorter-lived emission is fluorescence, and the longer-lived is phos- phorescence, regardless of their absolute values. It is virtually inevi- table, spectroscopically, that these values will be quite clearly sepa- rated in magnitude. This third distinction will probably serve as the clearest and most valuable answer to Dr. Mason's questions. Lfght Saturation of Delayed Light Production in Green Plants* William Arnold Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee Three years ago it was shown in this laboratory ( Strehler and Arnold, 1951) that green plants emitted light for some seconds after being illuminated. The action spectrum for the delayed hght was shown to be the same as that for photosynthesis, and thus it was chlorophyll which absorbed the light energy that was reemitted. Later experi- ments have shown (Arnold and Davidson, 1954) that the emission spectrum of the delayed light is the same as the fluorescent spectrum of chlorophyll in the living plant, and thus that it is chlorophyll which emitted the delayed light. Although photosynthesis and delayed light production seem to be closely connected, a fact that has been emphasized in the past, in the manner of light saturation they are very different. Figure 1 gives the two curves for continuous hght made on aliquots of the same Chlorella pyrenoidosa suspension. As can be seen, the delayed hght production saturates at an intensity very much lower than photosynthesis. In the hope of understanding this difference, experiments have been done with short flashes of light. A ChloreUa suspension is pumped from a darkened container through a glass tube, a length D cm of which is illuminated with intensity 7. From the rate of pumping, the diameter of the tube, and D, the time T that each cell spent in the light can be calculated. After flowing in the dark for a short distance, the suspension passes in front • Work performed under Contract No. W-7405-eng-26 for the Atomic Energy Commission. 47 48 DELAYED LIGHT PRODUCTION of a photomultiplier which measures the intensity S of the delayed Hght. It is found that if T is smaller than a few hundreths of a second, then S is a function of the product IT, that is, S depends only on total energy E in the light flash. 26-1 1 1 r 14 18 22 relative light intensity Fig. 1. Figure 2 gives the result of the experiment. The insert gives the beginning of the curve for energies expressed as quanta per square centimeter for light of wavelength approximately 6500 A. Let us assume that the saturation shown by the curve is due to exhaustion of some substance K, and that the light changes K into KB. The return of KB to K generates the excited chlorophyll which emits the delayed light. Let X = amount of K at any time .Yo = amount of A' at the beginning of flash a ^ cross section for the reaction K -^ KB k — constant. We further assume that S, the delayed light intensity at the end of the flash, is given by k times the amount of KB, that is, by WILLIAM ARNOLD 49 S - A-(Xo - X) Since it is known that the reaction KB ^ K is slow, as is shown by the decay curve of the delayed light, we have during the light flash, dX dt = -alX This has the solution X = X,e-'iT ^ Xoe- -aE 300- ^y/6, SLOPE = 7.7 X lo"'^ mv/cm^ lOxlo'* ENERGY (hx/Cm*) T r— 1 1 1 I 10 12 14 16 18 20 e (arbitrary units) Fig. 2. Therefore, the intensity of the delayed light will be given by ,S' = A-Xo(l - e-'E) The initial slope of the curve will be dS dE = A-Xctr and the saturation value given by o = kXi) 50 DELAYED LIGHT PRODUCTION Therefore, the cross section for the reaction can be found by dividing the initial slope by the saturation value. From the data given in Figure 2 it can be seen that ^ = ^-^ ^^^^~'' = 3.1 X 10-1^ cm2 Since the cross section for the absorption of light by chlorophyll in the living plant has been given (Arnold and Oppenheimer, 1950) as 0.5 - 1.4 X 10-i« cm2 it can be seen that the reaction K -^ KB has a cross section 200-600 times larger than a single chlorophyll molecule. This large cross section, as well as the early saturation of the de- layed light, can be understood in the following way. Let ch + hp -^ ch* ch* -\- A ^ B + ch K + B^KB be the reactions involved in the early steps of photosynthesis and delayed light production (here ch stands for chlorophyll and c/i* for excited chlorophyll ) . The large cross section is due to the ability of any B to combine with one of the K's, since the ratio of chlorophyll to K is supposed to be very large and the reaction K -» KB is thought to be very fast. If it is now believed that either the amount of B or the amount of B plus KB determines the rate of photosynthesis, it is seen that the saturation of photosynthesis can take place at a much higher light intensity than the saturation of the delayed light. References Arnold, W. A., and J. B. Davidson. 1954. The identity of the fluorescent and delayed light emission spectra in Chlorella. J. Gen. Physiol, 37, 677-84. Arnold, W. A,, and J. R. Oppenheimer. 1950. Internal conversion in the photosynthetic mechanism of blue-green algae. /. Gen. Physiol., 33, 423-35. Strehler, B. L., and W. A. Arnold. 1951. Light production by green plants, ;. Gen. Physiol, 34, 809-20. Fluorescence Spectrophotometry of Photosynthetic Pigments C. Stacy French Department of Plant Biology, Carnegie Institution of Washington, Stanford, California Absorption spectrophotometry has played a large part in the develop- ment of modern biochemistry. Fluorescence spectrophotometry, on the other hand, has not been used as widely because there are many more substances that can absorb light than can reemit fluorescent light; furthermore, the measurements are somewhat more difficult to make. However, in suitable cases the value of fluorescence spectro- photometry equals or exceeds that of absorption spectrophotometry for pigment identification in live cells and in extracts and for quanti- tative analysis. Also, it is ideally suited to the study of energy transfer between pigments. During the course of an investigation on energy transfer between the pigments of red algae it became apparent that the basic data avail- able on the fluorescence spectrophotometry of photosynthetic organ- isms and of their purified pigments is very meager. The spectrophoto- graphic location of fluorescence peaks determined largely by Dhere has been reviewed by Rabinowitch (1951). Vermeulen, Wassink, and Reman ( 1937 ) measured the fluorescence energy distribution curves of Clilorella, the purple bacterium Chromatium, a plant extract con- taining chlorophyll a and b and an extract of purple bacteria con- taining bacteriochlorophyll. Zscheile and Harris ( 1943 ) made precise measurements of the fluorescence spectra of pure chlorophylls a and b in various solvents. Van Norman, French, and Macdowall ( 1948 ) measured the fluorescence curves of two red marine algae and of a water extract from one of them. Duysens (1951) published the 51 52 FLUORESCENCE SPECTROPHOTOMETRY fluorescence spectrum of the red alga Porphyra lacineata in a study of energy transfer between photosynthetic pigments. This study was extended to pigments of blue-green algae, purple bacteria and Chlo- rella (Duysens, 1952). French and Young (1952) also measured fluorescence spectral energy distribution curves of photosynthetic pig- ments in live cells and in extracts. They have reviewed the work prior to 1951 on absorption, action, and fluorescence spectroscopy of photo- synthetic pigments ( Hollaender, Ed. ) . Although the experiments here described have been limited to the study of photosynthetic pigments, the possible uses of these methods in other fields of biology and biochemistry, particularly in the field of luminescence, are evident. This paper will describe briefly the appa- ratus constructed for automatically plotting the spectral energy dis- tribution of weakly emitting light sources such as fluorescing leaves and will illustrate the use of such measurements in studying the prop- erties of pigments in solution and in living cells. Most of the material has been taken from more detailed papers already published or now in preparation; only a superficial survey will be attempted here. Some progress reports in this field have been published (Carnegie Insti- tution of Washington Year Books, 1948-1953). Apparatus Quantitative measurement of a fluorescence excitation spectrum re- quires illumination of the sample with bright light in a narrow spec- tral range and of known total energy as measured by a thermopile. This incident light is converted in the sample, usually with a very low efficiency, into fluorescent hght that is emitted in all directions. To measure the spectral energy distribution of the weak fluorescent light, as much as possible of it must be put through a monochromator and the intensity of its various wavelengths determined. In order to make such measurements practical it is necessary to start with a bright light source. Our high intensity source is a Bol type of high-pressure mercury lamp made by the Huggins Laboratories in Menlo Park, California. This is similar to the GE H6, but it dissipates twice the power in a narrower capillary. This lamp gives a very high intensity continuous spectrum as well as the mercury lines which are very greatly broadened. Even in the red part of the spectrum this C. STACY FRENCH 53 source is brighter than a tungsten lamp. We run the lamp on alter- nating current and thereby get 120 light pulses per second. An image of the lamp is focused on the slit of a monochromator by a spherical mirror. The monochromator is usually set to isolate a total band width of 10 m/x. Stray light is removed by filters. As shown in Fig. 1, a block diagram of the apparatus, this incident light may be measured by means of a thermopile in a sliding mount or may be allowed to fall upon the sample. mCN riissuii Ml liar THiaitoriii ICIftlMT \^ \ -, iicm ^ ^ ■ eaocmoiiiToi i li— _- •^\\ *""^ \ HVOBISCIKT MiTiat ■ (ilOl ' IICHT ■ OIIOCNIOIIITOI : fNOIOKUITirilll M'CID- Mints 1 c tariDiii Color Sansitivity Correction Mechanism (la tOIUSIlltl ITKNUITOI 110 CTCtI lUIID ilPIIMII llCllllil AUt flllll (UBVI FOIIOWII Wavelength iSweep Mechanisni L fiiiteii SPUD aoToi *c^l Poper Drive Pan Drive ^ SIltT* ■ OTOI SlliT« MOTOt . SIISTN GIN. -) SIISTR cm. ± ••OWN •(COI»l^ Fig. 1. Block diagram of the apparatus for recording fluorescence spectra. Most of the fluorescent light given off by the sample escapes, but some of it is caught by another curved mirror and sent into an ana- lyzing monochromator, which also isolates 10 m^u,, total band width. Behind this monochromator is a filter to remove reflected incident light and one to give some correction for the wavelength sensitivity of the photomultiplier tube. The electrical output of this photomulti- plier tube is amplified and passed through an attenuator which is linked to the wavelength drive of the analyzing monochromator by a cam. The cam is cut to compensate for the variation of photomulti- plier sensitivity with wavelength and also to include the varying 54 FLUORESCENCE SPECTROPHOTOMETRY transmission of the monochromator with wavelength. The cam takes care of most of the corrections, but some residual errors are later removed by another attenuator drive by a photoelectric curve fol- lower. The residual correction curve is put on a frosted Incite drum with a soft pencil and can easily be adjusted. The corrected electrical signal, now proportional to the intensity of the light emitted from the sample at each wavelength, is separated from random noise by FIRST ORDER COUNTER WATER COOLED HIGH PRESSURE MERCURY LAMP SCALE.; ' 0 12 3 4 5 INCHES Fig. 2. Transmission grating monochromator used for producing the incident light. means of a 120^ tuned amplifier, rectified to direct current and used to drive the pen of a Brown recorder. The paper in the recorder is moved synchronously with the wavelength drive by means of a Selsyn motor. It is therefore possible to vary the speed of wavelength sweep through various parts of the spectrum during a run. The monochromators are designed for high light gathering power and are built with approximately 10 by 10 cm gratings. The monochromator. Fig. 2, used to isolate the incident light has a replica grating with a blaze angle producing a high eflBciency in blue. The wavelength setting of this instrument is read directly on a counter. Figure 3 shows the arrangement of the incident and the fluorescent beams as well as the calibrating lamp. The color sensitivity correction cam and the final correction curve are made with light from a stand- C. STACY FRENCH 55 ard lamp of known color temperature reflected off a magnesium oxide block in the sample holder and interrupted by a 120-cycle chopper. The apparatus retains its calibration for months. FLUORESCENT LIGHT ENTERING SECOND MONOCHROMATOR TO BE ANALYZED, LIGHT CHOPPER TUNGSTEN STANDARD LAMP FOR CALIBRATION FLUORESCENT SAMPLE ORM3O FOR CALIBRATION MIRROR BLUE LIGHT FROM FIRST MONXHROMATOR Fig. 3. Arrangement for illuminating the sample and for collecting the fluorescent light. Figure 4 shows the monochromator used to separate the spectrum of the fluorescent light into its different wavelengths. This analyzing monochromator uses a reflection grating. The grating is rotated at an adjustable speed by a motor which is also coupled to the cam and the curve follower as illustrated in Fig. 1. Fluorescence Spectra of Extracted and Purified Pigments Phycoerythrin Figure 5 shows the absorption and fluorescence spectra of a sample of phycoerythrin. Comparison of the absorption and fluorescence 56 FLUORESCENCE SPECTROPHOTOMETRY COUPIIHC C(»« fjjKi /vAI Fig. 4. Reflection grating monochromator used for analyzing the fluorescent light. T T -r T T Abittttiei I II 0 1 1 > I ( • ( t 400 500 600 WAVELENGTH ± 700 m)L Fig. 5. Absorption and fluorescence spectra of pure phycoerythrin in water. Meas- urements of Dr. Violet K. Young on a sample of pure phycoerythrin kindly given us by Professor Lawrence R. Blinks. (Redrawn from Rabinowitch, 1951, Fig. 23.9A.) C. STACY FRENCH 57 curves of this pure pigmeut illustrates two well-known facts. First, the fluorescence spectrum looks somewhat like a mirror image of the ab- sorption curve, at least two peaks in this case appear to be reflections of two of the absorption peaks. Second, the major fluorescence peak is universally located near an absorption peak, but at a longer wave- length. The situation that can cause a great deal of trouble in fluores- cence spectroscopy is the overlapping of the absorption and the fluorescence bands. Here we see that the left-hand part of the fluores- cence curve can be very strongly absorbed by the pigment itself 600 650 700 750 m;i. WAVELENGTH Fig. 6. Fluorescence spectra in ether of pure chlorophylls a and h of Smith and Benitez compared with the recalculated curves of Zscheile and Harris ( 1943 ) . whereas the right-hand part is outside of the absorption region of the pigment. This means that except in extremely dilute solutions the observed curve is very likely to be greatly distorted by internal reab- sorption of the fluorescent light. This difficulty can be avoided, at least in solutions, by using very low concentrations or thin layers. Reabsorption within the sample can be the cause of a great deal of trouble in interpreting the spectra of living organisms which have a high pigment content. Chlorophylls a and h Since Dr. James H. C. Smith and Mr. Allen Benitez have recently 58 FLUORESCENCE SPECTROPHOTOMETRY been redetermining the absorption constants of purified chlorophylls in this laboratory, we had an excellent opportunity to repeat the measurements of Zscheile and Harris on the fluorescence spectra of freshly purified chlorophylls a and h in ether. The new curves are compared in Fig. 6 with those of Zscheile and Harris which have been recalculated to correct for the variation of the effective slit width with wavelength due to their use of a prism. The need for such a correction of their curves was pointed out to us by Dr. L. N. M. Duysens. The agreement of the chlorophyll h curves is close, but the difference be- tween the two chlorophyll a curves seems to be well outside the 650 700 750 800 850 m^i VfAVElENGTH Fig. 7. Fluorescence spectrum in ether of "bacteriochlorophyll," measurements of Vermeulen, Wassink, and Reman ( 1937 ) and of this laboratory. experimental error of either laboratory. Dr. Smith's absorption meas- urements of his chlorophyll o also show a somewhat similar wave- length shift. It appears that the chlorophyll a of the two laboratories may actually have been composed of different isomeric fractions. Bacteriochlorophyll The dotted curve in Fig. 7 shows the fluorescence spectrum of bacteriochlorophyll published by Vermeulen, Wassink, and Reman (1937). The figure also shows the measurements of partially purified bacteriochlorophyll prepared by Dr. Smith and Mr. Benitez. Pre- sumably the differences between the curves are due to the use of C. STACY FRENCH 59 narrower slit widths in this laboratory. We were rather surprised to confirm the finding of a fluorescence peak at 687 m;a since bacterio- chlorophyll does not have a corresponding absorption band. This bacteriochlorophyll preparation was separated chromatographi- cally by Dr. Smith into three fractions — one blue, one green, and one I 436 mil. miidciil 4 05 m|i ifiddr nt / 650 750 m)i 700 WAVELENGTH Fig. 8. Fluorescence spectrum of the blue fraction of a bacteriochlorophyll prepa- ration in ether. ( Smith, unpublished. ) 650 750 m^ 700 WAVELENGTH Fig. 9. Fluorescence spectrum in ether of the green pigment accompanying bac- teriochlorophyll. ( Smith, unpublished. ) 60 FLUORESCENCE SPECTROPHOTOMETRY pink. The fluorescence spectrum of the blue fraction when illuminated by two different incident wavelengths is shown in Fig. 8. Now it is ordinarily found that a pure substance illuminated with different wavelengths gives the same fluorescence spectrum, but this prepara- tion gives different spectra for the two incident wavelengths. It is clear that the material fluorescing at 687 m/x absorbs at 436 m/A more strongly than it does at 405 m^u., since 436 m/x excites the 687-m;u, band 1 ^- r\ 1 \ A In ether / j I In Ulvo 1 1 tu 1 \ \^ Z I uj ^J 1 sy* 1 Ui 1 ae 1 1 O 1 1 3 1 1 1 kA. 1 / \ 1 1 / \ \ J / \ \ / / / / \ \ / / / / \ \ 1 >-: 600 6S0 700 WAVELENGTH 7S0ni^ Fig. 10. Fluorescence spectrum of a thin layer of a live green plant as compared with an ether solution of chlorophyll a. (Smith, unpublished.) more than it does the near infrared fluorescence, whereas 405-m/A incident light does the reverse. The green fraction, Fig. 9, of the bac- teriochlorophyll preparation, however, shows only the 687-m/x fluores- cence band. This green pigment appears to be very simflar to bacterio- viridin of the green photosynthetic bacteria, and it may be normally a minor constituent of purple bacteria or may be a decomposition product formed during extraction. C. STACY FRENCH 61 Fluorescence Spectra of Pigments in Live Plants Chlorophyll a The fluorescence spectrum of chlorophyll a in ether is compared in Fig. 10 with that of chlorophyll a in a very thin layer of plant mate- rial, sea lettuce (Ulva). The fluorescence spectra, like the absorption spectra, of chlorophyll pigments are shifted toward shorter wave- lengths when taken out of the plant by organic solvents. The extent of this shift is, however, different in plants which have been grown in the light and in etiolated albinos that have very recently had their chlorophyll formed from protochlorophyll by exposure to light. Table I shows that the freshly formed chlorophyll a has its peak between TABLE I Wavelengths of the Chlorophyll a Fluorescence Peak State of Chlorophyll Wavelength, m/x In solution Ether Olive oil" Methanol" 668 672 674 In etiolated albino plants just after trans- formation from protochlorophyll Barley* Corn 675-677 676 In pale light-grown plants Barley* Albino ivy leaf Albino grape leaf" Ulva 684 681-682 680 681-683 " Zscheile and Harris, 1943. * Todd, Virgin, and El Wakeel, unpublished. that of chlorophyll in solvents and of chlorophyll in light-grown plants. Perhaps this freshly formed chlorophyll is not yet completely built into its normal state in nature as a protein complex. Now we will see what happens to the fluorescence spectrum of chlorophyll in living material when it is present in higher concentra- tions. Figure 11 compares the fluorescence spectrum of a normally green grape leaf with that of a white grape leaf that is apparently com- pletely devoid of chlorophyll, as far as one can see by eye or by the 62 FLUORESCENCE SPECTROPHOTOMETRY microscope. In the normal leaf the much higher concentration of chlorophyll gives an increase in the height of the 730-m/A fluorescence band which is not absorbed to any great extent. On the other hand, the height of the main peak at 685 m/A is nowhere near as high as would be expected from the amount of chlorophyll present. This relative decrease in height is due to selective reabsorption of the fluorescent light near the chlorophyll absorption band by chlorophyll itself. The incident blue light does not penetrate very deeply into 600 7S0 HI*. WAVELENGTH Fig. 11. Fluorescence spectra of chlorophyll a in an albino and in a normal grape leaf. ( El Wakeel, Virgin, and Todd, unpublished. ) the leaf. If a very dark green leaf is illuminated with blue light, as in Fig. 12, we find a spectrum which is greatly distorted, but neverthe- less the two peaks are clearly recognizable. If, however, this same leaf is illuminated with green light which penetrates the leaf, then the fluorescence that is emitted within the leaf is largely reabsorbed on its way out. In this case the deeper penetration leads to a higher emission of light at longer wavelengths, but the main fluorescence peak has practically completely disappeared. The fact that the main peak does not show up is presumably due to the weak absorption of the incident light in the outside layers of the leaf. An alternative possible explanation of the shape of these curves might be that there was another pigment present in the leaf which preferentially ab- sorbed green light and fluoresced at longer wavelengths. It appears C. STACY FRENCH 63 impossible to use such curves to substantiate the existence of other pigments with long-wavelength fluorescence bands, although that would be the more obvious conclusion to be drawn from them. Virgin (1954) has found that the distortion of fluorescence spectra is very strongly influenced by the degree of light scattering within the leaf. This distortion can be greatly reduced by infiltrating the air spaces of the leaf with water. S4iu iaddcflt 436111^ iMidtst 600 650 700 WAVELENGTH 750 BA Fig. 12. Fluorescence spectrum of a dark green leaf, Photinia arbutifolia, when illuminated by strongly absorbed blue light as compared with weakly ab- sorbed green light. ( Hill, Young, and French, unpublished. ) Protochlorophyll The fluorescence spectrum of purified protochlorophyll in acetone is illustrated in Fig. 13. The fluorescence spectrum of something which might be called a chloroplast preparation from etiolated barley is also given. This curve bothered us greatly for some time, because the measurements of the action spectrum for the transformation of protochlorophyll to chlorophyll (Koski, French, and Smith, 1951) 64 FLUORESCENCE SPECTROPHOTOMETRY showed that the absorption maximum of protochlorophyll in barley was at 650 m/A. The fluorescence spectra are normally at longer wavelengths than the absorption maximum. This discrepancy is now satisfactorily cleared up. It is due to an effect which is shown in Fig. 14, the fluorescence spectrum of the inner seed coat of banana squash. There is a fluorescence peak at about 635 and another one somewhat be- yond 650 m/A while the major peak comes at about 700 m/x. Offhand, In otctone \ I \ ln*\hloroplosf$" hA4 1 \ 1^ 1 1 z 1 \ UJ t 1 1^ 1 iy% h«j ec O 3 ^ \ i 1 1 600 700 750m; 6S0 WAVELENGTH Fig. 13. Fluorescence spectrum of protochlorophyll in acetone and in a heated water suspension of etiolated barley leaf macerate. (Smith, unpublished.) this looks like a beautiful composite curve for a mixture of three pig- ments. The acetone extract, however, gives a fluorescence curve for pure protochlorophyll. The residue, after acetone extraction, shows a trace of protochlorophyll fluorescence and nothing else. The absorp- tion curve of the extract agrees with that of protochlorophyll. Here we have an excellent illustration of data which can be most mislead- ing. The squash seed coat is a very highly colored material in which intense reabsorption of the fluorescent light takes place. The 705-m/x peak we believe to be merely the unreabsorbed minor long-wave- length band of protochlorophyll. The other peaks are the major C. STACY FRENCH 65 fluorescence peaks of protochlorophyll which may exist in the plant in two forms — one having an absorption peak at 635, the other at about 650 m/x. This explanation would be hard to believe were it not for the recent experiments of Dr. Smith in which he has clearly shown by absorption measurements the conversion of the 650-m/i, form in freshly ground dark-grown barley leaves suspended in glycerin to the 635-m/x form in suspensions that have been heated or allowed to stand. 600 650 700 WAVELENGTH 750 mji Fig. 14. Fluorescence spectrum of the inner seed coat of a banana squash. ( Hill, Smith, and French, unpubhshed. ) The fluorescence spectrum of a dark-grown leaf exposed to light for a short period of time is given in Fig. 15. Here we can clearly distinguish the fluorescence spectrum of protochlorophyll in the pres- ence of a considerably larger amount of chlorophyll. Dr. Hemming Virgin is at present looking into the possibility of using such curves for the quantitative measurement of rates of protochlorophyll trans- formation to chlorophyll in living leaves. He has found that much more reproducible results may be obtained by infiltrating the leaves 66 FLUORESCENCE SPECTROPHOTOMETRY with hot water before measurements. It is of course necessary to heat the leaves to prevent further transformation by the Hght used for exciting the fluorescence and this heating itself shifts the fluo- rescence peak of protochlorophyll from 655 to about 638 m/i. Chlorophylls b and c The fluorescence band of chlorophyll b at 655 m/x in several different species of plant has been reported ( Rabino witch, 1951, p. 807). This o CMerofbTli a PrttecbUrophyll 600 7S0in» 650 700 WAVELENGTH Fig. 15. Fluorescence spectrum of a partially greened barley leaf. (Young, un- published. ) band has never been found in any of our curves of living plants. We have measured very pale leaves with the hope of finding chlorophyll b fluorescence which might have been reabsorbed in more highly colored material, but it has never appeared. There are several possible reasons for its lack. One may be that chlorophyll b is appreciably less fluorescent and that it occurs in smaller amounts than chlorophyll a. Another reason may be that the efficiency of energy transfer from chlorophyll b to chlorophyll a is very high. This transfer has been found by Watson and Livingston (1950), Young (1952, Carnegie Institution Year Book No. 51), and Duysens (1952) in ether solutions of the mixed pigments. A less interesting but also possible reason may C. STACY FRENCH 67 be that the chlorophyll b fluorescence band is near enough to the absorption maximum of chlorophyll a to be strongly reabsorbed. The fluorescence spectrum of two suspensions of the diatom Nitschia, kindly given us by Professor C. B. van Niel, which is presumed to contain chlorophyll c, is shown in Fig. 16 as compared with the fluorescence spectrum of the white part of an ivy leaf. Nitistliio svspention 600 iSO 700 WAVELENGTH 7S0in;i Fig. 16. Fluorescence spectra of a thick and a tliin suspension of Nitzschia clos- terium, minutissima, as compared with that of clilorophyll a in the white part of a variegated ivy leaf. The diatom culture was kindly gi\ en us by Professor C. B. van Niel. Instead of finding a chlorophyll c band at the anticipated position of approximately 640 mix, a new band appeared at about 705 m/x. The cause of this 705-m/ut fluorescence is not known. One possible interpre- tation might be that it is indeed due to chlorophyll c, but it is the longer wavelength subsidiary band which could reasonably be ex- pected to be at about this position and that the absence of the main band, anticipated at about 640 m/x, is due to its reabsorption by chlorophyll a within the diatoms. The cells themselves are quite dark so that very strong reabsorption of fluorescence, even within a single cell, might not be inconceivable. The action spectrum for the excita- tion of fluorescence in this material, Fig. 17. confirms again the participation of hght absorbed by fucoxanthin in the excitation of 68 FLUORESCENCE SPECTROPHOTOMETRY chlorophyll fluorescence. It also shows that other carotenoids absorb- ing at shorter wavelengths than fucoxanthin are less effective in transferring energy to chlorophyll. 400 500 600 WAVELENGTH 700 m>j. Fig. 17. The absorption spectrum of the thin suspension of Nitzschia in dilute agar (small circles), as compared with the action spectrum for fluorescence excitation (heavy line). The lower straight line is the estimated contribution of scattering to the measured absorption; it was used as the baseline above which the action spectrum was plotted. Phycobilins and Chlorophylls in Red Algae The spectra of red algae demonstrate some of the more complex fluorescence phenomena. They contain the fluorescent pigments phycoerythrin and phycocyanin in addition to chlorophylls a and d so that fluorescence spectra of mixtures are obtained. Haxo and Blinks (1950) have found that the hght absorbed by phycoerythrin and phycocyanin is used in photosynthesis even more efficiently than that absorbed by chlorophyll. Do these accessory pigments act directly in photosynthesis or do they participate by transferring their energy to chlorophyll a, which seems to be a more or less universal photosynthetic pigment? The fluorescence spectra for various incident wavelengths of measured intensity from Porphyri- diiim cruentum are given in Fig. 18. We see the typical peaks due to phycoerythrin, 678 mix, phycocyanin, 655 m/x, and chloroph>'ll a, at 685 m/x. Figure 19 shows the analysis of a similar curve in terms of the C. STACY FRENCH 69 600 650 700 750 WAVE LENGTH IN MU Fig. 18. Fluorescence spectra of Porphijridium cruentum excited by various inci- dent wavelengtlis. The curves are reduced to their relative sizes for the same number of incident quanta at all wavelengths. (French and Young, 1952.) 700 750 600 650 WAVE LENGTH IN MJU Fig. 19. Estimated contribution of three different pigments to the total fluores- cence of Porphijridium when illuminated by wavelengtli 530 m/^. (French and Young, 1952.) 70 FLUORESCENCE SPECTROPHOTOMETRY fluorescence contributed by each of the three pigments. A similar dissection of these curves and of five more of the same set with different incident wavelengths was carried out by means of a graphical computer (French et ah, 1954) so that we could calculate the efiFectiveness of each wavelength in exciting fluorescence of each of the three pigments separately. A plot of the effectiveness of different wavelengths in exciting chlorophyll fluorescence is given by the circles in Fig. 20, which also shows the absorption spectra of chloro- CO LU I— CC o (X o CO m 400 450 500 5 50 600 650 760 WAVE LENGTH IN MU Fig. 20. Action spectrum for the excitation of chlorophyll a fluorescence in Por- phijridium compared with the absorption spectra of water solutions of phyco- trythriii and phycocyanin and on ether solution of chlorophyll a. (French and Young, 1952.) phyll a, of phycoerythrin, and of phycocyanin. There is some fluo- rescence by direct absorption of blue light by chlorophyll a, but strangely enough the points rise along with the phycoerythrin absorp- tion curve in the green part of the spectrum where chlorophyll absorbs weakly. This shows clearly that phycoerythrin is an effective ligh*- absorber for the excitation of chlorophyll fluorescence. Furthermore, the phycocyanin fluorescence, as well as that of phycoerythrin, also C. STACY FRENCH 71 follows the same type of excitation curve, except that these two drop to a low minimum at about 435 m/x. It therefore appears in agreement with Duysens (1952) that phycoerythrin absorbs energy and transfers it to the phycocyanin which passes it along to chlorophyll. Presumably this same process may be followed by the energy used for photo- synthesis, although this conclusion of course involves several assump- tions. The effect of the intensity incident on a red alga upon the shape of the fluorescence curve is shown in Fig. 21. These curves were meas- I I -I — 1 — I I I — I — I I ' I I ' ' — r- LOW INCIDENT - INTENSITY . u^-^j. I ^^ I I 'DIFFERENCE ^^^ ^ /'between low and HIGfi- ' INTENSITY CURVES J 1 1 J I — 1 — 1 — I — 1- 600 650 700 WAVE LENGTH IN MJJ 750 Fig. 21. Fluorescence spectra of a red alga after illumination for 6 minutes with three different intensities. Only the chlorophyll peak is reduced in height by prolonged illumination at high intensity. (French and Young, 1952.) ured, as were all the others in this article, after the initial changes of intensity of fluorescence, the Kautsky effect, had come to completion, so that constant readings were obtained. These changes are connected with the induction period of photosynthesis. A great deal of specula- tion has been aroused by observations of these phenomena. Here we see that low intensity gives far more chlorophyll fluorescence in pro- portion to that of phycoerythrin and phycocyanin than do higher intensities. This is another manifestation of the Kautsky effect. Here 72 FLUORESCENCE SPECTROPHOTOMETRY it is evident that it is only the chlorophyll fluorescence and not the fluorescence of phycoerythrin and phycocyanin that is influenced by the Kautsky effect. It thus appears that the products of photosynthesis which somehow or other quench fluorescence act only upon chloro- phyll, either because only chlorophyll is sensitive to such quenchers or because the immediate products of photosynthesis are more closely associated with chlorophyll than with the other pigments. A New Leaf Pigment Found by Fluorescence A paper by Goodwin, Koski, and Owens (1951) describes a new leaf pigment, traces of which are present in the cells surrounding the I 1 - -1 1 — ' — I 100 - A - UJ III o II \ z 1 ' \ lij 80 V 1 I a> O 1 \ (/> 1 I UJ ' I §60 . ; \ - 3 1 1 / ^*\ .-1 1 \ / / 1 U. 1 i\ y / \ 40 - 1 *»v ^ y '' \ ~ Id \ x.^>< \ > *fc^ / ' \ ^"^ / \ ^^^^^ ^ I \ / \\ b \ / V. < 20 - 1 \ _-'-'' \ *" UJ 1 N v a: n / 1 1 1 , \ 560 600 640 680 WAVE LENGTH IN m/j 720 Fig. 22. Fluorescence of the methanol extracted epidermis of Vicxa (solid line), compared with uroporphyrin I octamethyl ester in methanol (dotted line). (Goodwin, Young, and Owens, 1952.) guard cells of certain species of Vicia. This pigment was discovered with a fluorescence microscope and was reasonably well identified solely by fluorescence spectroscopy. The guard cells making the stomata contain chlorophyll and show a red fluorescence. In the cells around them there are small bodies having a somewhat more orange fluorescence. A little piece of the epidermis peeled off from the rest of the leaf shows chlorophyll fluorescence and also another peak at ap- proximately 615 m/x. Some of this epidermis was treated with methanol to remove the chlorophyll, but the new fluorescing pigment was not extracted by methanol. Figure 22 shows the fluorescence spectrum of C. STACY FRENCH 73 the epidermis wet with methanol after extraction of the chlorophyll. In comparison with this is also drawn as a dotted line the fluorescence spectrum of uroporphyrin- 1-octamethyl ester showing a rather close agreement. The fluorescence spectrum of a solution of the pigment in hydrochloric acid also agreed with that of a hydrochloric acid solution of this porphyrin. Now it is well known that porphyrins generally have a very intense absorption band in the blue part of the spectrum known as the Soret band. The trace amounts of this new fluorescent pigment material and its insolubility made it impossible to measure its absorption spectrum. However, the effect of various wavelengths in exciting fluorescence of the pigment in epidermal peels was measured. This action spectrum showed points of equal height at about 405 and 410 m/x and dropped to lower values on both sides. By interpolation it was found that the Soret band of this pigment whose absorption could not be measured directly is at 408 m/*. ACKNOWLEDGMENTS Extended collaboration with Drs. Violet K. Young, James H. C. Smith, and Hemming I. Mrgin produced most of the results here described. Shorter periods of investigations with Drs. Robert Hill and Glenn Todd opened up new aspects of the field. The electronic components of apparatus were designed and constructed by Messrs. Bertram G. Ryland, Hemy G. Fatten, Jr., George H. Towner, and Dr. Donald R. Scheuch. The mechanical parts ,of the monochromators were built by Mr. Frank Schuster. The replica trans- mission grating was made available by Professor R. W. Wood, and the original reflection grating by Mr. Harold Babcock. References Carnegie Institution of Washington Year Books: 1948, No. 47, 96-97; 1949, No. 48, 93-94; 1950, No. 49, 86-88, 99-100; 1951, No. 50, 122-123; 1952, 51, 153-154; 1953, No. 52, 153-157. Duysens, L. N. M. 1951. Transfer of Light energy within the pigment sys- tems present in photosynthesizing cells. Nature, 168, 548. Duysens, L. N. M. 1952. The transfer of excitation energy in photosynthesis. Thesis. Utrecht. 74 FLUORESCENCE SPECTROPHOTOMETRY French, C. S., G. H. Towner, D. R. Bellis, R. M. Cook, W. R. Fair, and W. W. Holt. 1954. A curve analyzer and general purpose graphical com- puter. Rev. Sci. Inst., 25, 765-75. French, C. S., and V. K. Young. 1952. The fluorescence spectra of red algae and the transfer of energy from phycoerythrin to phycocyanin and chlorophyll. /. Gen. Physiol, 35, 873-90. Goodwin, R. H., V. M. Koski, and O. v. H. Owens. 1951. The distribution and properties of a porphyrin from the epidermis of Vicia shoots. Am. J, Botany, 38, 629-35. Haxo, F. T., and L. R. Blinks. 1950. Photosynthetic action spectra of marine algae. /. Gen. Physiol, 33, 389-422. Hollaender, A. (Ed.), Radiation Biology, Vol. III. In Press. Koski, V. M., C. S. French, and J. H. C. Smith. 1951. The action spectrum for the transformation of protochlorophyll to chlorophyll a in normal and albino corn seedlings. Arch. Biochem. and Biophys., 31, 1-17. Rabinowitch, E. I. 1951. Photosynthesis and Related Processes. Vol. II, Part 1, Interscience, New York-London. Van Norman, R., C. S. French, and F. D. H. Macdowall. 1948. The absorp- tion and fluorescence spectra of two red marine algae. Plant Physiol, 23, 455-66. Vermeulin, D., E. C. Wassink, and G. H. Reman. 1937. On the fluorescence of photosynthesizing cells. Enzymologia, 4, 254-68. Virgin, Hemming I. 1954. The distortion of fluorescence spectra in leaves and its reduction by infiltration. Physiologia Plantarum, 7, 560-70. Watson, W. F., and R. Livingston. 1950. Self quenching and fluorescence of chlorophyll solutions. /. Chem. Phys., 18, 802-09. Zscheile, F. P., and D. G. Harris. 1943. Studies on the fluorescence of chlorophyll. The effects of concentration, temperature, and solvent. /. Phys. Chem., 47, 623-37. Kinetics of Chemiluminescence of the 2,3-Dihydrophthalazine-l,4-diones Paul C. Wilhelmsen," Rufus Lumry,! and Henry Eyring Department of Chemistry, University of Utah, Salt Lake City, Utah The purpose of this paper will be to present some new data from experiments on the chemiluminescence of the 2,3-dihydrophthalazine- 1,4-diones (DPD) and in particular the 5-amino derivative (luminol). In addition, a brief discussion will be given of the known facts on which a mechanism for the chemiluminescence of the DPD's must be based. A tentative mechanism will be proposed which explains the available facts and points out possible directions for further investiga- tion. The most important data that this paper will discuss have to do with the kinetics of the reaction. Chemiluminescent reactions such as the oxidation of the DPD's are particularly well adapted for kinetic studies. The intensity of the hght emitted by a chemiluminescent reaction is the number of photons (I) emitted per second. The number of photons emitted is in turn related to the number of excited _ ^^.■Jf'>'^'^ molecules A* and the rate of emission k^: I = M* (1) The number of excited molecules is a function of their rate of produc- tion B ^ .4* (2) * A research fellowship granted by the University of Utah Research Com- mittee which greatly aided this author's contribution to this work is grate- fully acknowledged. The work was partially supported by the United States Atomic Energy Commission. f Present address: Institute of Technology, School of Chemistry, Uni- versity of Minnesota, Minneapolis, Minnesota. 75 76 KINETICS OF CHEMILUMINESCENCE minus the rate of internal quenching A*^A + heat (3) the rates of external quenching J^(A* + Qi -^ A + Q, + heat) (4) ^ • -t- Vt 1 = 4 and the rate of emission of light A*^A^ hv (5) The concentration of excited molecules as a function of time is given by dA* dt = hB - (^-1 + h + Yj ^'^Qi)^* (6) 1 = 4 The fraction of the excited molecules that produce light is « = '^- 1 ^ 71 ki + ^-3 + X) ^^Q- (7) This fraction is equal to the rate of emitting light divided by the total rate of destruction of the excited molecules. If one knows the relation- ship between these various reaction rates, it is possible to relate the intensity of the light to the rate of production of excited states and therefore to the rate of the reaction. However, for the purpose of this paper, it is assumed that (f>, the fraction of excited molecules that produce light, is independent of the concentration and nature of the oxidants employed. This would be true as long as the principal mode of quenching is by means of intramolecular transfer of energy or when the concentrations of the external quenchers remain constant. The effect of temperature on 4, will be determined by an experimental study of the quantum yield of a particular reaction as a function of temperature. Historical The DPD's and, in particular, luminol are well suited for kinetic studies. The light emitted on oxidation is relatively bright and luminol at least is readily available. Because of the availability of luminol, p. C. WILHELMSEN, R. LUMRY AND H. EYRING 77 most of the kinetic studies have been done on this compound. How- ever, there is evidence that the mechanism is the same in each case, and most information concerning the reaction of one of the DPD's can be extended to the remainder. This similarity has been found true in the investigations of Zellner and Dougherty (1937) where it was found that the rate of oxidation of many of the DPD's is the same when measured by the rate of evolution of nitrogen. The apparent difference in rates, when the intensities of the luminescence are compared, is probably due largely to the diflFerent efficiencies of the emitters produced in each case. An additional possibility is that there are two or more reaction routes and that different fractions of some of the DPD's proceed by the luminous and nonluminous reactions, in which case some forms of information about one of the DPD's could not be simply extended to the others in the series. Further investigations are needed in order to deterr.iine the extent to which this second possibility must be considered. The Role of Oxygen Many different substances and procedures have been employed in connection with the DPD's to produce chemiluminescence. Fre- quently, hydrogen peroxide is used either alone or with various other agents such as hemin, potassium ferricyanide, or sodium hypochlorite ( Albrecht, 1928 ) . Chemiluminescence can also be produced by adding solutions such as ferricyanide or hypochlorite to an alkaline solution of the DPD. Some investigators have employed electrolysis (Harvey, 1929; Spruit, 1950; Bremer, 1953) and supersonics (Prudhomme, 1949; Flosdorf et ah, 1936). However, in the final analysis, either oxygen or a compound that readily decomposes to form oxygen, e.g., hydrogen peroxide or sodium hypochlorite, appears to be indispensable. It might seem that the oxidation of the DPD's by ferricyanide does not appear to require oxygen. However, as shall shortly be seen, the chemilu- minescent reaction is largely dependent on dissolved oxygen. Inasmuch as oxygen is so intimately involved in the chemilumines- cence of the DPD's, it will be interesting to study the reaction between oxygen and luminol. Some authors (Harvey, 1929; Drew, 1938) have reported that no light was observed when oxygen was bubbled through a luminol solution. Their failure to observe the light 78 KINETICS OF CHEMILUMINESCENCE was probably due to lack of suitable light detection equipment and improper choice of concentrations. With the proper concentrations of reagents, it is possible to produce a light visible to the dark-adapted eye. A solution of one gram of luminol per liter of 2N NaOH has been found satisfactory in this connection although other concentrations will also produce light, especially when heated. This reaction between luminol and molecular oxygen has been studied briefly by Sveshnikov (1938), Bernanose et al. (1947), and Bremer (1953). However, the results contained in this paper are the first extensive kinetic studies that have been carried out on this interesting reaction. Some qualitative aspects of the reaction involving oxygen and the importance of dissolved oxygen on the emission of light in the case of oxidation by ferricyanide are shown in the following experiment. Experimental A series of A tubes was arranged as in Fig. 1. The first A tube arm contained a solution of potassium ferricyanide, the second arm con- FlG. 1. tained a solution of 0.05M luminol in O.IN NaOH. The first arm of the second A tube contained a dilute suspension of luminous bacteria such as A. fischeri or P. phosphoreum^ in a glucose media. The second arm of the second A tube was filled with water. The arm containing luminol was heated to 60° C. A readily visible light was produced by the luminol solution that was about ten times as bright as the lu- minous bacteria. However, when the oxygen was removed by flushing * These bacteria were kindly furnished by Dr. Frank Johnson. p. C. WILHELMSEN, R. LUMRY AND H. EYRING 79 the system with oxygen free hydrogen, the hght from the luminol was extinguished. It required less than a minute of bubbHng hydrogen through the system before the hght from the luminol faded below the visible threshold. The luminous bacteria continued to glow for a cou- ple of minutes after the luminol ceased luminescing. This indicates that a higher concentration of oxygen is required by the luminol to produce a visible light than is required by luminous bacteria. If the ferricyanide solution is now added to the luminol solution by tilting the A tube, the bright chemiluminescence characteristic of the reaction luminol and ferricyanide when saturated with air is produced. An hour of bubbling hydrogen through the solutions did not visibly affect the intensity of the light produced by luminol and ferricyanide. This means that a smaller quantity of oxygen is required in this case than is used with luminous bacteria — a very small quantity indeed considering that a partial pressure of 0.0007 mm Hg of O2 is all that is required by bacteria ( Harvey and Morrison, 1923 ) . However, when the hydrogen was bubbled through the system for twenty-four hours, the light produced by the addition of ferricyanide to luminol became faint and of short duration. Bremer (1953) performed a similar group of experiments where the oxygen was removed by repeated freezing of the solution and evacuation of the space above. She has reported results similar to those just described. The small residual amount of light produced with luminol and ferricyanide in the supposed absence of oxygen is possibly due to a small trace of oxygen still remaining. However, Bremer (1953) was not able to extinguish the residual glow by extended treatment with repeated freezing and evacuation of the system, and the present authors found no difference in the light obtained when the hydrogen was passed through the system for one day or for five days. The hydrogen employed in these experiments was purified by passing over platinized asbestos and might conceivably have contained a small trace of oxygen. The existence of a second, less efficient mode of chemiluminescence is the alternative. Another line of evidence that makes it troublesome to exclude the existence of this less efficient reaction comes from another source. It has been found that hydrazine and ferricyanide give out a feeble 80 KINETICS OF CHEMILUMINESCENCE but visible light.* The intensity of the chemiluminescence of hydrazine appears to be independent of oxygen concentration. This is interesting as luminol and the DPD's in general can be thought of as substituted hydrazine. This being the case, it should not be surprising if both were to undergo many of the same reactions, which suggests that the DPD's undergo a similar weakly luminous reaction with ferricyanide in the absence of oxygen. i 50 Oxygen (ml) Fig. 2. Intensity versus time curve for luminol being oxidized by ferricyanide in the presence of oxygen. It was found by Bremer ( 1953 ) that if oxygen was admitted after the ferricyanide and luminol have been allowed to react under anaero- bic conditions, the solution would emit a reasonably bright light. The present authors have been able to confirm this observation and, in addition, have found that if a few minutes is allowed to elapse between the mixing of the luminol and ferricyanide, no light is pro- duced. The importance of Bremer's observation will be discussed later in connection with the proposed mechanism. The second observation is important in that it indicates that ferricyanide does react with luminol in the absence of oxygen. This is proof that there is at least one other reaction route besides the chemiluminescent one involving oxygen. The experiments just described indicate that, in any case, * Unpublished work of the present authors. p. C. WILHELMSEN, R. LUMRY AND H. EYRING 81 very small quantities of oxygen greatly affect the efficiency of the chemiluminescence. A difference in the rates of the reaction with oxygen in the presence and absence of an additional oxidant might be inferred from the difference in quantities of oxygen required to produce a visible light in each case. However, a better idea of the relative rates might be obtained from a study of the length of times required for the reactions to proceed to a certain fraction of completion. The reaction involving only dissolved oxygen requires weeks or months to proceed even a small fraction of the way to completion. The reaction where an addi- tional oxidant such as ferricyanide is present is essentially complete in a matter of minutes. Figure 2 shows a typical plot of intensity versus time for a solution that was 0.05M luminol, 0.005A/ ferricya- nide and l.ON NaOH. The kinetic study of the reaction between the DPD's and oxygen would at best be very time consuming if the reaction were carried to completion. However, as was stated earlier, the reaction rate can be related to the intensity of the chemiluminescence if the temperature dependence of (f> is known. An experimental determination of this temperature dependence will now be discussed. Experimental Temperature Dependence of The reaction between ferricyanide and luminol in the presence of oxygen was chosen for a study of the temperature dependence of because of its relative freedom from complications and tlie ideal speed of the reaction for kinetic studies. Measurements were made by using the equipment diagrammed in Fig. 3. The solution of luminol and oxygen was mixed with the ferri- cyanide solution by means of a thermostated mixer, 5, similar to the one employed by Chance (1940). The solutions were thermostated and stored under pressure in flasks 2 and 3, and when the double stopcock 4 was opened, the solutions rapidly mixed and flowed along a 1-mm capillary. The solutions were allowed to flow until the light emitted reached a steady value. The flow was then stopped and the intensit>' of the light from a small quantity of solution in a short section of the capillary was studied. The Hght was measured by a photomultiplier, 7, which was in turn connected with an oscilloscope, 82 KINETICS OF CHEMILUMINESCENCE Mixer Photonnultiplier amplifier and milllvoltmeter 8 Fig. 3. 9. In this manner, the intensity of the hght was recorded as a vertical deflection and time was measured by means of the sweep along the X axis. A typical oscilloscope trace is given in Fig. 4. The final portion of the reaction is quite slow and was followed by periodic readings with a millivoltmeter 8. A typical intensity versus time curve is given in Fig. 3. Graphical integration of the total quantity of light emitted •♦••♦•.•^, •^•••..*^ •*t«M«« *«»• Fig. 4. Typical oscilloscope trace. The dots are %o second apart. p. C. WILHELMSEN, R. LUMRY AND H. EYRING 83 was carried out on several runs made at different temperatures, but with the same initial concentrations of reagents. These values are given in Table I and are plotted in Fig. 5. The variation in the quantity of light emitted was assumed to be proportional to the change in cp. To obtain an equation for the 950 875 800 \ \ C \ oV \q \ 2 30 3 10 Degrees absolute Fig. 5. Total light \ersus temperature. temperature dependence of it is necessary to return to equation 7. If it is assumed that one particular mode of quenching is the principal one, equation 7 becomes 0 = A-i ^-1 + k' (8) The rate of emission of light by an excited molecule is usually inde- pendent of temperature and therefore any temperature effect will generally be on k'. The rate fc' is given by the rate equation k' = h (9) 84 KINETICS OF CHEMILUMINESCENCE -G- S X ~e lO O lO i-O o I^ to t^ (N CO O i— I .-H (M CM CO cC' -V o liO rM CO CO CO CO Q 5^ 'vj iQ rs) ^ C: CTj 00 00 -^ O O O O d d d d d I> lO O lO to (M Oi O Oi O O GO 00 t"^ r^ o lo oo as o lO CO c-3 ^ >-; CO CO CO CO CO O-l Ci i-^ O lO -t^ ,-H ^ O O d> d d ci d> •^ CO O 00 O O LO c; o o o o o -^ lO lO CO 00 C2 d 00 -H CO d 00 C5 o ^ — (M CM CO CO CO OS OO -f CM CO 00 -f i> '-^ CO ■-H CM CM CO CO Oi 00 -t -f CO 00 ^ t^ '-' CO — CM CM CO CO d d d d d o I o 'S o .1— ( a; ^ H-1 p. C. WILHELMSEN, R. LUMRY AND H. EYRING 85 Substituting this expression for k' into equation 8 gives ^^ h n Rearranging and taking the logarithm of this expression gives l - (f) ki h (11) The terms /c,, k, h, k, and e^st/jj either are or may be assumed to be temperature independent. The plot of ln(l - )/T. This might justify making an approximation of (^ at some particular temperature and using it to calculate values at other temperatures. A lower limit for <^ in the case of luminol would be 0.003 which is the observed overall quantum yield for the chemiluminescence of luminol (Harris and Parker, 1935; Bernanose, 1951; and Bremer, 1953). Using this value at 20° C and making the appropriate calcula- tions lead to a value of 2.6 kcal for aHK This value assumes that all the luminol molecules become excited and ignores the existence of side reactions. Another value for (f> at 20° C which might be more reasonable is 0.1 which is based on the observed fluorescence yields of luminol (Spruit, 1950). Although the use of this value is open to the objection that the emitting molecules are not the same in the two cases, the molecules are probably similar and the fluorescence yield of a mole- cule is essentially cj>. Values are tabulated in Table I for <^'s based on being 0.10 at 20° C. The use of this group of <^'s give a value of 1 kcal for AH*. The value of 1 kcal and the value 0.1 for at 20° C gives an expression 1 1^ M 9^ '^"14- O.ieQTe^ooo/'^^ ; 86 KINETICS OF CHEMILUMINESCENCE for the temperature dependence of which fits the experimental data reasonably well. An alternative explanation for the variation in the efficiency of light production is that there is a change in the fraction of molecules passing along the reaction route favorable to light production. Inasmuch as there is only a relatively small change in the quantity of light emitted as the temperature is changed, the error in assigning the entire effect to changes in is independent of concentrations. Values for d In I/dt are given in Table I. If the standard state is taken as 1 mole per Hter, the values for d In I/dt divided by the concentration of the ferricyanide will give values for fc". These values are also tabulated in Table I. The rate equation for k' is k' = .^e-AZ/i/flreAsV/s (9) h If this is divided through by T and the logarithm is taken, the equa- tion becomes 1,^.1„,^-_A^ + ^ (15) ^'' T h RT ^ R As can be seen, a plot of In k'/T versus l/T should give a straight line p. C. WILHELMSEN, R. LUMRY AND H. EYRING 87 of slope —aH^/R. The value obtained from Fig. 6 for aH^ is about 3 kcal. Substituting of this value into equation 15 and assuming k is equal to 1 yields an entropy of activation. The value obtained at 20.3° C is 38 e.u. The standard state used for calculating the entropy in this case was 1 mole per liter. o 3.10 3.30 I/Txl03 Fig. 6. Kinetics of Reaction Involving Oxygen as Oxidant Equipment The investigations of the kinetics of the reaction involving only oxygen as the oxidant were made with the apparatus of Fig. 7. The reaction vessel 1 was a 3000-ml round bottom Hask with a graduation at the 3000-ml mark. It was placed in heating jacket 2 so that the temperature could be varied at will. The reaction vessel was stirred with a magnetic stirrer, 3. The flask was fitted with two con- nections, one of which was to the flask containing oxygen, 6, and the other to a manometer, a dry ice trap, 7, and a vacuum pump. 88 KINETICS OF CHEMILUMINESCENCE A flask of known volume was used to add oxygen to the system. Before adding oxygen the entire system was evacuated to as low a pressure as possible. Stopcock 4, connecting the reaction vessel with the vacuum pump, was then shut off, and the system allowed to come to equilibrium with respect to water vapor. Stopcock 5 was then closed, and oxygen was added to the flask 6 and the pressure was noted. By noting the pressure change when 5 was opened, a check could be obtained on the volume occupied by the oxygen in the manometer and reaction vessel. A small correction was found neces- Magnetic stirrer 3 Vi/ To vacuum Fig. 7. sary to compensate for the different volume displaced by the mercury in the manometer as the level changed. The temperature of the reac- tion solution was measured by a 0.1-degree thermometer. Procedure The distilled water used was boiled to remove as much dissolved oxygen as possible. Sodium hydroxide and luminol were added, and the solution was boiled under reduced pressure to remove as much of the residual oxygen as possible. The solution was concentrated to the correct volume by boiling off the excess water, and the temperature was adjusted. Pure oxygen was added, and the pressure was noted. The solution was stirred by means of the magnetic stirrer, and the rate p. C. WILHELMSEN, R. LUMRY AND H. EYRING 89 of absorption of oxygen was followed by the change in the pressure of the oxygen. Because of the slowness of the reaction, the quantity of oxygen absorbed can be taken as the concentration and can be pre- sumed uniformly distributed throughout the solution. Various tests were performed to test the validity of the various assumptions. As was stated earHer, the volume of the manometer and reaction vessel was checked by comparison with the vessel used to store oxygen. There was good agreement between the volume de- termined this way and by actual measurement. Distilled water was 150 100 50 X o_ 30 60 Seconds Fig. 8. added to some of the solutions and the experiment was repeated to check the assumption that the luminol concentration remained es- sentially constant throughout the experiment. There was no difference in repeated runs greater than the experimental error from other sources. The rate of absorbing oxygen was changed by varying the pressure of the oxygen, but no difference was noted when intensity was related to the quantity of absorbed oxygen. Results Measurements were made relating the intensity and the change in pressure of the oxygen above the reaction solution. The change in 90 KINETICS OF CHEMILUMINESCENCE pressure can in turn be expressed in milliliters at S.T.P. of oxygen absorbed. There was a small amount of light even before the oxygen was added. This varied from one experiment to another, depending on the thoroughness of the treatment to eliminate the dissolved oxygen. This was taken to mean that a small quantity of oxygen was still in the solution. The linearity of the plot of oxygen absorbed versus -0.40 -0.20 3.50 3.30 1/Txl03 Fig. 9. intensity permitted the extrapolation to zero concentration of oxygen. These results are given in Fig. 8. Tests were made at different temperatures. The plots of intensity versus milliliters of O2 absorbed in Fig. 8 make it possible to read intensities for any given concentration of oxygen as a function of temperature. Equation 12 permits the conversion of intensities into rates. Figure 9 gives the rates for solutions that have absorbed 4 ml p. C. WILHELMSEN, R. LUMRY AND H. EYRING 91 of O,. corrected to S.T.P. plotted as the In k/T versus 1/T. The slope of this line gives a AH* of 19 kcal using the rate equation. Without some way of obtaining absolute values for k\ it is impossible to cal- culate the entropy of formation of the activated complex. Information Available about Mechanism Kinetic Data The data that have been presented are suggestive with respect to the mechanism for chemiluminescence. Inasmuch as the reaction with oxygen alone proceeds very slowly and that with oxygen and an additional oxidant proceeds much more rapidly, it can be reasoned that the mechanism has as its slow step, an oxidation by oxygen which, however, proceeds more rapidly with ferricyanide. Furthermore, it has been demonstrated that oxygen is required for efficient chemilumines- cence of the DPD's even when ferricyanide is present. Bremer's experiments, in which she found that light was produced when oxygen was added to the anaerobic mixture of luminol and ferricya- nide, might indicate that the initial oxidation precedes the reaction with molecular oxygen (1953). The kinetic data also indicate that the reaction is first order with respect to both ferricyanide and oxy- gen concentration. Energy Considerations The energy of the light emitted by the DPD's during chemilumines- cence is another source of information about the mechanism. Luminol and all the DPD's emit radiation consisting of broad bands with the maxima of the intensities at about 4000 A (Bremer, 1953). This cor- responds to about 60 kcal per einstein. The radiation emitted includes some shorter wavelengths that correspond to 75 or more kcal per einstein. A valid mechanism for the reaction must provide a way for getting this sort of energy into the excited state of a molecule and certainly the reaction must evolve energy of at least this magnitude. However, inasmuch as visible hght can be produced from hydrazine and ferricyanide, it is probable that there will be sufficient energy available from the oxidation to form nitrogen of the DPD's to produce the observed light. 92 KINETICS OF CHEMILUMINESCENCE Structural Requirements All the DPD's that have been prepared produce light but many similar compounds do not. This information is useful in showing what structural features are necessary for the production of light. Drew and co-workers have made an extensive investigation of this phase of the problem of chemiluminescence. They have found that compounds such as 0 0 €s c^ -NCH3 and C' ^NCH3 .NCH3 O O do not chemiluminesce (Drew et al, 1937). It should be noted that these compounds cannot form the dienol form of the molecule. The ability to form this molecule is presumably tied in with the ability to produce chemiluminescence. Molecules such as OCHs X \/\c/ N I I N and OCH3 I N C r OH OCH3 have also been found not to produce Hght (Drew and Garwood, 1937). The substitution of the methyl group for the hydrogen in the enol forms can be acting to inhibit ionization or hydrolysis. Substitut- ing methyl for one of the ionizable hydrogens inhibits the chemilumi- nescence. This implies that both hydrogens are involved in steps leading to chemiluminescence. Substituents in the benzene ring have a marked effect on the intensity of the chemiluminescence ( Drew and Pearman, 1937; Zellner and Dougherty, 1937). The effect of these substituents on the intensity of the chemiluminescence could almost be predicted on the basis of p. C. WILHELMSEN, R. LUMRY AND H. EYRING 93 their effect on fluorescence yields. Groups like the amino group which greatly increase the intensity of the chemiluminescence generally in- crease the fluorescence yields of substituted aromatic compounds as can be seen in Table 11. This taken with the investigations of Zellner and Dougherty (1937) seems to indicate that the influence of the substituents on the aromatic nucleus is on 4> rather than on the reac- tion itself. TABLE II Fluorescence, Chemilu- Chemilu- Substituent onDPD Yield," % (in acid) minescent" Intensity minescent Fluor. Yields 5— OH 5— Br 6— NH2 5— NH2 5 NH-NH-SOsNa 2 2 2 10 0.02 20 1 4 100 0.2 10 .5 2 10 10 0.14 6 20 5 NH-COCHs 7 1 5— NH-CHs 10 60 5 NHCONHCOCH3 0.1 2 " These values are from Spruit (1950). Hydrazine is weakly luminous when oxidized by ferricyanide. How- ever, the mechanism seems to be different from that involved in the usual chemiluminescence of the DPD's in that oxygen apparently has no effect on the quantity of light emitted. The relatively high effi- ciency of the Hght produced by the DPD's appears to require the cyclic ring structure with the two hydrogens available to form the dienol form. End Products and Intermediates The identification of the intermediates and end products of a reac- tion is of great value in elucidating the mechanism. The efforts to isolate these from the chemiluminescent reaction of the DPD's have not been particularly successful. The conditions required for the chemiluminescence favor the subsequent reaction of the end product and the various substituted phthalic acids that have been isolated, eg 3-aminophthalic acid from luminol, might represent ultimate reaction products rather than the direct product of the chemilumines- 94 KINETICS OF CHEMILUMINESCENCE cent reaction. The compound 2,2'-dicarboxylbenzil has been obtained by Zellner and Dougherty (1937) and confirmed by Drew and Gar- wood (1938) from the oxidation of unsubstituted DPD. The fact that this compound can be further oxidized to phthahc acid without the production of Hght makes it probable that it is more closely related to the actual product of the light producing reaction than the phthalic acid derivatives. The alternative is that it is produced by a side reac- tion in which case there would be no relationship between it and the product of the chemiluminescence. The preparation of possible inter- mediates has been reported (Kautsky and Kaiser, 1950; Drew and Garwood, 1938). The compound prepared by Kautsky and Kaiser has not been proven to be related to the chemiluminescence of the DPD's. Drew and Garwood (1938) prepared a compound by dissolving lu- minol in strong alkali and adding a large quantity of hydrogen peroxide which they assumed to be an intermediate in the chemilu- minescent reaction. The subsequent adding of alcohol and cooling caused the precipitation of a crystalline compound. The analysis of this compound showed it to be the mono sodium salt. Drew supposed the compound was the hydrated form of NH2 o- I o 0 OH NH NH and did find that one molecule of water could be removed by drying in a vacuum over phosphoric oxide. The subsequent analysis was in substantial agreement with the proposed compound. However, it was found that some hydrogen peroxide was lost by heating and essentially all of it was lost when the compound was heated to 120° C. This seems to indicate that the compound isolated was one where hydrogen peroxide had replaced one molecule of water of crystallization. Drew has reported that the compound dissociated in water and that essen- p. C. WILHELMSEN, R. LUMRY AND H. EYRING 95 tially all the hydrogen peroxide could be titrated with potassium iodide and dilute sulfuric acid with ammonium molybdate as a catalyst. This would also be characteristic of the compound with hydrogen peroxide replacing water of crystallization. The compound containing the hydrogen peroxide in this form would have the same percentage composition as the one isolated. The evidence therefore seems to indicate that the compound isolated by Drew is not an intermediate in the reaction but a coordination complex with hydrogen peroxide. Further information is needed concerning intermediates and end products before the mechanism can be fully established. However, a knowledge of the general behavior of molecules makes possible a rather plausible supposition as to the mechanism for this reaction. Proposed Mechanism The DPD's probably exist in solution as a mixture of the diketo, mono keto, and dienol forms: 0 ^\ NH OH I OH .NH II O .NH X / N I N O c OH However, in basic solution, the enol forms probably ionize. This would serve to shift the equilibrium in favor of the enol forms. Stross and Branch (1939) have reported a first ionization constant of 10 ~'^ for luminol. They have also reported that there was no second ioniza- tion constant greater than lO"^^. It seems reasonable to suppose that in basic solutions such as are required for chemiluminescence, the major portion of the luminol is in the form of the singly ionized enolic form. In this connection it might be noted that the mono sodium salt is the one normally isolated from sodium hydroxide solutions. As has been explained earlier, there is evidence that indicates that preHminary oxidation takes place before the reaction with mo- lecular oxygen. Apparently this oxidation can be carried out by any 96 KINETICS OF CHEMILUMINESCENCE number of oxidants. It might well just involve the loss of the charge on the ionized portion of the molecule 0- 1 0- 1 1 ox. y\/^\N 1 1 OH 1 OH Oxygen then adds to form some such compound as: o- 0 I 6 ^N ,^N OH This could yield: 0- 1 I o o ^N .N N2 + OH 0- 1 c 0 C— OH 11 o The final molecule A* can be reduced to a phthalic acid derivative. It can also engage in other reactions, some of which might give rise to the 2,2'-dicarboxybenzil that was isolated by Zellner and Dougherty (1937). The excited molecule A* would emit light and undergo the reactions described in equations 3, 4, and 5. The proposed mechanism is speculative and is meant only to serve as a basis for further studies. An incidental result of the present study is the demonstration that p. C. WILHELMSEN, R. LUMRY AND H. EYRING 97 the chemiluminescent reaction of luminol and molecular oxygen can be employed as a means of measuring oxygen concentration. The intensity of the light emitted is a function of the amount of oxygen dissolved. The amount of oxygen dissolved can be related to the partial pressure of oxygen in the gas being analyzed. This procedure can be used to measure continuously the concentration of oxygen in flowing systems over reasonable periods of time as the luminol is only slowly con- sumed. The reaction of luminol with oxygen and an additional oxidant such as ferricyanide should be adaptable to detecting extremely small quantities of oxygen. References Albrecht, H. O. 1928. Ueber die Chemiluminescenz des Aminophthal saure hydrazide. Z. physik. Chem., 136, 321-30. Bernanose, A. 1951. La chimiluminescence des hydrazides. Mecanisme du phenomene. Bull soc. chini. France, 1951, 329-33. Bernanose, A., T. Bremer, and P. GoldHnger. 1947. Le mechanisme de la chemiluminescence en solution. Bull. soc. chim. Beiges, 56, 296-301. Bremer, T. 1953. Le mechanisme de la chemiluminescence en solution. Bull soc. chim. Beiges, 62, 569-610. Chance, B. 1940. The accelerated flow method for rapid reachons. /. Franklin Inst., 229, 455-76, 613-40, 737-66. Drew, H. D. K., and R. F. Garwood. 1937. Chemiluminescent organic compounds. /. Chem. Soc, 1937, 1841-46. Drew, H. D. K., and R. F. Garwood. 1938. The isolation of peroxide deriva- tives of phthalazine 1,4 diones. /. Chem. Soc, 1938, 791-93. Drew, H. D. K., H. H. Hatt, and F. A. Hobart. 1937. Chemiluminescent organic compounds. /. Chem. Soc, 1937, 33-37. Drew, H. D. K., and F. H. Pearman. 1937. Chemiluminescent organic com- pounds. /. Chem. Soc, 1937, 586-92. Flosdorf, E. W., L. A. Chambers, and W. M. Malisoff. 1936. Sonic activation of the chemiluminescence of luminol. }. Am. Chem. Soc, 58, 1069-76. Harris, L., and A. S. Parker. 1935. Chemiluminescence of 3-amino phthal- hyd'razide. /. Am. Chem. Soc, 57, 1939-42. Harvey, E. N. 1929. Luminescence during electrolysis. /. Phijs. Chem., 33, 1456-59. Harvey, E. N., and T. F. Morrison. 1923. The minimium concentration of oxygen for luminescence by luminous bacteria. /. Gen. Physiol., 6, 13-19. Kautsky, Hans, and K. H. Kaiser. 1950. Analysis of the luminescence of chemiluminescence of solutions. Z. Naturforsch., 5b, 353-61. Prudhomme, R. B. 1949. Luminescence produced in liquids by ultrasonics. /. chim. phys., 46, 318-22. 98 KINETICS OF CHEMILUMINESCENCE Spruit- van der Burg, A. 1950. Emission spectra of some chemiluminescent reactions. Rec. frav. chim., 69, 1536-44. Stross, F. H., and G. E. K. Branch. 1939. The chemiluminescence of 3-aminophthalhydrazide. /. Org. Chem., 3, 385-404. Sveshnikov, B. J. 1938. Acta PJnjsicochim. U. R. S. S., 8, 441. Zellner, C. N., and G. Dougherty. 1937. The chemiluminescence of phthalhydrazide derivatives. /. Am. Chem. Soc, 59, 2580-83, Spectroscopic Investigations of Luminescent Systems C. J. p. Spruit and A. Spruit-van der Burg Landbouwhogeschool, Wageningen, Holland The present discussion stems largely from the work carried out by the Biophysical Research Group at Utrecht during the period 1935- 1947. At the organization of this group, the subject of biolumi- nescence was chosen by the directors, Prof. Dr. A. J. Kluyver and the late Prof. Dr. L. S. Omstein, as a promising field for the application of physical methods to biology. During the preceding period, impor- tant conclusions had been drawn from the study of the spectroscopic properties of atoms and simple molecules. It was therefore considered advisable to make use of the experience which had been obtained in the Physical Laboratory at Utrecht with methods of spectral energy measurements, for the study of the emission spectra of bioluminescent phenomena. It was hoped that a detailed knowledge of the spectral energy distribution of bioluminescent emission might lead to conclusions as to the architecture of the molecules involved. It is true that during the following years, the increasing insight into the physics of hght absorp- tion and emission by more compUcated molecules has considerably tempered this initial optimism. The theoretical difficulties encountered during the attempt to for- mulate the connection between chemical structure and spectroscopic properties of organic molecules have been overcome only in some relatively simple cases even now. To these theoretical difficulties should be added the fact that in the case of organic molecules in solution, the ultraviolet and visible absorption spectra are of a diffuse 99 100 SPECTROSCOPIC INVESTIGATIONS nature, consisting of broad bands in which vibrational and rotational fine structure is almost always lacking. Moreover, in the case of emission spectra such as fluorescence spectra, the observed emission band corresponds to only one electronic transition in the molecule so that such spectra are even less detailed than the absorption spectra. This does not take away the fact that in the study of bioluminescence, the emission spectra are the only data obtainable that are immediately and uniquely connected with the bioluminescent compounds. In this connection, an accurate knowledge of spectra should be highly desirable. Li 0) ::=;: Abs riuor A /(i\ n\ 'Xrt \ Fluor. 'i Abs. Distemce Fig. 1. a and b. Energ>' states for a two-atom molecule with transitions in ab- sorption and fluorescence, c. The occurrence of mirror symmetry between absorption and fluorescence spectrum. It needs no emphasis that bioluminescence is a form of chemilu- minescence. The known chemiluminescent systems offer a means of studying the relation between the chemiluminescence emission spec- trum and other spectral properties of the compounds involved. During the chemiluminescent reaction, molecules are transferred to an excited state, and this is followed by the emission of light. Therefore, the process of light emission during a chemiluminescent reaction is re- lated to the light emission during fluorescence. It is therefore desirable to say a few words at this point about the nature of fluorescence and its relation to light absorption in organic molecules. We will consider the case of a two-atom molecule. In Fig. la, two electronic energy states of this molecule are given, together with some vibrational levels. The arrows indicate a transition from the ground level to a higher C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 101 energy level by absorption, and from the higher to the ground level by the emission of light. The same is represented in another way in Fig. lb. As long as a number of conditions are fulfilled as to transitional probabilities and spacing of vibrational levels in the two energy states, it may be predicted that there is a mirror symmetry between the emission and the fluorescence spectrum, if plotted versus frequency, as shown in Fig. Ic. The reemission of light by the excited molecule is not the only way v'^ 1 A\ / \ /, ^ \ > / \ 1/ \ \ / / \ .\ r \ / 1 \ \ / / \ \ J / / \ \ / / / \ \, / k 1 \ \ / / 1 ] \ \ \. /a / i \ ' / ji \ /f 1 \ V 1 A \ \ 16 18 20 22 24 25 x 10^ cm"' Fig. 2. Mirror symmetry as observed in practice. Fluorescence ( F ) and absorp- tion (A) spectra of riboflavin plotted against frequency. Dash line: mirror image of fluorescence spectrum. by which this energy can be dissipated, and, as a rule, the molecule returns to the ground level by way of other transitions not involving the emission of a quantum. In this case the energy is lost in the form of heat. Only if special conditions are fulfilled are such radiationless transitions of sufficiently low probability that fluorescence can occur with a measurable quantum yield, and only in relatively rare cases does this yield approach unity. It is an empirical fact that only for the first excited level are the probabilities of these radiationless transi- tions to the ground level sufficiently low to allow for the occurrence of fluorescence. For this reason, only those emission bands are en- 102 SPECTROSCOPIC INVESTIGATIONS countered that correspond to the long-wavelength absorption band in the absorption spectrum. In Figs. 2 and 3, examples are given of this mirror symmetry between absorption and fluorescence for some com- pounds of interest in connection with the following discussion. For a discussion of fluorescence of organic compounds the reader is referred to Forster (1951). Returning to the chemiluminescence spectra, the question arises as to the mechanism of excitation. We may visualize two possibilities: =EES::3 330 370 410 450 490 570 rryj Fig. 3. Mirror symmetry in the fluorescence and absorption spectra of methyl- acridone. during the reaction, intermediates or end products are directly formed in an excited state, or they may obtain their energy by transfer from other reactants containing sufficient energy. It is not easy to distinguish experimentally between these possibilities. Kautsky ( 1943 ) is believed to have shown that the emission spectra of various chemiluminescent reactions were very similar to the fluores- cence spectra of the parent compounds. This has led many investi- gators to the assumption that during the chemiluminescent reaction, the original molecules are regenerated, but in an electronically excited C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 3 103 94° —T "-••. '*•. '**••.. *... ■ / 51», _ ^-> '\ ■ 1 1 1 ^^ s 1 1 , /io" ^\ . 1 1 1 / / / X ^ ■••. / / Vs. ^ o / / ^v \ / / / \ < 1 1 1 1 S. 1 \ 1 ! 1 / rt 1 1 1 400 440 480 520 560 600 m^ Fig. 6. A comparison of the low-temperature chemiluminescence spectrum of dimetliylbisacridinium nitrate ( ) with the fluorescence spectrum of this compound ( ). seem to be sufficient to explain the broadening of the high-tempera- ture chemiluminescence spectrum. The nature of the second emitting molecule is not known. It cer- tainly is not dimetliylbisacridinium nitrate, as its spectrum extends much more to the violet than does the fluorescence spectrum of that compound, Fig. 6. x\t this point it appears advisable to point out some sources of con- fusion to those examining emission spectra of colored solutions. As a rule, the emission spectra, such as fluorescence and chemiluminescence 106 SPECTROSCOPIC INVESTIGATIONS spectra, overlap more or less with the absorption spectra. As it is advantageous to have as high a light intensity as possible for the study of the emission spectra, it is tempting to use highly concentrated solutions. This, however, has the result that in the case of fluores- cence practically the whole short-wavelength part of the emission spectrum may be lost by self-absorption. In chemiluminescence, this overlap may be even greater. Moreover, if the fluorescence yield of the solution is not very low, the absorbed radiation is reemitted for an important part as fluorescence. This means that in the examination of fluorescence spectra the short-wavelength part of the emission spec- trum will be suppressed, and in the examination of chemiluminescence spectra one may arrive at the erroneous conclusion that the emission spectrum is identical with the fluorescence spectrum of the starting material. In this way, Kautsky (1943), who first studied the chemi- luminescence spectrum of dimethylbisacridinium nitrate, may have arrived at his conclusion that at low temperatures the emission spec- trum is identical with the fluorescence spectrum. This conclusion certainly is erroneous as is demonstrated by Fig. 6. In the experiments of van der Burg, special attention has been paid to this source of error, and it can be stated that the fluorescence of dimethylbisacridin- ium nitrate certainly contributes less than 3% to the spectrum of Fig. 6. To avoid errors as those discussed above, it is therefore necessary to use relatively dilute solutions and to apply a correction to the emission spectrum for the absorption by the solution. In the case of the chemiluminescence of the derivatives of 2,3-dihy- drophthalazinedione ( DPD ) O NH .NH 6 the results were less clear-cut. We will discuss the most illustrative examples, namely the spectra of 5-amino-DPD and 5-methylamino- DPD, Fig. 7. It was found as a general rule that, as far as the wave- lengths of maximum emission are concerned, the chemiluminescence C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 107 spectra of these compounds are similar to the fluorescence spectra in acid medium, but different from the fluorescence spectra in alkaHne media, in which the luminescent reaction is carried out. It may be added that for practically all compounds investigated, the fluorescence 300 400 500 600 nyj 300 400 500 600 mu Fig. 7. The absorption, fluorescence and chemiluminescence spectra of two deriva- tives of dihydrophthalazinedione. Absorption and fluorescence spectra of acid solutions. Absorption and fluorescence spectra of solutions in 0.01 N KOH Chemiluminescence spectra. 108 SPECTROSCOPIC INVESTIGATIONS yield under the conditions of the reaction is practically nil. Notwith- standing the similarity, the chemiluminescence emission spectra are generally broadened with respect to the fluorescence spectra, just as in the case of the high-temperature spectrum of dimethylbisacridinium nitrate. If this broadening of chemiluminescence spectra relative to fluorescence spectra proves to be a general feature, it may be of con- siderable theoretical interest. Our only conclusion from the results discussed above is that the chemical structure of the emitting molecule is similar to that of the acid, diketo form of the derivatives of DPD. As the absorption spectra of these compounds are not very characteristic, a further identification is impossible, and in particular it is not possible to decide whether or not the heterocyclic ring is still intact in the emitting molecule. Its structure may be represented therefore by a general formula of the type O /C— R C— R' O The interpretation of chemiluminescence spectra touches upon the question of the reaction mechanisms and the occurrence of interme- diate products. Attractive as this aspect may be, it is surprising that so little is known about it with certainty. We will now turn to the examination of the emission spectra of luminous organisms. From the foregoing we may expect them to be related to fluorescence spectra of reaction products of the molecules entering into the luminescent reaction. Hence they should show vibrational structure only in so far as the fluorescence spectra and in particular also the longest-wavelength bands in the absorption spectra of the compounds involved show such vibrational structure. The occurrence of secondary maxima, such as those found by Eymers and van Schouwenburg (1937), in emission spectra of various types should be of special interest. We regret to say that more careful measurements have failed to confirm the claims of these authors. C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 109 In making spectral energy measurements by the photographic method, one should always take averages of a considerable number of separate exposures, as the plates are never quite homogeneous. Other points to be observed in the measurement of emission spectra consisting of diffuse bands are absorption and scattering. In this case it is not possible to apply a correction for these effects to the observed emission spectrum, and the only solution is therefore to make use of such dilute suspensions that they may be neglected. The absorption M ^—Ph phosphoreum — — Ph. splendidum Ph fischeri A. mellea iOO 420 ««0 ~i6d i80 500 520 5iO 560 580 600 ' 620 6«0 nyi Fig. 8. Bioluminescence emission spectra of three species of luminous bacteria and one fungus. (From Bioch. et Biopliys. Acta 5 (1950), by pemiission of Elsevier Publishing Company. ) of the emitted light by the emitting cell itself is unknown, however, and the spectra may be still in error by this amount. In the emission of luminous bacteria, a calculation shows that in all probability this absorption by only one cell is negligible. In luminous fungi, however, one is compelled to make use of mycelial mats, consisting of several layers of hyphae. Here absorption may not be without importance. The spectra obtained by van der Burg (1950c) for various organ- isms are reproduced in Figs. 8 and 9. What conclusions can be drawn from these spectra? In the first place, it is likely that the three fungi have identical emission spectra and hence identical bioluminescent 110 SPECTROSCOPIC INVESTIGATIONS systems. These appear to differ from those of the three bacteria. As regards the latter, Ph. phosphoreu7n certainly deviates much from the others. Although Ph. splendidum and Ph. fischeri have different spec- tra, it is perhaps not advisable to take their difference for granted as we have to account for the absorption of the light by the bacteria themselves. Although this is probably negligible, Ph. fischeri has a distinctly yellow color and in this case the absorption may well be of some importance. This should affect the short-wavelength part of the spectrum in particular. ">, Mycena polygramma // \S -ump nana riavia« \\ i !\ "■■■^ ^^^^ \ J / \-s> ^ / 580 460 500 540 Fig. 9. Emission spectra of three luminescent fungi. 620 m^ Occasionally spectra of bioluminescent processes have been pub- lished. It is difficult to compare them with those just discussed, as it is uncertain and even unlikely that the precautions we have mentioned have been observed during their measurement. For example, two spectra of the emission of Cypridina have come to our knowledge, one by Coblentz (1926), the other by Eymers and van Schouwenburg ( 1937 ) , showing maximum emission at about 480 and 464 nxfi respec- tively. Both spectra have been obtained by moistening Cypridina powder before the slit of the spectrograph. This powder is a strongly C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 111 colored substance with considerable absorption in the blue and violet regions of the spectrum. From visual observation we feel reasonably sure that the emission maximum of Cypridina is at shorter wavelengths than that of Ph. phosphoreum (472m/A). Three spectra of the emission of Photinus pyralis are known to us. They do not agree among each other save that the maxima of emission are rather close together (563, 565, and 568 m/i ) . As this is far in the green, where there is not much self-absorption, this maximum should be quite reliable. At any rate we may conclude that the emission spectra do prove that the emitting molecules in these various organisms are definitely different. As all these spectra are related to one band in the absorption spectra of the compounds involved, it will be clear that little infor- mation can be drawn from them without the assistance of further independent observations. It has been suggested on several occasions that bioluminescence is connected in some way or other with the presence in the cell of de- rivatives of riboflavin. It is therefore advisable to point out that none of the emission spectra obtained so far show much resemblance with the fluorescence of riboflavin (Fig. 2), the maximum of which is sit- uated at 534 m/x. If the luminescent system occurs in the cell in the form of an en- zyme with a specific substrate, the concentration of this compound in all probability will be so low that direct observation by absorption spectroscopy will be impossible. Accidentally, during a study of car- bon monoxide inhibition of luminescence, van Schouwenburg and van der Burg came across an indirect method for the measurement of the absorption spectrum of the luminescent system in Ph. phosphoreum. The idea has been the basis of an investigation by van der Kerk and van der Burg. Whereas irradiation with red light has no effect upon luminous bacteria, irradiation with comparable intensities of blue light tends to dim the light emission of the bacteria. A quantitative study of this effect, which is fully described in the thesis of van der Kerk (1942) (Kluyver et al, 1942) led to the establishment of an action spectrum for the quenching of bacterial luminescence which should be proportional to the absorption spectrum of the system which emits the light. Later observations ( Spruit, 1946, 1949a ) have necessi- tated a correction to be applied to this spectrum at wavelengths below 112 SPECTROSCOPIC INVESTIGATIONS 300 m/x. The corrected spectrum is given in Fig. 10. Unfortunately, this spectrum is not very detailed, and it is not immediately obvious to what class of compounds it has to be attributed. The only safe conclusion to be drawn from it is that it does not belong to any of the well-known, generally occurring cellular components such as caro- 240 \ \ / / \ 1 \ \ - \ \ ••h / \ \ V \ / \ \ \ \J \ \ \ \ \ \ \ 1 1 i \- -18 -19 •20 320 400 mL Fig. 10. The photochemical inactivation spectrum of P/i. phosphoreum ( ) and the absorption spectrmn of l,4-dihydroxynaphthyl-2-hydroxymethyl- ketone ( ). tenoids, flavine enzymes, hematin compounds, or to systems contain- ing any of the colored vitamins. This supports van der Kerk's con- clusion that it really represents the spectrum of a system, closely connected to the bioluminescent reaction and not merely the spectrum of a sensitizer. Another important piece of information lies in the recovery of lumi- C. J. p. SPRUIT A\D A. SPRUIT-VAX DER BURG 113 nescence after the end of the irradiation, which is a first order reac- tion. This recovery of luminescence also exists in those bacteria which have lost their reproductive capacity by exposure to wavelengths shorter than 300 mfx.. Ultimately, such "killed" bacteria emit light many times more intense than a nonirradiated part of the same culture. This observation is in perfect harmony with the old views of van Schouwenburg that bacterial luciferin has a dual function and that part of it is involved in a reaction other than the luminescent one. Obviously in such damaged bacteria more luciferin is accessible to the luminescent reaction after recovery is complete than in normal , bacteria. This may well be due to the fact that that part of the respira- tory system responsible for the reversible dehydrogenation of luciferin (see van Schouwenburg, 1938) is inactivated by irradiation with light of short wavelengths. Although definite proof is lacking, these facts support the conclusion that van der Kerk's inactivation spectrum in- deed is the absorption spectrum of the bacterial luciferin or its imme- diate precursors. The same authors also made a few observations on the inactivation spectrum of Ph. splendidum and Ph. fischeri, though only in the visi- ble region. It is highly regrettable that they have not found it pos- sible to include in their study the ultraviolet parts of these spectra, but from what has been published it is clear that, whereas the inactiva- tion spectra of Ph. splendidum and Ph. fischeri cannot be distinguished on the basis of the material available, they are certainly not identical with the inactivation spectrum of Pli. phosphor eum. The situation is therefore somewhat similar to that of the emission specti"a. This ob- servation raises the question of mirror symmetry'. As was explained earlier, chemiluminescence emission spectra, being related to fluores- cence spectra, should show qualitative mirror symmetry with the long-wavelength bands in the absorption spectra of the compounds in question. If we observe a certain amount of symmetry between the emission and the inactivation spectra of a number of organisms, we may therefore take this as a fair indication that both types of spectra belong to the same, or at least to pairs of closely related compounds. Now, this is what is really observed. Fig. 11. The form of the inactiva- tion spectra really is rather different, but the important point is the small overlap for all pairs of spectra. This comparison therefore is a 114 SPECTROSCOPIC INVESTIGATIONS material support for the attempt to attribute both the emission and inactivation spectrum to the same system. Although the information that can be obtained from the detailed form of the photochemical inactivation spectrum of Ph. phosphoreum is not great, it has more weight if the additional information we have about the chemical nature of bacterial luciferin is taken into account. As van Schouwenburg has amply discussed (1938), bacterial luciferin y'"4 / / /\ /r V inflr»ii'«'inn - / \; \\ / f I 1 / / \ \ / 1 \ / / , \ \ / t \ ■7 ''■■ M / 1 \ .7 1 •• \\ / f \ . / 1 / \, // \J /: s. ■ / y /.- S^ / / ^ / \ / /; N V / / \/ / / \ / / 1 / .■ \ .. / 1 \A // \ ' 1 1 1 — Ph. phosphoreum — Ph. splendidum ..'/ / 1 \- \ I.- / / V ^ ( -Ph. fischeri ^ / / \ ^^ y / \ ^* / \ / V >» I / \ "* 15 17 19 21 23 25 27 xlO'cm- FiG. 11. Qualitative mirror symmetry between the emission and photochemical inactivation spectra of three luminous bacteria. may be reversibly oxidized and reduced in the bacterium, just as was demonstrated previously by Harvey for Cypridina luciferin. At the time the inactivation spectrum was measured, we still believed it a reasonable assumption that the various luminescent systems found in nature were at least related. It is probably for this reason that we have thought it permissible to take into account the results of ob- servations with other organisms in making speculations about the chemical nature of the compound whose inactivation spectrum had been measured. One of the very few bits of American literature which came to us during World War II was a publication by Chakravorty and Ballen- C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 115 tine (1941), which contained results of the resynthesis of luciferin from its irreversible luminescent oxidation product. We were not in a position to check the results reported and we may have attached too much value to them. Taking all information together, van der Kerk suggested that the compound whose absorption spectrum is the inactivation spectrum had as part of its structure: COCH2OH From available data on the absorption spectra of naphthoquinones (Macbeth et al, 1935) the assignment of a naphthoquinone structure to the compound appeared justifiable. Although the situation at the moment no longer makes it especially attractive to look for a relationship between the various forms of bioluminescence, at that time it appeared promising to test this hypothesis further by a study of the absorption spectra of naphtho- quinones, and in this connection an attempt was also made to prepare the compound mentioned by van der Kerk. Accepting his idea as a working hypothesis, it appeared necessary to establish with more cer- tainty the state of reduction of the hypothetical compound. Accord- ing to van der Kerk, this should be the quinone form. This conclusion he based mainly on the observation that several quinones (e.g., 2-methylnaphthoquinone ) are known to be photolabile. A closer anal- ysis of the data, especially van der Kerk's experiments on the photo- inactivation under low oxygen pressure, have convinced us that it is much more likely that the reduced form of luciferin is the photolabile component ( Spruit, 1949a ) . This is in harmony with the observation that l,3,4-trihydroxynaphthyl-2-methylketone, a derivative of naphtho- hydroquinone, is also photosensitive in the presence of low pressures of oxygen (Spruit, 1947). It was therefore quite an exciting observa- tion that the absorption spectrum of l,4-dihydroxynaphthyl-2-hydroxy- nV ^^\Cai >»* 'rS^'t, I 1 M 116 SPECTROSCOPIC INVESTIGATIONS methylketone, the hydroquinone of van der Kerk's hypothetical com- pound OH -COCH2OH OH is very similar to the corrected inactivation spectrum, Fig. 10. These observations were made in 1947. We immediately tested the compound with luminous bacteria and with Cijpridina luciferase (Spruit, 1949a), but without success. In view of the number of hy- potheses upon which our attempt was based, this was not very sur- prising, and as there is no immediate way of further testing the idea, this is where our story could come to an end. However, during the years that followed, new contributions were made by other workers to the subject of bioluminescence, and apart from van der Kerk's inactivation spectrum we have now two more spectra of compounds involved in bioluminescent reactions. One is the absorption spectrum of purified Cijpridhui luciferin, measured by Chase and Brigham (1951); the other is that of the stimulating factor in the luminescence of the firefly Photimis pyralis, measured by Strehler and McElroy ( 1949 ) . A comparison of these spectra serves to extinguish our last hope that the various forms of bioluminescence might be related. It is certain that these three spectra belong to widely different classes of compounds. We also have to conclude that most of the arguments in favor of van der Kerk's proposal are of doubtful value, and the nature of the compound involved in the photochemical inactivation is therefore still unknown. Notwithstanding these criticisms, the remarkable similarity between the inactivation spectrum of the bacteria and the absorption spectra of certain derivatives of naphthohydroquinone is striking. It is hard to believe that this similarity is completely meaningless. We must recall here also the magnificent work by van Schouwenburg (1938), which has given a firm basis to the theory that in the bacterium Ph. phosphoreum the luciferin is reversibly oxidizable and reducible, a C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 117 conclusion that has received considerable support from the work of Spruit and Schuiling ( 1945 ) . So we know at least one chemical prop- erty of this compound which is in agreement with the chemical struc- ture of the naphthoquinones. It may be argued that although certain derivatives of naphtho- hydroquinone possess absorption spectra quite similar to van der Kerk's inactivation spectrum, it might well be that other unrelated classes of compounds exist with similar spectral properties. During later years we have had opportunity to study the ultraviolet absorption spectra of various compounds in more detail, and the results of this work make it appear justifiable to reconsider the matter of an identification of the bacterial luciferin. We should like to present some of the relevant material here. How typically characteristic is the absorption spectrum mentioned of this particular class of compounds? To test this, one can investigate the influence upon the absorption spectrum of alterations in the sub- stituents attached to the naphthalene nucleus. The second method is to change the character of the aromatic nucleus itself. If we take naphthyl-2-methylketones as a starting point, it is clear from theoreti- cal considerations that substitution in the methyl group should not alter the absorption spectrum profoundly. This is confirmed by the experiments (Spruit, 1949c). On the other hand, keeping in mind that the long-wavelength absorption band in the inactivation spec- trum is somewhat more to the red than the corresponding band in the naphthohydroquinone compound, we are interested in the effects of other substituents in the nucleus. Figure 12 gives examples of what happens if one shifts around with the OH groups. The result demon- strates that the exact position of the long-wavelength band should not pose much of a problem. Further examples can be found elsewhere (Spruit, 1949b). Quite another problem is posed by the nature of the aromatic nu- cleus. As far as the quinones are concerned, the absorption spectra of anthraquinones are very similar to or almost identical with those of the analogous naphthoquinones. The quinone nucleus forms an effec- tive barrier between the conjugated systems on both sides of it. This observation is amply confinned by the spectra of a number of other tricyclic paraquinones containing heterocyclic ring systems (Spruit, 118 SPECTROSCOPIC INVESTIGATIONS unpublished). On the other hand, the tricycHc hydroquinones show a type of absorption spectrum, quite different from those of the naphthohydroquinones. A / 1 '\ ^ o \ \\ 11 1 \ \ \ \ r ' \ \ 1 1 240 320 400 tny f1 A \ 1 \ — \ 1 / 1 *C~^> \ 1 1 / 1 1 \ \ ^^/ i 1 •■•. \ ■ >•— • Vj / \ i \ \ W 1 ■: 1 1 1 \ 240 320 400 4 80 mu Fig. 12. The effect of substituents upon the absorption spectra of naphthyl-2- methylketones. Left: l-hydroxynaphthyl-2-methylketone, 1,5- dihydroxynaphthyl-2-methyIketone. Right: l,4-dihydroxynaphthyl-2- methylketone, - - - - l,5,8-trihydroxynaphthyl-2-methylketone, l,3,4-trihydroxynaphthyl-2-methylketone. In how far are the absorption spectra determined by the properties of the aromatic ring system? To answer this question one should like to have a means of continuously altering these properties. Obviously this is impossible, and the next best one can do is to compare a series of analogous quinones containing different heterocycHc ring systems, instead of the naphthalene nucleus. If the hetero atoms show different electron affinities, we have more or less what we want. C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 119 Accordingly we have examined paraquinones derived from benzo- furan, thianaphthene, selenanaphthene, and isoindazole. In the first place it was found that the properties of the pyrazole 9 ■p, A 9 e 1 \ /^ ' /". <; \ ^ \ \ \J t \ 3 \ \ V \ \ 2 A ' / \ ■ \// A / i / 4 A H / ?^^ \ \\ / . \ ^ r \ \\ J^ — 3 V \ \ \ \ vy — / — / / / \ v> \ 5 V 230 310 390 m^ 230 310 390 470 550 ftyj Fig. 13. Alterations of the nucleus. Similarity between the absorption spectra of isoindazolequinones and of analogous naphthoquinones. Left: isoin- dazole-4,7-quinone, 1,4-naphthoquinone. Right: alkaline solu- tion of 2-hydroxynaphthoquinone, alkaline solution of 5-hydroxyisoin- dazolequinone. nucleus are very close to those of the benzene ring, as the absorption spectra of the isoindazolequinones O 120 SPECTROSCOPIC INVESTIGATIONS are very similar to those of the analogous naphthoquinones, Fig. 13. On the other hand, if we compare with 2-hydroxynaphthoquinone the series: O O /OH 1 /OH O s o 0 5-Hydroxybenzofuranquinone o 5-Hydroxythianaphthenequinone .OH Se 0 5-Hydroxyselenanaphthenequinone we have such a series of compounds with decreasing electron affinity of the hetero atom. This leads to interesting changes in the absorption spectra, Fig. 14. The bands marked with a, a', and a" appear to form a group related to each other and to the short-wavelength bands of the naphthoquinones. The same holds for the groups marked b and c respectively which compare with the 330-m/x and long-wavelength bands of the naphthoquinones. The less the electron affinity of the hetero atom, the more "aromatic" is the character of the nucleus and the more are the bands in the absorption spectra shifted toward longer wavelengths. This holds least for the long-wavelength bands but we will not go into the reason for this here. As is well known, furan has only very weak aromatic properties, and this agrees with the observation that the absorption spectrum of 5-hydroxybenzo- furanquinone is very similar to that of dihydroxybenzoquinones. So far we have not yet had the opportunity to investigate the absorption spectra of the corresponding hydroquinones in detail, owing to their unstability. There is no reason to expect a deviation from what has been discussed above, however. C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 121 This brief summary may suffice to show that it is certainly possible to prepare compounds with spectra duplicating those of the naph- thoquinone derivatives. It is at the same time clear that in order to obtain compounds with absorption bands more or less at the same il f Q i Y 1 \ 1 \ 1 /lb } / 1 {/ 1 \ 1 I ■r > if 1 i. % ' U ' \ y ■I- \ \ \ \ r^ V \ / ^ \ \\ c' \ / \ \ \ \ \ \ \ \ \ \ \ \ \ -V -t- \ \ 270 350 430 510 nyi Fig. 14. Alterations of the nucleus: — 5-hydroxybenzofuranquinone, 5-hydroxythianaphthenequinone, .... 3,6-dimethyl-5-hydroxyselenanaphthene- quinone. position as those in the inactivation spectrum, certain restrictions are imposed as far as the nature of the nucleus is concerned. Its properties should be close to those of naphthalene. The results of the investiga- tions outlined above are therefore certainly not in disagreement with the view that bacterial luciferin has a chemical structure which may be symbolized as follows: 122 SPECTROSCOPIC INVESTIGATIONS R4 R5 Ar. = ring system with aromatic properties. Ri . . . R5 may represent other groups, not contributing materially to the absorption spectrum, but required in connection with the other functions of the molecule. If we now attempt to summarize the conclusions that can be drawn from the spectroscopic investigations of bioluminescent phenomena, we are forced to admit that the results have not come up to the origi- nal expectations. It is not possible to draw definite conclusions as to the structure of the emitting molecules from the bioluminescent emis- sion spectra, and it is not likely that further theoretical developments will make such a conclusion possible. On the other hand, the diver- gence between the emission spectra of various organisms is a demon- stration that the emitting systems are considerably different. Only in the case of the three fungi examined, has the conclusion to identify the bioluminescent processes seemed justified. At the same time, the study of chemiluminescence spectra has led to the important observation that in this case, emission spectra are closely related to fluorescence spectra of molecules involved in the luminescent reactions. This makes it attractive to compare bioluminescence emission spectra with fluores- cence spectra of known biologically important compounds. This com- parison does not point to an identity of the emitting molecules with such generally occurring substances. This same conclusion follows from an examination of the three absorption spectra of bioluminescent compounds, obtained so far. Together, these facts seem to indicate that bioluminescence is the result of certain specific and, as yet, not fully known metabolic steps. Moreover, it is likely that different organ- isms have fundamentally different types of luminescent systems which may not be even superficially related. C. J. p. SPRUIT AND A. SPRUIT-VAN DER BURG 123 References Chakravorty, P. N., and R. Ballentine. 1941. On the luminescent oxidation of luciferin. /. Am. Chem. Soc, 63, 2030-31. Chase, A. M., and E. H. Brigham. 1951. The ultraviolet and visible absorp- tion spectra of Cypridina luciferin solutions. /. Biol. Chem., 190, 529-36. Coblentz, W. W., and C. W, Hughes. 1926. Spectral energy distribution of the light emitted by plants and animals. Sci. Tap. U. S. Bur. Standards, 21, 521-34. Duysens, L. N. M. 1952. Transfer of excitation energy in photosynthesis. Thesis. Utrecht. Eymers, J. G., and K. L. van Schouwenburg. 1937. On the luminescence y of bacteria. Enzymologia, 3, 235—41. ^^orster, Th. 1951. Fluoreszenz Organischer Verbindungen. Gottingen. Kautsky, H., and H. Kaiser. 1943. Naturwissenschaften, 31, 505. Kautsky, H., and O. Neitzke. 1925. Spektren emissionsfahiger Stoffe bei Erregung durch Licht und chemische Reaktionen. Z. Physik, 31, 60. Kluyver, A. J., G. J. M. van der Kerk, and A. van der Burg. 1942. The effect of radiation on light emission by luminous bacteria. Proc. Konink. Akad. Wetenschap. (Amsterdam), 45, 886-94, 962-67. Macbeth, A. K., J. R. Price, and F. L. Winzor. 1935. The colouring matters of Drosera whittakeri. Part I. /. Chem. Soc, 325-33. Spruit, C. J. P. 1946. Naphthochinonen en bioluminescentie. Thesis. Leiden. Spruit, C. J. P. 1947. Carbonyl-substituted naphthoquinones. Part I. Rec. trav. chim., 66, 655-72. Spruit, C. J. P. 1949a. The chemical nature of the luciferins. Enzymologia, 13, 191-200. Spruit, C. J. P. 1949b. Absorption spectra of quinones. Part I. Rec. trav. chim., 68, 309-24. Spruit, C. J. P. 1949c. Absorption spectra of quinones. Part II. Rec. trav. chim., 68, 325-35. Spruit, C. J. P., and A. L. Schuiling. 1945. The effect of naphthoquinones on light emission and respiration of Ph. phosphoreum. Rec. trav. chim., 64, 219-28. Strehler, B. L., and W. D. McElroy. 1949. Purification of firefly luciferin. /. Cellular Comp. Physiol, 34, 457-66. van der Burg, A. 1943. Spektrale onderzoekingen over chemo- en biolumi- nescentie. Thesis. Utrecht. van der Burg, A. 1950a. Emission spectra of some chemiluminescent reac- tions. Part I. Rec. trav. chim., 69, 1525-35. van der Burg, A. 1950b. Emission spectra of some chemiluminescent re- actions. Part II. Rec. trav. chim., 69, 1536-44. 124 SPECTROSCOPIC INVESTIGATIONS van der Burg, A. 1950c. Emission spectra of luminous bacteria. Biochim. et Biophys. Acta, 5, 175-78. van der Kerk. G. J. M. 1942. Onderzoekingen over de bioluminescentie der lichtbacterien. Thesis. Utrecht. van Schouwenburg, K. L. 1938. On respiration and light emission in lumi- nous bacteria. Thesis. Delft. Discussion Dr. Mason: One of the major purposes of Dr. Spruit's work has evi- dently been to characterize Cijpridina luciferin by means of system- atic study of the absorption spectra of certain chromophores. Since very small quantities of substance are required for the measurement of absorption spectra, the method has obviously a wide application to the determination of structure of natural products, and it is of interest to consider what its limitations may be. Aside from technical problems, such as the establishment of homo- geneity, the principal limitation lies in the presupposition that it is possible to assign a structure to a light-absorbing molecule with a high degree of confidence when only the absorption spectrum is avail- able. In so far as the ultraviolet and \ isible regions of the spectrum are concerned, very large classes of compounds tend to absorb light in certain well-defined regions such as 260 to 280 millimicrons. If the substance we wash to characterize displays a single absorption band in this region, the confidence with which we can assign a structure to it is very low indeed, even if the correspondence between the absorption spectra of the known and unknown substances is high. If the unknown substance displays two absorption maxima in the ultraviolet and visible region and the correspondence between the absorption spectra of the known and unknown substances is again relatively exact, the degree of confidence in the structural identifica- tion rises correspondingly. Extending this hue of thought, the larger the number of corresponding uniquenesses between the spectra of known and unknown substances, the higher the degree of confidence we may have in the identification. Characteristic absorption maxima for a single compound can be multiplied by observing spectral changes which can be induced by changes in environment or by frank reac- tion, that is, alterations in pH, in solvent type, or by autoxidation. There is one important reservation which should be made in regard to the amount of structural detail which is revealed by absorption spectra in the ultra\dolet and visible regions. In order to be spectro- scopically detectable, each atom or system of atoms in an absorbing molecule must have the capacit\' to affect, detectably and character- istically, the levels of excitation to which the molecular orbitals can be raised. Inasmuch as there are very large numbers of possible alter- 125 126 SPECTROSCOPIC INVESTIGATIONS ations in chromophoric molecules which will have virtually no detect- able effect upon the absorption spectrum of the substances in these regions, the degree of confidence in identification by means of corre- spondence between electronic spectra must be correspondingly lim- ited. This general line of argument may be readily extended to identifica- tions derived by correspondences between the spectra of substances in the infrared region. Recent Studies on the Chemistry of Cypridina Luciferin* Frederick I. Tsuji, Aurin M. Chase, and E. Newton Harvey Biological Laboratories, Princeton University, Princeton, New Jersey Historical Introduction The luminescent system in the marine ostracod crustacean Cypridina hilgendorfii is one of the simplest among luminous animals. Only lu- ciferin, luciferase, and dissolved oxygen appear to be necessary for light production in aqueous solution. In this respect the Cypridina system differs from that of the firefly, where luciferin, luciferase, adenosine triphosphate, magnesium ions and oxygen are all essential ( see McElroy and Coulombre, 1952, and Hastings, McElroy and Cou- lombre, 1953, for latest results). Structurally, the luminous organ of Cypridina, the submaxillary gland, is also simple, made up of elongated gland cells, whose contents are squeezed into the sea water through individual openings on the tip of the upper lip by contraction of muscle fibers. Observation of the living animal indicates that two varieties of granules pour out of the submaxillary gland, one large (diameter about 10 ix) and yellow, undoubtedly luciferin, and another small (diameter about 2 fi) and colorless, probably luciferase. Both types of granules dissolve on meet- ing the sea water, and at the same time the luminescence appears (see Harvey, 1952, for details of gland structure). * This research has been supported in part by a grant from the National Science Foundation, in part by a contract between the Office of Naval Re- search, Department of the Navy, and Princeton University, NR 165-167, and in part by funds of the Eugene Higgins Trust allocated to Princeton University. 127 128 CHEMISTRY OF CYPRIDINA LUCIFERIN Whole Cypridinae can be dried, and if the drying process is rapid and the material preserved with a drying agent, the abihty to lumi- nesce brightly on adding water will last indefinitely (at least 32 years). This dried material has been the starting point for most of the chemical studies on luciferin and luciferase. The early work of Harvey (1917-19) established that the Cypridina luciferin-lucif erase reaction was comparable to that of the mollusc, PJiolas dactylus, de- scribed by Dubois (1887), and that luciferin could be oxidized by oxidants or would oxidize spontaneously without light emission in absence of luciferase; that it was dialyzable and insoluble in ether and hydrocarbons, but soluble in lower alcohols and some other non- aqueous solvents. It was found that after oxidation, luciferin could be recovered, at least partially, by various procedures which add hydro- gen to a molecule. Consequently the oxidation of luciferin was com- pared to the oxidation of a leuco-dye, such as methylene white to the dye, methylene blue. It was also established (Harvey, 1925-26) that luciferin undergoes photochemical destruction without light emis- sion, an effect subsequently studied in detail by Chase and Giese (1940). In the 1920's, additional properties of the Cypridina luminescent reaction were studied by Harvey and by Kanda, but the solutions were impure and nothing would be gained by reviewing the tests applied, or the procedures used in fractionation. One difficulty in test- ing for chemiluminescent substances is the extreme sensitivity of the luminescence test as compared with ordinary chemical reactions for organic substances. Luciferin may give a bright light with luciferase in concentrations where chemical tests are negative. It is possible to see the light from 1 part of dried powdered Cypridinae added to 400,000,000 parts of water, whereas a chemical test is considered sensitive if one part in a million can be detected. A great advance in Cypridina. luciferin chemistry was made by Anderson (1933), who introduced a quantitative method of measur- ing luciferin by the total light produced under standard conditions, and by the development of a method of purification ( Anderson, 1935 ) , which has been used ever since, and which has been the starting point for the more recent work to be described in this paper. The compHcated procedure, mostly carried out in a hydrogen atmosphere F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 129 or at low temperature, involves extraction of dried Cypridinae with benzene to remove lipids, extraction with methanol to dissolve the luciferin, replacement of methanol with n-butanol, treatment with benzoyl chloride, partition between ether and water, hydrolysis of the benzoylated luciferin with aqueous HCl, partition of the freed luciferin with 71-butanol, and storage of the butanol solution under hydrogen. At least two cycles of benzoylation must be carried out for greatest purity. In this paper, material so treated will be referred to as doubly cycled luciferin. Removal of the butanol in vacuo leaves a resin-like amorphous brownish-yellow film with high luminescent activity. Each milliliter of the butanol luciferin solution contains from 0.05 to 0.115 mg. of solid and the purification of the material as measured in hght units per unit of weight is about 2000 times that of the dried powdered Cypridinae. The purified luciferin is much more stable against spon- taneous oxidation than are crude aqueous solutions, especially in dilute acid. Anderson (1936) also demonstrated that only the product of spon- taneous oxidation of luciferin, or the oxidation carried out with oxi- dants like ferricyanide, was reversible with reducing agents, and that the yield of luciferin on reduction of oxidized luciferin is only high (70%) if the reduction is carried out immediately after oxidation. If allowed to stand, little reduction occurs. When luciferin is oxidized with luciferase, and reduction is imme- diately attempted, a slight amount of luciferin can be obtained, at- tributed by Anderson to reduction of a small amount of spontaneously oxidized luciferin. This type of oxidation proceeds simultaneously with the light-emitting oxidation in presence of luciferase and com- phcates the interpretation of reaction kinetics (Chase and Lorenz, 1945). Anderson (1937) also studied the relation between salt con- tent and pH of the medium and total hght emitted. Salts are not necessary for luminescence, although NaCl in 0.01 M concentration increases the total light 2.3 times, while KI decreases total light to 0.08 of that in distilled water. Kinetics of the Ctjpridina luminescent reaction, involving quantitative studies of the effects of temperature, light, pH, and various inhibitors have been carried out by Chase and collaborators during the years 1940-1952. A resume of this work and 130 CHEMISTRY OF CYPRIDINA LUCIFERIN references to the literature will be found in the book of Harvey (1952). Recently the effect of hydrostatic pressure has been studied quantitatively by Bronk, Harvey, and Johnson (1952); the relation of oxygen concentration to Hght intensity by Hastings (1952). Since Anderson's studies, a number of workers have published cer- tain statements concerning the chemistry of luciferin, not all of which have been confirmed by later investigations. The earlier experiments have been reviewed by Harvey (1940) and later findings by Chase (1948), whose work has been especially concerned with the absorp- tion spectrum, to be considered in a later section. It is not yet possible to specify the exact chemical structure of Cypridina luciferin. Early statements have little meaning since they were based not only on very impure material but also on reasoning from one luminous animal to another. It now seems certain that the luciferin of Cypridina is quite different from that of the firefly and from that of luminous bac- teria; a considerable number of luminescent systems have probably arisen independently in evolution ( Harvey, 1953 ) . Historically it is interesting to note that at various times Dubois regarded Pholas luciferin as a proteose, a nucleoprotein, or an albu- min with acid properties. Cypridina luciferin has been called a peptone (Harvey, 1919), a phospholipid (Kanda, 1930), a polyhydroxy- benzene derivative (Anderson, 1936; Korr, 1936), a hydroquinone- like compound with a ketohydroxy side chain (Chakravorty and Bal- lentine, 1941), a flavoprotein and pyridine nucleotide (Johnson and Eyring, 1944), and a chromopolypeptide (Mason, 1952b). The hydroquinone type of structure was championed by the Dutch group of investigators (Kluyver, van der Kerk, and van der Burg, 1942; van der Kerk, 1942), particularly for bacterial luciferin, in the form of l,4-dihydroxynaphthyl-2-hydroxymethylketone. It seems quite certain that this compound cannot be Cypridiiw luciferin, since no light appears when it is mixed with luciferase (Johnson, Rexford, and Harvey, 1949; Spruit, 1949), although related naphthohydroqui- nones have not been ruled out. On the other hand, a number of investigators have suggested that flavins are important in bacterial luminescence. It has recently been found that luminescence results when reduced riboflavin or reduced riboflavin phosphate is added to cell-free extracts of luminous bac- 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 132 CHEMISTRY OF CYPRIDINA LUCIFERIN quinonoid structure are ruled out, but the amino group appears to be present, and it is possible that the oxidation-reduction change in luciferin involves the NH group. Anderson (1936) and Korr (1936) by an indirect method have placed the redox potential of luciferin at about +0.26 at pH 7. The rapid oxidation by f erricyanide has been made use of by Chase ( 1949 ) in estimating that the combining weight of luciferin is between 250 and 570, while a similar calculation from combination with cyanide indicated 800 to 2400 (Giese and Chase, 1940). Such calculations are subject to revision, depending on the amount of impurity present. Recent methods of purification and approach to the structure of Cypridina luciferin such as infrared spectroscopy, chromatography, and acid hydrolysis have been initiated by Mason ( 1952b) and Mason and Davis (1952). Together with a fourth approach, paper electro- phoresis, they will be discussed subsequently. Infrared spectroscopy, carried out with thin films of dried doubly cycled luciferin, has re- vealed strong absorption bands at 3250, 2825, 1680, 1625, and 1510 cm"~^, collectively indicating the amide bond as it occurs in peptides or in cyclic ureides. Acid hydrolysis of paper chromatographed lu- ciferin has yielded a yellow pigment and some eight amino acids, a result which led Mason (1952b) to the designation of Cypridina luciferin as a chromopolypeptide of a cyclic nature. A polypeptide structure of luciferin would place it in the group with a number of naturally occurring biological compounds, particularly the antibiotics gramicidin and polymixin, but with the addition of a color group. The solubilities of Cypridina luciferin are surprisingly similar to the cyclopeptide gramicidin, with the one exception that luciferin is more soluble in water. The primary objective of the current work has been to isolate pure Cypridina luciferin and establish a criterion of purity, while the ulti- mate goal is the isolation of luciferin in sufficient quantity to permit the elucidation of the structure by physical and chemical means. Three main lines of investigation have been pursued in an attempt to iso- late pure luciferin: (1) paper chromatography, (2) separation on ion-exchange resins and other adsorbents, and (3) filter paper elec- trophoresis. Spectral absorption measurements and fluorescence ob- servations have been made on the highly purified luciferin, as well F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 133 as quantitative determinations of the amino acids present after hy- drolysis by acid. Methods and results are described below. Chromatographic Purification The starting point for chromatographic procedures was Cijpridina luciferin prepared by two cycles of benzoylation according to the method of Anderson (1935). The final butanol solution of luciferin was stored under an atinosphere of hydrogen and aliquots were with- drawal as required. The butanol could then be removed by evacuation and the residue redissolved in methyl alcohol for work with paper chromatography and in the appropriate buffers for work with paper electrophoresis and ion-exchange resins. All these methods, as well as measurement of the absorption spectrum of doubly cycled luciferin, have revealed the presence of at least several impurities. Mason and Davis (1952) were able to chromatograph Cypridina luciferin on No. 1 Whatman paper, using the ascending technique. They worked at room temperature in a hydrogen atmosphere satu- rated with the vapor phase of the developing solvent. They reported that doubly cycled luciferin was separable by several solvents into what they designated as alpha and beta luciferins possessing the fol- lowing Rf values, respectively: chloroform, 0 and 0.25; n-butanol, 0 and 0.55, and n-butanol saturated with water; 0.8 and 0.8. Both luciferins were yellow but did not fluoresce. Their chromatographic studies did not reveal any impurities, either by ordinary or ultraviolet light, or by any of various tests tried. The preceding evidence may not necessarily indicate the existence of two kinds of luciferin, for in most of the present work only one luciferin has been observed, as manifested by a single, intense lumi- nous area on the paper when treated with luciferase. Occasionally after paper electrophoresis and more often with paper chromatog- raphy, a less intense luminous spot was detected at the origin, but this might be ascribed either to luciferin strongly adsorbed by the paper or to undissolved particles of luciferin, or both. At other times extremely weak luminescence has been observed at the solvent front after paper chromatography, but this is probably due to luciferin bound to impurities moving at the front. Luminescence at the origin was nearly always observed in paper chromatography where separa- 134 CHEMISTRY OF CYPRIDINA LUCIFERIN tions were poor as evidenced by either high or low luciferin Rf values accompanied by considerable tailing. At no time, however, have two luminous regions of the same area and intensity been observed; when two regions have been observed, one has always been at the origin. The work of Mason and Davis was repeated in every detail. For chloroform, the principal luminescent area was at Rf — 0, with a broad, unbroken band extending forward to about Rf = 0.14. When n-butanol was used, the principal luminescent region was at Rf = 0.77 and a very small, less intense spot occurred at Rf = 0. The diameter of the spot at R/ = 0 was practically the same as that of the luciferin spot at the origin before chromatography. With n-butanol saturated with water, the Rf of luciferin was about 0.88; there was no other luminous region. In addition to these solvents two other solvents, phenol and collidine ( 2,4,6-trimethylpyridine ) , each saturated with water, were also tried with little success. With the former, the residual phenol remaining in the paper after drying strongly inhibited the luminescent reaction with luciferase, whereas with the latter, rapid destruction of luciferin occurred. With Whatman No. 3 filter paper, the ascending technique was again employed and the chromatography run in the cold room at 2.0-2.5° C for a period of four or more hours. Doubly cycled luciferin in methanol was used to spot the paper. Solvents tried in various com- binations were benzene, ?j-propanol, water, acetone, n-butanol, eth- ylene glycol, n-amyl alcohol, n-hexyl alcohol, and ethyl ether. They gave from very poor to fair separations and considerable streaking occurred in most of them. The best chromatography was obtained with a developing solvent consisting of ethyl acetate, ethyl alcohol, and water (5,2,3 by volume, upper layer used). The paper, spotted with doubly cycled luciferin, stood in the developing chamber, in contact with the vapor, for one-half hour before lowering into the developing solvent. After the solvent front had moved a distance of about 26 cm (4 hours), the paper was removed from the chamber and dried in air, this process taking only a few minutes. On visual examination, the paper showed two main areas: a yellow-brown one at the solvent front and a yellow one between the front and the origin (see Fig. 1). When examined under a Mineralite lamp, two brightly fluorescent areas were observed: one of yellow fluorescence, F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 135 identical in position to the yellow color (R; = 0.5-0.6), and one of blue fluorescence (R/ = 0.2). The yellow-brown area at the solvent front showed only a weak, diffuse yellow fluorescence. On moistening the paper with luciferase solution, a single intense luminous area was observed which corresponded exactly to the area of the yellow color and yellow fluorescence. The yellow color, yellow fluorescence, the luminescence with luciferase and the Rf of 0.5-0.6 in this developing solvent can, therefore, be considered characteristic properties of Cy- pridina luciferin. Of these, luminescence with the enzyme is the ulti- mate criterion. The other materials on the paper are presumably im- FiG 1 Diagram showing paper chromatogram of doubly cycled luc.fenn devel- ■ oped with ethyl acetate-ethyl alcohol-water mixture. A represents origm; band B, an impurity, shows only blue fluorescence; band C. luciferm, shows yellow color and yellow fluorescence; D, a yellow-brown impurity; and E represents solvent front. purities. In similarly developed chromatograms, the yellow luciferin area of the paper was cut out and eluted with methyl alcohol. The yellow methanol solution was evacuated to dryness and the residue used for spectrophotometric studies (to be discussed in a later sec- tion) and for amino acid analysis by the chromatographic method ot Moore and Stein (1951). Efforts have recentlv been made to isolate luciferin by ion-exchange resins in amounts larger than heretofore obtainable with paper chro- matography and paper electrophoresis. Fine mesh Dowex 50 of vari- ous cross-Hnkages was tried in a column 2.26 sq cm X 9.0 cm. Lucif- erin was taken up readily from 0.1 N HCl at room temperature (23° C) in a system in which air had been replaced by pure hydro- gen. Elution was attempted, using successively oxygen-free 0.1 N HCl 1 N HCl, 3 N HCl, saturated NaCl, phosphate buffer pH 6.8, 136 CHEMISTRY OF CYPRIDINA LUCIFERIN 0.2 M disodium hydrogen phosphate (pH 8), and O.l N NaOH. Elu- tion occurred only with alkahne solutions of pH 8 or greater, but here luciferin lost its activity so rapidly that alkaline eluants appeared undesirable. A carboxyhc acid resin, XE-97 (Rohm and Haas), an analog of IRC-50, was also tried in fine mesh form in a column 2.26 sq cm X 17.5 cm at 2.0-2.5° C. Luciferin was readily taken up from pH 6.4 phosphate buflFer, forming a thin yellow fluorescent band at the top of the column. One-cubic centimeter fractions were collected with a fraction collector. The following eluants were tried without success: pH 6.4 phosphate buffer, pH 5.0 citrate buffer, and pH 3.4 citrate buffer. None of the fractions, adjusted to pH 6.8, gave light when tested with luciferase, and the yellow fluorescent band remained at the top of the column. When pH 3.4 citrate buffer containing 25% by volume of ethyl alcohol was used, the yellow fluorescent band moved slightly, but the effluent fractions gave no light when tested with luciferase (after making the pH 6.8 and diluting to minimize lucif- erase inhibition by the alcohol ) . The yellow band of the resin column was carefully removed and transferred to a 15-ml centrifuge tube where, after centrifuging off the buffer, the resin was washed with methyl alcohol and then cen- trifuged. A clear yellow supernatant was obtained which gave bril- liant luminescence with luciferase. Some of this was evacuated to dry- ness and the yellow residue redissolved in 0.1 N HCl. The absorption spectrum of this solution resembled that of those obtained by paper chromatography and also those from paper electrophoresis, although it showed slightly greater density at the short wavelengths. Using spectral absorption as a criterion, it is obvious that chromatographic procedures have resulted in removal of many extraneous substances and the preparation of luciferin in a high degree of purity. Further attempts have been made to isolate luciferin in larger amounts by employing column-partition chromatography techniques. Preliminary experiments with powdered cellulose columns operated in the cold room at 2.5° C have yielded rather good separation of luciferin from crude extracts of the material. Water represented the stationary phase, and a mixture of ethyl acetate and ethyl alcohol, saturated with water, made up the mobile phase. The cmde luciferin solution F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 137 was prepared by extracting fat-free Cypridirm powder with methyl alcohol, then removing the alcohol by evacuation, extracting the luciferin residue with butyl alcohol, removing this alcohol by evac- uation, and finally redissolving the luciferin in the mobile phase mix- ture for placement on the column. The developed band of luciferin could then be followed on the column by its yellow color and yellow fluorescence and could be collected at the appropriate time with a fraction collector. The luminescence with luciferase served to identify this luciferin. There is some evidence from spectral measurement that the luciferin isolated by the above procedure from the crude extract is different from the luciferin prepared by the Anderson's benzoyla- tion method. Electrophoretic Purification Filter paper electrophoresis of doubly cycled material was also used to isolate luciferin for spectrophotometric comparison with that obtained by paper chromatography. The electrophoretic method was essentially that of Kunkel and Tiselius (1951). A 0.1 M NaHoP04- H3PO4 buffer of pH 2.55 and ionic strength 0.1 was used to dissolve the luciferin. A potential gradient of approximately 5.9 volts/cm and 6.4 ma was appHed to a double layer of Whatman No. 3 paper for 14 hours at 10° C. The acid pH was selected because luciferin is more stable at low pH (Anderson, 1936). Dextran and Armour's crystalline bovine albumin were run concurrently and localized in a strip of the paper after electrophoresis with 1% bromophenol blue in absolute ethyl alcohol saturated with mercuric chloride. After dyeing the paper was washed with absolute methyl alcohol, leaving the albumin blue. Luciferin is presumably washed out by the methyl alcohol in which it is extremely soluble. Dextran indicated the degree of electro-osmotic flow of the buffer. After electrophoresis, a definite area on another strip of the paper showed a yellow color, a yellow fluorescence in the ultraviolet light of a Mineralite lamp, and luminesced with luciferase. As in the case of chromatography with ethyl acetate, only one intense luminescent area was observed. Luciferin, like albumin, was positively charged at this pH. The ratio of the distance moved by luciferin to that moved by albumin was approximately 0.34 (3.4 cm for luciferin, 10 cm for 138 CHEMISTRY OF CYPRIDINA LUCIFERIN albumin). The luciferin area was eluted with metliyl alcohol, the yellow solution evacuated to dryness, and the residue dissolved in 0.1 M HCl for measurements of the absorption spectrum (to be dis- cussed in the next section ) . Paper electrophoresis of luciferin was also carried out at room tem- perature and pH 4.0 (acetic acid/sodium acetate buffer), pH 5.8 (acetic acid/pyridine buffer), and pH 8.9 ( veronal/ veronal sodium buffer). At these pH values, two intense blue fluorescent spots, prob- ably the same blue fluorescent substance seen in paper chromatog- raphy, were observed moving toward the negative pole. At pH 4.0 and pH 5.8, the luciferin was still positively charged, although bo- vine albumin was positively charged at the lower pH but negatively charged at the higher pH. At pH 8.9, the inhibition of lucif erase by sodium barbiturate and the auto-oxidation of luciferin (resulting in low light production) were too great to localize the position of the migrating luciferin. Absorption Spectrum The first indication that luciferin possessed specific spectral absorp- tion which might prove useful in its isolation and identification was the observation (Chase, 1940) that neutral aqueous solutions had an absorption band at about 435 m/x which was replaced during exposure to air by another at about 465 m/x. The latter subsequently disap- peared, leaving the solution colorless to the eye. A yellow color had, indeed, been associated with Cypridina luciferin since it was first studied by Harvey (1917), and Anderson (1935) had never been able to dissociate color from the ability to give light with luciferase, even after as many as three cycles of purification by his method. The visible absorption band of an aqueous luciferin solution* and the changes which it underwent during exposure to air made it evi- dent that information on luciferin structure might come from ultra- violet absorption spectrophotometry, and that the spectrum would certainly be useful as a criterion of purity. Chase and Brigham (1951) obtained the first reliable measurements of the ultraviolet absorption " The luciferin used in the work to be described in this section was in all cases obtained by two cycles of Anderson's (1935) purification pro- cedure before being subjected to further purification methods. F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 139 spectrum of doubly cycled Cypridina luciferin, which had been washed with benzene to minimize ultraviolet-absorbing impurities. Luciferin itself is insoluble in benzene. The spectra were measured in pH 6.8 phosphate buffer solution, using the Beckman spectrophotom- eter, and are shown in Fig. 2. The heavy line represents the solution immediately after dissolving the luciferin. As shown earher (Chase, 1940), an initial absorption maximum in the visible region at about o o Ul o 0.20 0.15 - 0.10 0.05 0.00 0.3 HRS A 1.0 HRS 1 r \" 2 5 HRS 4 0 HRS 9 5 HRS pH S.8 "V X V w> ^ ^-•>.. ^-'■^ 1 250 300 350 400 WAVELENGTH - mjj 450 500 Fig. 2. Absorption spectra of a solution of doubly cycled and benzene-washed luciferin, measured in pH 6.8 phosphate buffer. The heavy line spectrum was obtained as soon as possible after the solution had been prepared. The solu- tion, standing exposed to air, at room temperature, was then remeasured at the times indicated. Characteristic of a luciferin solution under these condi- tions is the shift of the visible absorption band from a maximum at 435 to one at 465 mii. Solutions of pH 1.0 do not show this shift (see Fig. 3). 435 m/x is rapidly replaced by one at about 465 m^u and the latter then almost entirely disappears during exposure of the solution to air. In the ultraviolet the initial spectrum has a pronounced absorption peak at 265 m/x, and a shoulder at about 310 m^i. On exposure to air the absorptions at 265 and 310 m/j. decrease and, simultaneously, a new band appears at about 365 m/x. Isosbestic points exist at 330 and 140 CHEMISTRY OF CYPRIDINA LUCIFERIN 395 mfi. The absorption peak at 265 m/x could be responsible for photochemical changes which have been observed in luciferin after exposure to ultraviolet light (Chase and Giese, 1940). The stability as well as the form of the absorption spectrum are very dependent upon the hydrogen ion concentration. The spectrum is relatively stable at pH 1.0 but becomes rapidly less so at higher pH's. For this reason many measurements of the visible and ultra- violet spectra have been made under various experimental conditions in an effort to establish a cui"ve which would with certainty represent luciferin. None of these measurements has previously been published. Recently, application of the techniques of paper chromatography and paper electrophoresis has produced luciferin that is very much purer than any available previously. The details of the methods have al- ready been described. By means of micro attachments for the Beck- man spectrophotometer (Lowry and Bessey, 1946), we have now measured and compared absorption spectra of luciferin isolated by these methods. Since these purification procedures yield products having one and the same absorption spectrum which, furthermore, undergoes identical changes on exposure to air and treatment with base or acid, it would seem justified to assume a high degree of purity for the luciferin. In Fig. 3, five sets of absorption spectra are shown. Two different, doubly cycled luciferin preparations are represented, made independ- ently by two of us, using different batches of Cypridinae and slightly different procedures. These preparations will be designated as I and 11. The absorption spectra shown in Figs. 3A, 3B, and 3C are from preparation I; those in Figs. 3D and 3E, from preparation II. In all cases the solid line represents the spectrum measured as soon as pos- sible after dissolving the material in 0.1 N hydrochloric acid, with minimal exposure to air. The dash line and dotted line show the spec- tra of the same solutions after exposure to air for the times indicated. Figure 3B gives spectra of a luciferin solution which was subjected to no further purification than two cycles of Anderson's (1935) pro- cedure. Considerable general ultraviolet absorption is evident, but inflections in the curves and definite changes during exposure to air are quite apparent. The spectra in Fig. 3C are from a solution of luciferin isolated from preparation I by paper chromatography, while F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 141 r\ c 0.6 ^ J ^ A *\ 0 HR --- 12 HR 59 HR 0.4 « ' k \ 1.4 z o 1.2 en en i «n 1.0 < tr t- 0.8 • ft I 1 N k ^ ^h>^ i \ ^ \ * 1 ^ \ N , 1 V\ o \ I r 1 0.6 \ 0 HR 0.4 \ --- 23 HR 49 HR A 0.2 0.0 • ""- ::>^ V 1 0.8 0.6 0.0 250 300 350 400 450 500 550 250 300 350 400 450 500 550 WAVELENGTH IN m\A Fig. 3. Absorption spectra of 0.1 .V HCl solutions of doubly cycled luciferin, purified further in \ arious ways as described in the text. In each set of spectra the solid line represents the solution measured as soon as possible after being prepared, with minimal exposure to air, whereas the other two curves were measured after the solution had been exposed to air for the times indicated. A, B, and C are from one luciferin preparation, and D and E from another. It is apparent from inspection of these fi\e sets of absorption spectra (par- ticularly C, D and E ) that tliey are similar in form and in the changes which they undergo on exposure of the luciferin solutions to air. They may be presumed to represent luciferin. See the text for further details. those of Fig. 3A represent a solution of luciferin isolated from the same preparation by paper electrophoresis. Figure 3E shows the spectra of preparation II, washed with benzene after the last step in Anderson's procedure, but not otherwise treated. 142 CHEMISTRY OF CYPRIDINA LUCIFERIN In Fig. 3D are drawn the spectra of this same preparation after fur- ther purification by paper electrophoresis. It is at once apparent on inspection of the five sets of absorption spectra in Fig. 3 that they all show, not only the same absorption maxima, but also the same changes when the solutions stand in con- tact with air at room temperature. Even in the presence of consider- able impurity, as in Fig. 3B, the same absorption maxima ( represented by inflections) and the same changes in the spectrum are easily dis- cernible. The definite difference between the initial spectrum in Fig. 3A and those of Figs. 3C, 3D, and 3E is undoubtedly due to the fact that, in the first case, a longer time elapsed between eluting the material from the paper and measuring the initial spectrum. It is practically certain that the changes observed in the absorption spectra of luciferin solutions during exposure to air are the result of oxidation of some sort. Disregarding Fig. 3B, the initial spectra in Figs. 3A, 3C, 3D, and 3E are practically identical. They have an absorption maximum at 265 m/jL, a smaller peak (or shoulder) at about 310 m/i, and a broad absorption band in the visible region, centering at about 465 m^n. On standing, exposed to air at room temperature, the 265-m/x peak be- comes less, the peak at 310 m^ becomes much more pronounced (is possibly unmasked ) , a new band appears at 380 m/A and much of the visible absorption disappears. Finally, as the dotted line curves show, the 380-m/x band and practically all visible absorption disappear and, at the same time, there is a further decrease in the absorbance of the bands at 265 and 310 m/x. By this time the spectrum has become practically stable. It seems reasonably certain that Cypridirm luciferin has now been isolated in a rather high state of purity and that, consequently, the absorption spectra of Fig. 3 may be characteristic of luciferin at pH 1.0 and of the changes which it undergoes at this pH on exposure to air. A comparison of Figs. 2 and 3 shows that the spectrum of a lucif- erin solution is not only less stable at pH 6.8 than at pH 1.0, but also that the changes on exposure to air are different. Clearly, the hydro- gen ion concentration affects one or more components of this system. Furthermore, dilute hydrochloric acid solutions of doubly cycled lucif- F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 143 erin are definitely yellowish to the eye but become colorless in excess of sodium hydroxide. The disappearance of the visible absorption band at 465 mfi is responsible for this color change and even more striking changes occur in the ultraviolet region of the spectrum. If the solution is immediately made acid again, the yellow color partly returns. There are evidently acid-base changes in absorption, compli- cated by rapid oxidation changes especially marked in alkahne solu- tion, and these will now be described. 0.7 0.6 z o ^0.5 (n ^0.4 < tr. t-0.3 o -"0.2 1 \ — freshly dissolved in 0.1 N HCI — pH adjusted to 13 -J [ \ pH then readjusted to 1.0 \ ..- ■{ \V '''~ -N fi \ v_ \ ^ 0.1 '■" \ \ 1 '•.\ 250 300 350 400 450 500 550 250 300 350 400 450 500 550 WAVELENGTH I N ITl JU Fig. 4. In A, the dash hne shows the absorption spectrum of a luciferin solution adjusted to pH 13 soon after having been prepared in 0.1 N hydrochloric acid and before significant oxidation has occurred in the acid solution. An unstable absorption peak appears at 380 mix, which shifts rapidly toward shorter wavelengths. The dotted line curve shows the spectrum of the same solution after having been readjusted to pH 1.0. In B, the dash line represents the spectrum of a luciferin solution adjusted to pH 13 after having stood exposed to air at pH 1.0 for twelve days, until presumably the luciferin was completely oxidized. The alkaline spectrum, which is relatively stable, shows a new absorption peak at 330 m/i. See the text for details and dis- cussion. As shown in Fig. 3, a freshly prepared 0.1 N hydrochloric acid solu- tion of luciferin, isolated by paper chromatography, has a pronounced absorption peak at 265 m^t and a shoulder at about 310 m/x. There is no specific absorption in the long-wavelength region of the ultraviolet. On adjusting the pH of such a solution from 1.0 to 13 with sodium 144 CHEMISTRY OF CYPRIDINA LUCIFERIN hydroxide, a striking change occurs in the spectrum. Much of the far ultraviolet absorption and practically all in the visible disappear and a new peak is seen, maximal initially at about 380 m/x. This new peak apparently shifts when the solution is exposed to air at room tempera- ture and in about thirty minutes becomes stable, with a maximum at about 355 m/x. The initial shift is quite rapid. These changes are shown in Figs. 4A and 5. In Fig. 4A, dash line curve, the first measurement was made at 550 m/x., and the measure- 0.6 2 0.5 o 10 V) 2 0.4 iD Z < H0.3 o _l ' 0.2 0.1 0.0 \ u. / \ v^ / V, peak maosurtd 8 minults ofttr V j\ odiusling pH lo 13 r \ offer 25 minufei of pH 13 \ \, -^\ \ offer 34 minute* of pH 13 \ ;ioble condition finally raoc>, \ \. Xxn \ ^\^ ^^""■■"^-i \. ^~~--- 250 300 350 400 WAVELENGTH 500 550 450 IN m;j Fig. 5. An experiment designed specifically to show the rather rapid change in the absorption spectrum of a freshly prepared 0.1 N hydrochloric acid solu- tion of luciferin on adjusting the pH to 13 with sodium hydroxide. The most prominent feature of the new alkaline spectrum is appearance of an absorp- tion peak at 380 m/x, which rapidly shifts until it becomes stable at about 355 mn. The form of the stable alkaline spectrum is shown by the light solid line curve. ments were then continued toward shorter wavelengths, to 365 m/x. The next was then made at 400 m/x, after which they were again continued in the direction of the shorter wavelengths. The rapid shift of the 380-m;u absorption peak is quite evident from this sequence of measurements. F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 145 On adjusting the pH of such an alkahne luciferin solution back to pH 1.0 from 13, the absorption band at 355 m/A largely disappears and some of the specific absorption of tlie acid solution in the visible region and at 265 and 310 irifi is restored. A spectrum measured under such conditions is shown in Fig. 4A (dotted line curve), and it is apparent that the final acid absorption spectrum resembles qualita- tively that of the acid luciferin solution as originally prepared, al- though the densities are considerably decreased. An obvious interpre- tation of these results (although not necessarily the only one) might be that during the interval when the luciferin was exposed to dis- solved oxygen at pH 13 an appreciable amount was oxidized, so that upon making the solution acid once more, and thereby restoring the original form of the absorption spectrum, much of the absorbance of the peaks had been lost. The slight qualitative differences between the initial and the final spectra might be attributed to contribution by the spectrum of the oxidized form of the material. It is evident from the following experiment that reaction with dis- solved oxygen certainly plays a role in the above changes. If all oxy- gen is displaced from an acid luciferin solution immediately after its preparation, by passing pure hydrogen through the solution in the absorption cell, and if such an oxygen-free solution is then adjusted to a pH of 13, an absorption peak centering at 380 m/x is found and this peak is perfectly stable as long as the solution remains oxygen free. As has already been demonstrated in some detail, the absorption spectrum of aqueous luciferin solutions undergoes specific changes on standing exposed to air, and these changes are probably due to oxidation of the luciferin. If a luciferin solution of pH 1.0 is allowed to stand, exposed to air, until the spectrum has reached a practically stable condition, and if the pH is then adjusted to 13 with sodium hydroxide, a different absorption spectrum is obtained than when freshly prepared luciferin solutions of pH 1.0 are adjusted to pH 13. It will be recalled that fresh solutions show a new band at 380 m^n which rapidly shifts to a stable position at about 355 m/i,. On the other hand, the solution which has stood at pH 1.0 until its spectrum is stable, and is then adjusted to pH 13, shows a band which is relatively stable at 330 m/A as shown in Fig. 4B. This result indicates that a 146 CHEMISTRY OF CYPRIDINA LUCIFERIN change in the structure of the molecule may have occurred during its slow oxidation at pH 1.0. If the solution now be made acid again, some of the original acid form of the spectrum is restored ( dotted line curve of Fig. 4B). It is quite clear that solutions of luciferin purified in several different ways possess very specific absorption in the visible and ultraviolet regions of the spectrum. Furthermore, the spectrum undergoes pronounced changes under various experimental condi- tions. Were this absorption spectrum less complicated, one might feel optimistic about the identification of Cypridiim luciferin from its spectrum alone. Actually, the interpretation of an absorption spectrum in terms of chemical structure can be very difficult and frequently impossible unless other physical and chemical data are available. To date we have not been able to find in the literature any known compound whose absorption spectrum possesses the properties de- scribed here, and at the present time it is not possible on the basis of these measurements to make definite statements regarding the type of organic molecule to which Ctjpridina luciferin belongs. The spectrum may well prove of value in the ultimate identification, in conjunction with other information which may become available. Certainly, the ultraviolet absorption spectrum is proving an excellent criterion of purity and, in this way at least, is aiding greatly in the isolation of this extremely unstable compound. Fluorescence It has usually been stated that Cypriditia luciferin is nonfluorescent. Certainly there is no marked fluorescence in solution characteristic of such substances as quinine sulfate, oxidized flavins, acid luminol solution, and various dyes. In this respect Cypriditia luciferin differs from that of fireflies, which is highly fluorescent, bluish violet in weak acid and yellow-green in alkali (Strehler and McElroy, 1949). However, it had been observed (Harvey and Chase, unpublished, 1941) that the brownish, solid, doubly cycled luciferin fluoresced weakly yellow in the light from a mercury arc filtered through a Wood nickel oxide filter. This light consists principally of radiation in the near ultraviolet, the 365-m/x line. There was, however, no certainty that the fluorescence came from luciferin. F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 147 It has been pointed out in previous sections that paper chroma- tography and paper electrophoresis techniques demonstrate that in the doubly cycled luciferin there are at least two fluorescent sub- stances, one of which is luciferin, yellow fluorescent, and the other an impurity, blue fluorescent. For further study of luciferin fluorescence, the yellow fluorescent region of the filter paper was eluted with methanol, the solvent removed in vacuo, and the residue dissolved in 0.1 N HCl. In a beam of white light, observed from the side, no fluorescence can be detected, indicating a relatively weak intensity, but in the Wood light (maximum energy 365 mju) and also in a Mineralite (ultraviolet without the visible; maximum energy 253.6 m^) the luciferin solution is yellow fluorescent. In the light from a "Purple ultra" incandescent bulb the fluorescent color was green to bluish green, probably due to a combination of the yellow fluorescence and the blue light transmitted by this bulb. Dilution of the 0.1 N HCl luciferin with 9 parts of water nearly abolishes the fluorescence, in- dicating its low intensity. Paper chromatographed luciferin is also yellow fluorescent when dissolved in methanol and butanol. It is clear that Cypridina luciferin does exhibit fluorescence,* although it is far less bright than the fluorescence of many other chemiluminescent sub- stances. The effect of acid-base change and oxidation of chromatographed luciferin in aqueous solution was studied by preparing three tubes with luciferin in 0.1 N HCl: leaving one tube acid, neutrahzing another with NaOH, and making a third alkaline with NaOH to about pH 13. Table I shows the relative fluorescence of the fresh, active luciferin and of the material allowed to stand for 14 days until the luciferin had completely oxidized. It is apparent from the observations noted in the table that the fluorescence of a 0.1 N HCl solution of luciferin is very much less after it has been exposed to air until completely oxidized than when freshly made up. This can quite probably be interpreted in terms of the decrease of the ultraviolet absorption of such a solution which ° In certain solvents (e.g., ethanol) at room temperature, Cypridina luciferin exhibits an apparent "phosphorescence" after irradiation with ultraviolet light. This effect iS believed to be a sustained chemiluminescence induced by the radiation (see Tsuji and Harvey, 1954). 148 CHEMISTRY OF CYPRIDINA LUCIFERIN was demonstrated in the last section (see Fig. 3, for example). The fact that, even in the freshly prepared acid luciferin solution the fluorescence is only slight, compared with that of such compounds as riboflavin, may be in part due to absence of pronounced absorption bands in the near ultraviolet region of the spectrum (i.e., 330-400 m/t). On the other hand, quenching phenomena may be involved, as in the case of flavins combined with protein (see Weber, 1950). TABLE I Fluorescence of Chromatographed Luciferin and Completely Oxidized Luciferin Color Fluorescence in Preparation Mineralite Wood's ultraviolet Luciferin On paper Yellow Bright yellow Yellow 0.1 N HCl YeUow Yellow Yellow Neutral Yellow YeUow Yellow 0.1 N NaOH Colorless" Deep yellow Yellow* Oxidized Luciferin On paper Colorless Pale yellow Pale yellow 0.1 N HCl-^ Colorless Pale yellow Pale yellow Neutral'' Colorless Pale yellow Pale yellow 0.1 A^ NaOH'' Colorless Very pale yellow Very pale yellow " The color returns to a less intense yellow on immediate neutralization. ' On first making alkaline, the yellow fluorescence is relatively more intense than after standing exposed to air for 15 minutes. " Studied in a nonfluorescent cellophane tube after 3 weeks when no light appeared on mixing with luciferase. When a freshly prepared 0.1 N hydrochloric acid solution of paper chromatographed luciferin is made alkaline (pH 13) with sodium hydroxide and is then irradiated with the Wood light, there is at first a slight enhancement of the fluorescence compared with that given by the freshly prepared acid solution. The fluorescence rapidly de- •creases, however, as the alkaline solution stands, and becomes rela- tively faint. This effect can perhaps be explained by the changes which occur in the absorption spectrum of such a luciferin solution upon adjusting its pH to 13. It will be recalled that a prominent, new absorption band appears at 380 mix when such a luciferin solution is F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 149 first made alkaline, but that on standing exposed to air, this band rapidly shifts to a new maximum at about 355 m/t passing through the region 365 m/* of great intensity in the Wood light (see Fig. 4A and Fig. 5 ) . It is quite possible that the initial enhancement of the fluores- cence when the solution is made alkaline is due to absorption in the 365-m/x region, and its subsequent diminished intensity is because of the appearance of the 355-m^ band at a shorter wavelength where less energy is emitted by the Wood light. On the other hand, an oxidized acid luciferin solution that is then adjusted to pH 13 shows very faint fluorescence when irradiated with the Wood hght. This seems quite understandable in view of the near ultraviolet absorption spectrum of such a solution, as described in the last section and shown in Fig. 4B. It will be recalled that such a solution exhibits a rather narrow absorption band centering at about 330 millimicrons. Very little energy would be available at this wave- length in the Wood light for fluorescence excitation. There does, therefore, appear to be consistency between the results of the experi- ments on fluorescence and those involving the absorption spectrum of luciferin solutions under various conditions. A quantitative determina- tion of the spectral distribution of the fluorescence of solutions of paper chromatographed luciferin should be of value for the identifica- tion of this compound and such measurements are contemplated. Of course, it is always possible, as was stated earlier in this section, that if luciferin is a chromopolypeptide, as suggested by Mason (l&52b), the spectral distribution of fluorescence might be so affected by the presence of the constituent amino acids as to make any interpretation in terms of chemical structure extremely difficult. Hydrolysis of Luciferin Mason ( 1952b ) has reported that his alpha and beta kiciferins are chromopolypeptides, on the following grounds: (1) alpha luciferin was convertible to beta luciferin under certain experimental condi- tions; (2) the infrared spectrum of beta luciferin indicated amide bonds as they occur in peptides and cyclic ureides; (3) hydrolysis of beta luciferin yielded a number of amino acids, including an unidenti- fied ninhydrin positive substance and a yellow pigment; (4) beta luciferin gave a positive N-chloroamide test (Rydon and Smith, 1952), 150 CHEMISTRY OF CYPRIDINA LUCIFERIN thus indicating the presence of either an amino acid, protein, cycHc peptide, diketopiperazine, acylated amino acid, or acylated peptide; and (5) beta luciferin gave a positive dye retention test (Robinson and Fehr, 1952) for proteins and peptides. The amino acids obtained from 24-hour hydrolysis of beta luciferin with 4 N HCl at 135° C were identified by two-dimensional paper chromatography and micro- biological assay as follows: glycine, threonine, proline, lysine, aspartic acid, glutamic acid, and either leucine, isoleucine, or phenylalanine. In view of Mason's findings, the following experiments were under- taken, using 28 mg of doubly cycled luciferin. The absorption spec- trum of this material indicated the presence of some impurity, but the following tests were negative: biuret, ninhydrin, anthrone (for carbohydrates), and molybdenum blue (for phosphorus after com- plete digestion). Quantitative protein determination by the method of Lowry et al. ( 1951 ) indicated that this luciferin preparation was 28% protein or polypeptide, calculated in terms of equivalence of crystalline bovine albumin. This doubly cycled luciferin was hydro- lyzed by refluxing with 6 N HCl (ninhydrin negative) for 16 hours according to the method of Stein and Moore (1948).* The hydrolyzate after filtration had a clear golden-orange color, was moderately fluorescent (ultraviolet excitation) and gave weak luminescence with luciferase. It also showed a strong positive ninhydrin test according to the method of Moore and Stein ( 1948 ) . The ninhydrin color value obtained indicated an equivalence of 35% leucine. The hydrolyzate was analyzed for amino acids, using both the long and short Dowex 50 columns, by the method of Moore and Stein (1951). Very good separations of the amino acids were obtained. Recovery of ninhydrin color value, expressed as equivalence of leucine introduced into both columns, was 107%. By this technique, the following substances were found in the luciferin hydrolyzate: taurine, aspartic acid, threonine, serine, sarcosine, glutamic acid, proline, glycine, alanine, valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, lysine, ammo- nia (from degradation of amino acids and amides), histidine, and arginine. The presence of beta alanine, tryptophane, ethanolamine, hydroxylysine, and ornithine was uncertain. On the other hand, there ' It is a pleasure to acknowledge the collaboration of Dr. Harold H. Williams of Cornell University in the amino acid analysis. F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 151 was no cysteic acid, urea, hydroxyproline, citrulline, cystine (nor cysteine), alpha-amino-n-butyric acid, and glucosamine. The relative abundance of the amino acids is shown in Figs. 6 and 7. In addition, hydrolysis studies were carried out on luciferin isolated by both paper electrophoresis and paper chromatography. However, 2.50 - t.ti- t.oo- I.T» I. SO ■ O i.oof- to O.TS < q: e> o Alanine (4 «7I 5 «»TT...i.. ""•' (O o.ts Tkrttnint Clalinile ClfCint (0.3» Aci4IS.IT) (l.tS) Swln«(0.t9) " frtlllt S«rc«>ln» I \ i\ (4.l«) 10. St) AtMrtic Aci4 <0.I7| 90 litltuclii (l.»») VtllM (I 4*1 Mtlllionint (O.SO ISO -»HS.S7, ST.S'e- citratt Llnciat (2 1*) PhtnylaUnin* (OtI) ITS ZOO 229 2S0 2TS JOO S2S f. ISO '»H4.2S,S0 C- cilrttt ^^»M4.2»-. TS*C l.tf I.I oieo- o.te Aaatii* < / )ti«ill>«n« T LftlM ArfUli*t ' - ■ jts 400 4tS 4S0 47S ioO tti »»0 JT9 Soo MS Sso iT» TOO ^»H4.tI->^»H«.T0>- r»*c tf'c •itt4t* 4itt(tt -»n t.io- -»H •.tO.tS'C' ->HII.O, is'c- (•r»*4- pM6 8,25*C|pnosphat« -.|. a. -1- 150 175 200 225 PH 6 5,25*C, citiole ■ EFFLUENT cc. Hv^ Fig. 7. Isolation of basic amino acids from same luciferin hydrolyzate as shown in Fig. 6. Chromatography was carried out on Dowex 50 column 0.9 X 15 cm according to the method of Moore and Stein (1951). The column was op- erated in the sodium form. The pH, the buffers used as eluants, and tem- perature of operation of the column are indicated. The first peak may be taurine, while the second peak represents all the amino acids emerging be- fore tyrosine. paper, one on each side of the luciferin band. The amount of luciferin isolated by paper electrophoresis was not sufficient to carry out amino acid analysis of the hydrolyzate on Dowex 50 columns. Luciferin isolated by paper chromatography was also hydrolyzed with 6 N HCl for 16 hours and evacuated to dryness in a vacuum desiccator. The residue gave a very strong ninhydrin test, whereas before hydrolysis only a faint test was obtained. This hydrolyzate also showed some luminescence with luciferase and a characteristic yellow F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 153 fluorescence in ultraviolet light. The paper areas on both sides of the luciferin region were also hydrolyzed, both hydrolyzates giving a very strong ninhydrin test. When ahquots of the hydrolyzates from paper chromatography and from paper electrophoresis were repeatedly made alkaline and evacuated each time to dryness in a desiccator, they still gave strong positive ninhydrin tests, indicating that the ninhydrin- positive material was not a volatile substance like ammonia. From the ninhydrin color value, 25% of the luciferin hydrolyzate from paper chromatography could be expressed in terms of equivalence of leucine. Following these tests, the remainder of the hydrolyzate of luciferin from the paper chromatography isolation was analyzed for amino acids in the long Dowex 50 column according to the method of Moore and Stein (1951). The following compounds, listed in approximate order of abundance, were found to be present in the hydrolyzate: either aspartic acid, threonine, or serine; ammonia, arginine, leucine, isoleucine, glutamic acid, alanine, vahne, sarcosine, lysine, taurine, and methionine. Of these compounds ammonia and either aspartic acid, threonine or serine were found in very high concentration, although ammonia, arginine, and lysine may be present in higher concentration than indicated. The presence of proHne and histidine was uncertain, whereas tyrosine and phenylalanine were absent. There is no doubt but that doubly cycled luciferin contains amino acids. The positive ninhydrin reaction given by hydrolyzates from other regions of the paper chromatogram may have come from oxidized luciferin or poly- peptide impurities. References Anderson, R. S. 1933. The chemistry of bioluminescence I. Quantitative determination of luciferin. /. Cellular and Comp. Physiol, 3, 45-59. Anderson, R. S. 1935. Studies on bioluminescence II. The partial purifica- tion of Cypridina luciferin. /. Gen. Physiol, 19, 301-305. Anderson, R. S. 1936. Chemical studies on bioluminescence III. The re- versible reaction of Cypridina luciferin with oxidizing agents and its relation to the luminescent reaction. /. Cellular and Camp. Physiol, 8, 261-76. Anderson, R. S. 1937. Chemical studies on bioluminescence IV. Salt effects 154 CHEMISTRY OF CYPRIDINA LUCIFERIN on the total light emitted by a chemiluminescent reaction. /. Am. Chem. Soc, 59, 2115-17. Bronk, J. R., E. N. Harvey, and F. H. Johnson. 1952. The effects of hydro- static pressure on luminescent extracts of the ostracod crustacean, Cypridina. J. Cellular and Comp. Physiol., 40, 347-65. Chakravorty, P. N., and R. Ballentine. 1941. On the luminescent oxidation of luciferin. /. Am. Chem. Soc, 63, 2030-31. Chase, A. M. 1940. Changes in the absorption spectrum of Cypridina lucif- erin solutions during oxidation. /. Cellular and Comp. Physiol., 15, 159- 71. Chase, A. M. 1942. The reaction of Cypridina luciferin with azide. /. Cellular and Comp. Physiol, 19, 173-81. Chase, A. M. 1948. The chemistry of Cypridina luciferin. Ann. N. Y. Acad. Sci., 49, 353-75. Chase, A. M. 1949. Studies on cell enzyme systems I. The effect of ferricya- nide on the reaction of Cypridina luciferin and luciferase and the com- bining weight of luciferin. /. Cellular and Comp. Physiol., 33, 113-22. Chase, A. M., and E. H. Brigham. 1951. The ultraviolet and visible ab- sorption spectra of Cypridina luciferin solutions. /. Biol. Chem., 190, 529-36. Chase, A. M., and A. C. Giese. 1940. Effects of ultraviolet radiation on Cypridina luciferin and luciferase. /. Cellular and Comp. Physiol., 16, 323-40. Chase, A. M., and J. H. Gregg. 1949. Analysis of Cypridina luciferin for nitrogen. /. Cellular and Comp. Physiol., 33, 67-72. Chase, A. M., and P. B. Lorenz. 1945. Kinetics of the luminescent and non- luminescent reaction of Cypridina luciferin at different temperatures. /. Cellular and Comp. Physiol., 25, 53-63. Dubois, R. 1887. Fonction photogenique chez le Pholas dactylus. Compt. rend., 105, 690-92. Giese, A. C, and A. M. Chase. 1940. The effects of cyanide on Cypridina luciferin. /. Cellular and Comp. Physiol., 16, 237-46. Harvey, E. N. 1917. Studies on bioluminescence. IV. The chemistry of light production in a Japanese ostracod crustacean, Cypridina hilgendorfii, Miiller. Am. J. Physiol, 42, 318-41. Harvey, E. N. 1919. Studies on bioluminescence IX. Chemical nature of Cypridina luciferin and Cypridina lucfferase. /. Gen. Physiol, 1, 133-45, 269-93. Harvey, E. N. 1925. The inhibition of Cypridina luminescence by light. /. Gen. Physiol, 7, 679-85; 10, 103-10, 1926. Harvey, E. N. 1940. Living Light. Princeton University Press, Princeton, N.J. Harvey, E. N. 1952. Bioluminescence, p. 302. Academic Press, New York, N. Y. F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 155 Harvey, E. N. 1953. Bioluminescence: Evolution and comparative biochem- istry. Federation Proc, 12, 597-606. Hastings, J. W. 1952. Oxygen concentration and bioluminescence, 1 and 11. ]. Cellular and Comp. Physiol, 39, 1-30; 40, 1-9. ^ , r Hastings J W W. D. McElroy, and J. Coulombre. 1953. The effect ot oxygen upon 'the immobilization reaction in Brefly luminescence. /. Cel- lular arid Comp. Physiol, 42, 137-50. . , , -r ■ , -r Johnson, F. H., and H. Eyring. 1944. The nature of the luciferm-luciferase system. /. Am. Chem. Soc., 66, 848. ,, ,. i Johnson F. H., D. R. Rexford, and E. N. Harvey. 1949. The hypothetical structure of luciferin. /. Cellular and Comp. Physiol, 33, 133-36. Kanda, S. 1930. The chemical nature of Cypridina lucifenn. Science, 71, 444 Kluyver, A. J., G. L. M. van der Kerk, and A. van der Burg. 1942. The effect of radiation on light emission by luminous bacteria. Proc. Ned. Akad. Wetenschap., 45, 886-95, 962-67. Korr, I. M. 1936. The luciferin-oxyluciferin system. /. Am. Chem. boc, a^, 1060-61. , f . • fiu^v Kunkel, H. G., and A. Tiselius. 1951. Electrophoresis of proteins on hltei paper. /. Gen. Physiol, 35, 89-118. , u t. v Loww O. H., and O. A. Bessey. 1946. The adaptation of the Beckman spectrophotometer to measurements on minute quantities of biological materials. /. B/o/. C/iem., 163, 633-39. j n m-i td Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J Randalk 19ol. Pro- tein measurement with the Folin phenol reagent. /. Eiol Chem., 193, MasonT H. S. 1952a. Cypridina luciferin. Arch. Biochem. and Biophys., 35, 472 Mason, H. S. 1952b. The ^-luciferin of Cypridina. ]. Am. Chem. Soc, 74, Mason, H. S., and E. F. Davis. 1952. Cypridina luciferin. Partition chroma- tographv. /. Biol. Chem., 197, 41-45. r j • McElroy, W. D., and J. Coulombre. 1952. The immobilization of adenosine triphosphate in the bioluminescent reaction. /. Cellular and Comp. Physiol, 39, 475-85. Moore, S., and W. H. Stein. 1948. Photometric "i^l^y^--;" "^t.^^^^"' "'" in the chromatography of amino acids. J. Biol Chem 176, 367-88. Moore S and W. H. Stein. 1951. Chromatography of amino acids on sul- phonated polystyrene resins. /. Biol Chem., 192, 663-81. ■ Robinson, F. A., and K. L. A. Fehr. 1952. Estimation of protamine and in- sulin in protamine zinc insulin. Biochem. J., 51, 29«-.iU.i. Rydon, H. N., and P. W. G. Smith. 1952. A new method for the detection of peptides and similar compounds on paper chromatograms. Nature, 169, 922-23. 156 CHEMISTRY OF CYPRIDINA LUCIFERIN Spruit, C. J. P, 1949. The chemical nature of the luciferins. Enzymologia, 13, 191-200. Stein, W. H., and S. Moore. 1948. Chromatography of amino acids on starch columns, separation of phenylalanine, leucine, isoleucine, methionine, tyrosine, and valine. /. Biol. Chetn., 176, 337-65. Strehler, B. L., E. N. Harvey, J. J. Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Proc. Natl. Acad. Sci. U. S., 40, 10-12. Strehler, B. L., and W. D. McElroy. 1949. Purification of firefly luciferin. /. Cellular and Comp. Physiol, 34, 457-66. Sumner, J. B. 1944. A method for the colorimetric determination of phos- phorus. Science, 100, 413-14. Tsuji, F. I., and E. N. Harvey. 1954. Apparent "phosphorescence" of Cypridina luciferin solution. Arch. Biochem. and Biophys., 52, 285-86. van der Kerk, G. J. M. 1942. Onderzoekingen over de bioluminescentie der lichtbacterien. Doctorate thesis. N. V. Kemink en Zoon. Utrecht. Weber, G. 1950. Fluorescence of riboflavin and flavinadenine dinucleotide. Biochem. }., 47, 114-21. Discussion The Infrared Spectrum of Anderson's Luciferin Dr. Mason: In Figs. 8 and 9 are depicted infrared spectra of Cypri- dina luciferin prepared by the method of Anderson. To obtain these spectra dry flakes of amorphous luciferin were placed between rock salt surfaces and absorption was measured by manual adjustment of wavelength setting and null point with a Beckman infrared spectro- WAVE NUMBERS IN cm"' 4000 3000 2000 1500 12 50 1000 2 0 40 6.0 8.0 100 WAVELENGTH IN MICRONS Fig. 8. 12 0 14.0 meter. Strong bands occur at 3250, 2800, 1680, 1630 and 1510 cm-^. The absorption of infrared energy in the 2700 to 3100 cm"^ (Fig. 9) region by a luciferin film was measured by Dr. R. C. Gore of the American Cyanimid Company, Stamford. Using a lithium fluoride prism, ab- sorption bands were found at 2860, 2920, 2960 and 3060 cm-^ It was the comparison of these results with the published spectra of proteins and polypeptides that led to the inference that the peptide bond is present in Anderson's preparations of Cypridina luciferin. As it seems evident from Dr. Tsuji's new work that Anderson's luciferin may yet be of a low order of homogeneity, these spectra are being ofiFered as a physical characterization of the material obtained at that stage and as 157 158 CHEMISTRY OF CYPRIDINA LUCIFERIN a basis for comparison with preparations of Cypridina luciferin of higher degrees of homogeneity. In our hands, the amino acids, glutamic acid, glycine, threonine, proline and leucine, isoleucine, or phenylalanine were identified in acid hydrolyzates of Cypridina luciferin. Microbiological tests were also performed on luciferin hydrolyzate by Dr. B. D. Davis of the U. S. Public Health Service using a series of amino acid auxotrophes of 2700 2800 2900 WAVE NUMBERS IN cm Fig. 9. 3000 I 3100 B. coll; the presence of all the above amino acids was confirmed and, in addition, evidence was found for the presence of methionine, argi- nine, histidine, valine, and tyrosine. Dr. Kauzmann: Have you tested for the presence of sulfhydryl and disulfide groups in Cypridina luciferin? Dr. Tsuji: Sulfhydryl groups do not appear to be present in so far as tests with alkaline nitroprusside and p-chloromercuribenzoic acid are concerned. Mason also was not able to detect sulfhydryl groups with alkaline nitroprusside. However, we have observed that when the sulfhydryl blocking agent, N-ethyl maleimide, is allowed to stand F. I. TSUJI, A. M. CHASE AND E. N. HARVEY 159 with a water solution of luciferin for one-half hour, the yellow color of luciferin is replaced by a pink color and luminescence is markedly decreased when luciferase is now added to the solution. In answer to the second part of your question, we have not tested for disulfide groups specifically. Dr. Mason: Is it possible to estimate the concentration of solute in the solutions with which you measured your absorption spectra and to compare the concentrations with those of the solutions of Ander- son's luciferin with which you and Brigham measured your earlier spectra? Dr. Chase: Because luciferin in minute amounts was eluted from filter paper and then finally dissolved in very small volumes of hydro- chloric acid for measurement of the absorption spectrum, it was not possible actually to determine the weight of the solute used in the more recent studies. In the measurements of Chase and Brigham (1951), where larger amounts of material were involved, weighing of solid residues was possible and the actual concentration of solute could therefore be determined. It amounted to about 0.02 mg of dry material per milliliter of solution. Assuming that the absorption spectrum which has been presented really represents luciferin, it is possible, of course, to estimate con- centrations from relative extinction values. If this is done, the con- centration of solute used in the present measurements would be approximately three times that used in those reported by Chase and Brigham (1951; Fig. 2, p. 532, and Fig. 3, p. 534). Biochemistry of Firefly Luminescence W. D. McElroy and J. W. Hastings McCollum-Pratt Institute and Department of Biology, Johns Hopkins University, Baltimore, Maryland, and Department of Biological Sciences, Northwestern University, Evanston, Illinois Judging from the extensive literature on fireflies and glowworms it is apparent that these luminous forms have excited the interest and imagination of scholars and laymen alike. The folklore, history, and scientific accomplishments connected with these luminous forms have been reviewed in detail by Harvey (1952). The first definitive experi- ment regarding the nature of the components necessary for light production was reported by Dubois in 1885. He found that the lumi- nous organs of Pyrophorus, a luminous beetle, would cease to emit light if immersed in hot water. He observed, however, that a cold water extract which had ceased to luminesce could be stimulated to emit hght by adding the hot water extract. On the basis of this type of experiment Dubois proposed the theory that there was, in the hot water extract, a substance stable to heat which was destroyed during its luminescent oxidation through the action of a catalyst present in the cold water extract. He named the heat stable material luciferin and the enzyme which catalyzed its oxidation luciferase. This observa- tion which has been greatly extended and clarified by Harvey and associates is the classical luciferin-luciferase test which is routinely made on all new luminous forms discovered by workers in this field. From a comparative biochemical viewpoint the dissipation of energy in a biological system in units of 40-60 kcal appears quite unique. Most of our information on the generation and transfer of useful energy in biological systems suggests that the process takes 161 162 BIOCHEMISTRY OF FIREFLY LUMINESCENCE place in a stepwise manner and that each unique step involves energy changes not greater than 10-15 kcal. The two important and obvious exceptions involve either the absorption of radiant energy (photosyn- thesis) or the emission of light (bioluminescence). In an effort to determine the nature of the oxidative reaction which presumably liberated 40-60 kcal the senior author in 1947 attempted to isolate luciferin from firefly lanterns. The component which was finally isolated as the barium salt was ultimately identified as adenosine triphosphate (McElroy, 1947). Thus by the classical definition ATP was firefly luciferin. Since ATP was known to be the immediate energy source for mechanical activity as well as other processes, this seemed to be a logical extension of the idea of energy transport and utilization in a biological system. Three important questions, however, remained unanswered: (1) Does all the energy for light emission come from phosphate bond energy? (2) What is the importance of the oxidative reaction? (3) What is the nature of the light emitting molecule? Subsequent work on the fireflies indicated that two addi- tional factors were required. One of these turned out to be magnesium ion and the second was a fluorescent substance with an emission spectrum similar to the luminescent spectrum ( see Fig. 1 ) . Since this substance is destroyed during luminescence and because of its fluores- cent properties, it was called luciferin, the phosphor of fireflies. All subsequent work has indicated that the basic reaction in fireflies can be described by the following reaction: Luciferin + luciferase + Mg + ATP + oxygen — > light + products Both luciferin and luciferase have been highly purified, and the pres- ent but brief review will attempt to describe some of the properties of the individual components as well as the entire system. Purification of Luciferase and Luciferin Luciferase The following procedure is based upon the original report of McElroy and Coulombre (1952). Five grams of the dried lanterns of Photinus pyralis are ground with sand and extracted three times with a total volume of 100 ml HoO. The pH of the extract is adjusted to 8 w. D. Mcelroy and j. w. Hastings 163 with NaOH, and the sokition is placed in the deep freeze. After freezing and thawing, the inactive precipitate is removed by cen- trifuging. Twenty-five milhhters of a calcium phosphate gel (16.7 mg/ml) is centrifuged, and the supernatant is discarded. The extract is then thoroughly mixed with the gel and the pH adjusted to 8. After 15 minutes the mixture is centrifuged and the gel is discarded. >- (9 Z 50- < o a. CO 500 550 600 MILLIMICRONS 650 Fig. 1. The emission spectrum of the light emitted from firefly extracts (McElroy and Rainwater, 1948). The supernatant (prep. II) is considerably more active than the crude extract. Ninety milliliters of the calcium phosphate gel is centrifuged and subsequently mixed with 90 ml of preparation II. The pH is maintained at 8. After 15 minutes the mixture is centrifuged and the supernatant is discarded. In this latter step most of the luciferase was adsorbed onto the gel while the majority of the luciferin remained in the supernatant. To remove the residual luciferin as well as inactive 164 BIOCHEMISTRY OF FIREFLY LUMINESCENCE protein the gel is washed twice with cold alkaline water and then with a 2% solution of (NH4)2S04 at pH 8. Elution of the enzyme is ob- tained by washing the gel twice with a 7% solution of (NH4)2S04 at pH 8 (prep. III). The final volume of combined eluates of preparation III is 95 ml. Preparation III is then fractionated with (NH4)2S04 in successive steps of 10% saturation up to 50% saturation and then in units of 2 to 3% saturation up to 65%. The pH during this procedure is main- tained at 8.0, The major part of the active enzyme is recovered between 57-65% (NH4)2S04 saturation. The latter precipitate is dis- solved in 25 ml of water (prep. IV) and the enzyme is readsorbed onto calcium phosphate gel as described above. The supernatant is discarded. The enzyme is eluted with 7% (NH4)2S04 at pH 8 (prep. V) and precipitated by adding sohd (NH4)2S04 to 70% saturation (pH 8). The precipitate is dissolved in 5 ml of H2O and the pH is adjusted to 8 (prep. VI). A further treatment of preparation VI with the low concentration of calcium phosphate gel removes some inert protein (prep. VIII). The activity of the various fractions is summarized in Table I. In this procedure the enzyme was purified approximately 70 times on a protein basis with a total recovery of 15%. In addition, the preparation is completely free of luciferin, and under these conditions no light is emitted upon the addition of ATP. Preparation VI (Table I) remained stable for several weeks at temperatures below 0°, but was rapidly inactivated at room tempera- ture, especially in dilute solutions. At 40°, approximately 50% loss of activity was encountered in 10 minutes. The preparation was colorless with a sharp absorption peak at 278 millimicrons. The total amount of light which could be obtained from preparation VI was greatly re- duced by purification. Within a few minutes after the addition of ATP, the light disappeared completely and reappeared only when additional purified luciferin was added. The results demonstrated that under these conditions a definite, but limited amount of light was obtained for a given amount of luciferin, indicating an irreversible inactivation such as observed in the Cypridina system. Although preparation VI responded normally to ATP, no light was emitted with ADP, in contrast to the crude extract. Similarly, ATP which had been treated with hexokinase and glucose failed to initiate light emission in preparation VI. The results demonstrate that the terminal phosphate w. D. Mcelroy and j. w. Hastings 165 TABLE I Purification of Firefly Luciferase Light, Protein, Specific activity, Preparation units/ml-volts mg/ml volts/mg protein I. Crude 57 10.1 5.7 II. Supernatant 1st Ca3(P04)2 gel 166 6.3 26 III. Eluate 2nd Ca3(P04)2 gel 80 0.82 98 IV. 57-65% (NH4)2S04 precipitate 105 0.60 175 V. Eluate 3rd Ca3(P04)2 gel 92 0.33 279 VI. 70% (NH4)2S04 precipitate 210 0.627 335 VII. Supernatant 4th Ca3(P04)2 gel 182 0.465 391 on ATP is the only immediate labile phosphate group which can be used in the luminescent reaction. These results, as well as others which will be reported later concerning the hexokinase reaction, indicate that an active myokinase had been removed during purifica- tion. Luciferin Most of the firefly luciferin remains in the supernatant after the calcium phosphate gel treatment. The supernatant is adjusted to pH 3.5 and extracted twice with an equal volume of redistilled ethyl acetate. All the active luciferin passes into the ethyl acetate. The ethyl acetate is removed by vacuum distillation and the active luciferin is dissolved in a small volume of water. This crude prepara- tion can be used for enzyme assay. Further purification is achieved by adsorbing the luciferin on an acid (2N HCl) treated Dowex 50 column (mesh size less than 80). The column is washed thoroughly with 2N HCl and finally with water. The luciferin is slowly developed on the column by a weak solution of NH4OH (1.5%). The luciferin migrates down the column in a sharp band and is finally eluted. The luciferin can be readily followed on the column by its brilliant yellow- green fluorescent band. The eluates containing the active luciferin 166 BIOCHEMISTRY OF FIREFLY LUMINESCENCE are again extracted with ethyl acetate, and the latter is concentrated by vacuum distillation. The luciferin is finally concentrated in water. At neutral pH luciferin has two characteristic absorption maxima, one major peak at 330 and a secondary peak at 263 ni/x. The exciting wavelength for fluorescence corresponds to the adsorption peak at 330 m/x. The concentrated luciferin is slightly yellow in alkaline solu- tion, but it changes to a colorless solution in weak acid. In the former case the fluorescence upon ultraviolet activation is an intense yellow- green, whereas in the latter case the fluorescence changes to a pale 5.5 60 65 70 75 80 85 90 P H Fig 2. Effect of pH and buffers on firefly luminescence (McElroy and Strehler, 1949). red. The luciferin can be maintained for several weeks without appreciable loss of activity either frozen in the aqueous solution or in the dried state. In aqueous solution at pH 3.5 and 100° C complete inactivation occurs in 15 minutes, and there is approximately 50% loss of activity in 5 minutes. At pH 10 less than 5% inactivation occurs in 20 minutes at 80° C. The inactive luciferin can be removed from the active by extraction with ethyl acetate at pH 3.5. Under these condi- tions only the latter is removed from the aqueous phase. The physico- chemical properties of firefly luciferin are discussed in the following paper (Strehler and McElroy). w. D. Mcelroy and j. w. Hastings 167 Properties of the Luminescent System Effect of pH, Temperature, Activators, and Inhibitors. The lumi- nescent reaction has a rather sharp pH optimum at approximately 8.0 under conditions where none of the other diffusible cofactors are limiting. The results in Fig. 2 indicate that this optimum can be shifted by changing the ionic environment. In a sodium phosphate buffer at pH 7.5 the temperature optimum is approximately 25° C (see Fig. 3) with an experimental energy of activation of 18 kcal. 20 25 TEMP-C* Fig 3. Effect of temperature on firefly luminescence (McElroy and Strehler, 1949). Although several divalent ions will stimulate the luminescent reac- tion when crude extracts are used, only Mg and Mn will function in the purified system. In this its behavior seems similar to that of other phosphate transfer systems in which ATP participates. Calcium is a potent inhibitor of the light reaction, competing with the Mg++ ion. The results in Fig. 4 indicate that the flash height as well as the decay to the baseline level is affected by calcium. With a decreased rate of light intensity decay a greater total light output is obtained with 168 BIOCHEMISTRY OF FIREFLY LUMINESCENCE increasing concentrations of calcium ion. It is also significant that the response to inorganic triphosphate is similar to the initial response to ATP with and without calcium present. The results are in keeping with the proposed scheme for luminescence ( see below ) . The luminescent reaction is essentially insensitive to azide, cyanide, and fluoride although it is strongly inhibited by p-chloromercuroben- zoic acid. The latter inhibition can be reversed by glutathione. The results indicate a definite involvement of a — SH group for the lu- TIME-MINUTES CoCLX 10 Fig. 4. The eflFect of calcium on the hght intensity and the total Hght emitted by the purified luminescent system. Curve A, control; B and C contain a final concentration of CaCL of 4 X 10* M and 6 X 10"^ M respectively. Curve E relates to total light emitted when ATP is added initially, while D refers to the secondary response to inorganic triphosphate (McElroy and Coulombre, 1952). minescent reaction. Inhibitors of the light reaction, which are of particular interest with respect to the structure of luciferin, are those listed in Table II. All compete with firefly luciferin. Of the various substituted compounds 2-phenyl benzothiazole is the most effective. Although 5-methyl- and 5,6-dimethylbenzimidazole are effective com- petitive inhibitors neither crystalline vitamin B12 nor 1-a-D-ribofurano- sido-5,6-dimethylbenzimidazole influence light emission. Substitution in the one position of the parent compounds completely eliminates inhibitory effects, whereas substitution in the two position potentiates. w. D. Mcelroy and j. w. Hastings 169 The product of the oxidation of luciferin is, likewise, a potent in- hibitor of the Hght reaction. Preparations of oxidized kiciferin when added to reduced luciferin completely and apparently irreversibly inhibit light emission. The inhibitory effects can be prevented by add- ing crude protein preparations to the luciferin mixture before mixing with luciferase. ApparenUy the protein selectively adsorbs oxidized L TABLE II COMPOUND I BENZIMIOAZOLE 5-METHYL 5,6 - DIMETHYL |-=t-0-RIBOFURANOSIOO- 5,6 -DIMETHYL — 2 INDOLE 3 -METHYL 1,3-DIMETHYL ETHOX Y INDOLE ACETIC ACID TRYPTOPHANE 3. BENZOTHIAZOLE 2 -CHLORO — 2- AMINO 2 - PHENYL 2 - MERCAPTO 4 8ENZ0TRIAZ0LE 5 BENZOXAZOLE PARENT STRUCTURE /N^'^c 2 II 3n /V">c 2' II 3C /\/N CONC 50 PERCENT INHIBITION MOLAR 1 9 X 10 6 5 X lO'* 2 4 X 10"* NON- INHIBITOR 3 5 X 10 ^ 4 0 X 10'* NON- INHIBITOR NON - INHIBITOR NON - INHIBITOR NON - INHIBITOR /V°\ 26 X 10"* 4.7 X ID'S BOX 10'* 3 3 X 10'^ NON ■ INHIBITOR I 6 X 10 -3 67 X 10 -4 luciferin. Once the inhibition occurs, however, it cannot be reversed by adding the protein to the reaction mixture. Coenzyme A appears to be able to remove the oxidized luciferin from the luciferase since the addition of this cofactor after light initiation will stimulate luminescence (McElroy, unpubhshed). The secondary response to coenzyme A, under the conditions described, appears to be specific and has been used in the study of Co A synthesis. 170 BIOCHEMISTRY OF FIREFLY LUMINESCENCE Quantitative Response to ATP and Mg Concentration One of the interesting characteristics of the crude luminescent system was the rapidity with which ATP was apparently utilized. In crude extracts, the light emission was correlated with the disappear- ance of ATP and the appearance of inorganic phosphate. When the two equivalents of phosphate from ATP were liberated, the light emission of the preparation ceased. A second addition of ATP restored the luminescence (McElroy, 1947). On the other hand, when ATP was added to any of the partially purified preparations, there 120 i t tn 110 o '00 ^ 90i V 1 1 L CONTROL S BO 5 70 »- ? 60 ^s \ 1 • M^ SO4 I 0 Mn S04 ft. H 50 D 40 ^x \ V 30 — ^v^^ 20 - \ X^'^"""^'--^^ 10 1^ V " r> ^- T , 1 1 1 1 1 1 III" • r^-^ 5 6 7 8 9 10 TIME-MINUTES 11 12 13 14 15 Fig. 5. Light intensity-time relationships for firefly luminescence using partially purified preparations. Secondary addition of ATP at 5 and 12 minutes (McElroy and Strehler, 1949). was initially a very bright light response, which rapidly decreased to a basal level of approximately 10 to 15^ of the maximum (Fig. 5). The low residual luminescence has been observed to last over a period of more than 15 hours. The light intensity of this preparation may be increased by a second addition of ATP. The results definitely indicate that ATP concentration was the hmiting factor, as far as hght in- tensity was concerned. Contrary to observations in the crude prepara- tions, however, was the fact that no inorganic phosphate appeared as the ATP was apparently consumed. Several different experiments w. D. Mcelroy and j. w. Hastings 171 were performed in an effort to determine the fate of the labile phos- phate groups. Analysis of samples of the reaction mixture at various times after addition of ATP for labile phosphate estabhshed the fact that prac- tically all the labile phosphate groups were still present. Since hexokinase catalyzes the transfer of the terminal phosphate of ATP to glucose, it was possible to determine whether the labile phosphate was still available for this reaction. As shown in Table III, the addi- tion of hexokinase and glucose 5 minutes after ATP caused a rapid decrease in the labile phosphate level. In addition to the evidence cited earlier concerning the presence of myokinase, the results in Table III clearly establish the fact that the enzyme was present, since the phosphate groups of ADP were made available for the phosphoryla- tion of glucose upon the addition of hexokinase, TABLE III Effect of Hexokinase on Labile Phosphate in Luminescent System (McElroy, 1951b) (Values in micrograms per milliliter) Reaction A Reaction B Reaction C Reaction D 2 ml 1 ml V 2 ml 0.015 M ATP, 0.01 M ADP 0.015 M ATP, 1 ml 1 ml 2 ml 1 ml 0.1 Af glucose, 0.1 M glucose Time, 0.015 M ATP 0.1 M glucose 1 ml he.xokinase 1 ml hexokinase min. P AtP P A7P P A;P P A:P 0 8 116 11 126 11 122 5 36 15 10 118 15 78 6 6 30 12 110 10 123 14 45 5 4 45 10 113 14 38 60 11 HI 12 119 16 32 Fractionation of the luminescent reaction mixture after the addition of ATP with barium acetate showed that over 90% of the labile phos- phate was precipitated as the barium salt. This barium-precipitable labile phosphate could also enter into the hexokinase reaction. The results indicate that the ATP was still present in the reaction mixture, even though the light intensity was rapidly decreasing. 172 BIOCHEMISTRY OF FIREFLY LUMINESCENCE The piifified enzyme results in Fig. 6 demonstrate that the initial light intensity varies, both with the Mg++ and ATP concentration in a manner suggesting a complex between the two molecules. This is particularly noticeable with low Mg++ concentrations where it is possible to demonstrate an inhibition with high concentrations of ATP. For the intermediate range of concentrations the maximum initial light response is obtained with a ratio of ATP to Mg++ of approxi- 70 OOlMg CONC. 4 6 8 10 12 14 ATP CONC. MOLAR X 10* Fig. 6. The effect of varying ATP and Mg concentration on initial light intensity .(McEbroy et al, 1953). mately 1. In addition to the effect of Mg++ on the initial light intensity, there is also a noticeable influence on the decay and final baseline level. The results of such an experiment are presented in Table IV. In purified enzyme preparations, light emission cannot be elicited by a variety of other phosphorylated compounds which have been tested. The initiation of light emission in the crude extracts by ADP is due to the presence of an effective myokinase. Inosine triphosphate, uridine triphosphate, acetyl phosphate, creatine phosphate, inorganic pyrophosphate, and a variety of other phosphorylated and nonphos- phorylated cofactors fail to initiate light. The specificity of response \ of the firefly extracts has proved useful as a tool for ATP assay in various systems (Strehler and Totter, 1952; McElroy, unpublished). w. D. Mcelroy and j. w. Hastings TABLE IV Effect of Varying Mg++ Ion Concentration on Light Response to ATP (McElroy, 19ola) 173 MgS04 concentration, Initial maximum response Basal light intensity M X 105 to ATP, volts at 3 min., volts 0 8.4 4.7 2.0 10.5 3.8 3.5 12.5 3.2 5.0 15.0 2.9 10.0 18.0 3.0 25.0 21 3.0 50.0 25 2.8 100.0 26 2.9 Light Response to Luciferin Concentration In the presence of excess ATP and Mg++ ions, the intensity of light from the purified enzyme preparation depends upon the luciferin concentration. Figure 7 illustrates this relationship. The initial maxi- i/) o 3.0 > > 25 1- /^ (O 2.0 f z / UJ 1 H 15 - 1 z J 1.0 _ / »- d X 1 ^ 05 - _l ,1,1,1,1 - t 70 tn 2«° h- Z 50 — K ^ H- 40 _ V ^ *-^ ■— <>-^-^-_^ X \V k "* -0-- _ 2 30 - V ^ ^ _J w \ 20 - >^ ^ "^^:;:>-^ ,o|- -1— ^s, """"^"^^^^^^^^^^T"^ 0 L till 10 20 30 40 50 60 TIME- SECONDS Fig. 9. The effect of pyrophosphatase on the complexing reaction. Purified inor- ganic pyrophosphatase ( 10, 20, 40, and 80 micrograms protein ) was added to the reaction mixtures ( B, C, D, and E respectively) prior to initiating the reaction with ATP (McElroy et al, 1953). reaction has been observed to vary with different enzyme prepara- tions which suggests that a second factor may be involved. By further purification of the luciferase preparation it is possible to show that a second protein as well as Mg++ are essential for the rapid decay. The most effective protein for accelerating this decay reaction which is present in the firefly lanterns in high concentration is inorganic 176 BIOCHEMISTRY OF FIREFLY LUMINESCENCE pyrophosphatase (McElroy, Coulombre, and Hays, 1951). The results in Fig. 9 illustrate the effect of adding pyrophosphatase to a purified luciferase. As discussed below, the results indicate that luciferase reacts with the second protein to form an inactive complex which effectively removes most of the enzyme from active participation in the light reaction. Although there is some rapid complexing of the luciferase during the first few seconds, it is apparent that the rapid decay to the low light intensity is not observed unless pyrophosphatase is added. The rate of decay to the steady-state level of luminescence is proportional to the added pyrophosphatase. It is not clear why one obtains an initial rapid decrease of the light intensity to approximately 50% of the flash height even in the absence of pyrophosphatase. It may mean that a second luciferase molecule may t unction in the complex- ing reaction. Effect of Secondary Addition of Polyphosphates In preliminary experiments designed to determine whether pyro- phosphate would influence the utiHzation of ATP in the Hght reaction, it was observed that the addition of the former, after light production by ATP had decreased to a small value, stimulated light production in much the same way as additional ATP (Fig. 10). Since pyrophosphate by itself or pyrophosphate plus adenylic acid failed to initiate light production, it seemed likely that this compound was in some way making the ATP available for light production. A second luminescent response to pyrophosphate occurred in the presence of ADP as well as ATP. With ADP there was a slight lag in the initial light emission in contrast to the instantaneous response to ATP. Furthermore, the maximum light intensity obtained by the addition of pyrophosphate was greater than that obtained with the initial addition of ADP. With ATP the maximum response obtained with a secondary addition of pyrophosphate was approximately 50% of the initial intensity (Fig. 10). The response of the luminescent system to pyrophosphate was only temporary, as the results in Fig. 10 demonstrate. In many respects it simulates the effect of ATP. The rapid drop in Hght intensity in the experiments described above is due to a rapid hydrolysis of pyrophos- phate by inorganic pyrophosphatase. An analysis for orthophosphate W. D. McELROY AND J. W. HASTINGS 177 after the addition of pyrophosphate indicates the presence of a large amount of pyrophosphatase ( Fig. 10 ) . The pyrophosphatase has been partially purified by means of (NH4)oS04 fractionation and a number 24 — W 20 ' # 1- _l O > ' lA _ 32 _ >- '" K ^^ • "" (/> ] ^^ ■ ' I — x*^ Z JT \ / |i! .2 -^ \ ^^ -Ol / Z 1 CO / i .2 / 1- 1 « \ 1 '^ < -O O / -1 X \ / 4 r "^'^v.^.,,,*^ V 8 ^ / -V _ ' -^- 1— 35 - \x/M ^^---^"^^ en UJ 30 - ^J Mg t Mn Q — ^- 25 - ^v.^^^ 1 ^^ ■*■ '^^^^*^"^*'SL »- X 20 — Mg-^ r ^^-^ 0 1 \ _J 15 10 5 I \ 1 1 1 ' 1 1 12 3 4 5 6 7 Tl ME - MINUTES Fig. 19. The effect of inhibitors on the huninescent response to pyrophosphate (McEIroy et al, 1953). The pyrophosphate and inhibitors were added at 2 minutes. reaction is initiated with ATP. Likewise pyrophosphatase greatly affects the response of the light reaction to the secondary addition of inorganic pyrophosphate and triphosphate. The results in Fig. 19 illustrate one type of response which can be obtained by the secondary addition of pyrophosphate to a partially purified luciferase system which contains pyrophosphatase. Upon addi- tion of pyrophosphate (final cone. 10~* M), there is at first an imme- w. D. Mcelroy and j. w. Hastings 187 diate increase in the light intensity which reaches a maximum within a second. The intensity decreases during the next few seconds to a minimum, then begins to rise, and finally reaches a secondary maxi- mum, from which it rapidly decreases to the low baseline level. Ions that are known to inhibit pyrophosphatase activity greatly influence the response to added pyrophosphate. Mn++, in the concentration used, should inhibit the action of pyrophosphatase approximately 90%. The results in Fig. 19 indicate that the light intensity in the presence 6 8 10 12 14 TIME- MINUTES 18 20 Fig. 20. The effect of pyrophosphate on the luminescent response to ATP ( McEl- roy et ah, 1953). Sodium pyrophosphate was added initially to reaction mix- tures B, C, D, and E to give a final concentration of 5 X 10 *, 2 X 10 "^ 5 X 10 '^ and 10 * respectively. of Mn++ is maintained at a high steady-state level for considerable time. Intermediate concentrations of Ca++ and fluoride give similar effects. Pyrophosphate in low concentrations is a potent inhibitor of the luminescent reaction if added prior to the ATP. Apparently it can compete with the latter to form an inactive intermediate with the luciferin-lucif erase system. As the results in Fig. 20 indicate, however, this inhibition is slowly reversed if pyrophosphatase is present. The higher the pyrophosphate concentration, the longer is the time re- quired to reach the second peak of luminescence. The quantitative relationships between initial inhibition by various concentrations of 188 BIOCHEMISTRY OF FIREFLY LUMINESCENCE pyrophosphate and the time to reach the secondary peak are shown in Fig. 21. If inhibitory concentrations of pyrophosphate are incubated with the luciferase preparations containing pyrophosphatase for various time intervals prior to the addition of ATP, a similar recovery is observed, i.e., ATP is not required for this effect. The results of such an experiment are shown in Fig. 22. In curve A the reaction was started in the usual manner with ATP at zero time. After the steady- I 23456789 PYROPHOSPHATE CONC. M x 10^ Fig. 21. The relationship between pyrophosphate concentration and light emission (McEIroy et ah, 1953). The black circles represent initial light intensity while the white circles represent the time required to reach the secondary peak of luminescence as recorded in Fig. 20. state luminescence had been reached, an inhibitory concentration of pyrophosphate was added. The response is at first a depression fol- lowed by an increase in the light intensity, which reaches a maximum 2 minutes later. In the other experiments, ATP was not added ini- tially; however, the pyrophosphate was added at the same relative time after mixing enzyme, buffer, luciferin, and Mg+ + . At various intervals after the addition of pyrophosphate, the ATP was introduced to initiate light emission. The results clearly indicate that the longer the incubation time the greater is the initial effect of ATP, in so far as light intensity is concerned. One obtains the secondary peak of lu- minescence only at certain critical times of incubation. This diphasic w. D. Mcelroy and j. w. Hastings 189 response appears to be directly related to the amount of free and pyrophosphate-bound luciferase, the latter being released by the action of pyrophosphatase. The maximum and normal response to ATP is obtained when the incubation time is extended to a point where all the added pyrophosphate has been hydrolyzed. The phosphate anal- -300 -250 -200 TIME-MINUTES Fig. 22. The effect of delayed addition of ATP on the pyrophosphate response. In curve A, the reaction was started with ATP; 2 minutes later 0.5 ml of 0.01 M sodium pyrophosphate was added. In curves B, C, and D, the ATP addition was delayed until 2.5, 3.5, and 4 minutes respectively. Pyrophos- phate (0.5 ml of 0.01 M), however, was added at 2 minutes to all the reaction mixtures. Curve E represents the average inorganic phosphate con- centration in the reaction mixtures. ysis of the reaction mixture indicates that over 95% of the pyrophos- phate is decomposed at the time the secondary peak in luminescence has been reached. The phosphate analysis is also recorded in Fig. 22 and was approximately the same for all the reaction mixtures. It is apparent from this curve that no detectable phosphate is released 190 BIOCHEMISTRY OF FIREFLY LUMINESCENCE from ATP during the rapid complexing reaction, a point that has been estabhshed and discussed previously. If pyrophosphate is added to a highly purified enzyme mixture containing relatively little or no pyrophosphatase activity, the light intensity does not rapidly decrease from the initial flash height to the en 2 UJ t- X o TIME - MINUTES Fig. 23. The effect of delayed addition of pyrophosphatase. The luciferase used was free of pyrophosphatase activity. Pyrophosphate was added initially to give a final concentration of 5 X 10* M. The luminescent reaction was allowed to proceed for 5 minutes before the experiment recorded in the graph was started. The arrow indicates when two different concentrations of pyrophosphatase were added to the reaction mixture. To reaction mixtures B and A were added 0.1 and 0.2 ml of partially purified pyrophosphatase ( 1.0 mg protein/ml) respectively. The initial depression, as well as the steady- state baseline level, after hydrolysis of pyrophosphate, are proportional to the protein added. There was no phosphate liberated during the 6.5 minute"^ prior to the addition of pyrophosphatase. low baseline level. The results in Fig. 23 show that the luminescence is maintained at a high steady-state level. Under these conditions no phosphate is liberated, and the secondary rise in light emission is not observed. If, however, a partially purified pyrophosphatase is added, there is at first a depression which varies directly with the amount of protein added, followed by an increase in the light intensity. The rate w. D. Mcelroy and j. w. Hastings 191 of light-intensity increase, as well as the time required to reach the maximum intensity, depends upon the amount of pyrophosphatase added. The curves A and B of Fig. 23 illustrate this point. Phosphate liberation parallels the light-intensity curve in a manner similar to the results presented in Fig. 22. The initial depression of the light intensity, as well as the depression of the steady-state baseline level, by the addition of pyrophosphatase is additional evidence for the im- portance of a second protein in the complexing reaction. That pyro- phosphatase is the important protein for the complexing reaction is indicated by other evidence than was presented above. The ability of a preparation to complex the luciferase parallels the pyrophosphatase activity during purification of the latter. Factors such as temperature and a variety of inhibitors depress the complexing and pyrophospha- tase activity in a similar manner. Inorganic triphosphate is not broken down even by the partially purified luciferase preparations, and it might be expected that a differ- ent type of response would be observed with this compound. Several different experiments have supported this viewpoint. In none of these experiments has there been observed the secondary rise after the initial response to added inorganic triphosphate. There is, however, a flash which rapidly decays to a steady-state level. The level of this baseline luminescence is, however, higher with each successive addi- tion of triphosphate, in contrast to pyrophosphate response in the presence of pyrophosphatase. The results of one such experiment are recorded in Fig. 24. Curves A, B, and C represent three successive additions of triphosphate at 2-minute intervals. Although the initial flash response is progressively decreased, the steady-state basehne increases in proportion to the triphosphate concentration. Similar results with different concentrations of triphosphate have been re- ported previously. Curves 2, 3, and 4 show the effect of successive additions of pyrophosphate to a similar reaction mixture, while curve 1 represents a similar experiment, but with one-haff the amount of pyrophosphatase added. Although the preparation containing pyrophosphatase does not catalyze the hydrolysis of inorganic triphosphate, it influences the luminescent response to the compound. During purification of the lu- ciferase the maximum flash elicited by the secondary addition of tri- 192 BIOCHEMISTRY OF FIREFLY LUMINESCENCE phosphate decreases and may, as in the purest preparations, give no more than 15% of the flash obtained initially with ATP. Addition of the pyrophosphatase to these preparations accelerates the complexing reaction and raises the triphosphate response to as high as 85% of the initial flash. This evidence is additional support for the idea that a second protein is involved in the complexing of luciferase and that both triphosphate and pyrophosphate accelerate the breakdown of this inactive complex. The increase of the baseline level of lumines- 80- 05 I 0 0 05 10 TIME - MINUTES 15 2 0 Fig. 24. Light emission with successive additions of triphosphate and pyro- phosphate in tlie presence of pyrophosphatase. In the reaction on the right curves 1,2,3,4 show the effect of successive additions of pyrophosphate. After the first pyrophosphate addition ( 1 ) the pyrophosphatase concentration was doubled before additional pyrophosphate was added. Curves A, B, C show tlie effect of successive additions of triphosphate. The initial response to ATP was 90 light units. cence with triphosphate is taken as evidence for the complexing of the protein with triphosphate, thus shifting tlie equilibrium between active and inactive intermediate. Additional protein in this case will restore the original low baseline level of luminescence. The fact that the triphosphate response cannot be completely eliminated may be due to the fact, as mentioned above, that luciferase itself may serve as the second protein in the complexing reaction. The results summarized above demonstrate that the rapid decrease of light intensity after ATP addition to firefly extracts is due to the w. D. Mcelroy and j. w. Hastings 193 reversible complexing of the enzyme luciferase. The evidence suggests that an active intermediate composed of luciferin, luciferase, Mg++, and ATP is first formed, which, in the presence of oxygen, will react to give rise to an excited state that subsequently decomposes to emit a quantum of light. Most of the active intermediate is normally con- verted into an inactive complex which effectively immobihzes the enzyme. Except for oxygen, all the components necessary for light emission are also essential for the complexing process. The formation of the inactive complex is accelerated by Mg++ and by additional luciferase or other proteins, particularly inorganic pyrophosphatase. The level of the low light intensity appears to be a measure of the steady-state equihbrium between active intermediate and inactive complex. Removal of the inorganic pyrophosphatase by various means affects both the basal light intensity level, as well as the rate of decay. The relationship between Mg++ and ATP for maximum lumines- cent activity (Mg++ to ATP ratio of approximately 1) suggests that the true substrate for the formation of the active intermediate with the luciferin-luciferase is a specific Mg-ATP complex. Similar sug- gestions have been made by Hers for liver fructokinase and by Kielley and Kielley for mitochondrial adenosinetriphosphatase (ATPase). The relationships between Mg, inorganic pyrophosphatase, and active intermediate for the formation of the inactive complex indicate a similar type of reaction. The action of inorganic pyrophosphate and triphosphate in stimulating light production, after its initiation by ATP, is attributed to the rapid breakdown of the inactive complex by these agents. In the case of pyrophosphate the evidence shows that it competes with ATP in the formation of the active intermediate. Thus the addition of pyrophosphate before ATP strongly inhibits light emission. The duration and extent of this inhibition depends, however, on the concentration of both pyrophosphate and pyrophosphatase. On the other hand, the delayed addition of pyrophosphate leads to an initial stimulation followed, in many cases, by a decline and then a rise to a secondary peak which rapidly decreases to the baseline level as the pyrophosphate is hydrolyzed by the action of pyrophosphatase. Such results are to be expected if pyrophosphate splits the inactive complex to form initially some active intermediate. The amount of active intermediate formed would depend upon both the nature of the 194 BIOCHEMISTRY OF FIREFLY LUMINESCENCE inactive complex or complexes and the mode of the splitting by pyro- phosphate. The simplest inactive complex would presumably be formed of luciferin-luciferase-Mg-ATP-Mg-pyrophosphatase. Pyrophos- phate could split such a complex to give rise to both active and in- active intermediates containing luciferase. The initial light intensity obtained with the addition of pyrophosphate would then be a meas- ure of the active intermediate formed. The secondary peak of lumin- escence, after the addition of pyrophosphate, would represent the slow release of luciferase from an inhibitory complex with pyrophosphate. The schematic relationships are shown in Fig. 25. LHg+E + Mg *• ATP 1 1 LHg-E'Mg-ATP (ACTIVE INTERMEDIATE) » + 0, » LIGHT Mg Pr P; +Mg-Pr t LHj-E-Mg-ATP + PQP-Mg-Pr LH,-E-Mg-ATP'Mg-Pr -f p-q-p °' (INACTIVE COMPLEX) « L Hj- E' Mg- POP + ATP- Mg- Pr [INHieiTORvl COMPLEX CUNIKUL -■ MECHANISM ACETYLCHOLINE CHOLINE + AC ♦ CoA ♦ POP + Ac'Co A + Ad. Fig. 25. Scheme for the coniplexing reaction and the function of pyrophosphatase and pyrophosphate in firefly luminescence. The effect of various pyrophosphatase inhibitors, such as Mn, Ca, and F, on the luminescent response to pyrophosphate can be ex- plained on such a hypothesis. The quantitative relationships between pyrophosphate, pyrophosphatase, luciferase, and Mg++ presented support this interpretation. The results also indicate that triphosphate has an action similar to that proposed for pyrophosphate. However, since triphosphate is not hydrolyzed by the enzyme preparations, the steady-state luminescence remains higher after the addition of this compound. The results suggest that the equilibrium between active intermediate and inactive complex is shifted in favor of the former when triphosphate is added. Since additional pyrophosphatase can w. D. Mcelroy and j. w. Hastings 195 restore the original steady-state level, it appears that the triphosphate is effectively removing at least some of the protein essential for the complexing reaction. Control of Firefly Luminescence The existence of an inactive complex as proposed above wherein all the components of the system are present without appreciable reaction occurring may be of some significance in considering pos- sible mechanisms for controlling enzyme-catalyzed reactions. The evi- dence presented suggests that such an inactive complex may be im- portant in controlling the flash of the firefly. Pyrophosphate, which is known to be liberated in several reactions can serve as a trigger in the present reaction after the decay of the initial response to ATP. The pyrophosphatase which is present in high concentrations in the firefly lantern would rapidly decompose the released pyrophosphate, which allows the reformation of the inactive complex leading to the extinction of the light. The scheme in Fig. 25 illustrates one possible way in which nervous stimulation could lead to the rapid liberation of inorganic pyrophos- phate. Since the flash of the firefly is under nervous control this would appear to be a plausible mechanism. It is suggestive that the firefly lanterns contain a very high concentration of coenzyme A. One cannot fail to be impressed, however, with the striking re- semblance between the in vitro anaerobic flash and the flash exhibited by the live firefly. As mentioned previously, we believe that anaerobic conditions allow the accumulation of an active intermediate from an inactive one which is rapidly oxidized by molecular oxygen when air is readmitted to give an excited molecule which emits light. Most workers have felt that the firefly regulates flashing by liberating oxy- gen into the photogenic cells. ( See Harvey, 1952, for a review of this literature.) Implied in this statement and in keeping with the fact that "resting" fireflies are not emitting light, is that the photogenic cells are under anaerobic condition. This seems most unusual for insects. Buck (personal communication, see later) has suggested that the firefly must do work in order to remain dark and that only a brief anaerobic period may be essential to bring about the flash response. This suggestion is in keeping with all the biochemical data on the 196 BIOCHEMISTRY OF FIREFLY LUMINESCENCE light reaction. One difficulty remains, however, and that is that even with an excess of inorganic pyrophosphatase the light is never en- tirely extinguished in the in vitro reactions. It is not unlikely, how- ever, that in the intact firefly the low concentration of active inter- mediate can be rapidly oxidized by components other than molecular oxygen. A brief period of anaerobic conditions may allow the accu- mulation of an active intermediate to sufficient concentration that it can react significantly with molecular oxygen to give light. Further work on the physiology and biochemistry of the flash reaction will be necessary before a decision can be reached with respect to these various hypotheses. Mechanism of Luminescence — Function of ATP The response of crude extracts of numerous species of fireflies to ATP has now been tested (McElroy and Harvey, 1951). All have given a positive response. In some cases it is necessary to supplement even the crude extracts with additional luciferin before significant amounts of light can be obtained with ATP. It would appear from these experiments that the luciferins of the various firefly species are identical. Additional experimental work is necessary to establish this point definitely. Numerous other luminous forms have been tested for their ability to respond to ATP and thus far positive results have been obtained only with the lympyrid and elaterid beetles (Haneda and Harvey, 1954). The significance of this fact is not clear, but it may be related directly to the mechanism of control of the light emit- ting reaction. In Cijpridina, for example, the luminescent components are liberated from separate glands into the surrounding sea water where they mix, react, and emit light. In other words, the luciferin must be chemically prepared to react immediately without previous dark reactions. This instability in the luciferin molecule is apparent when it is isolated from the dried Cypridina. In the presence of oxy- gen but in the absence of luciferase it rapidly oxidizes without light emission. This is in conti-ast to firefly luciferin which is relatively stable when purified. Apparently certain dark reactions are essential before it can be oxidized, even in the presence of luciferase. Adeno- sine triphosphate is essential for this activation process. It now seems likely that firefly luciferin and ATP in the presence of luciferase and w. D. Mcelroy and j. w. Hastings 197 Mg form an active intermediate, possibly a nucleotide, which can now undergo rapid oxidation or peroxidation to form an excited mole- cule. Whether the ATP contributes, along with the oxidation, activa- tion energy for light emission cannot be answered at present because of the extremely small amounts of the products formed during the reaction. One point which seems clear, both for Cypridina and fire- flies, is that the luciferins are irreversibly utilized during the light reaction. Earlier studies on the preparation of crude luciferases from dried and butyl alcohol-treated fireflies suggested the formation of an active intermediate. Some of these preparations would emit light when ground with water but failed to respond when ATP was added. It is possible that this represents an active derivative of luciferin which was preserved under these conditions. Further work will be necessary before a definite conclusion can be made with respect to the function of ATP in the light reaction. References Dubois, R. 1885. Note sur la physiologie des pyrophores. Compt. rend. (Ser. 8), 2, 559-62. Haneda, Y., and E. N. Harvey. 1954. Additional data on the adenosine triphosphate and the luciferin-luciferase reactions of various luminous organisms. Arch. Biochem. and Biophys., 48, 237-38. Harvey, E. N. 1952. Bioluminescence, Academic Press, New York. Hastings, J. W., W. D. McElroy, and J. Coulombre. 1953. The effect of oxygen upon the immobilization reaction in firefly luminescence. /. Cellu- lar and Comp. Physiol., 42, 137-50. McElroy, W. D. 1947. The energy source for bioluminescence in an isolated system. Proc. Natl. Acad. Sci. U. S., 33, 342-45. McElroy, W. D. 1951a. Properties of the reaction utilizing adenosine triphosphate for bioluminescence. /. Biol. Chem., 191, 547-57. McElroy, W. D. 1951b. Phosphate bond energy and bioluminescence. In Phosphorus Metabolism, Vol. I, The Johns Hopkins Press, Baltimore. McElroy, W. D., J. Coulombre, and R. Hays. 1951. Properties of firefly pyrophosphatase. Arch. Biochem. and Biophys., 32, 207-15. McElroy, W. D., and J. Coulombre. 1952. The immobilization of adenosine triphosphate in the bioluminescent reaction. }. Cellular and Comp. Physiol, 39, 475-86. 198 BIOCHEMISTRY OF FIREFLY LUMINESCENCE McElroy, W. D., and E. N. Harvey. 1951. DifiFerence among species in the response of firefly extracts to adenosine triphosphate. /. Cellular and Comp. Physiol., 37, 1-7. McElroy, W. D., J. W. Hastings, J. Coulombre, and V. Sonnenfeld. 1953. The mechanism of action of pyrophosphate in firefly kiminescence. Arch. Biochem. and Biophxjs., 46, 399-416. McEhoy, W. D., and C. S. Rainwater. 1948. Spectral energy distribution of the hght emitted by firefly extracts. /. Cellular and Comp. Physiol., 32, 421-25. McElroy, W. D., and B. L. Strehler. 1949. Factors influencing the response of the bioluminescent reaction to adenosine triphosphate. Arch. Biochem. and Biophijs., 22, 420-33. Strehler, B. L., and J. R. Totter. 1952. Firefly luminescence in the study of energy transfer mechanisms. I. Substrate and enzyme determination. Arch. Biochem. and Biophys., 40, 28. Firefly Luciferin Bernard L. Strehler Department of Biochemistry (Fels Fund) University of Chicago, Chicago, IlHnois In the course of the early work on luminescence of firefly extracts (McElroy and Strehler, 1949) one of the factors which was demon- strated to be necessary for hght production, in addition to ATP, mag- nesium ion, and oxygen, was a compound which we termed "firefly luciferin." Subsequently, some of the properties of this compound were described (Strehler and McElroy, 1949). Further physical and chemical studies were undertaken independently by Dr. McElroy at the Johns Hopkins University and by the author at the Oak Ridge National Laboratory. The results presented here represent work soon to be published jointly with Dr. McElroy (Strehler and McElroy, 1954), and Mr. John Sites (Strehler and Sites, 1954) of the stable isotopes division, Oak Ridge. The luciferin used in these experiments was isolated by two dif- ferent procedures, one involving partition chromatography on a Celite column with water as a stationary phase and butanol-chloroform as the moving phase. The other involved chromatography on a Dowex 50 column and adsorption and elution from a fuller's earth column. The materials prepared in these two ways appeared to be identical in both the ultraviolet and infrared regions of the spectrum. Elemen- tary analysis (qualitative) indicated the presence of sulfur and nitro- gen and the absence of phosphorus. The material is soluble in polar organic solvents and water at neutral pH and partitions into ether from strongly acid aqueous solution. 199 200 FIREFLY LUCIFERIN Physical and Chemical Properties The absorption spectrum of luciferin in solutions of various pH's in the ultraviolet visible region of the spectrum appears in Fig. 1. The shift in absorption was studied as a function of pH as was the fluores- cence intensity. The change in fluorescence with pH is illustrated in Fig. 2. From both these measurements it appears that luciferin has a 220 240 260 280 300 320 340 360 38 0 400 Fig. 1. Ultraviolet absorption spectrum of firefly luciferin at pH 1 and pH 13 in water solution (0.1 M HCl and 0.1 M NaOH). pKa in the neighborhood of 8.5-8.6 (perhaps due to an imine group- ing). The infrared absorption spectrum (Fig. 3) of a dry sample of the purified luciferin was determined with a double-beam infrared spectrophotometer (Perkin-Elmer) through the courtesy of Dr. Cam- eron of the K-25 laboratories in Oak Ridge. Many materials which were considered possible relatives of luciferin on the basis of some of its other physical properties were also examined. Among them were various pteridines, pyrimidines, purines, nucleosides, and nucleotides, as well as riboflavin. Only the riboflavin and luciferin spectra exhib- ited any marked resemblance from the 2- to 15-micron region. Of particular note is the striking similarity in the absorptions of the two compounds in the 10- to 15-micron region, some bands being BERNARD L. STREHLER 201 nearly identical, others being slightly displaced from each other (see Fig. 3). These similarities would seem to indicate that luciferin and riboflavin have a predominantly similar general architectural design and that they differ considerably in the nature of the substituents on this framework. Fig. 2. Fluorescence intensity as a function of pH. The fluorescence was excited with the 365-millimicron Hg hne. The pH was varied gradually and pH and fluorescence measured simultaneously. The electrophoretic mobility of luciferin as a function of pH was determined, and the results are indicated in Fig. 4. Here a pKa in the region of pH 3-4 and pH 8-9 are apparent. The more acidic dissociation is probably due to a carboxyl group. In Fig. 5 is illustrated the polarographic half-wave of luciferin. The £o suggested by this half-wave is quite far on the reducing side of the hydrogen zero, al- 202 FIREFLY LUCIFERIN Fig. sNM V'^ '^--^ 235 I II I I 208 II I I. '95 ^^ c-M/ c c !t -' '■" I I ■'' 179 H N I I ■'" II I ,52 V ,/*\/^\ 91 B I c c Fig. 9. Proposed structure of firefly luciferin and fragments arising from it through pyrolysis. BERNARD L. STREHLER 207 of the degree of unsaturation on an empirical basis. An attempt was then made to derive structural formulas consistent with the remain- ing empirical formulas, taking into account the other physical and chemical evidence available. Although many different combinations and permutations were attempted, only those arising from the dipyri- midopyrazine nucleus could be fitted into the restrictions of the empirical formulas. Subsequently, we synthesized several dipyrimido- pyrazines and found, on the part of several of them, physical prop- erties (such as fluorescence and absorption spectra) strongly remi- niscent of luciferin, although the precise structure postulated on the basis of the mass cracking pattern was not capable of synthesis by the methods we employed. Figure 9 shows a structure for luciferin consistent with the above data and a proposed scheme of pyrolytic degradation to account for the major mass fragments observed. Note that this cleavage involves the breaking only of single bonds. It should be emphasized that this structure is perhaps not the only one which could be derived to fit the many properties observed. It is, however, the only one we have been able to formulate which is consistent with all the data available. Note that the calculated masses are always one mass unit higher than the observed masses. If the degradation product is a dimer through the sulfur grouping (disul- fide) the rupture of the S — S bond would give an observed mass one less than that calculated. ACKNOWLEDGMENTS The author wishes to express his appreciation to Dr. Russell Baldock of the Oak Ridge National Laboratory for his generous contribution of time, equipment, and assistance in certain phases of this work, to Mr. John Sites for his collaboration in certain phases, and to Miss E. Brigham and Mr. B. Bottoms for their expert assistance in portions of this work. References McElroy, W. D., and B. L. Strehler. 1949. Factors affecting the response of the luminescent system to ATP. Arch. Biochem., 22, 420-33. 208 FIREFLY LUCIFERIN Strehler, B. L., and W. D. McElroy. 1949. Purification of firefly luciferin. /. Cellular Comp. Physiol, 34, 457-66. Strehler, B. L., and W. D. McElroy. 1954. Further physical and chemical studies on firefly luciferin (in press). Strehler, B. L., and J. R. Sites. 1954. The application of mass spectrographic pyrolysis to luminescent molecules: Riboflavin and firefly luciferin {in press). Factors and Biochemistry of Bacterial Luminescence Bernard L. Strehler" Department of Biochemistry and Institute for Research in Biophysics, University of Chicago, Chicago, lUinois The blue-green glow emitted by various species of luminous bac- teria has been a recurrent object of curiosity among biologists, chem- ists, and physicists for several scores of years. For workers in each of these fields, the phenomenon of "cold light" emission poses a special set of problems. The biologist is more interested in the evolution, selective advantage, and relation of light production to other functions of organisms; the chemist concerns himself with the mechanism of excitation and the chemical identity of the reacting molecules; while the physicist is primarily interested in the energetics and kinetics of the process and the effect of well-defined environmental variables on light emission. This discussion is an attempt to assess present knowl- edge of bacterial luminescence touching on all three of these fields. Of necessity the main emphasis here will be placed on the newly available information on the biochemistry of the process, both because this is the least alien to the writer and because it furnishes a starting point for discussion of the other aspects of the problem. A prodigious amount of work has been expended in studying the effect of various factors on in vivo bacterial luminescence, and a co- gent summary and analysis of these works is presented in Harvey's Bioluminescence (1952). Among the more instructive earlier findings, mainly on salt water species, bearing on this present discussion are the following: " Fels Fund. 209 210 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE 1. Bacterial luminescence is a respiratory phenomenon that has an absolute requirement for Oo, although luminescence is less sensitive to low [O2] than respiration (Eymers and van Schouwenburg, 1937; Shoup, 1929). 2. Tliis "light respiration" is essentially cyanide insensitive ( Harvey, 1920), although various organic compounds, particularly naphtho- quinones, are strongly inhibitory to it (Spruit and Schuiling, 1945; McElroy and Kipnis, 1947). 3. Ultraviolet light inhibits luminescence and shows a discrete "in- activation spectrum" (Gerretsen, 1915). 4. The light emitted is blue-green in color, showing a band with a maximum at ca. 500 millimicrons for a number of species investigated (Spruit-van der Burg, 1950). 5. The yield of luminescence/02 consumed is ca. 1/100 to 1/1000 (van Schouwenburg and Eymers, 1936). 6. Luminescence shows a temperature dependence similar to that of many other respiratory processes (Johnson et al., 1942). 7. Pressure-temperature studies indicated that the luminescent sys- tem behaves as a typical protein enzyme ( Johnson, 1947 ) . 8. Until recently, attempts to extract the system and demonstrate a luciferin-luciferase reaction in vitro have been unsuccessful or not capable of confirmation ( Harvey, 1952; Gerretsen, 1920; Korr, 1935 ) . Our success in obtaining brightly luminous extracts (Strehler 1953a), since confirmed in other laboratories (McElroy et al., 1953), must be ascribed to the superior light detecting equipment we em- ployed which enabled us to follow very dim luminescences (Strehler, 1951), quantitatively and for considerable periods, and to the fact that high purity biochemical reagents and intermediates are now available cheaply and in quantity. Thus, the quantum counter of nearly ultimate sensitivity and low noise level made it possible to measure the dim luminescence ex- hibited by acetorized extracts of A. fischeri while available supplies of DPN, FMN, etc. made it possible to study the effects of these com- pounds easily and rapidly. The rapidity with which developments have been forthcoming, the general good fortune which attended critical phases of this work, and the complexity of some of the results have made it difficult to keep interpretation abreast of experi- BERNARD L. STREHLER 211 ment. It is hoped that discussion at this conference will deal critically with those aspects of the work which we have handled unsatisfac- torily because of limitations of time and background. Extraction of Luminescent System It had been noted repeatedly (Harvey, 1952; Korr, 1935) that the luminescent powders obtained by a variety of quick-drying methods will emit light when they are suspended in water. Similarly, in un- published experiments performed in early 1951, Dr. Charles S. Shoup and I were able to obtain light for a few minutes when we added acetonized bacterial powders to water. Addition of boiled extracts did not result in measurable effects once the luminescence had disap- peared. However, with the possibility in mind that the oxidant rather than 'luciferin" might be limiting, we added hydrogen peroxide to the dark extracts and obtained considerable light. Further studies indicated that one of the compounds responsible for this chemilumi- nescence in the presence of peroxide is a flavin (Strehler and Shoup, 1953). Partly because of the press of other work and partly because of our skepticism of the eventual success of further work, these powders re- mained for nearly two years in a deep freeze before we again attempted to extract the luminescent system. A systematic study was again undertaken in the fall of 1952 with almost immediate success (Strehler, 1953; Strehler and Cormier, 1953). The crucial finding from our point of view was the fact that the duration of luminescence and its restoration by added agents depended on the concentration of powder used. Low concentrations were incapable of luminescing for longer periods or of responding to added biochemical reagents, while a tenfold increase in the ratio of powder to water sustained a con- tinual luminescence and exhibited a "luciferin-luciferase" reaction. This effect is illustrated in Fig. 1 while Fig. 2 illustrates the effect of dilu- tion on the system. Nature of Diffusible Requirements DPNHo and FMN Once a sustained and renewable luminescence was attainable a wide variety of shelf biochemical reagents were tested, including 212 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE DPN, TPN, DPNHo, riboflavin, FMN, FAD, Co A, lipoic acid, thia- min pyrophosphate, ATP, ADP, AMP, glucose-6-P, glucose, fructose, phosphoglyceric acid, a-ketogkitarate, acetate, acetyl phosphate, cit- rate, malate, succinate, lactate, ethanol, acetaldehyde, vitamin A, II 13 15 17 19 21 23 25 TIME(MIN.) Fig. 1. Time course of luminescence emitted by bacterial extracts. A, 2 mg/ml acetonized bacterial powder added at zero time; B, 20 mg/ml acetonized bacterial powder added at zero time; C, 20 fig/ml DPNH; added at zero time; D, 16 Mg/ml DPN added at zero time. Water solution, temperature, 23° C. pyridoxine, ascorbic acid, glutathione, cysteine, glycine, versene, vita- min K, 2-methyl-l, 4-naphthoquinone, vitamin E, benzoquinone, hy- poxanthine, xanthine, vitamin D, numerous steroid hormones, hemo- globins, a variety of vitamin-rich concentrates, yeast extract, Mg+ + , Fe++, Mn++, Pb++, Zn++, Co,+ +, etc., and cytochrome c. Most of these were tested not only alone, but in combination with other com- ponents. Of the compounds examined only the following showed marked effects: DPN, DPNH2, riboflavin, FAD, FMN, naphthoquinone, Co A, BERNARD L. STREHLER iiO-i ^ 213 90- o LlI O o >• if) 70- z 50- I o 30- .10- 5 ! — T" 2x 4x -T 8x I6x 32x DILUTION FACTOR Fig. 2. Effect of dilution of bacterial enzyme extract on luminescence. Each tube contains: 600 /xg DPNHo, 0.2 ml phosphate buffer, 0.01 M, pH 7.0; 0.3 ml of stock extract or of diluted extract (diluted just before measure- ment). Total volume, 0.6 ml. Stock extract: 13 g acetonized powder in 100 ml of water, centrifuged 1 hour at 0° C on Serval centrifuge at 12,000 rpm. Gross luminescence was multiplied by dilution factor to obtain luminescence per unit weight of enzyme. malic acid, thiamin pyrophosphate, and Mg+ + . Riboflavin, naphtho- quinone, and FAD were "powerful inhibitors of luminescence, the Others either initiating or increasing light production under various conditions. In the presence of DPNH2, only, FMN, of the defined compounds, had a potentiating effect (Strehler and Cormier, 1953). DPN or 214 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE DPNH2 were invariably necessary for luminescence (see Fig. 3). All these studies suggested that the luminescent system is a DPN-coupled respiratory pathway, while the two- to threefold increase obtainable with FMN in the presence of excess DPNH2 suggested that the oxi- dase was coupled to oxygen via flavin. That such is indeed the case was clearly demonstrated by the work of McElroy, Hastings, and CONCN OF DPNHj (moles « 10" ) Fig. 3. Effect of DPNHo concentration on bacterial extract luminescence: 36 fig of palmitic aldehyde plus 4 /j.g of FMN were added to 0.3 ml of an aqueous extract of acetonized powders (4.0 g/100 ml) dissolved in 0.1 M phosphate buffer (pH 7.0); total volume, 3.0 ml. DPNH2 added as indicated. BERNARD L. STREHLER 215 co-workers (1953), who were able, through combined acid precipi- tation and ultraviolet treatment, to obtain a flavin-free system which exhibited luminescence only in the presence of added flavin mono- nucleotide. The effect of flavin is illustrated in Fig. 4. These workers also reported the preliminary finding of a component in the extracts which they called bacterial luciferin. As is the case with firefly lucif- erin, this component was destroyed during luminescence. -I- 3 5 CONCN. OF FMN ( moles x lO"^) Fig. 4. Effect of FMN concentration on bacterial extract luminescence. To 0.6 ml oi a 27c (2 g acetonized powder/ 100 H2O, resolved by precipitation at pH 4-4.5), aqueous extract of A. jischeri in 0.1 M phosphate buffer (pH 7.0) was added 500 /xg of DPNH2 and 36 Mg of pabnitic aldehyde. Total vol., 2 ml FMN added as indicated. 216 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE KCF (Long-Chain Aldehydes) We had also noted an increased luminescence when a boiled bac- terial extract was added to a system containing excess DPNHo and FMN. (The effect of this factor is shown in Fig. 5.) However, we were never able to separate this material from the heat precipitable COMCM (molesilO"^) 12 20 28 36 MOLARITY « lO'' Fig. 5. Effect of aldehyde concentration on A. fisclieri extract luminescence. 0.1 ml of a 2% aqueous extract of A. fischeri was mixed with 1.5 ml of 0.1 M phosphate buffer, pH 7, plus 1000 ^g of DPNH= plus 4 ixg of FMN, and light intensity was determined. Palmitic aldehyde was then added in concentration indicated and light intensity was measured immediately after each addition. fractions, although a variety of procedures was employed. Since it seemed possible that some nonspecific effect of the boiled extracts might be involved, we used another source of crude protein, in this case kidney cortex powders, and obtained a large increase in lumines- cence both from the residue and from the supernatant of this boiled tissue powder (Cormier and Strehler, 1953). We called this com- ponent the kidney cortex factor or KCF. A series of attempts were then made to purify this component by various chemical and physical methods (Strehler and Cormier, 1954). BERNARD L. STREHLER 217 It was clear that the compound was lipoidal in nature and that it could be purified by partition between organic solvents. Finally, a procedure yielding a material of considerable purity was devised. It consisted essentially of the following steps: (1) extraction of ace- tonized powders with chloroform, (2) partition between hexane and 5% HoO-95% methanol, (3) precipitation of impurities with acetone, (4) precipitation of impurities from small volumes of hexane, chloro- form, and methanol, and finally (5) suspension in IN NaOH and pre- cipitation with HCl. The material at this point was a yellowish-white sludge, sparingly soluble in water, but relatively more soluble in strong base and organic solvents. Its activity was such that a small fraction of a microgram per milliliter produced a five- to tenfold increase in luminescence. TABLE I Comparison of Apparent Dissociation Constants and Maximum Rates of Luminescence in the Presence of Various Long-Chain Aldehydes Aldehyde K^ X 10« M Max. Rate (relative) Ct 84.0 28 Cs 20.8 67 C9 13.3 59 Cio 4.1 133 Cii 2.36 52 C16 3.56 81 A variety of qualitative organic tests was applied in order to elimi- nate certain groups of lipids as possibilities. The nitrogen content was very low (ca. 1-2%), as was phosphorus, while a fuchsin aldehyde test was strongly positive. Treatment of the material with 2,4-dinitro- phenylhydrazine gave a good yield of the dinitrophenylhydrazone in crystalline form. Attempts to decompose this derivative and recover activity were unsuccessful. The recrystallized dinitrophenylhydrazone exhibited a sharp melting point at 104-105° C. Since this melting point is in the region ascribed to the derivatives of long-chain fatty alde- hydes, we obtained some aldehydes from the stockroom and, although benzaldehyde, acrylaldehyde, and butyraldehyde were not active, syn- thetic heptaldehyde produced a marked increase in luminescence, one 218 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE thirtieth as great on a weight basis as KCF. Finally, KCF was iden- tified as the Ci6 (palmitic) aldehyde by elementary analysis, mixed melting points of the derivative, and molecular weight determination. All aliphatic aldehydes from C7 to Cie are active although the lower homologs are relatively less efiFective. See Table I. General Properties of System Color of Light The spectral distribution of the luminescence of extracts and intact bacteria were compared and found, within experimental error, to be identical (see Fig. 6). This finding along with the other parallelisms between in vivo and in vitro luminescence makes it likely that the 500- < q: o 3 o o 300- I o 100- 400 700 500 600 WAVE LENGTH ( m^ ) Fig. 6. Emission spectra of intact A. fischeri and extracts obtained from A. fischeri. 0.5-mm slit width Farrand quartz monochromator. # Emission of bacteria. O Emission of extracts. A Extract emission normalized to bac- terial emission at 490 ni/i. The gross reading was not corrected for changes in dispersion of monochromator or sensitivity of photomultiplier. BERNARD L. STREHLER 219 same reactions are occurring under both conditions. Although it seems probable that flavin in association with luciferase and perhaps long- chain aldehyde, is the light-emitting complex, the considerable dif- ference between the emission maxima exhibited by riboflavin fluores- cence and chemiluminescence on the one hand, and the emission of the bacterial extracts on tlie other, is a puzzling and possibly crucial point at issue. At least four solutions to this diflRculty may. be suggested. The first is simply that the flavin is not the emitting molecule, even though the biochemical evidence is fairly convincing that flavin is directly con- nected with light emission. The second possibility is that the broad flavin fluorescent and chemiluminescent emission is made up of tran- sitions from two different excited states and that only the more ener- getic of these is formed during the enzymatically catalyzed chemi- luminescence. The third alternative is that the binding of the flavin to the enzyme hinders the occurrence of certain vibrational modes, thus largely preventing transitions from the excited state, to higher vibrational levels of the ground state and consequentl}' shifting the emission closer to the absorption band. The final possible solution to the difficulty is that the emission in the longer region of the spectrum is quenched by a pigment in the bacteria. Sir.ce there is no evidence that such compounds occur in appreciable amounts in luminous bac- teria, the process would of necessity involve a mechanism analogous to sensitized fluorescence, with the condition that there is a \ery low yield of fluorescence from the absorbing entity. Nature of Enzyme — General Properties The enzyme will not pass through a dialysis membrane, is destroyed by heat, and is nonparticulate (as evidenced by its lack of precipita- tion under ultracentrifugation ) . It is sensitive to dilution, inhibited by sulfhydryl inhibitors (Hg++ and p-chloromercuribenzoate), and can be frozen and thawed repeatedly without destroying its activity'. It can be separated from some impurities by precipitation with acid, acetone, and (NH4)oS04. Dr. Green in Dr. McElroy's laboratory is presently engaged in preparing this enzyme in purified form. Whether or not the various activities exhibited by this enzymatic extract in the luminescent sequence of reactions are due to a single component or 220 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE to a multiplicity of factors must await its isolation for a definitive answer. Effect of Physical and Chemical Environment The pH dependence of the reaction is illustrated in Fig. 7. The double optima may be due to the semipurified nature of the system, 40-1 in O in o 20- z »- z I o r 6 7 pH 8 9 Fig. 7. Effect of pH on dialyzed bacterial extract luminescence: Extract dia- lyzed for 15 hours against distilled water at 0° C. Each vessel contained 50 fig DPNHo, 0.2 ml enzyme, 0.2 ml NaHo POi (0.01 M) titrated to desired pH with 1 N NaOH; total volume 0.5 ml; temperature 23° C. but are readily duplicated with crude acetonized extracts in the presence of added DPNHo. Temperature profoundly affects the rate of the luminescent reac- tion in vitro and Fig. 8 illustrates a typical result with the crude ex- tract. Particular care must be exercised in making such measurements, since the instantaneous effect of temperature is somewhat different from its delayed effects, even in the lower temperature range. Per- haps some of this effect is due to different pool sizes of intermediates if differential effects are not measured rapidly. The extremely high apparent activation energies (ca. 25 to 31 kcal) observed cast some doubt on the meaning of this measurement, the presence of some arti- BERNARD L. STREHLER 221 90-1 CO in o > in z X o 25 34 TEMPERATURE CO Fig. 8. Effect of temperature on bacterial extract luminescence. 1.0 ml bacterial extract (2%); 1.0 ml phosphate buffer, pH 7.0, 0.01 M; 2.0 mg DPNH^. The sample was cooled to 0° C, and its temperature was then raised rapidly with stirring by immersing in hot water (ca. 40° C). Readings were made for 10 seconds at each temperature. Total time for eight determinations, 10 minutes. fact being indicated, since such a high activation energy would hardly permit the reaction to proceed at significant rates. Possibly a large entropy factor is involved, although another alternative, in such a complicated concatenation of steps as seems to exist, is that nonlumi- nous alternative pathways compete more effectively at lower tempera- tures. Various agents have been examined for inhibitory action. Some of those tested are indicated in Table II, while Fig. 9 illustrates the effect of ultraviolet light (Strehler and Cormier, 1953). According to McEl- roy ( private communication ) , this initial stimulatory effect of ultra- BIOCHEMISTRY OF BACTERIAL LUMINESCENCE I 20 40 DURATION OF IRRADIATION (MIN) 60 Fig. 9. Effect of ultraviolet illumination on luminescence of bacterial extracts. Two per cent centrifuged extract diluted (1:1) with 0.01 M phosphate, pH 7.0, and illuminated with stirring (12 in. distant) by a Keese ultraviolet spot lamp (365 m/x). Samples withdrawn at times indicated (0.4 ml) and tested widi 300 iig of DPNHo. Boiled bacterial extract did not restore ability to luminesce to irradiated extracts. violet may be due to the photo-induced formation of aldehyde. The completely ultraviolet-inactivated system cannot be reactivated in our hands either by addition of the known necessary components or by the addition of boiled active extracts, suggesting that the protein is destroyed under these conditions. Comparative Biochemistry Although little is known concerning the evolution of biolumines- cence in luminous bacteria or other luminous forms, it may be pos- sible to draw some tentative conclusions from the comparative bio- BERNARD L. STREHLER 223 TABLE II Effect of Inhibitors on Ultracentrifuged Undialyzed Extracts Per Cent Inhibition Concentration (molar) Inhibitor 10-5 10^ KCN 23 47 NaNs 30 47 1 ,2-Naphthoquinone 53 100 2-Methyl-l,4-naphthoquinone 33 100 lodoacetate 37 26 jo-Chloromercuribenzoate 40 71 FAD 46 Riboflavin 76 HgCla 19 73 (Ag)2S04 26 100 chemistry of the ten strains of luminous bacteria which have been examined in vitro (Cormier and Strehler, 1954). Identical require- ments were found for luminescence in terms of diffusible factors among all these strains, e.g., reduced DPN, FMN, and long-chain aldehydes are necessary for the luminescence of their extracts. Cer- tain striking differences were also obvious. These included a wide variation in apparent Michaehs constants and a considerable range of temperature optima and temperature dependences (apparent activa- tion energies ) for the luminous reaction. Inasmuch as there was a great diversity in the morphology of the organisms studied and in the luminescence per unit dry weight of acetonized powders (whose luminescence yield paralleled the in vivo brightness of the strains ) , it would seem unlikely that they are closely related to each other evolutionarily. On the other hand, the identical biochemical requirements for extract luminescence indicate that the mechanism operating in the various strains is probably identical. These differences and similarities can all be accounted for if one assumes that the luminescent pathway in all the luminescent bacteria studied is derived from a normal respiratory pathway (Harvey, 1940). The ability to produce light, then, can be visualized, as arising from a special mutation in a flavin auto-oxidase pathway, permitting the energy liberated on the oxidation of one or more flavins to be lib- 224 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE erated in a light-producing reaction. Such mutations, since they prob- ably involve at most only a few steps, could arise repeatedly in un- related strains. This thesis necessitates no postulation of a special luciferin since flavins, coenzyme I and perhaps aldehydes are ubiq- uitous among living things. Evidence Concerning Mechanism of Interaction of Various Components Required for Bacterial Extract Luminescence A number of lines of evidence are available which bear on the relationship between the various factors necessary for in vitro lumi- nescence. These studies are concerned with kinetic measurements ( Strehler and Cormier, 1954b ) with the effect of various added com- ponents on the respiratory rate (Strehler and Cormier, 1954a), with the effect of reduced flavins on the luminescence with and without added aldehyde (Strehler et ah, 1954) and some pressure effects ex- amined in cooperation with Dr. Frank Johnson of Princeton Univer- sity (Strehler and Johnson, 1954). Respiration Attempts to determine an effect of long-chain aldehydes on respira- tion were at first unsuccessful, perhaps because the limiting reaction either in the presence or absence of aldehyde is the DPNH2-FMN reaction. However, it was possible, by varying the concentration of oxygen in the medium, to show that the aldehyde did in fact affect the level of luminescence and in a parallel manner the rate of respira- tion at low oxygen tensions. The results of such studies are indicated in the accompanying Fig. 10. From this figure it can be seen that the luminescence as well as respiration — not shown) is accelerated at a low oxygen tension, if aldehyde is added. The presence of aldehyde changes the apparent Michaelis constants for oxygen (both for lumi- nescence and respiration) by about a factor of 4. "Rise Time" Experiments Another type of kinetic study which was useful in determining the relationships of the various components was the so-called % rise-time experiment based on similar studies with Cijpridina extracts and in- tact bacteria reported by Chance et al. (1940). The general design BERNARD L. STREHLER 225 50- K^(PAL.l=0.55mm Km(N0W4L.)=2.03mm -200 o 150| 5 3 lOOS O z o 50 o o 0 5 10 15 CONCN OF OXYGEN (mm) Fig. 10. Effect of o.xygen concentration on luminescence of bacterial extracts in the presence and absence of palmital. To 1.0 ml of a 10% aqueous extract of A. jischeri in 0.1 M phosphate buffer (pH 7.0) was added 2 mg of DPNHa and 8 fig of FMN. Oxygen tension was measured with a dropping mercury electrode simultaneously with the light emission. 60 ng of palmitic aldehyde was then added to the same aliquot of extract and oxygen tension and lumi- nescence were again measured. Total \olume 4.0 ml. # No palmital. O with palmital. of these present experiments was to add all except one of the com- ponents necessary for a bright luminescence to the enzyme and then, at zero time, to add the remaining necessary component. The light output as a function of time after mixing was measured and recorded electronically and the approach to steady-state luminescence was plotted in such a manner as to give an index of the % rise time, i.e., the time required for the luminescence to become half maximal after mixing the various components. It was found that the terminal addi- 226 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE tion of factors which, on the basis of the effect of the aldehydes on respiratory activity or other work with nonluminous organisms, might be expected to react with each other, resulted in rise times of very similar or identical magnitudes (Strehler and Cormier, 1954a). For example, if malic acid were used as a hydrogen doner to reduce DPN and the time course of the luminescence was plotted according to the equation log (R max. — R obs.)/time = —a, the slope of the line ob- tained was the same whether malate or DPN was the last added com- ponent. Similarly, oxidized flavin or reduced DPN when added last exhibited similar rise times, and reduced flavin, oxygen, or aldehyde produced increases in luminescence, whose rate of approach to steady state conditions were comparatively close to each other. These results are summarized in Table III. TABLE III Time Required for Half-Maximal Luminescence in Bacterial Extracts When Various Essential Components Are Added Last Component Added Last Factors Present with Enzyme Half-Rise Time O2 FMN, DPNH2, KCF 0.05 KCF O2, FMN, DPNH2 0.08 RFH2 O2, KCF 0.14 RFH2 O2 0.27 FMNH2 O2, KFC 0.075 FMNH2 O2 0.41 FMN O2, DPNH2, KCF 2.7 DPNH2 O2, FMN, KCF 2.7 DPN O2, Malate, FMN, KCF 108. Malate O2, DPN, FMN, KCF 108. In any system consisting of a number of consecutive steps, it might be expected that the steps further separated from the final reaction would require longer to reach steady-state rates than those separated by one or a few steps. This expectation is borne out in the data pre- sented in Fig. 11. Studies with Reduced Flavins In cooperation with Dr. Harvey and Mr. Chang of Princeton Uni- versity a number of experiments were performed to test the hypothe- sis that the sole function of DPNHo was to reduce FMN. It was BERNARD L. STREHLER lO-A 227 SHORT SCALE O 02ADD'N • PAL. ADD'N A DPNH2 ADD'N | LONG SCALE A FMN ADD'N TEMP. =22°C 4 6 8 TIIVIE AFTER MIXING (SeC.) Fig. 11. Plot of data on half-rise times for oxygen, palmital, DPNHj, and FMN when each of these factors is added last to an enzyme preparation contain- ing an excess of the other factors. The extrapolated maximum value for luminescence was taken from the records (Rmax). The values for lumines- cence at various times after mixing the last necessary component were sub- tracted from Rni.ix and plotted on a logarithmic axis against the time after mixing. 228 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE u a5 T^ u M O to D -b o . O o '^ CO 3 -I Q, -C 4-1 _r o T3 13 T3 C H q; f— < 1) ■'-' fa u 3 i S ■S fa o 3 0) i-i Li o ° ^ U •g ^ W 0) 3 03 cs _c; n on T3 b 3 « 1^ :^ ? u c O T-j 4) ^ •3 *- ^ O ^ o oj .5 a > - =« C q3 (D o •^ -Q bo "C a _£- 2 U 2 '% 8 .S O 03 « O _5 rt 3 r-" "^ 5 rn T3 -G !-§ &:^ •3 c 0) f_ 03 S . ^ lU oi 'S S c -^ fa o M I UJ -1 UJ > tsi 2 UJ O • X + O X r?- CO UJ r X o o Six UJ >- N z UJ , o z> o o tr Q. UJ o CD UJ -J I 4 + X o o \l/ UJ >- M UJ a s o o o C3 •4—' E o c o to o 03 ■c o 03 2 O OJ C 4-' o 03 0) s 'IS CD T3 OJ O a o 1— ( ci p a c o T3 ^ 0) O a D >> C a l-H K O O I u X 0) t« C o > o u o E o c bO o o c E c v o to a u ° y a o E u fl • o to to 3 .s ^ a to -tS s o cq d-^ CD a - » 3 y ID 2 •" >, "-^ -w UJ -c .5 13 . "^ c c ^ 03 o *j '-* eg to 3 T3 3 0) O 2^ E O "* CD o3 s C 03 -w _^ %n 3 )H d-l 0) 'pical Embden-Meyerhof components. Reduced DPN thereupon reacts with oxidized FMN in the presence of a diaphorase 236 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE type enzyme and the flavin then is oxidized in the luminescent flavin oxidase. According to this picture aldehyde functions in this sequence by forming an oxygen adduct ( Jockusch, 1949; McDowell and Thomas, 1949; Wittig and Pieper, 1941), which oxidizes the flavin forming in essence a peroxide aldehyde addition product.* This peroxy aldehyde then oxidizes another flavin or, as mentioned alternatively, peroxide with the subsequent emission of light. The slowest half-rise times are elicted by the DPN-substrate reactions; the intermediate rise times occur when the DPNH2-FMN reaction is the starting point; and the more rapid rise times are characteristic of the main luminescent reactions involving a somewhat complicate mechanism of FMNH2 oxidation. The pressure studies give some clue as to the site of action of the aldehyde. Among the observations which must» be fit into any scheme which purports to present a unified picture are included the fact that reduced flavin requires aldehyde for maximal luminous oxidation, that the pressure effects are not observable in the absence of aldehyde, and that limiting amounts of aldehyde produce an anomalous type of pressure response. Since reduced flavin is easily auto-oxidizable by oxygen the lack of effect of KCF on the rate of respiration at high oxygen tensions is simply due to the fact that auto-oxidation of the flavins maintains the respiratory rate. When aldehyde is added, however, the pathway is considerably diflFerent. The flavin is oxidized by the aldehyde oxygen addition product, and the pool size of reduced FMN is diminished. The inter- mediate "D" represents the hypothetical aldehyde peroxide addition product, which is formed only in the presence of KCF and reduced flavin. If the concentration of aldehyde is nil or vanishingly small the amount of D which accumulates is minescule and, therefore, since the reaction B to D is limiting, the rate of light output cannot be accelerated by pressure. Moreover, since the pool size of FMNHo will remain large, any B to D enzyme remaining will be essentially * We have recently synthesized in our laboratory the crystalline peroxide- nonaldehyde addition product (alpha-oxynonylhydroperoxide) and find its luminescence potentiating effect to be several times that of the free aldehyde on a molar basis. Although this observation does not necessarily support the scheme as set forth, it is interesting that this derivative is more effective in promoting luminescence than the free aldehyde. BERNARD L. STREHLER 237 saturated with respect to FMNHo. At intermediate concentrations of aldehyde, the concentration of FMNH2 drops considerably, but the amount of D which accumulates is relatively small and the slowest step is B to D. When pressure is applied therefore, there is an instantaneous small increase in luminescence which soon decays to its original rate, under which conditions the rate of D to C is again determined by B to D. However, since pressure affects the amount of reaction A to B (/ci) the pool size of B soon drops and the reaction B to D slows as does D to C. This proposed scheme is consistent with the fact that the presence of aldehyde seems to be required for maximal respiration and luminescence at low oxygen tension, which effect can be viewed as a result of Oo binding by the aldehyde. Finally, this scheme furnishes a plausible mechanism for the con- servation of energy in intermediates preparatory to the final light- emitting step. One of the weaknesses of the scheme which has been presented revolves about the energetic aspects of the process. Although peroxide dismutation to water and oxygen liberates about 50-55 kcal and the oxidation of reduced flavin by peroxide would presumably yield (37 to 40 plus 25) or 62 to 65 kcal, neither of these processes by itself would seem to be exergonic enough to support luminescence of a maximum energy per einstein roughly equivalent to 62 kcal. This cautious view is based on the fact that chemiluminescent reactions in general would be expected to require for their occurrence a consider- able excess of energy over the energy stored temporarily in the excited state, since no energy transfer process is likely to proceed without incidental losses. On the other hand, it is possible that the net free energy change in the chemical reaction prerequisite to an excited molecule need not represent the total energy available for the formation of the excited state. Conceivably thermal energy could also contribute substantially to the total energy budget in at least two ways. Since experimentally it is known that the luminescent reaction proceeds with an appreciable activation energy, all or a part of this energy may be available and be added to the energy supplied concurrently by the net energy release during reaction. Thus, if the activation process for oxidation by the luminescent path involves an atomic configuration in which the elec- 238 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE trons are already displaced toward the excited level, this thermally derived energy would be available for the formation of the excited state, in addition to the energy liberated on reaction. Another possible mechanism involves the formation of an unstable oxidizing free radical (e.g., enzyme • RCHO • HO2 • ) at the expense of thermal energy. Formation of such a compound might represent a major portion of the observed activation energy and, upon reaction with a flavin, give rise to a much more exergonic step reaction than would be indicated by the overall nonluminous process. The energy thus derived would in all likelihood be sufficient to excite the fluores- cent molecule. Unfortunately, evidence is not at present available to assay critically the merits, faults, or pertinence of the various possibilities discussed. The biochemical evidence can, in a large part, be molded into a consistent picture. Whether the physical objections to the biochemical deductions are fatal, must be resolved by further experiment. ACKNOWLEDGMENTS The author wishes to point out that a large portion of this work would not have been carried out so promptly and efficiently were it not for the able collaboration of Mr. Milton CoiTnier of the Oak Ridge National Labora- tory Biology Division. Dr. John R. Totter and Dr. William A. Arnold are to be thanked for invaluable assistance and advice during the course of the work. The latter and Dr. James Franck, of the University of Chicago, contributed liberally of time and advice on certain points in the manuscript as did Dr. Frank Johnson, of Princeton University, although the ideas ex- pressed do not necessarily coincide with theirs. Much of the work with bacterial extracts on which this review is based was perfoiTned at the Biology Division of the Oak Ridge National Labora- tory and in collaboration with Drs. E. Newton Harvey, Joseph Chang, and Frank Johnson of Princeton. References Chance, B., E. N. Harvey, F. H. Johnson, and G. Millikan. 1940. The kinetics of bioluminescent flashes. A study in consecutive reactions. /. Cellular and Comp. Phijsiol., 15, 195-215. Cormier, M. J., and B. L. Strehler. 1953. The identification of KCF: Re- quirement of long-chain aldehydes for bacterial extract luminescence. ;. A771. Chem. Soc, 75, 4864. y BERNARD L. STREHLER 239 Cormier, M. J., and B. L. Strehler. 1954. Some comparative biochemical aspects of the bacterial luciferin-luciferase reaction in vitro. J. Cellular and Comp. Physiol, (in press). Drew, H. D. K. 1939. Chemiluminescence in the oxidation of certain organic substances. Trans. Faraday Sac, 35, 207. Eymers, J. G., and K. L. van Schouwenburg. 1937. On the luminescence of bacteria. Enzymologia, 1, 328-40. Gerretsen, F. C. 1915. Die Einwirkung des Ultravioletten Lichtes auf Leuchtbakterien. Zbl. Bakt. (Abt. 2), 44, 660-61. Gerretsen, F. C. 1920. Uber die Ursachen des Leuchtens der Leuchtbakte- rien. Zbl. Bakt. (Abt. 2), 52, 353-73. Harvey, E. N. 1920. Is the luminescence of Cypridina an oxidation? Am. J. Physiol, 51, 580-87. Harvey, E. N. 1940. Living Light. Princeton University Press, Princeton, N.J. Harvey, E. N. 1952. Biohirninescerice. Academic Press, New York, N. Y. Jockusch, H. 1949. Angew. Chem., 61, 318. Johnson, F. H. 1947. Bacterial luminescence. Advances in Enzymol., 7. 215-264. Johnson, F. H., D. E. S. Brown, and D. A. Marsland. 1942. A basic mecha- nism in the biological effects of temperature, pressure and narcotics. Sci- ence, 95, 200-03. Johnson, F. H., H. Eyring, and M. J. Polissar. 1954. The Kinetic Basis of Molecular Biology. John Wiley and Sons, N. Y. Johnson, F. H., H. Eyring, R. Steblay, H. Chaphn, C. Huber, and G. Gherardi. 1945. The nature and control of reactions in bioluminescence. With special reference to the mechanism of reversible and irreversible in- hibitions by hydrogen and hyroxyl ions, temperature, pressure, alchohol, urethane and sulfanilamide in bacteria. /. Gen. Physiol, 28, 463-537. Korr, I. M. 1935. The relation between cell integrity and bacterial lumines- cence. Biol. Bull, 68, 347-54. McDowell, C. A., and J. J. Thomas. 1949. Oxidation of aldehydes in the gaseous phase. I. Low-temperature oxidation of autaldehyde. /. Chem. Soc, 2208. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1953. The requirement of riboflavin phosphate for bacterial luminescence. Science, 118, 385-86. McElroy, W. D., and D. M. Kipnis. 1947. The mechanism of inhibition of bioluminescence by naphthoquinones. /. Cellular and Comp. Physiol, 30, 359-80. McElroy, W. D., and B. L. Strehler. 1954. Bioluminescence. Bacteriol Revs., 18, 177-94. Shoup, C. S. 1929. The respiration of luminous bacteria and the effect of oxygen tension upon oxygen consumption. /. Gen. Physiol, 5, 269-75. 240 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE ;8pruit, C. J. P., and A. L. Schuiling. 1945. On the influence of naphtho- quinones on the respiration and hght emission of Photobacterium phos- phoreum. Rec. trav. chim., 64, 220-28. Spruit-van der Burg, A. 1950. Emission spectra of luminous bacteria. Biochim. et Biophtjs. Acta, 5, 175-78. Strehler, B. L. 1951. The luminescence of isolated chloroplasts. Arch. Bio- chem. and Biophys., 34, 239. Strehler, B, L. 1953. Luminescence in cell-free extracts of luminous bacteria and its activation by DPN. /. Am. Chem. Soc, 75, 1264. Strehler, B. L., and M. J. Cormier. 1953. Factors affecting the luminescence of cell-free extracts of the luminous bacterium, Achromobacter fischeri. Arch. Biochem. and Biophtjs., 47, 16-33. Strehler, B. L., and M. J. Cormier. 1954a. Isolation, identification and func- tion of long-chain fatty aldehydes affecting the bacterial luciferin-luciferase reaction. /. Biol. Chcm. (in press). Strehler, B. L., and ^L J. Cormier. 1954b. Luciferin-luciferase Reaction in vitro. Arch. Biochem. and Biophys. (in press). Strehler, B. L., E. N. Harvey, J- J- Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Troc. Natl. Acad. Sci. U. S., 40, 10-12. Strehler, B. L., and F. H. Johnson. 1954. The temperature-pressure-inhibitor relations of bacterial luminescence in vitro. Proc. Natl. Acad. Sci. U. S., 40, 606-17. Strehler, B. L., and C. S. Shoup. 1953. The chemiluminescence of riboflavin. Arch. Biochem. and Biophys., 47, 8-15. van Schouwenburg, K. L., and J. G. Eymers. 1936. Quantum relationship of the light-emitting process of luminous bacteria. Nature, 138, 245. Wittig, G., and G. Pieper. 1941. tJber gehemmte und beschleunigte Autoxy- dation der Benzaldehyds in Gegenwart von Dibiphenylen-athylen Ann. Chem. Justus Liebigs, 546, 142-71. Discussion Bacterial Luciferin Dr. Harvey: The striking luminescence which appears when reduced flavin mononucleotide (reduced riboflavin phosphate, FMN-H2) is added to cell-free luminous bacterial extracts containing a long-chain aliphatic aldehyde is easily visible to the dark adapted eye and re- minds one of the luciferin-luciferase reaction of other organisms. The experiment raises a number of questions: (1) whether the FMN-Ho is reducing some other compound in the bacterial extract which is responsible for the light; (2) whether other reduced flavin compounds will luminesce when mixed with bacterial extracts; (3) whether FMN-Ho may be regarded as bacterial luciferin. The first question has been answered in the negative by showing that a reduced com- pound with a lower redox potential than FMN-Ho (£'o at pH 7 = — 0.185), when added to the bacterial extract, does not cause light emission. Strehler et al. (1954) found that reduced anthraquinone-2-6- disodium sulfonate (E'o at pH 7 — —0.192), a substance harmless for luminous bacteria (Harvey, 1929), evokes no luminescence when mixed with the bacterial extract. However, if oxidized riboflavin phos- phate (FMN) has been previously added, then the reduced anthra- quinone reduces the FMN and a bright light appears. The experiments suggest that reduced FMN may be designated bacterial luciferin. In the same paper it was reported that riboflavin (Merck) emitted light when mixed with cell-free bacterial extracts of Achromobacter fischeri. Recently I have tested a number of other flavin compounds with a crude cell-free extract of a luminous bacterium having a low (12° C) temperature optimum. The cell-free extract was prepared by Dr. B. L. Strehler from a strain of luminous bacteria obtained from Dr. C. B. Van Niel and known as "Gest." The active acetone bacterial powder had been extracted with water and centrifuged at 70,000 X g for one hour. Although the temperature rose somewhat during centri- fuging, the water extract always gave a bright light when mixed with reduced riboflavin phosphate, indicating that the enzyme essential for light production had not been denatured. The extract was stored at — 10" C before use. The dilute flavin solutions in distilled water were reduced in small centrifuge tubes in a stream of hydrogen after adding platinized asbestos. The tubes were then stoppered, centrifuged at 241 242 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE 1200 X g for 5 minutes to throw down the asbestos, and the reduced flavin supernatant removed with a 1-cc tubercuHn syringe and 0.2 cc squirted under 0.6 cc of the bacterial extract containing a small amount of palmitaldehyde. This method is simple and has many advantages for handling solutions of rapidly oxidizable substances. Observation of fluorescence in ultraviolet light indicates that the flavin oxidizes only at the surface of such a solution in contact with air and not in the syringe. TABLE IV Luminescence with centrifuged cell-free bacterial extract (Gest) plus Reduced flavin Source palmitaldehj'de Riboflavin phosphate (FMN) Sigma Chem. Co. Bright Flavin adenine dinucleotide (FAD) Sigma 15 None Moccasin snake venom, a source of L-amino-acid oxidase H. R. Mahler None DPNH cytochrome c reductase (liver) H. R. Mahler None Lumichrome (dimethylalioxazine) Merck & Company Very faint Lumiflavin (trimethylisoalloxazine) Bios Labs. None L-araboflavin L. Michaelis None Giucoisoalloxazine L. Michaelis None N-methylalloxazine L. Michaelis Fair Riboflavin Merck & Co. Faint Riboflavin Gen. Biochem. Inc. None The results of adding various flavins to the crude Gest extract are shown in Table IV. The FMN experiment served as a control to indicate that the extract was active. It will be observed that with the exception of riboflavin phosphate and N-methylalloxazine, practically no luminescence appears with any flavin. Samples of riboflavin from two different sources gave different results. To what extent contami- nants may be responsible for luminescence with riboflavin and with lumichrome is impossible to say. The negative results with the flavoproteins, both of which contain FAD ( flavinadeninedinucleotide ) or an FAD like prosthetic group, should be especially noted. These flavoproteins were lyophilized and kindly supplied by Dr. H. R. Mahler, of the University of Wisconsin Institute of Enzyme Research. It will be most important to test a flavoprotein containing FMN. BERNARD L. STREHLER 243 Regarding the question as to which substance among a number necessary for luminescence is to be regarded as luciferin, I would like to quote from a paper by Harvey and Tsuji, entitled "Luminescence of Cypridina luciferin without luciferase, together with an appraisal of the term, luciferin," now in press in the Journal of Cellular and Comparative Physiology: Rather than placing the emphasis on a limiting factor, or on heat stability or dialyzability or even oxidizability [as an indication of luciferinl, as has been done previously, light emission should be the criterion. In the case of luminous organisms requiring dissolved molecular oxygen for lumines- cence, luciferin may properly be defined as the oxidizable substance sup- plying molecules capable of absorbing enough excess emergy to emit in the visible region. Such a definition implies that some form of luciferin molecule — either free base or acid, either dissociated anion or cation, in reduced or oxidized form, either free or combined with protein, like a prosthetic en- zyme group — can pick up the energy of the oxidative reaction in which it is involved. Such a definition does not mean that luciferin is the same sub- stance in different luminous animals, nor does it necessarily designate luciferin molecules themselves as the ones which emit, but it does imply that a related molecule, such as a luciferin-luciferase combination, or an oxidized luciferin molecule, or a molecule of an intermediate step, is the emitter. The molecule actually emitting might be referred to as the photo- gen. ... It has long been recognized that a substance, whose molecules are readily excited to fluoresce by the energy of radiation, is most likely to be chemiluminescent from the energy of a chemical reaction. . . . Therefore it seems most logical to regard the reduced form of the fluorescent flavin as bacterial luciferin rather than the non-fluorescent aldehyde. Al- though riboflavin and FMN are not fluorescent in the reduced form, the oxidized flavins fluoresce yellow green over a wide pH range, from pH 1 to 11. An analogous situation is to be observed among the ctenophores, where a striking phenomenon is the fluorescence of the luminous organ after, but not before bioluminescence has occurred, as if the final product of lumi- nescence was a fluorescent molecule (Harvey, 1925). Reduced FMN is comparable to Cypridina luciferin in that it under- goes spontaneous oxidation by dissolved oxygen without luminescence and only emits light in the presence of what may be called bacterial luciferase. References Harvey, E. N. 1925. Studies on bioluminescence XVII. Fluorescence and inhibition of luminescence in ctenophores by ultraviolet light. /. Gen. Physiol, 7, 331-39. 244 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE Harvey, E. N. 1929. A preliminary study of the reducing intensity of lumi- nous bacteria. /. Gen. Physiol., 13, 16-20. Strehler, B. L., E. N. Harvey, J. J. Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Proc. Natl. Acad. Sci. U. S., 40, 10-12. Dr. Johnson: The role of peroxide in luminescent reactions is an interesting problem. A theory is given in the book (Johnson, Eyring, and Polissar, 1954) that would account for the luminescence of lu- minol, with destruction of H2O2 but without destruction of the luminol, the peroxide acting to induce a quinone type of electronic structure which then radiates on returning to the freely resonating structure of the substituted or unsubstituted ring. In arriving at this theory, how- ever, it was necessary to assume that four molecules of H2O0 are decomposed, in the reactions leading to light emission, in order to account for the energy of the emitted light. In the scheme of reactions presented by Dr. Strehler, the decomposition of only two molecules of H2O2 is assumed, and the free energy made available thereby is considerably short of the free energy needed for the observed light emission. To this extent, the hypothesis, as it stands, does not appear to be thermodynamically sound. The activation energy for the change between normal and activated states cannot be added to the free energy difFerence between the initial and final states to increase by much the available free energy of reaction. Dr. Eyring: One restriction that thermodynamics places on any mechanism is that one cannot get more free energy in the form of quanta than is used up in the reactions. Thus, riiAF > n-Jiv (1) Here ni is the total number of molecules reacting and aF is the free energy used up per individual process; 712 is the number of quanta emitted and hv is the free energy of a quantum. The absolute maxi- mum efficiency is thus * = ^' = t^ (2) rii hv Now such an efficiency can be approached if you use the reaction to do work reversibly storing it in a battery and then use the battery in an efficient fluorescent lamp. In principle this can also be realized BERNARD L. STREHLER 245 in chemical reactions as indicated in Fig. 15, similar to a diagram in our book (Johnson, Eyring, and Polissar, 1954). Thus thermodynamics will be satisfied if kn W2 ^ AF A'l + A'2 fli fiv The problem is to find actual systems where ^2/(^1 +^2) is this large. I know of none. I would rather believe this high efficiency is obtained by processes in which the free energy of additional reactions is used Fig. 15. to pile up sufficient potential energy for the emission without absorb- ing kinetic energy from the heat reservoir by a not thermodynamically impossible but by a nevertheless unknown specialized mechanism. If one thinks of a Planck energy density versus frequency distribution curve transformed so that each abscissa, representing a frequency, is lengthened by the free energy donated by the chemical reaction to the emission process, the ordinates give the maximum intensity to be expected for the changed frequencies. The actual distribution curve can only be predicted from a detailed knowledge of reaction mechan- ism. One qualitative argument against this type of efficient mechanism is to be found in the fact that kinetic energy of reaction is never passed along in an efficient manner to make thermal chains. It seems enormously more likely, as I have already stated, that enough reactions 246 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE pool their potential energy so that the kinetic energy = (hv — AF ) picked up from the heat reservoir is negligible. You may find the papers of Audubert (1936, 1937) and of Evans, Eyring, and Kincaid (1938) interesting in this connection. References Audubert, R. 1936. Emission de rayonnement par les reactions chimiques. /. Chim. Phys., 33, 507-25. Audubert, R. 1937. Etude de I'emission de rayonnement ultra-violette au cours de la decomposition lente des azotures. /. Chim. Phys., 34, 405—15. Evans, M. G., H. Eyring, and J. F. Kincaid. 1938. Nonadiabatic reactions. Chemiluminescence. /. Chem. Phys., 6, 349-58. Johnson, F. H., H. Eyring, and M. J. Polissar. 1954. The Kinetic Basis of Molecular Biology. John Wiley & Sons, New York. On Light Energy versus Free Energy Changes in Bioluminescence Dr. Strehler: In attempting to evaluate the plausibility of any overall chemical reaction as a source of the energy in a quantum emitted in a coupled luminescent reaction, the question arises: Must the AF liberated per mole of the proposed reaction be greater than the energy per einstein of the emitted light? If the answer is affirmative, only those reactions furnishing more energy than that in the light need be considered as possible sources of energy for bioluminescent processes. If, on the other hand, the answer is negative, then even reactions somewhat less energetic than the light emitted cannot be ruled out as possibilities. A case in point is the oxidation of peroxide by peroxide which liberates about 54 kcal during the dismutation of 2 moles of H2O2 and thus might be considered as a source of energy for bacterial or riboflavin chemiluminescence (max. energy/einstein = 60 kcal). The following discussion is a qualitative series of arguments and is limited by the author's training and experience. A quantitative treatment of the thermodynamic and kinetic aspects of this problem has kindly been made by Dr. Joseph Mayer in the following portion of this discussion. Premises ( 1 ) The chief premise is simply that thermal energy may contribute BERNARD L. STREHLER 247 to the energy prerequisite to the emission of a light quantum. Thus: b.Fi"hv > 1 or AF/hv < 1 but (AF^ + AF)/hv > 1 (2) The second premise is that partial reactions will proceed irre- spective of the source of the reactants (if they are thermodynamically possible ) , whether the reactants are chemically or thermally generated. (3) The third premise is that the first premise does not violate the first or second law of thermodynamics, since not all of the energy in liv is available to do work. Mechanism Consider a reaction: ROH + XH2 ^ RH + H2O + X + hv or ROH + XH2 -^ RH + H2O + X + AF^ in which the partial reactions are: 1. ROH + heat->R- + -OH 2. R- +XH2^ -XH + RH 3. -OH + -XH ^ X + HOH + (heat or hv) The reactions are diagrammed in Fig. 16. ^•♦•OH^XHz. RH ♦ ;c >- HOH Fig. 16. 248 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE Such a reaction, if efficient, would convert thermal energy into a portion of the energy in a light quantum and AF < hv. No objection could be raised if R- and -OH were generated by independent exergonic reactions and there seems no reason for assuming that reac- tion 3 can distinguish the source of the reactants. Objections and Conclusions Despite the above arguments, if it could be shown that such a mechanism violated either the first or second law of thermodynamics, it would perforce be ruled out. It clearly does not violate the first law since the total energy is constant. That the proposed mechanism does not violate the second law is also clear on the following qualitative grounds and is developed quantitatively in the succeeding quantita- tive treatment. If it were possible to build a 100% efficient photoelectric or photo- chemical device to convert the total light energy emitted by such a proposed reaction mechanism into useful work, the second law would be violated. Whether such a device could in principle be constructed is the point at issue. Suppose a photoelectric device were to be operated on light emitted as black body radiation by another body at its own temperature. Since thermal energy is here converted into light energy as an intermediate ( some, though a minute amount in the wavelength region of bioluminescences!), if such a machine were possible it would itself violate the second law. It follows from the above and following that the overall reactions possibly leading to luminescence are not restricted to those having a greater AF than hv. One cannot rule out as participants in and energy sources for bioluminescences certain reactions less energetic than the light emitted. On the Maximum Efficiency of a Photochemical Reaction* Dr. Mayer: We consider the photochemical reaction: A + B^C -^hv (1) and consider that the free energy change -AF(°) = Fn(C) - F°(B) - F°(A) (2) * Dr. Joseph E. Mayer, of the Institute of Nuclear Studies, University of Chicago, kindly contributed this paper, by invitation, for additional discus- sion on the question at issue [Ed.h BERNARD L. STREHLER 249 is known for the chemical reaction with the reactants and products at unit concentration. The reaction may actually go to emit a range of frequencies, hv, but we suppose the mechanism for the emission of any single fre- quency vo to be such that one molecule of A plus one of B follow some successive steps to emit one quantum, /7i'o, uncoupled with any parallel steps by which molecules A and B react, either without the emission of light, or with emission of lower frequencies /. The rate Ro of photons of frequency r-o, per unit frequency range, per cubic centimeter, at unit concentration of A, B, and C is now measured: Rodi'dV — number of photons of frequency between vo and vo + dv emitted by the reacting mixture of volume dV, at temperature T, with A, B, and C at unit concentration. The question now is: What is the maximum value of Ro for any frequency, vo at T, in terms of the free energy liberation, — aF"", of the chemical reaction? The question can be answered by considering the following hypo- thetical reversible machine. We place the reacting mixture, at temperature T, unit concentration of A, B, and C in a vessel and, with a negative catalyst, inhibit any reaction which emits other frequencies than vo, or which does not emit light. Since the reaction which emits vo is supposed to be un- coupled to any other reaction, this fiction is permissible and is equiva- lent to the assumption of microscopic reversibility. We surround the reaction vessel with an insulating wall, transparent to the frequency vo, but to no other radiation. The space outside of this contains black body radiation enclosed in completely reflecting walls, fitted with a completely reflecting piston, so that by compression or expansion, the temperature T* of the radiation can be altered. An auxiliary machine can be used to remove radiation from the reservoir at T*, by adiabatic expansion to the temperature T of the reacting mixture it can obtain useful work, and deliver the remaining radiation to a reservoir at T. This process can be carried out reversibly, the useful work AW, obtained from the radiant energy a£* taken out at T*, is ^={T*-T)jT* (3) By adjusting the temperature T* of the surrounding black body radiation, we can keep the radiation density C7(vo) in the reaction 250 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE vessel, which will be equal to that in the black body surroundings: U{vo) = {8Trhvo'/c')[e''-'^''^T* - 1]-' (4) just such as to balance the forward and backward rate of reaction We thus have a reversible machine by which the maximum useful work obtainable from the chemical reaction ( 1 ) can be evaluated through the light emitted. We have _AF(°) = -A.4° - APV > NohvoiT* - rj/T*-^^'' ^ No hMT* - T)/T* (5) since we can assume the APV of the reaction to be negligible. The temperature T* of the surrounding black body radiation bath, which would just reverse reaction (1), is related to j'o and the rate Ro of the emission of photons in the absence of the reversing bath. Let o- be the absorption cross section of molecules C for light of frequency lo to reverse Eq. ( 1 ) . The reverse reaction will go at a rate Rr = [U{vo),hpo]CaNo/V (6) with No = Avogadro's number, and V the volume (in cubic centime- ters) of one mole of material at the unit concentration used in com- puting - AF<°' of Eq. (2). We use (4) for U(vo) in (6) setting Rr equal to Ro to compute T* as with \ = C 27ri'o (8) The dimensionless quantity, Xo = a/Xo- = 4xVoV/c^ (9) has a maximum value of unity at resonance, but would normally be expected to be very much smaller for visible light and ordinary molecules, say of the order of Xo = 10 ~- to 10 ~^. If the concentration units of Eq. (2) are moles per liter, then Nn/V — 6 X 10-'\ The rates Ro, photons per cubic centimeter per second per unit frequency range may be of order 10''\ The quantity Xo(No/V)/Ro is of order 10^^ to 10'^»1. We can neglect the unity under the logarithm of Eq. (7). Setting -AF^°^/Nohvo > 1 - {T/T*) (10) BERNARD L. STREHLER 251 from (5), and from (7), we arri\e at T/T* = (JcT/hvo) In - .Yo No/VRo (11) IT In I Xo No/VRo > 1^ - ^ J^^ (12) Ro<- Xo ^ e-f''--(-^^(°)/^»)]/^-^ (13) It appears that values of XoNq/V of 10^^ are not unreasonable, integrated o\ er an>- small frequency range. Even if [hvo - (-AFn/iVo)]/A-r^30 Nohvo - (-AF(°)) ^ 18 kcal, T = 300° values of R,, (integrated over a short frequency range) of the order of 10'' photons per cubic centimeter per second would not seem to violate the second law in any way. Equation (13) refers to the rate Ro, at /u'o, of emission of photons at the concentration of reactants used in the computation of AF^°'. If the actual concentrations are lower either the rate should be corrected to unit concentration, or the aF^°' should be computed from the actual concentrations emplo>'ed. In either case the equation will be the same. Chemiluminescence from Thermally Activated Intermediates Dr. Kauzmann: Suppose that a substance can exist in two states, C and D, and that the change from state C to state D proceeds through thermal activation to any one of a series of intermediates, Ci*, C2*, . . . , each of which then produces state D by emitting a quantum of radiant energy lin, /h'2, . . . , respectively, according to the following scheme: C ^ Ci* -> D + /ivi C ^ C2* ^ D + hv.2 It is convenient to denote the energy change per molecule of the overall reaction C -» D by hvo- This is the energy of the quantum which would be emitted if a molecule in state C could be transformed 252 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE directly into one in state D with complete conversion of the energy difiFerence into light. Similarly, let us indicate by hvi*, /iv2^ the energy absorbed when Ci*, Co are formed from state C. As seen from Fig. 17, the frequencies which are actually emitted are I'l = I'O + Vi* V2 = vo + P2*, etc. Fig. 17. ■ V The i'*'s represent the contributions of the thermally activated inter- mediate states to the quanta that are actually emitted, while I'o is the contribution of the overall chemiluminescent reaction, C ^ D + light. It is obvious that there must be some thermodynamic limit to the increase in the energy of the emitted quantum above /iio that is ob- tainable from the emission produced in this way by thermally acti- vated intermediates. One might expect that the intensity of light emitted at a frequency v, greater than lo, would decrease with increas- ing frequency in proportion to exp( — h{i'—i'o)/kT), where T is the temperature of the system. If vo corresponded to a wavelength of 5000 A, this means that the intensity should decrease by a factor of 10 for each 120-A decrease in the wavelength — which is a somewhat sharper edge for an emission band than is usually observed with luminous bacteria (Spruit-van der Burg, 1950). An exponential factor of this kind must indeed operate, but as we shall now show, the upper limit set by the laws of thermodynamics on the intensity is surprisingly BERNARD L. STREHLER 253 high, so this factor may not become important until frequencies are reached that are well above vo- Let us place one mole of the substance in state C in a vessel at temperature T whose walls are perfect reflectors of radiation, so that all the radiation emitted by the reaction is trapped inside the vessel. Eventually a steady state ought to be reached in which the rates of transitions from the states C* to state D are balanced by absorption of quanta by molecules in state D which return them to state C*. The density of radiation present in this steady state leads to a thermo- dynamically defined upper limit to the luminosity that is attainable from the conversion of C to D through the intermediates C*. Transitions from state C* to state D can occur in two ways: in the first place, there is a probability, Aidt, that a molecule in state Ci* will in the absence of radiation undergo a transition to state D in a small time interval dt. Secondly, if radiation is present, the probability of a transition is increased by an amount p,B,(if where p, is the energy per unit volume of radiation whose frequency lies in the range between n and n + 1 cycles per second. The fundamental quantum theory of electromagnetic radiation ( Mott and Sneddon, 1948 ) shows that A,/B, = S-Trhvi^/c^, where c is the velocity of light. Furthermore, the probability of a reverse transition, D -^ Q*, occurring in time interval dt is equal to piBidt. (Since state D is assumed to have less energy than states C,*, there can be no transitions from D -^ d* in the absence of radiation. ) Writing the concentrations of C;* and D as (Ci*) and (D), we see that when the steady state is reached, (p.B, + A,)(C,*) = p,B,(D) (1) Since the states Cj* are assumed to be populated from C by thennal activation, we may make use of the equihbrium constant, Ki* = (C.*)/(C) = exp(-hui*/kT) (2) From (1) and (2) and the value of Ai/Bi we find for the steady state radiation density ^' ^ (D) (3) (C)' ^ where a = Stt/i/c^. Note that no steady state is possible if (D)/(C) ^ exp(— hvi/kT). Let us compare this result with the ordinary thermal 254 BIOCHEMISTRY OF BACTERIAL LUMINESCENCE radiation ("black body radiation"). Planck showed that this is av/ ^' ^hvJkT „ \ (4) If C and D are in thermodynamic equilibrium, (D)/(C) = exp(/ii/o/ fcT) and Eq. (3) becomes identical with Planck's formula — as of course it must. (Indeed, this is a simple and often used way of deriving Planck's formula.) If C and D are present in comparable concentrations, as they must be during most of the course of a chemi- luminescent reaction, we can write (C) = (D), giving av? *'^ ghv,*/kT _ 1 We see that the radiation density thus obtained is related to the black body radiation density p;* associated with the quantum /ivi* in the following way: Pi ~ iVi/v^*Y Pi* That is, during a chemiluminescent reaction involving an intermediate which is activated thermally by an amount equivalent to a quantum of energy /ivi*, the radiation density may be greater than the ordinaiy black body radiation of photons of energy hvi* at the same tempera- ture by a factor [(vo + vi*)/n*]^. This factor might be quite large; for instance, if vo corresponds to a wavelength of 5300 A or an energy of 54,000 cal/mole, and vi* corresponds to an energy of 6000 cal/mole (or about lOfcT), the steady state chemiluminescent radiation density at 4750 A ( corresponding to an energy of 60,000 cal/mole ) would be 1000 times greater than the black body radiation at 47,500 A (cor- responding to 6000 cal/mole). The physical reason for this somewhat surprising result is that the number of states available to a photon is proportional to the square of its frequency, while its energy is pro- portional to its frequency. By attaching the energy /ir,* to a more energetic photon of energy /jvo we increase the probability of finding the energy /ivi* by a factor of [(vo + vi*)/vi*]^. In order to find the numerical value of the limiting intensity of emission, let us assume that the states Cj* form a continuum above a threshold frequency vi* = v*. Let us also assume that there is a small hole in the side of the vessel containing the chemiluminescent reaction, through which radiation may leak out. The rate of emission of energy associated with a frequency in the range from vi to vj + dvi from such a hole is simply (c/2)/3idvi per unit area of the hole per unit time. For BERNARD L. STREHLER 255 the total rate of emission of all frequencies we find Total rate = (c/'2) I pidv, = ^ If /u'* is considerably larger than kT but still small compared with vo and if (C) = (D), we can obtain the approximate result, Total rate ^ ^ a ^ j^o' e-'^'^/^^ 2 h = 5.7 X 10-''5vo^e-'»'*/^-^ watt/cm2 where lo is expressed in sec.~^. Let us take vq = 0.57 X 10^^ sec.~^ (equivalent to a wavelength of 5300 A) and hv* = lOkT. At 300° K this corresponds to emission at a wavelength less than 4800 A, which is 500 A beyond the expected threshold. We find a possible rate of emission of about 5 X 10 ~^ watt/cm-, equivalent to about 10^^ quanta per second per square centimeter. This is to be compared with the normal emission by a black body of radiation having quanta whose energy is more than lOfcT, which comes to only 5.8 X 10" ^ watt/cm- at 300° K. It corre- sponds to a relatively bright light, though not very much brighter than the observed maximum brightness of a concentrated suspension of luminous bacteria or other bioluminescent organisms, being of the order of the brightness of a white surface one meter distant from a 40-watt tungsten lamp. The total possible rate of emission beyond 4690 A, however, is 5 X 10 ~*^ watt/cm-, and that beyond 4600 A is 5 X 10 ~^ watt/cm-. The luminescence of luminous bacteria does not drop off this rapidly with decreasing wavelength, and it would be interesting to see if the actual emission at short wavelengths surpassed the theoretical upper limit. References Mott, N. F., and I. N. Sneddon. 1948. Interaction of radiation with matter. In Wave Mechanics, chapter 10, pp. 247-90. Oxford University Press, London. Spruit-van der Burg, A. 1950. Emission spectra of luminous bacteria. Bio- chim. et Biophys. Acta, 5, 175-78. Pur/'fication and Properties of Bacterial Luciferase J. W. Hastings and W. D. McElroy Department of Biological Sciences, Northwestern University, Evanston, Illinois, and McCollum-Pratt Institute and Department of Biology, The Johns Hopkins University, Baltimore, Maryland The demonstration of luminescence in cell-free bacterial extracts, which had long been sought by workers in the field, was finally achieved by Strehler early in 1953. After the demonstration that di- phosphopyridine nucleotide (DPN) markedly stimulated lumines- cence, work progressed rapidly. Within the year this bioluminescent reaction was the first in which all known stimulatory factors had been chemically identified. Although many unanswered questions concern- ing the mechanism of the reaction remain, it seems worth while to review some of our studies on the system, particularly where our results differ from or amplify those reported by Strehler. Preliminary experiments with relatively crude extracts had indi- cated that ( 1 ) DPN was functional in the luminescent reaction only when reduced and might be replaced by other reducing compounds (TPNH), and (2) that a heat stabile fraction of the crude extract contained a stimulatory factor not identical with DPNH. The puri- fication of the enzvme and a studv of other factors was thus un- dertaken. Conventional enzymological methods resulted in a forty- to sixty-fold purification of the enzyme (see Table I).* Bacteria " Repeated (NH4)oS04 fractionations of the active fractions have resulted in even greater purity. Green and McElroy (unpublished) find that their best fractions are homogeneous by ultracentrifugation (mol. wt. ca. 85,000) and contain three components with slightly different electrophoretic mo- bilities at pH 7.6. The active fraction constitutes over 70% of the total protein. 257 258 PURIFICATION AND PROPERTIES TABLE I Purification of Bacterial Luciferase Specific Activity, Light Protein/ml, Light units/mg Fraction nig Units/ml protein Recovery, % Crude 1.2 17.5 14.1 — Acid precipitate 11.2 193. 17.3 110 (NH4)2S04 fractions 40-50% 0.67 0.85 1.27 0.3 50-60% 1.36 26. 19. 10.0 60-70% 1.58 244. 155. 94.5 70-80% 0.50 15.5 31. 6.0 { Achromobacter fischeri) grown on Farghaly's (1950) liquid medium with 1% peptone added were harvested by centrifugation after 12-18 hours and lysed in distilled water (3 g wet weight per 100 ml H2O). After removal of the cell debris by centrifugation the protein was pre- cipitated by acidification and resuspended in a small volume of water at pH 7.0. Further purification was achieved by fractionation with (NH4)2S04, pH 7.0. All procedures were carried out in the cold. Light intensity was measured with a photomultiplier apparatus (Hastings, McElroy, and Coulombre, 1953). The reaction mixture consisted of 0.5 ml of buffer (0.135 M KH2PO4, 0.135 M NaoHP04, and 0.01 M NaCl) pH 6.8, 0.1 ml of 2.5 X 10-^ M DPNH, 0.1 ml of 5 X 10"^ M riboflavin-5-phosphate, 1.0 ml of a satin-ated water solution of dodecyl aldehyde, and enzyme, other reagents, and water to a total volume of 2.5 ml. The reaction was usually initiated by adding DPNH. The enzyme preparation from the 60-70% (NH4)2S04 fraction is colorless, rapidly destroyed at 80° C, and stable for several months in the frozen state. It will emit a weak light with DPNH alone, but for appreciable activity FMN (McElroy, Hastings, Sonnenfeld, and Cou- lombre, 1953) and aldehyde (Cormier and Strehler, 1953) must be added. FMN reduced by bubbling hydrogen in the presence of a catalyst will dispense with the requirement for DPNH ( Strehler, Har- vey, Chang, and Cormier, 1954). Other reduced compounds (reduced riboflavin, reduced dyes) will function in lieu of DPNH, but FMN is J. W. HASTINGS AND W. D. McELROY 259 a specific requirement (McElroy and Green, unpublished). The puri- fied enzyme acts as a typical diaphorase (McElroy and Green, 1954). The overall kiminescent reaction may be written: FMNH2 + RCHO + O2 + enzyme -^ light + products None of the products has been identified as yet, although we have evidence that the aldehyde is destroyed during the reaction. The pH optimum for this reaction is 6.9 (Fig. 1). The luminescent reaction Fig. 1. Effect of pH upon bacterial luminescence. Buffer as described in text. does not involve free hydrogen peroxide. The level of luminescence is unaflFected by added hydrogen peroxide or crystalline beef catalase. This of course does not rule out the possibility that organic peroxides may be functional in the system. In the course of our investigation of the FMN requirement we ob- served that the purified enzyme without added FMN still emitted appreciable (ca. 5% max.) hght when DPNH was added. Microbio- logical assays using Lactobacillus casei confirmed the presence of flavin in the enzyme, and numerous unsuccessful attempts were made to separate it from the enzyme. Irradiation at 5° C with a Keese lamp produced a marked effect in this respect. Figure 2 illustrates the type of result obtained. During an initial period of irradiation added FMN does not give additional stimulation. With continued irradiation, how- ever, the stimulation by FMN rapidly increases until the requirement may be considered essentially absolute. The more striking effect of irradiation is illustrated in Fig. 3. Sam- ples of the enzyme were removed at intervals and assayed. Both with 260 PURIFICATION AND PROPERTIES and without added FMN the total activity of the enzyme at first in- creases and subsequently declines with continued irradiation. It was thought that this effect might be due to the destruction of riboflavin, I 2 IRRADIATION TIME-HOURS Fig. 2. Stimulation of luminescence by added FMN, using enzyme irradiated for time intervals indicated. which is a potent inhibitor of the system (Table II). This was shown not to be true when it was found that the stimulatory factor could be produced by irradiation of the 20-^30% (NH4)oS04 fraction, which it- self had no inhibitorv effect and no luciferase activity. The relation- TABLE II Inhibitors of Bacterial Luminescence in Vitro Inhibitor Concentration, molar Riboflavin 2 X 10-« p-Chloro Hg benzoate 4 X 10-« KCN 1 X 10-^ Version e 7 X 10-^ Inhibition, % 43 58 47 47 J. W. HASTINGS AND W. D. McELROY 261 ship between irradiation dosage and degree of stimulation when an irradiated 20-30% (NH4)2S04 fraction is added to the reaction mix- ture is shown in Fig. 4. Studies concerning the utiHzation of this factor ( McElroy, Hastings, Sonnenfeld, and Coulombre, 1954) indicated that it is destroyed dur- ing luminescence and could therefore be considered as bacterial lucif- erin, analogous to Cijpridina and firefly luciferins. IRRADIATION TIME -HOURS IRRADIATION TIME - HOURS Fig. 3. Effect of irradiation dosage Fig. 4. EflFect of ultraviolet dosage upon activity of enzyme fraction, upon stimulation of luminescence both with and without added by 20-30% (NH4)2S04 fraction. FMN. Since long-chain aldehydes will substitute for the factor produced by ultraviolet treatment, it seems possible that a photochemical pro- duction of aldehyde occurs during irradiation. This possibility has not been investigated. However, it has been found (McElroy and Green, 1954) that the aldehyde is utilized for light production in the same way as is the factor derived from irradiation. The total light emitted by a reaction mixture is proportional to the amount of alde- hyde added (Fig. 5). This evidence is not in accord with Strehler's results which indicate (personal communication) that aldehyde acts only by accelerating the luminescent oxidation of FMNH2. It is pos- sible that the different results may be ascribed to the fact that Strehler has used crude acetonized powders rather than a purified enzyme fraction. With the crude powders it is possible that products of the aldehyde reaction may be reconverted into active aldehyde or that 262 PURIFICATION AND PROPERTIES other enzymatic pathways for the utihzation of flavin may exist. This aspect of the problem has not been clarified. It is important to emphasize that aldehyde destruction occurs only under conditions where luminescence appears, i.e., the enzyme prep- aration will not destroy aldehyde unless both DPNH and FMN are present. We agree, however, that the aldehyde accelerates the oxida- tion of FMNHo. It does not, however, accelerate the oxidation of DPNH. Apparently the rate-limiting step is the reduction of FMN by DPNH. ML OODECYL ALDEHYDE Fig. 5. Effect of varying amount of aldehyde added to reaction mixture upon total light emitted during the reaction (McElroy and Green, unpubUshed). The effect of various inhibitors is shown in Table II. The inhibition by p-chloro Hg benzoate is reduced from 58% to 17% by 2 X 10" ^ M glutathione. Although the earlier indication that iron was involved in the system ( McElroy, Hastings, Sonnenfeld, and Coulombre, 1954 ) has not been confirmed, the effect of cyanide and versene suggests that some metal may be functional in the reaction. In spite of the fact that the peak of bacterial emission is at a shorter wavelength than is riboflavin fluorescence, there is good rea- son (Strehler, Harvey, Chang, and Cormier, 1954) to suppose that FMN is the light emitter in the bacterial reaction. Moreover, on the basis of its probable role as light emitter, the above authors designate FMN as bacterial luciferin. On the other hand, we have proposed that aldehyde be classified as bacterial luciferin, since it is destroyed during light emission. Although a complete evaluation of the roles of FMN and aldehyde must await additional experimental studies, the question of the criteria by which luciferin may be defined is raised. J. W. HASTINGS AND W. D. McELROY 263 DuBois ( 1887 ) described lucif erin as a substance in the heat stabile fraction of Pholas extracts which produced hght when an unheated dark fraction containing the enzyme was added. McEboy and Harvey (1951) showed that more than one substance could produce this "luciferin" reaction in fireflies. Harvey (1920) defined luciferin ex- tracted from Cypridum as "the heat resistant dialyzable substance which takes up oxygen and oxidizes with light production in the presence of . . . luciferase." Although he emphasized the chemical dissimilarities of luciferins extracted from different organisms, Cypri- dina luciferin serves as a model for the general mechanism of biolumi- nescent reactions, where the overall reaction is written: LH2 + iOs -> L* + HoO L* ^ L + light Luciferin in this scheme (LHo) embraces at least two features: (1) it is (or forms) the Hght-emitting molecule, and (2) it is the reduced compound which is split in the terminal reaction with oxygen, releas- ing sufficient energy for light emission. Although no compound with a completely analogous role has yet been demonstrated in either the firefly or bacterial systems, the essen- tial feature of the above scheme is the designation of luciferin as the substrate in the terminal oxidation. Studies by McElroy and colleagues indicate that in the firefly system the compound as isolated is more strictly a proluciferin and must undergo preliminary dark reactions, possibly involving energy transfer from ATP, before its final oxidative reaction. Similar reactions may also occur in the bacterial system. In addition, it is possible that luciferin might not be directly involved in light emission. That is to say, the energy from the oxidative split of luciferin might be transferred to and activate another molecule in the system, which would itself be designated as the light emitter. These ideas are in accord with the fact that it has not been pos- sible to reverse the luminescent oxidation of luciferin to reform active luciferin. That the reaction involves a rather drastic split of luciferin might be expected on energetic grounds, since energy sufficient for the emission of a quantum of light is derived from the reaction. It would therefore be expected that, if the general mechanism of bio- luminescence involves the irreversible oxidation of luciferin, the total 264 PURIFICATION AND PROPERTIES light in the isolated system would be proportional to the amount of luciferin present, analogous to the situation in Cypridiiia (Chase, 1948) and the firefly (McElroy, 1951). This has been found to be the case for the long-chain aldehyde in bacterial luminescence. References Chase, A. M. 1948. The chemistry of Cypridina luciferin. Ann. N. Y. Acad. Scl, 49, 353-75. Cormier, M. J., and B. L. Strehler. 1953. The identification of KCF: Re- quirement of long-chain aldehydes for bacterial extract luminescence. ;. Am. Chem. Soc, 75, 4864. DuBois, R. 1887. Note sur la fonction photogenique chez les Pholades. Compt. rend. (Ser. 8), 3, 564-65. Farghaly, A. H. 1950. Factors influencing the growth and light produc- tion of luminous bacteria. /. Cellular and Comp. Physiol, 36, 165-84. Harvey, E. N. 1920. The Nature of Animal Light. J. P. Lippincott, Phila- delphia, Pa. Hastings, J. W., W. D. McElroy, and J. Coulombre. 1953. The effect of oxygen upon the immobilization reaction in firefly luminescence. J. Cellu- lar and Comp. Physiol., 42, 137-50. McElroy, A. D. 1951. Phosphate bond energy and bioluminescence. In Phosphorus Metabolism. Johns Hopkins Press, Baltimore, Md. McElroy, W. D., and A. Green. 1954. Utilization of aldehydes for light pro- duction by bacterial luciferase. Bacteriol. Proc, 96-97. McElroy, W. D., and E. N. Harvey. 1951. Differences among species in the response of firefly extracts to adenosine triphosphate. /. Cellular and Comp. Physiol, 37, 1-7. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1953. The requirement of riboflavin phosphate for bacterial luminescence. Science, 118, 385-86. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1954. Partial purification and properties of bacterial luciferin and luciferase. /. Bacteriol, 67, 402-408. Strehler, B. L. 1953. Luminescence in cell-free extracts of luminous bacteria and its activation by DPN. /. Am. Chem. Soc, 75, 1264. Strehler, B. L., E. N. Harvey, J. J. Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Proc. Natl. Acad. Sci. U. S., 40, 10-12. Inhibition and Activation of Intracellular Luminescence* Frank H. Johnson Department of Biolog> ', Princeton University, Princeton, New Jersey For the purposes of this discussion we may define intracellular lumi- nescence as visible light emission resulting from metabolic processes within intact, living cells, such as bacteria, independently of nervous or other physiological mechanisms that regulate the luminescence of tissues and organs in complex organisms, such as the firefly. An under- standing of how various factors operate to inhibit or to activate intra- cellular luminescence is obviously an important aspect of the total problem of bioluminescence in particular. Of no less importance are the contributions that advances in the understanding of this particular process have made, or may make, toward the understanding of the general problem of inhibition and activation in various other biologi- cal processes, both relatively simple and highly complicated. It seems appropriate, therefore, to consider both these aspects, and in the dis- cussions which follow, certain principles or concepts, which were either first recognized clearly, or whose establishment has been especially aided through studies of bioluminescence, are hsted, together with a few examples indicating the applicability of these concepts to processes other than luminescence. Intracellular luminescence is limited first of all, of course, by genet- ically determined potentialities: the catalytic machinery as well as other factors essential for fight emission must be present. Among lumi- nous bacteria, mutant strains that emit no visible light, or that exhibit * This paper has been aided by a contract between the Office of Naval Research, Department of the Navy, and Princeton University. 265 266 INTRACELLULAR LUMINESCENCE various degrees of brightness as compared to the parent strain under similar conditions, have long been known (Beijerinck, 1912, 1915) and have been studied further in recent years (Doudorolf, 1942; Giese, 1943; McElroy and Farghaly, 1948; Miller, Farghaly, and McEl- roy, 1949; McElroy and Friedman, 1951). With the genetic potentiality for luminescence, the intrinsic bright- ness, i.e., the average intensity of light emitted per cell, may vary enormously at different stages of the growth cycle, or with environ- — 50 1 1 1 1 1 \ T 2 4 6 8 10 12 14 TIME IN HOURS LUM. INT. • • CELL COUNT »o 28* C — t Fig. 1. Rate of increase in cell numbers and in luminescence intensity during aerobic growth of a species of luminous bacteria in 3% NaCl-nutrient broth at 28° C. (After Baylor, 1949.) mental factors such as temperature of incubation and composition of the medium. Beijerinck (1915) noted that, at relatively high tem- peratures, growth of Pliotobacteriiim splendidum will take place, accompanied by very little luminescence. When dark cultures, grovm at above optimal temperatures, are placed at a lower temperature favorable for luminescence, light does not immediately appear, al- though subcultures grow and luminesce at the lower temperature. Along with these studies Beijerinck made the interesting observation FRANK H. JOHNSON 267 that a suspension of brightly kiminescent cells could be momentarily exposed to a temperature high enough practically to extinguish the light, followed by a more or less complete recovery of luminescence on cooling. Much later, this phenomenon was independently redis- covered (Johnson, Brown, and Marsland, 1942) and provided a key to the interpretation of certain temperature-pressure relationships of luminescence as discussed presently. Quantitative studies on the intensity of luminescence during growth 1 1 1 1 1 1 T I 2 3 4 9 6 T TIME IN HOURS 50 LUM. INT. > • CELL COUNT • O 35 * C k 30 •20 ^ • ^ o .10 X is -I z ui 5 - e J. Fig. 2. Rate of increase in cell numbers and in luminescence intensity during aerobic growth of a species of luminous bacteria ( same as Fig. 1 ) in nutrient broth at 35° C. (After Baylor, 1949.) of a luminescent species at different temperatures have shown that the rate of development of luminescence may differ markedly from the rate of production of cells (Figs. 1 and 2; Baylor, 1949). With other things the same, the temperature coefficient for the overall process of growth and cell division is evidently different from the temperature coefficient for the overall process of the production and functioning of the luminescent system. As a result, the intrinsic bright- ness per cell may vary many-fold within a growth period of only a few hours. 268 INTRACELLULAR LUMINESCENCE At a given temperature, the composition of the medium, both with respect to inorganic salts and nutrient constituents, can also pro- foundly modify the expression of genetic potentialities for lumines- cence. With minimal amounts (about 0.5%) of NaCl to permit growth of A. fischeri in nutrient broth, no visible luminescence occurs (War- ren, 1945; Johnson, 1947; Farghaly, 1950), although subcultures in approximately isotonic (3%) NaCl luminesce normally. Complicated 140- A LUMINESCENCE J \ \ L GROWTH-—, U- LUMINESCENCE i I \ L 20 100 SO 60 40 20 TIME (hours) Fig. 3. Rate of growth and of increase in luminescence intensity of A. fischeri on a basal medium consisting of inorganic salts plus glycerol (A), and on the same medium with the addition of 3 micrograms of methionine per ml (B). The points on the ordinate are expressed in terms of per cent, with the highest value of luminescence and the corresponding amount of growth on the basal medium without the added methionine arbitrarily taken as 100%. (After Farghaly, 1950.) relationships have been observed between the luminescence of cul- tures and the types as well as concentrations of salts in the medium (reviewed by Harvey, 1940, 1952). The same is true of specific nutri- tive substances, as shown in quantitative studies by Farghaly (1950); for example, the addition of small amounts of methionine to a basic medium practically eliminates the lag in light production as com- pared to growth ( Fig. 3 ) . Among different species of luminous bacteria, the maximum bright- FRANK H. JOHNSON 269 ness of a culture, under the most favorable conditions known, varies widely. Psychrophilic, marine species, such as Photohacteriiim phos- phoreum, exhibit perhaps the most brilHant luminescence. The lumi- nescence of certain mesophihc, marine species, such as Photobac- terium splendidum or its very close relative, A. harveyi (which should perhaps be considered a subspecies and designated "Photohacteriiim splendidum harveyi") is greatly enhanced by the addition of fish ex- tract to the medium. The significance of the fish extract is not yet known, but the occurrence in fish of vitamin A in the form of the palmitate suggests a possible interpretation, viz., that of providing a source of palmitic aldehyde, very small concentrations of which have recently been found to increase by more than 100-fold the intensity of luminescence in certain extracts of luminous bacteria (Strehler, 1953; Cormier and Strehler, 1953). Turning now to the problem of inhibition and activation of lumi- nescence in "mature" cells, after they have grown and developed their light-emitting system, the following list includes some of the chief principles that are more or less of general interest. Since the evidence and theory are discussed at length elsewhere (Johnson, Eyring, and Pohssar, 1954), they need not be considered in detail here. Reversible Denaturation of Intracellular Enzymes Intracellular luminescence, Hke essentially all biological processes, exhibits an optimum temperature, or temperature for maximum, over- all reaction rate. The actual temperature varies somewhat among dif- ferent species, and within a single species it may be reversibly raised or lowered by physical and chemical changes in the environment of the cells. Under given conditions, a reaction of fundamental impor- tance in determining the optimum temperature, and in part the tem- perature activity curve, is the reversible thermal denaturation of an essential enzyme. Qualitative evidence for this reaction resides in the ready reversibilit>% by cooling, of the diminution in luminescence in- tensity during momentary exposures to temperatures well above the normal optimum. Analyses of quantitative data, relating the amount of reversible diminution in intensity to various temperatures above the optimum, indicate that a single reaction, characterized by the high heat and entropy typical of protein denaturation, is primarily 270 INTRACELLULAR LUMINESCENCE responsible for the thermal diminution. The rapidity of the changes in intensity, on either heating or cooling, is indicative of a mobile equilibrium. The simplest interpretation is that an equilibrium exists between the native (active) and denatured (inactive) form of an enzyme essential to the overall process of hght emission. This equi- oo ^ 5 AH^= 34,000 A// = 80,000 AS = 266.12 0.587 e 66 66 •C35 30 25 20 I I I I, I I 15 I I I I, 10 5 I 1,1 I I I 0.0032 0.0033 0.0034 0.0035 0.0036 Reciprocal of the absolute temperature Fig. 4. Influence of temperature on the intensity of luminescence in A. fischeri. The smooth curve was calculated in accordance with the equation and con- stants as given in the figure. (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons; data of Johnson, Eyring, and Williams, 1942.) FRANK H. JOHNSON 271 librium reaction, together with the catalytic reaction of a limiting enzyme, are sufficient to account for a major part of the temperature- activity curve. For convenience of discussion, we may designate the equilibrium constant for the reversible denaturation as Ki, and the specific rate constant of the catalytic reaction as ki. Figure 4 illustrates the observed intensity (by visual photometry) of steady-state luminescence in a suspension of A. fischeri cells during brief exposures to various temperatures, above as well as below the optimum. The soHd line is a curve calculated in accordance with the equation and constants given in the figure, assuming only the two reactions, with constants Ki and ki, referred to above. Although the theory is obviously oversimplified, the curve fits the data within the limits of experimental error, except at the highest temperatures where destructive reactions with high-temperature coefficients complicate the simplified picture. With some other species of bacteria, it has not proved possible to describe corresponding data with the same accu- racy, showing again that the theory is oversimplified in not including additional reactions which appreciably influence the quantitative var- iation of the overall process with temperature. Among other processes, the simplified theory describes with con- siderable accuracy the rate of reproduction of Escherichia coli as a function of temperature, from somewhat above to well below the normal optimum of 37° to 39° under the conditions involved (John- son and Lewin, 1946). Bacteriostasis occurs at about 45° C, but growth is immediately resumed on cooling to 37° C. Analysis of the data indicates that more than one equilibrium reaction is involved in the reversible bacteriostasis at high temperatures, although most of the temperature-activity curve can be accounted for on the same basis as that of bacterial luminescence. Volume Changes of Activation (AV*) in Enzyme-Catalyzed Reactions Reactions involving only small molecules are not likely to be accom- panied by volume changes of activation exceeding a few cubic centi- meters per mole (cf. Stearn and Eyring, 1941). With large molecules, such as proteins and enzymes, there is the possibility of large volume changes of activation, of the order of 50 to 100 cc per mole, depend- 272 INTRACELLULAR LUMINESCENCE ing upon the mechanism of reaction. When volume changes of this magnitude occur, the reaction rate constant is markedly affected by moderately increased hydrostatic pressures of a few hundred atmos- pheres. The steady-state luminescence of P. phosphoreum at a low temperature, where Ki is neghgible, is reversibly reduced by increased 20 P. phosphoreum atm68 136 204 272 340 408 — - 7'=25°C — P. phosphoreum >^ A. fischeri s ^"■^^^ * — ^^.^4. harueyi — *^N^ * atm68 136 204 272 340 40S 2,000 4,000 6,000 0 2,000 4.000 Hydrostatic pressure (psi) (A) (B) 6.000 Fig. 5. (A) Intensity of Inminescence of P. phosphoreum as a function of pres- sure at different temperatures. The intensity at normal pressure is arbitrarily taken equal to 100 at each temperature in order to show the per cent change in intensity with change in pressure. (B) Influence of increased pressure on the luminescence of three different species of bacteria at 25° C. (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons; data of Brown, Johnson, and Marsland, 1942. ) pressure (Fig. 5A). The data indicate that the reactions leading to light emission proceed with a net volume increase of activation amounting to about 50 cc per mole at 0° C. There are, as yet, veiy few data pertaining to pressure effects on well-defined, purified enzyme systems, for comparison to the lumines- cence data. Studies with crystallized chymotrypsin and crystallized FRANK H. JOHNSON 273 trypsin, however, have shown that the value of AV^ in the hydrolysis of synthetic peptides varies with the substrate, and may be negative or zero (VVerbin and McLaren, 1951a,b). The observed effect of pressure on the steady-state level of lumines- cence becomes less as the temperature is raised toward the optimum, and at still higher temperatures the level increases, rather than de- creases, under pressure. The family of curves in Fig. 5A resembles a family of curves obtained by Professor D. E. Brown (cf. Johnson, Eyring and Polissar, 1954) for the tension of auricle muscle as a function of temperature and pressure. Figure 5B shows that the observed effect of pressure at a given tem- perature depends upon the specific biological system involved. The three species represented in this figure normally exhibit different tem- perature optima for luminescence, i.e., around 20° in P. phosphoreum, 26° in A. fischeri, and 30° C in A. harveyi. In each case, the effect of pressure is in accordance with what one would expect on the basis that pressure diminishes the intensity of luminescence at temperatures below that of the specific optimum and augments the intensity at temperatures above the specific optimum, with little effect at the op- timum. A pressure-temperature relationship of this kind was first rec- ognized by Brown (1934; 1934-35) in studies with muscle. Volume Change of Reaction (aV) in Reversible Denaturation The opposite effect of pressure at high and low temperatures, re- spectively, as illustrated in Fig. 5A, shows that the limiting reactions are not the same at these different temperatures. Since the increase in steady-state luminescence under pressure becomes greater as Ki of the reversible denaturation becomes greater with rise in temperature, the simplest interpretation is that pressure causes a partial reversal of the denaturation of the enzyme involved. In other words, the equilibrium change from native to reversibly denatured forms of the enzyme is accompanied by a volume increase of reaction. The data indicate that the value of aV in this reaction amounts to about 65 cc per mole at 35° C. If we assume as before that there are two reactions primarily limit- ing the overall luminescent reaction, viz., the catalytic reaction of a limiting enzyme and the reversible denaturation equilibrium, tem- 274 INTRACELLULAR LUMINESCENCE perature activity curves can be computed for various pressures. In order to obtain a close fit to the data, however, it is necessary to take into account a temperature dependence of aV* and of AV respectively (Fig. 6), showing again that the simple theory does not include all the reactions that are of importance. It is interesting to note ( Fig. 6 ) that the maximum intensity of luminescence remains practically the same at different pressures, but the temperature at which the maxi- 0.00324 0.00330 0.00336 0.00342 0.00348 0.00354 0.00360 0.00366 Reciprocal of the absolute temperature Fig. 6. The brightness of luminescence in P. phosplioreum as a function of tem- perature at three different hydrostatic pressures. The points represent data from experiments by Brown, Johnson, and Marsland (1942). The smooth curves were calculated by Eyring and Magee (1942). (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons.) mum intensity occurs is higher under increased than at atmospheric pressure. Although data pertaining to the influence of pressure on the re- versible denaturation of isolated enzyme systems are still scarce, there is evidence that highly purified trypsin, with casein as substrate, un- dergoes a reversible, thermal denaturation accompanied by a consid- erable volume increase of reaction (Fraser and Johnson, 1951). FRANK H. JOHNSON 275 Volume Changes of Activation in Irreversible Protein Denaturation At temperatures above the normal optimum, the luminescent sys- tem is destroyed at rates that increase very rapidly with rise in temperature. These rates have the characteristics of a first order reac- tion, and their increase with temperature resembles the thermal de- naturation of proteins. At a given temperature, however, the rate decreases with rise in pressure (Fig. 7) by an amount indicative of a volume increase of activation of about 70 cc per mole. Thus, it ap- pears that pressure not only acts to reverse the equilibrium change from native to denatured states of an essential enzyme, as discussed above, but also pressure retards the rate of thermal destruction. When these interpretations, based entirely on kinetic data pertain- ing to intracellular luminescence, were first expressed, they had no parallel in previous experiments with isolated proteins. Subsequently, it has been shown that the thermal denaturation of highly purified human serum globulin is retarded by pressures up to 10,000 psi (John- son and Campbell, 1945; 1946), and likewise, tobacco mosaic virus (Johnson, Baylor, and Fraser, 1948). The inactivation of bacterio- phage T5 at 68° C (Foster, Johnson, and Miller, 1949) and the disin- fection of Bacillus subtilis spores at 92.5° to 93.6° C (Johnson and ZoBell, 1949a) are also retarded by increased pressure. The heat inac- tivation of bacteriophage T7 at 60° C, however, is accelerated under pressure. Moreover, there is essentially no effect of pressure on the salicylate denaturation of methemoglobin (Schlegel and Johnson, 1949), so it is not a general rule that pressures up to about 10,000 psi retard protein denaturation at increased temperatures or in the pres- ence of denaturants. Promotion of Reversible Protein Denaturation by Narcotics The sensitivity of luminescence to narcotics has long been known. In 1672, Robert Boyle reported that the light of "shining flesh" was quickly extinguished by "pouring on it a little pure spirit of wine." Modern studies of the temperature relationships in the inhibition of bacterial luminescence by alcohol, urethan, or other members of the lipid-soluble group of narcotics have provided evidence that the mechanism of their physiological effect involves, in part, promotion 276 INTRACELLULAR LUMINESCENCE 100' 1 fW ▼j^j^ 90 -Vv B A ^^^^ ^^^-^^^^^^^^^ Pressure (psi) 80 — \» \^ I^J^-^^ 5.000 4.000 70 ^^ ^ ^ 3.000 60 — \ ^ 50 c 9) *^ c £ 40 °0^ ( \ * * >• c E 3 -1 . 30 ^\^ » • \ 25 — o \ ^ \ 2,000 20 34'C \ \ 1.000 \ CO 15 1 1 1 1 ^ Control I 1 0 12 3 4 5 6 7 Minutes Fig. 7. The influence of hydrostatic pressure on the relati\e rate of destruction of the luminescent system of P. plwsplioreum suspended in phospliate-buffered sodium chloride of neutral pH at 34° C. ( From Johnson, Eyring, and Polis- sar, 1954, courtesy of John Wiley & Sons; data of Johnson et al., 1945.) of the reversible thermal denaturation of an essential enzyme. In effect, the drug combines with the reversibly denatured form, and the amount of denatured enzyme then becomes significant at a lower FRANK H. JOHNSON 277 temperature. Thus, with alcohol, for example, the inhibition caused by a given drug concentration increases with rise in temperature (Fig. 8). Stimulatory' efiFects of low concentrations of the drug at low tem- peratures, as seen in Fig. 8 or encountered under other conditions, must involve other reactions. Two interesting results of the denaturation-promoting mechanism of action are that, in the presence of alcohol, ( 1 ) the measured activa- tion energy for luminescence decreases, and (2) the temperature of 0.0033 0.0034 0.0035 Reciprocal of the absolute temperature 0.0036 Fig. 8. The influence of temperature on the inhibition of luminescence in P. phosphoreum by different concentrations of ethyl alcohol at neutral pH. Alcohol concentrations from uppermost to lowermost curve: 0, 0.4, 0.5, 0.6, 0.8 M. (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons; data of Johnson et al., 1945.) 278 INTRACELLULAR LUMINESCENCE maximum luminescence intensity is lowered. Similar effects have been obtained with the urethan inhibition of oxygen consumption and methylene blue reduction by Rhizobium trifoUi (Koffler, Johnson, and Wilson, 1947). The action of higher members of a homologous series of carbamates on luminescence, however, apparently involves a mech- anism different from the one just discussed (Johnson, Flagler, Simp- son, and McGeer, 1951). Reversal of Narcosis by Increased Hydrostatic Pressure According to the theory briefly described above, narcotics such as alcohol and urethan promote a reversible denaturation that involves a large volume increase of reaction in going from the initial (native) state to the final (denatured) state of an essential enzyme. It is not surprising, on this basis, to find that the inhibition caused by these drugs is partially counteracted by increased pressure. Figure 9 illus- trates the results of applying hydrostatic pressure to aliquot portions of a suspension of luminous bacteria containing different concentra- tions of alcohol, at the normal optimum temperature where pressure has little effect on the intensity of luminescence in the absence of the drug. At low concentrations of alcohol, the inhibition is virtually abolished by pressure. With higher concentrations, the effects of pres- sure are more complicated, possibly because of multiple equilibria involved in the action of alcohol (vide infra), but in each case the amount of inhibition is reduced by pressure. It is interesting to note also that the "optimum pressure" in the presence of alcohol is rever- sibly shifted by as much as several thousand pounds per square inch, depending upon the concentration of alcohol. Although the narcosis of animals is a much more complicated phe- nomenon than the inhibition of bacterial luminescence, the mech- anisms involved at the molecular level are not fundamentally different. It is not surprising, therefore, that salamander and frog larvae, when narcotized in water containing appropriate concentrations of alcohol or urethan, recover their activity and swim in apparently normal man- ner when the pressure is increased from atmospheric up to between 2000 and 5000 psi (Johnson and Flagler, 1951). Moreover, observa- tions made thus far indicate a close correlation between the action of pressure on the inhibition of bacterial luminescence and on the FRANK H. JOHNSON 279 Control 100' • ^..-^-^ — • ^^*" 90 < J^^M y^^ ^ 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 Pressure (psi) Fig. 9. Luminescence intensity of P. phosplioreum as a function of hydrostatic pressure at 17.5° C, in the presence of various concentrations of alcohol in the phosphate-buffered sodium chloride solution, pH 7, of the suspending medium. ( From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons; data of Johnson et ah, 1945. ) narcosis of these aquatic animals, respectively, by various drugs, i.e., the effects of different dmgs on both luminescence and narcosis ex- hibit varying degrees of sensitivity to pressure that are at least quali- tatively the same in the two phenomena. Multiple Equilibria in Reversible Effects of Narcotics At a given temperature and pressure, the relation between drug concentration and amount of inhibition of luminescence by alcohol or urethan often conforms rather closely to the relationship which would be expected if the drug entered into a single equilibrium combination with the enzyme affected. The relationship should be 280 INTRACELLULAR LUMINESCENCE linear, when plotted in the manner illustrated in Fig. 10, the slope of the line indicating the ratio of drug to enzyme molecules in the equilibrium established. In Fig. 10, it is evident that the relationship is not linear throughout a wide range of drug concentrations, for the slope increases at the higher concentrations. Moreover, the numerical 2.0 1.0 Jo.o o 10 Luminescence P. phosphoreum pH7 1 — Slope = s = / / / yC. / / y ^j^ Slope = 2.8 Slope = 1.0 o--'*'-s = 13 Yeast oxygen consumption 2.0 L 2.5 00 10 15 log jg molar concentration urethan Fig. 10. Relation between concentration of urethan (abscissa) and amount of inhibition (ordinate) of luminescence in P. phosphoreum at 5°, 20°, and 30° C, respectively. The symbol Ti for inhibition represents [(L/Iw) — 1], where /<■ is the intensity of luminescence in a control suspension of bacteria, and /u is the intensity in a corresponding suspension containing a given concentration of the drug. (Data of Johnson et ah, 1945.) The broken line repres*ents data replotted from Fisher and Stearn (1942) concerning the urethan inhibition of oxygen consumption in yeast. (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons.) values of the slopes are not integers, and they change with tempera- ture. They also change with pressure. It follows that more than a single equilibrium is involved in the total effect, and the measured slopes give only the average ratios of the combining molecules. Figure 10 indicates a similarity in the action of urethan on bacterial luminescence and on yeast respiration. The data on the latter process were interpreted to mean that two different enzyme systems are af- FRANK H. JOHNSON 281 fected (Fisher and Steam, 1942). The same results could occur, how- ever, if two or more different equilibria, characterized by different ratios of combining molecules, and very likely also by different heats and entropies of reaction, were established with the same enzyme system. In living cells it is difficult to distinguish between two such possibilities. Action of the drug through multiple equilibrium reac- tions with a single enzyme is the simpler explanation, however, and there is convincing evidence that urethan catalyzes the thermal de- naturation of tobacco mosaic virus through more than a single reac- tion with this protein (vide infra). The denaturation of tobacco mosaic virus in urea at 0° to 40° C also involves multiple reactions between urea and the protein ( Lauffer, 1943 ) . Catalysis of Irreversible Protein Denaturation by Narcotics On general considerations it would appear likely that the same drugs which promote a reversible denaturation of enzymes or other proteins would also under appropriate conditions be found to promote, or catalyze, an irreversible denaturation of the same proteins. Urethan, in fact, has been shown to promote the denaturation of egg albumin and serum proteins at room temperature ( Hopkins, 1930 ) . In concen- trations of physiological interest, urethan catalyzes the destruction of the bacterial luminescent system, at a rate dependent upon the con- centration of the drug at a given temperature (Fig. 11). As a result, the initial inhibition, just after adding the drug, increases progres- sively with time. There are numerous other instances of such a dual action of a drug, which is in marked contrast to the action of others, e.g., of sulfanilamide on bacterial luminescence, wherein the inhibi- tion remains essentially constant with time. The inhibition of respira- tion of rat brain slices by phenobarbital, for example, indicates a dual action similar to that of urethan on bacterial luminescence (cf. analy- sis of the data of Jowett (1938) in Johnson, Eyring, and Polissar, 1954). Pressure Retardation of Narcotic-Catalyzed Denaturation On the basis of the foregoing observations, it would be expected that the irreversible denaturation of one or more enzymes essential to bacterial luminescence, in the presence of certain narcotics, would be 282 INTRACELLULAR LUMINESCENCE 100 80 60 50 40 30 TT Molar cone. Molar cone, sulfanilamide 1 urethan ^0.003 ■♦• 0.10 ■5-» ^ 20 - 10 I 8 0.30 2 - 22.5''C 0.45 10 15 20 Time (mm) 25 30 35 Fig. II. The time course of the inhibition of luminescence in P. phosphoreum at 22.5° C by sulfanilamide (broken lines) and urethan (solid lines), re- spectively, in concentrations that cause initially similar diminutions of inten- sity. (From Johnson, Eyring, and Polissar, 1954, courtesy of John Wiley & Sons; data of Johnson et ah, 1945.) retarded by pressure. Data are not yet available in this regard. It has been shown, however, that pressures of the order of 10,000 psi mark- edly retard the precipitation of human serum globulin at 65° C in the presence of small concentrations of ethyl alcohol which accelerate the rate of precipitation (Johnson and Campbell, 1946). Similarly, FRANK H. JOHNSON 283 small concentrations of urethan catalyze the precipitation of tobacco mosaic virus at 68.8° C in a manner that is retarded by pressure (Fraser, Johnson, and Baker, 1949). The same is true of the thermal disinfection of bacteriophage T5 in the presence of urethan (Foster, Johnson, and Miller, 1949). Finally, the accelerated disinfection of B. subtilis spores in the presence of urethan is retarded by pressure (Johnson and ZoBell, 1949b). Although further examples of this sort of phenomenon may be expected it is not to be looked upon as a general rule, inasmuch as eflEects of pressure depend upon the mech- anism of the reaction, and among a variety of drugs as well as of experimental conditions (temperature, pH, dissolved salts, etc.) the mechanisms of denatiuation, even with the same protein, may be expected to differ in detail. Multiple Reactions in Narcotic-Catalyzed Protein Denaturation Here again, data with respect to intracellular luminescence are not yet available, but it is reasonable to expect certain similarities to the results that have been obtained with other systems in studies stemming from work on bacterial luminescence. A particularly clear example of multiple reactions is afforded by tlie action of urethan in catalyzing the thermal denaturation of tobacco mosaic virus, wherein at least two, and probably three, distinct reactions between urethan and the protein are involved (Fraser, Johnson, and Baker, 1949). Now, since many of die interpretations discussed above have been based primarily on kinetic data pertaining to the luminescence of intact, living cells, additional evidence is needed before these inter- pretations can be considered fully convincing. Such need is met in part by finding, and in some instances anticipating, fundamentally similar phenomena in various processes other than luminescence, i.e., denaturation of purified proteins, reactions of purified proteins, activity of purified enzymes, rates of growth and disinfection, and other exam- ples mentioned. Until quite recently, however, it has not been possible to investigate the influence of different factors — temperature, pressure, narcotics, etc. — on bacterial luminescence in cell-free extracts. Streh- ler's success in obtaining luminescent extracts (Strehler, 1953) has provided a long sought means not only of getting more direct evi- dence than is possible with intracellular luminescence but also of 284 INTRACELLULAR LUMINESCENCE identifying biochemically certain components of the luminescent sys- tem, as well as of purifying the enzymes involved. Rapid progress has already been made toward the latter objectives ( Cormier and Strehler, 1953; McElroy et al, 1953, 1954; Strehler et al, 1954; Strehler and Cormier, 1953). For the remainder of the present discussion, it is appropriate to dwell on results that have been obtained in a joint study ( Strehler and Johnson, 1954) of pressure-temperature-inhibitor relationships of bac- terial luminescence in cell-free extracts. Kinetics of Bacterial Luminescence in vitro The observations described in the following paragraphs were made with extracts of A. fischeri cells, prepared by treating first with cold acetone, and then taking up the dried powder in distilled water, fol- lowed by high-speed centrifugation to remove all particulate matter. The supernatant provided a stock enzyme solution which was diluted with phosphate buffer at neutral pH as needed. The addition of fla- vine mononucleotide (FMN), didhydrodiphosphopyridine nucleotide (DPNH2), and decaldehyde in final concentrations of approximately 0.3 microgram, 1 mg, and 7 micrograms per milliliter, respectively, gave a "saturated system" that emitted a bright luminescence over extended periods of time at room temperature. Figure 12 illustrates the changes in intensity of the saturated system after the sudden application and sudden release, respectively, of hy- drostatic pressure at temperatures below, above, and near that of the optimum (about 26° C) of the system used. One curve (16°, Cells), obtained with a suspension of cells of the same species, is included for comparison to the data obtained with the extracts. Among the several points of interest shown by these data are the following: (1) the changes in luminescence intensity after applica- tion of pressure are reversible upon release of pressure, (2) a fairly sudden but transitory increase in intensity ( "spike" ) occurs on sudden application of pressure, and a fairly sudden but transitory decrease ("dip") occurs on sudden release of pressure, (3) the spikes and dips are followed by relatively slow changes to a new steady-state level under increased or at nomial pressure, respectively, (4) the new steady-state level is lower, nearly the same, or higher than the initial FRANK H. JOHNSON 285 level before applying pressure, according to whether the temperature is below, near, or above the normal optimum temperature of the system, and (5) the steady-state level of luminescence intensity after releasing pressure is the same as the steady-state level before apply- 250 34' 6000 psi ON < 6000 psi OFF \ I I I Fig. 12. Influence of pressure at different temperatures on the luminescence of cell-free extracts of A. fischeri with added DPNH2, FMN, and decaldehyde (saturated system). Arrows pointing upwards indicate the time when pres- sure was applied; arrows pointing downwards, when it was released. The cur\es for rising intensity after release of pressure (except the curve for 34° C) have been arbitrarily displaced to the right on the abscissa, to avoid intersecting hnes. The intensity at normal pressure, just before raising the pressure, is arbitrarily taken equal to 100 at each temperature, and all other points computed relatiNe to this value. Allowance was made for the amount of decay when it was significant during the periods involved, as at the higher temperatures. The increased pressure was 6000 psi at each tempera- ture except 3° C where it was 6500 psi. Data obtained with intact cells of A. fischeri at 16° C are included for purposes of comparison. (After Strehler and Johnson, 1954.) 286 INTRACELLULAR LUMINESCENCE ing pressure, provided correction is made for the decay of lumines- cence during the intervening period of time, and at the higher tem- peratures, provided correction is made also for a rate of thermal inactivation of the system. Evidence of the transitory spikes and dips had been observed in earlier studies with intact cells ( cf . discussion in Johnson, Eyring, and Polissar, 1954), but quantitative records had not been made of them or of the changes in intensity between steady-state levels. Data are not yet available whereby, at various temperatures, quantitative com- parisons can be made between spikes and dips in cells and extracts, respectively. It appears that the transitory changes occur more rapidly in the intact cells, as judged by the curves shown for 16° in Fig. 12. The true magnitudes of the spike and dip in cells at 16° C are uncer- tain because the half-second period of the recording instrument was not fast enough to measure them accurately. Qualitatively, the results obtained with extracts are remarkably similar to those obtained in previous studies with intact cells. Quantitatively, the agreement is close, at the one temperature for which data are presently available, except that the time relations differ, i.e., changes in intensity follow- ing changes in pressure take place faster in cells than in the extracts. Steady-state levels of intensity as a function of pressure at various temperatures are plotted in Fig. 13, from the data of Fig. 12 and similar experiments. For comparison, the pressure-temperature data obtained with cells of P. phosphoreum in the initial study of this relationship in luminous bacteria- (Brown, Johnson, and Marsland, 1942) are replotted in this figure. Corresponding effects of pressure occur at temperatures a few degrees higher in extracts of A. fischeri than in cells of P. phosphoreum, a difference that would be logically expected on the basis of the difference in normal optimal tempera- tures for luminescence in the two species, i.e., the normal optimum is several degrees higher in cells and extracts of A. jischeri than in cells of P. phosphoreum. As indicated earlier in the discussion, the bio- logical effects of pressure may be expected to be related to the tem- perature-activity curve of the specific system and conditions involved. Thus, the increase in steady-state levels under pressure becomes pro- gressively greater as the temperature is increased above the normal optimum of the specific system. The most likely intei-pretation is that FRANK H. JOHNSON 1 287 T T — r— I — I — I — r EXTRACTS. A FISCHERI 1 — I — I — I — I — I — I — r- LIVING CELL^RPHOSPHOREUM _ •<* J I UJ. J I L I I I L_ 2000 4000 I .L 1 6000 90 hy^ 2000 4000 8000 8000 0 PRESSURE (psi) Fig. 13. Steady-state levels of luminescence in tlie saturated system of A. fischeri extracts (left) and in living cells of P. phosphoreum (right). The data on the right are replotted from Brown, Johnson, and Marsland (1942). The initial intensity at normal pressure is arbitrarily taken equal to 100 at each tem- perature, and allowance is made for decay, when significant, in the lumi- nescence of extracts. The lower of the two curves for 26° C was obtained with an enzyme solution that had stood for several days at room tempera- ture. (After Strehler and Johnson, 1954.) which has ah-eady been expressed, namely, that at these relatively high temperatures, pressure acts to reverse an equilibrium change from native to denatured states of an essential enzyme (or enzymes), thereby increasing the overall velocity of the light-emitting reaction. Qualitative evidence for a reversible thermal denaturation of the extiacted enzyme system was found by visual observation: diminu- tion in intensity occurs on momentary exposures of luminescent solu- tions to relatively high temperatures, followed by recovery on cool- ing. At low temperatures, where the amount of thermally denatured enzyme is negligible, pressure merely reduces the steady-state level. 288 INTRACELLULAR LUMINESCENCE The straightness of the Hne for S° C in the left-hand part of Fig. 13 indicates that a single reaction is primarily affected, and the slope of this line indicates that this reaction proceeds with a volume in- crease of activation of 86 cc per mole. With the extracted system, it is possible to interpret in further detail the influence of pressure on the reactions involved in lumines- cence. Omitting the complications introduced by pressure-sensitive equilibria between native and thermally denatured forms of essential enzymes, and considering only the effects observed at temperatures below the optimum, the evidence favors the view that the large vol- ume increase of activation is associated with the reduction of FMN by DPNHi;, and it is for this reason that the steady-state level at low temperatures is reduced by pressure. In the absence of added DPNH2, a rapidly decaying luminescence occurs on addition of reduced FMN to a solution of enzyme plus aldehyde. The rapidity of decay makes it difficult to determine clearly whether or not there is any change under pressure corresponding to the change in steady-state levels of the complete system under pressure. The data clearly reveal, how- ever, that sudden increases and sudden decreases in intensity accom- pany the sudden application and release, respectively, of pressure during the rapid decay of luminescence. The magnitude of these sudden increases and decreases corresponds to that of the spikes and dips, and careful analyses of the data failed to reveal evidence of changes which could be interpreted as corresponding to effects of pressure on steady states. Thus, it appears that the spikes and dips are associated with the luminescent oxidation of the flavin component, which appears to proceed with a small volume decrease of activation, of the order of —10 cc per mole. Further interpretations require taking into consideration the influ- ence of pressure on the rate of change in lumnescence intensity be- tween two different steady-state levels due to a change in pressure. With the saturated system at temperatures below the normal opti- mum, the rates of change, after application cf pressure, between the spike peaks and the lower steady-state levels, conform to first order kinetics. The same is true for the rates of change, after release of pressure, between the bottoms of the dips and the higher steady-state levels. In Fig. 14 are plotted representative data which show that FRANK H. JOHNSON 289 TIME (MIN) Fig. 14. Rate of change in intensity of luminescence in the saturated system following application (on) and release (off) of pressure at different tem- peratures. The half time (f^^) for the change between steady states is indi- cated below the respective lines. The lines, plotted relative to an arbitrary value of 100 at t = 0, represent the progressive difference in intensity, with time, between the peak of the spike and steady-state level under pressure, and the progressive difference, with time, between the lowest point of the dip and the steady-state level after pressure, respectively. although the rates increase with rise in temperature, they do not vary significantly with the amount of pressure at a single temperature. The diflFerences in rates under increased as compared to normal pressure are small but perhaps significant. - The following simplified diagram is helpful in picturing the rela- tionship of some of the reactions involved in luminescence of the saturated system at low temperatures: A ki B — -> C + light C In this diagram, A stands for DPNHo + FMN, B stands for DPN + FMNHo, and C and C stand for the products of the luminescent and 290 INTRACELLULAR LUMINESCENCE of all nonliiminescent reactions, respectively, of FMNH2. The k's stand for specific rate constants of the reactions as indicated. With DPNH2 present in excess in the experiments described, its concentration remains essentially constant over short periods of time. When the pressure is changed, the rate constants change immedi- ately, in accordance with the amount of pressure and the respective values of the volume change of activation constants. On the basis of the evidence at hand, fci is characterized by a large volume increase of activation, whereby it becomes smaller, and the steady-state inten- sity therefore lower, under increased pressure. Similarly, ^2 is charac- terized by a small volume decrease of activation, whereby it becomes slightly larger under pressure than at normal pressure, thus giving rise to transitory spikes and dips. (At elevated temperatures the much greater magnitude of the spikes and dips is presumably due to the relation of denaturation equilibria of one or more enzymes to pres- sure.) The rate of change between steady states, with change in pressure, is given by the exponential factor in the following equation, derived in line with the theory of consecutive first order reactions dis- cussed elsewhere (Johnson, Eyring, and Polissar, 1954; Strehler and Johnson, 1954): / = um = M.{rBo] - ^>-<'-"' + jffi; In this equation, I represents the intensity of luminescence, b is a. proportionahty constant. Bo is the initial concentration of B at the time when pressure is applied or released, and the k's refer to the reactions as diagrammed above. It will be noted that reactant con- centration does not enter into the exponential factor; the rate of change between steady states is exp. - {ko + h)t equals the slopes of the lines in Fig. 14. Since kn is sensitive to pressure, whereas (^2 + ^3) is not appreciably sei^sitive to pressure, it follows either (1) that ^3 is very much larger than k. or that (2) compensatory changes in fco and ^3 occur under pressure. Of these alternatives, the former appears far more probable. Among other reasons, unless h were much larger than ko, the quantum efficiency would be extraordi- narily high. The nonluminescent, auto-oxidation of reduced FMN probably accounts, in large part, for ^3. FRANK H. JOHNSON 291 Although extensive studies concerning the action of inhibitors on the luminescence of bacterial extracts have not been undertaken as yet, exploratory experiments with the saturated system have yielded some interesting results. Thus, sulfanilamide reduces the steady-state level of luminescence, and the amount of inhibition at a given con- centration of the drug decreases with rise in temperature, whereas it is only slightly affected with rise in pressure, as in cells. The actual concentration required for a given per cent inhibition in extracts of A. fischeri, however, is many times higher than that required in cells of this species, possibly because of the presence, in these extracts, of large amounts of inert material on which the sulfanilamide adsorbs. Ethyl alcohol or urethan also inhibit the steady-state luminescence of extracts, at concentrations that are only 2 or 3 times greater than those required for corresponding per cent inhibitions of cellular lumines- cence under similar conditions of temperature and hydrostatic pres- sure. The inhibitory concentrations of alcohol or urethan, however, are considerably higher than those of sulfanilamide, and it is reason- able to suppose that adsorption on inert substances has less effect in altering the concentration initially added to the solution. As in cells, the inhibitory effect of a given concentration of alcohol or urethan on steady-state luminescence increases with rise in temperature and de- creases with rise in pressure. These drugs also affect the spikes: at optimum temperature and normal pressure, the spike is markedly higher in the presence of inhibitory concentrations of either drug. Subinhibitory concentrations of alcohol increase the steady-state in- tensity of luminescence in extracts. Increases in the steady-state inten- sity of cellular luminescence, in the presence of alcohol under certain conditions, have also been noted (cf. Fig. 8). The role of the aldehyde remains to be clearly established. In the absence of added aldehyde, the pressure-temperature relationships are different, but the intensity of luminescence is so weak that these rela- tionships have not been determined with a desirable accuracy by means of the methods available in the initial study. The evidence at hand indicates that unless adequate concentrations of aldehyde are present, steady-state levels of luminescence are not significantly low- ered by pressure at low temperatures, and the magnitude of the spikes is somewhat less. Thus, the limiting reactions are not the same 292 INTRACELLULAR LUMINESCENCE in the absence of aldehyde. With the addition of minimal concentra- tions to cause a fairly bright luminescence, the kinetics of the chang- ing intensity due to pressure is again different. It is difficult at the moment, to arrive at an explicit hypothesis that provides a uniquely probable explanation of these observations. The correspondence between the pressure-temperature-inhibitor relationships of luminescence in living cells and in the saturated extracted system argues for the fundamental similarity of the lumines- cent process in cells and in these extracts. Moreover, this correspond- ence shows that the earlier observations, made with intact cells and interpreted in terms of simple systems, can actually be attributed to the operation of relatively simple systems, rather than to extremely complicated relationships existing in the highly organized chemical environment of living cells, and resembling simple reactions only by chance. Thus, justification is provided for the conclusions reached earlier in regard to intracellular luminescence, and at the same time, for further studies with the living cells, which in a number of respects are more convenient to use than extracts. It is only with extracts, of course, that the biochemistry of the system can be firmly established, and that detailed analysis of the physical chemistiy of the system can be made under fully controllable conditions. A great deal can be learned also, however, from studies of the relatively simple process within the immensely complex setting of a living cell. The study of luminescence, in vivo as well as in vitro, is of general as well as of particular interest. ACKNOWLEDGMENTS The author is indebted to Professor W. J. Kauzmann for the derivation of the equation on page 290, and for critical discussions of the kinetics of luminescence in bacterial extracts. The views expressed in this paper are those of the author, and they do not necessarily coincide with those of all the authors and co-authors of the papers referred to in this paper. FRANK H. JOHNSON 293 References Baylor, E. R. 1949. The growth cycle of luminous bacteria on Hmited sub- strate. Thesis. Princeton University. Beijerinck, M. W. 1912. Mutation bei Mikroben. Folio Microbiol, I, 4-100. Beijerinck, M. W. 1915. Die Leuchtbakterien der Nordsee im August und September. Folio Microbiol, 4, 1-26. Boyle, R. 1672. Some observations about shining flesh, both of veal and of pullet, and that without any sensible putrefaction in those bodies. Phil. Trans. Roy. Soc. London, No. 89, 5108. Brown, D. E. 1934. The pressure-tension-temperature relation in cardiac muscle. Am. J. Physiol, 109, 16. Brown, D. E. 1934-35. Cellular reactions to high hydrostatic pressure. Annual Report of Tortugas Laboratory, Carnegie Institution of Washing- ton, pp. 76, 77, issued Dec. 13, 1935. Brown, D. E., F. H. Johnson, and D. A. Marsland. 1942. The pressure- temperature relations of bacterial luminescence. /. Cellular and Comp. Physiol, 20, 151-68. Cormier, M. J., and B. L. Strehler. 1953. The identification of KCF: require- ment of long-chain aldehydes for bacterial extract luminescence. /. Am. Chem. Soc, 75, 4864-65. > ,Doudoroff, M. 1942. Studies on the luminous bacteria. I. Nutritional require- ments of some species, with special reference to methionine. /. Bacteriol, 44, 451-59. Eyring, H., and J. L. Magee. 1942. Application of the theory of absolute reaction rates to bacterial luminescence. /. Cellular and Comp. Physiol, 20, 169-77. Farghaly, A. 1950. Factors influencing the growth and light production of luminous bacteria. /. Cellular and Comp. Physiol, 36, 165-83. Fisher, K. C, and J. R. Stearn. 1942. The separation of an "activity" metabo- lism from the total respiration of yeast by the effects of ethyl carbamate. J. Cellular and Comp. Physiol, 19, 109-22. Foster, R. A. C, F. H. Johnson, and V. K. Miller. 1949. The influence of hydrostatic pressure and urethane on the thermal inactivation of bacterio- phage. /. Gen. Physiol, 33, 1-16. Eraser, D., and F. H. Johnson. 1951. The pressure-temperature relationship in the rate of casein digestion by trypsin. /. Biol. Chem., 190, 417-21. Eraser, D., F. H. Johnson, and R. S. Baker. 1949. The acceleration of the thermal denaturation of tobacco mosaic virus by urethan at normal and increased pressure. Arch. Biochem., 24, 314-20. Giese, A. C. 1943. Studies on the nutrition of dim and bright strains of a species of luminous bacteria. /. Bacterial, 46, 323-31. 294 INTRACELLULAR LUiMINESCENCE Harvey, E. N. 1940. Livnng Light. Princeton University Press, Princeton, N.J. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Hopkins, F. G. 1930. Denaturation of proteins by urea and related sub- stances. Nature, 126, 328-30, 383-84. Johnson, F. H. 1947. Bacterial luminescence. Advances in Enzymol, 7, 215-54. Johnson, F. H., M. B. Baylor, and D. Eraser. 1948. The thermal denatura- tion of tobacco mosaic virus in relation to hydrostatic pressure. Arch. Biochem., 19, 237-45. Johnson, E. H., D. E. Brown, and D. A. Marsland. 1942. A basic mechanism in the biological effects of temperature, pressure and narcotics. Science, 95, 200-03. Johnson, F. H., and D. H. Campbell. 1945. The retardation of protein denaturation by hydrostatic pressure. J. Cellular and Comp. Physiol., 26, 43-46. Johnson, F. H., and D. H. Campbell. 1946. Pressure and protein denatura- tion. /. Biol. Chem., 163, 689-98. Johnson, F. H., H. Eyring, and M. J. Polissar. 1954. The Kinetic Basis of Molecular Biology. John Wiley & Sons, New York. Johnson, F. H., H. Eyring, R. Steblay, H. Chaplin, C. Huber, and G. Gherardi. 1945. The nature and control of reactions in bioluminescence. With special reference to the mechanism of reversible and irreversible inhibitions by hydrogen and hydroxyl ions, temperature, pressure, alcohol, urethane, and sulfanilamide in bacteria. /. Gen. Physiol, 28, 463-537. Johnson, F. H., and E. A. Flagler. 1951. Activity of narcotized amphibian larvae under hydrostatic pressure. /. Cellular and Comp. Physiol, 37, 15-25. Johnson, F. H., E. A. Flagler, R. Simpson, and K. McGeer. 1951. The in- hibition of bacterial luminescence by a homologous series of carbamates. /. Cellular and Comp. Physiol, 37, 1-14. Johnson, F. H., and I. Lewin. 1946. The growth rate of E. coli in relation to temperature, quinine and coenzyme. /. Cellular and Comp. Physiol, 28, 47-75. Johnson, F. H., and C. E. ZoBell. 1949a. The retardation of thermal disinfec- tion of Bacillus suhtilis spores by hydrostatic pressure. }. Bacterid., 57, 353-58. Johnson, F. H., and C. E. ZoBell. 1949b. The acceleration of spore disinfec- tion by urethan and its retardation by hydrostatic pressure. /. Bacterid, 57, 359-62. Jowett, M. 1938. The action of narcotics on brain respiration. /. Physiol (London), 92,322-35. Koffler, H., F. H. Johnson, and P. W. Wilson. 1947. Combined influence of FRANK H. JOHNSON 295 temperature and urethan on the respiration of Rhizobium. }. Am. Chem. Soc, 69, 1113-17. Lauffer, M. A. 1943. The denaturation of tobacco mosaic virus by urea. II. Kinetic aspects. /. Am. Chem. Soc, 65, 1793-1802. Linderstr0m-Lang, K. U. 1952. Lane Medical Lectures. Stanford University Press, Stanford, Cahf. Linderstr0m-Lang, K. U., and C. F. Jacobsen. 1941. The contraction accom- panying enzymatic breakdown of proteins. Compt. rend. trav. lab. Carls- berg, Ser. chim., 24, 1-46. McElroy, W. D., and A. Farghaly. 1948. Biochemical mutants affecting growth and light production in luminous bacteria. Arch. Biochem., 17, 379-90. McElroy, W. D., and S. Friedman. 1951. Gene recombination in luminous bacteria. J. Bacteriol, 62, 129-30. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1953. The requirement of riboflavin phosphate for bacterial luminescence. Science, 118, 385-86. McElroy, W. D., J. W. Hastings, V. Sonnenfeld, and J. Coulombre. 1954. Partial purification and properties of bacterial luciferin and luciferase. /. Bacteriol, 67, 402-408. Miller, H., A. Farghaly, and W. D. McElroy. 1949. Factors influencing the recovery of biochemical mutants in luminous bacteria. /. Bacteriol., 57, 595-602. Schlegel, F. McK., and F. H. Johnson. 1949. The influence of temperature and hydrostatic pressure on the denaturation of methomoglobin by urethanes and sahcylate. /. Biol. Chem., 178, 251-57. Simpson, R. B., and W. Kauzmann. 1953. The kinetics of protein denatura- tion. I. The behavior of the optical rotation of ovalbumin in urea solutions. 7. Am. Chem. Soc, 75, 5139-52. Steam, A. E., and H. Eyring. 1941. Pressure and rate processes. Chem. Revs., 29, 509-23. Strehler, B. L. 1953. Luminescence in cell-free extracts of luminous bacteria and its activation by DPN. /. Am. Chem. Soc, 75, 1264. Strehler, B. L., and M. J. Cormier. 1953. Factors affecting the luminescence of cell-free extracts of the luminous bacterium, Achromobacter iischeri. Arch. Biochem. and Biophijs., 47, 16-33. Strehler, B. L., E. N. Harvey, J. J. Chang, and M. J. Cormier. 1954. The luminescent oxidation of reduced riboflavin or reduced riboflavin phos- phate in the bacterial luciferin-luciferase reaction. Proa. Natl. Acad. Sci. U. S., 40, 10-12. Strehler, B. L., and F. H. Johnson. 1954. The temperature-pressure-inhibitor relationships of bacterial luminescence in vitro. Proc Natl. Acad. Sci. U. S., 40, 606-17. 296 INTRACELLULAR LUMINESCENCE Warren, G. H. 1945. The antigenic structure and specificity of luminous bac- teria. /. Bacteriol, 49, 547-61. Werbin, H., and A. D. McLaren. 1951a. The effect of high pressure on the rates of proteolytic hydrolysis. I. Chymotrypsin. Arch. Biochem. and Biophys., 31, 285-93. Werbin, H., and A. D. McLaren. 1951b. The effect of high pressure on the rates of proteolytic hydrolysis. II. Trypsin. Arch. Biochem. and Biophys., 32, 325-37. Discussion Dr. Kauzmann: Among a few examples wherein denaturation is accelerated by pressure is the denaturation of ovalbumin in urea at 0° C, which is accelerated by a factor of about 10 on applying 9000 psi (Simpson and Kauzmann, 1953). On the other hand, at 40° C under the same conditions, there is no effect of pressure. One might therefore expect that at still higher temperatures pressure would tend to reverse or retard the denaturing effect of urea, in accordance with the other examples of retardation of protein denaturation by pressure, just discussed by Dr. Johnson. The work of Linderstr0m-Lang's group on the volume changes which occur in the enzymatic hydrolysis of native beta lactoglobulin also has some bearing on this point ( Linderstr0m-Lang, 1952); Lin- derstr0m-Lang and Jacobsen (1941) ). These workers have found that during the initial stages of degradation there is a volume decrease amounting to several hundred cubic centimeters per mole. They have ascribed this to a disruption of the native protein structure and have given reasons for believing that this is akin to denaturation by agents such as heat and urea. Dr. Mason: A good deal of evidence, especially with oxidative sys- tems, has demonstrated that many enzymes tend to be aggregated, possibly in a highly ordered way, upon intracellular particulates and that their activities are dependent upon or modified by their order in this aggregation. In considering the relationship between the response of intact bioluminescent cells and extracts of those cells to changes in environmental conditions such as pressure and temperature, is it not possible that particulate suspensions and solutions of enzymes may display qualitatively the same effects, yet be responding to the changes in conditions by essentially different mechanisms? Dr. Johnson: That would seem possible, of course, but it would be a more complicated interpretation of the results discussed. I do not believe that the similarities of the pressure-temperature-inhibitor rela- tionships of luminescence in the extracts to those in living cells are merely coincidental, involving different mechanisms; in my opinion, the evidence favors the view that they are fundamentally the same. 297 Physiological Control of Luminescence in Animals J. A. C. Nicol' Marine Biological Laboratory, Plymouth, England Luminescence appears to be almost universally subject to some degree of regulation in animals (Harvey, 1953). As an initial approach w^e can distinguish two main categories of luminescence, namely extrinsic or bacterial, in which the light is produced by bacteria harbored by the animal; and intrinsic, in which photogeny depends upon the animal's own biochemical accomplishment. In either event, overt ap- pearance of light, i.e., light emission to the exterior, is usually con- trollable. If luminescence is considered in terms of effector systems, then I think we can concede the following classification: (1) continuous production of light by symbiotic bacteria; (2) discharge of luminous secretion into a surrounding aqueous medium; (3) intracellular lumi- nescence by the animal's photocytes. This is the foundation of my discussion, and in this paper I shall review certain aspects of lumi- nescence from the viewpoint of neuro-effector control. It would appear that symbiosis with luminescent bacteria is rela- tively rare. The most interesting and best authenticated instances of symbiosis are encountered in certain marine teleosts, in which lumi- nescent bacteria are harbored in special circumscribed organs. By the use of screening devices the light emitted by the bacteria can be occluded or revealed. A few instances are also known of continuous light emission on the part of the animal, e.g., in pelagic squid Spirula, a millipede Luminodesmtis, the larva of the beetle Phengodes, and * Guggenheim Canadian Fellow. 299 300 PHYSIOLOGICAL CONTROL IN ANIMALS eggs and pupae of lampyrids. Apart from these instances, in which control appears to be lacking, some form of regulation of light emis- sion is the rule (Buck, 1948; Harvey, 1952; Davenport et al., 1952). In the evolution of animals, effectors have preceded neural regu- lation, as emphasized by Parker ( 1919 ) . This condition is still existent in two phyla, the Protozoa and Porifera. Regulation of Luminescence in Absence of Differentiated Nervous System The Protozoa are sensitive to a multiplicity of environmental agents, to which they respond mechanically, by changes in body shape, movement of organelles, and chemically, by secretion and by light production. In Noctihica, which has been most studied, lumi- nescence is evoked normally by mechanical stimulation, resulting in a brief flash; and with interrupted induced current the animals flash on the first shock and remain glowing thereafter. These observations pose several problems, which are not neces- sarily peculiar to Noctihica, nor to the luminescent reaction. Local and gentle tactile stimulation results in luminescence only in the region of the cell affected. A generalized luminous response, there- fore, depends upon agitation of the whole cell, or transmission of excitation. Photogenic granules are believed to lie superficially in the cell. However, propagation of an excitatory wave may well be other than a surface phenomenon; and an analogy is at hand in transmission of excitation across a muscle fiber, the contractile proteins of which lie within. It is, perhaps, relevant that punctured and collapsed Noc- tihica, with injured membranes, give normal luminescent responses to mechanical and electrical stimuli (Robin and Legros, 1866; Allman, 1872; Massart, 1893; E. B. Harvey, 1917). Another approach to this problem, full of interest, may be indicated here. There are several metazoan groups in which eggs and larvae are luminescent, preceding or following establishment of innervation. In ctenophores, eggs and early segmenting stages light up when stimulated, either by a mechanical disturbance or an electrical shock. Larvae of the polychaete Cliaetopteriis also luminesce when agitated, but these are metamorphosing trochospheres in which a larval nerv- J. A. C. NICOL 301 ous system is already defined (Peters, 1905; Yatsu, 1912; Enders, 1909). With the highly sensitive multiplier phototubes now available, the luminescent responses of some of these forms may be amenable to photoelectric recording. They afford additional instances of direct responses to environmental changes before the advent of nervous con- trol and provide experimental material for the study of excitation and effector activity at the cellular level. Luminescence and Nerve Net Many coelenterates, both pelagic and benthic, are luminescent. It is characteristic of these animals that their responses to environmental changes are controlled by a nerve net. This is to a large extent a meshvvork of discrete neurones, across which two-way conduction takes place. The quality and magnitude of response are governed by various factors, prominent among which is neural facilitation. This is a selective condition governing behavior by which seriated impulses bring about responses which are out of all proportion to the appar- ent effect of a single impulse. Luminescent responses in certain pennatulids will illustrate the functioning of the nerve net. The response in these animals takes the form of a flash of light which sweeps over the surface. In sea pens Pennatida, a stimulus apphed at any point excites a wave or waves which sweep away from the affected area. Similarly, simultaneous stimulation at the two extremities excites convergent waves. Con- firmatory experiments, showing that conduction is nonpolarized, have been carried out on Renilla, in which animal it has been shown that complex cuts, producing devious pathways, still allow transmission (Panceri, 1872a; Parker, 1920a, b). In many coelenterates several impulses are necessary to evoke a response owing to the intervention of facilitation. Harvey noted that Cavermilaria usually failed to respond to a single shock, but gave a flash after three shocks in rapid succession. In Renilla it is found that several shocks are necessary to elicit a luminous wave, and with continued stimulation the consecutive flashes increase progressively in intensity (Fig. 1). Both the aforementioned features (ineffectiveness 302 PHYSIOLOGICAL CONTROL IN ANIMALS of a single pulse, augmentation of intensity in successive responses) are typical of facilitation. Observation shows that transmission is non- decremental, i.e., a wave once initiated, courses over the whole surface of the rachis. There are cogent reasons for believing that facilitation B wvnniun Fig. 1. Luminescent responses of the sea pansy Renilla. A, series of responses to a burst of 10 electrical shocks at a frequency of 1 per second. The first response appears after the third shock, and subsequent responses increase progressively in intensity (facilitation). B, responses to a burst of shocks at high frequency (3 per second). Note prolonged after-discharge. C, flashing induced by tactile stimulation. Time scale above, 72 per minute. Recording from the entire rachis. Luminous responses shown as downward deflections of middle trace; electrical stimuli showai on lower line. Photomultiplier + cathode-ray oscilloscope recording. occurs peripherally, at the neuro-effector junction, and not in the synapses of the nerve net. A single tactile stimulus gives rise to a wave or a series of waves, and this type of response is usually ascribed to facilitation, resulting from a volley of impulses set off by mechanical stimulation of receptors ( Fig. 1 ) (E.N. Harvey, 1917; Pantin, 1935; Buck, 1953, 1955). Well-defined synapses have been recognized in medusae, and^ these exhibit nonpolarized conduction (Bozler, 1927). Using contractions of the umbrella as indicative of transmission phenomena, Bullock (1943) finds that the nerve net of scyphomedusae is in a state of permanent facilitation and transmits each impulse. Continued stimula- tion, however, produces staircase, indicative of neuromuscular facilita- tion. In the luminescent scyphomedusan Pelagia noctiluca tactile and electrical stimulation evokes a glow which may spread in some animals as a wave over the surface of the bell (Panceri, 1872b; Heymans and Moore, 1924; Moore, 1926). By stimulating with con- J. A. C. NICOL 303 denser shocks, I have found that the animal gives a flash to each stimu- lus (Fig. 2). The total duration of a local flash is 3 seconds, and time taken to reach maximal height occupies 0.2 second. There is no evidence for facilitation in my records, since each shock brings up Fig. 2. Flashes of the scyphomedusan Pelagia noctihica to electrical stimulation (condenser shocks). Burst of 5 shocks at about 12 per minute. Time scale 1 per second. Stimuli added to this record represent number but not abso- lute position of pulses. a response, and the intensity of consecutive flashes decreases in an exponential fashion (Figs. 2 and 3). Indeed, this progressive decay of photogeny is the most obvious feature of the records and may well be obscuring facilitation in the transmission system. 0 1 2 3 4 5 6 CONSECUTIVE RESPONSES Fig. 3. Fatigue of consecutive responses in Pelagia noctihica. The animal was stimulated with a burst of stimuli (condenser shocks) at a frequency of 1 per second. Upper curve, flash duration (ordinates, on right, in seconds). Lower curve, decrease in intensity of Hash and in total light emitted in successive flashes. © mean flash intensity. + integration of total light emitted during a flash. 304 PHYSIOLOGICAL CONTROL IN ANIMALS Only limited information is available for the nerve net of cteno- phores. On tactile stimulation, light appears under the combs, and conductile pathways are likewise restricted to these areas. Nervous transmission along the meridians is evidently unpolarized, since a luminescent wave is capable of oral or aboral passage (Panceri, 1872b; Peters, 1905; Moore, 1924, 1926). With electrical stimulation, a flash appears on each stimulus after the first. These flashes are of very brief duration, about 0.2 second, and summate at frequencies above 5 per second (Fig. 4). Facilitation of response is also clearly recog- nizable at slow frequencies of stimulation, the first few responses increasing progressively in height to plateau level, after which fatigue sets in. With repetitive stimulation, rhythmic flashing is also induced {Beroe, Fig. 4). Fig. 4. Responses of the ctenophore Beroe to electrical stimulation (condenser shocks). A, 2 shocks, 1-second interval. Note repetitive discharge on second shock. B, 2-second burst at 7 per second. C, burst of 1. 1-second duration at 18 per second. D, burst of 3''/^ -second duration at 20 per second. E, 1-second burst at 44 per second. Time scale, 1 per second. Luminescence in Animals with Central Nervous Systems As illustrative of nervous regulation in animals with a differentiated central nervous system, I shall refer to two groups with which I have had personal experience. These will provide examples of intracellular luminescence at invertebrate and vertebrate levels. Polynoids In polynoid or scale worms, flashes are produced by the dorsal elytra under appropriate stimulation. The photocytes involved are innervated by fibers which emerge from a ganglion in the center of the elytrum. The normal response to a single shock is a series of J. A. C. NICOL 305 flashes, with intervals of 0.1-1 second, and flashing continues for a minute or more (Fig. 5). Each flash is very brief, having a latent period of some 19 milliseconds, and lasting 83 milliseconds; and fusion and summation attend rapid stimulation, above 8 per second (Bonhomme, 1942; Nicol, 1953). Repetitive flashing is regulated by the peripheral elytral ganglion and fails to occur in its absence, a single shock then inducing a single flash. Of particular interest is the fact that peripheral facilitation is well displayed. The first few responses show a progressive augmen- tation of maximal intensity, and the increment in intensity between successive impulses is inversely related to the stimulation interval (Fig. 5B). There is thus evidence for accumulation and decay of facilitator (Nicol, 1954a). H rn-nn^nnrTTTTirrTT 1 1 1 1 ■ 1 r-l 1 i-l 1 r i-l 1 1 1 Fig. 5. Luminescent responses from single elytra of Acholbe astericola (polynoid worm). A, prolonged flashing following a single electrical stimulus. This is a continuous record. B, single flashes induced by repetitive stimulation at a slow rate (42 per minute). Note facilitation. Time scale of these records, 1 per second (Nicol, 1953, 1954). Luminescence is normally evoked reflexly by tactile stimulation: afferent impulses enter the central nervous system and are relayed peripherally in efferent pathways to the elytra. Excitation is also transmitted up and down the nerve cord and causes adjacent segments to flash. Another notable featiue of the response is that it is provoked by autotomy of the scales. The elytrophore possesses a visible breaking plane, and when a scale is cast off under tactile stimulation, it begins to flash rhythmically. This is occasioned either by the discharge of impulses from the central nervous system at the same time as autotomy is induced, or by rupture and stimulation of nerve fibers during autotomy. In any event the resultant impulses excite the elytral gan- glion, provoking rhythmic discharge and flashing. 306 PHYSIOLOGICAL CONTROL IN ANIMALS Teleosts From the viewpoint of the evolution of the vertebrate nervous system, one of the most intriguing aspects of photogeny in fishes is the character of nervous regulation. Some pelagic teleosts, notably myctophids, can emit very brief flashes with short latent periods. Receptors are tactile and visual. The innervation of the photophores has been worked out in three forms, viz., Argijropelecus, Cyclothone, and Lampanyctus (Handrick, 1901; Gierse, 1904; Ray, 1950). None of these workers traced the neural pathways involved, but from the pat- tern of distribution of efferent nerves I hazarded the suggestion that they might be autonomic (Nicol, 1952a). Earlier work on the midshipman Porichthys showed that the photo- phores could be excited by electrical stimulation of the whole fish, and by injecting adrenaline or pituitrin. Since nerve fibers to the photophores were little in evidence, it was concluded that regulation of luminescence was hormonal (Greene, 1899; Greene and Greene, 1924). I find that electrical stimulation of the spinal cord (condenser shocks) causes all the photophores to luminesce, both those of the head and trunk. Transection of the cord, anterior to the electrodes, and ligaturing the heart fail to prevent the response. The latent period of the response under electrical stimulation is only 7-10 seconds, a period far too short to be explained by hormonal excitation. The circulation time of fish is not known, but is likely to exceed 10 seconds. Moreover, when adrenaline is injected directly into the heart (0.1 mg in a 1.5-kg fish), it takes 2 minutes for the luminescent response to appear, and this at a concentration which can be considered highly potent. It is apparent that the luminescent responses of Porichthys and certain other teleosts appear far too quickly to be governed primarily by endocrine mechanisms. These observations can be explained most easily by postulating that the photophores of Porichthys are innervated by the sympathetic nervous system. The ability to respond in the absence of circulation shows that the endocrine system is not essential to the response. Teleosts generally possess well-defined sympathetic trunks which connect with cranial nerves, including the facial (Young, 1931). The latter provide avenues for peripheral distribution of sympathetic fibers J. A. C. NICOL 307 in the head. In Lampanyctus, and other teleosts, all the photophores of the head are innervated by the facialis, and those of the trunk are innervated by rami of spinal nerves. The sympathetic system is the only longitudinal efferent pathway, traversing the entire length of the fish, that is available after transec- tion of the cord. By assuming that photophore innervation in Porich- thys is similar to that worked out for other teleosts, we can form a picture of the photophores of the head receiving their sympathetic fibers via the facialis, and those of the trunk receiving sympathetic fibers which traverse recurrent gray rami and spinal nerves in each segment. The fact that the photophores are excited by adrenaline is compatible with sympathetic innervation, since this is a normal chemical transmitter of sympathetic fibers, and the photophore nerves may well be, nay probably are, adrenergic in character. Since supra- renal tissue is well represented in teleosts, the possibility remains that the secretion of adrenaline into the bloodstream is a contributory factor in a prolonged response. Regulation of Light Emission by Screening Devices Screening devices for regulating light emission make use of mus- cular movement, chromatophore movement, or a combination of both. Two fishes, Pfwtoblepharon and Anomalops possess a bacterial lu- minescent organ under each eye. That of Photoblepharon is provided with a fold of opaque tissue which can be raised over the light organ. The organ of Anomalops, on the other hand, can be rotated on a hinge at its anterodorsal end, so as to turn the light surface inward. Steche (1909) has described a pair of antagonistic muscles which serve these organs, but their innervation is unknown. Similar devices are also known in other teleosts and in cephalopods (Harvey, 1922, 1952). Another screening device in teleosts involves movement of chroma- tophore pigments. Certain fishes such as Acropoma and Leiognathus possess internal glands containing luminous bacteria. The light from these organs shines through the translucent body wall, and the pres- ence of chromatophores in tissues overlying the photogenic organs suggests that they may be involved in regulating light emission. These organs shine continuously, and the chromatophores, by concentration 308 PHYSIOLOGICAL CONTROL IN ANIMALS or dispersion, intensify or weaken emission without occluding it entirely. The light of Secutor and Gazza is said to increase suddenly in brilliance when the fish are strongly stimulated (Haneda, 1950). There is still some uncertainty about the mode of regulation obtain- ing in teleost chromatophores, but it is at least certain that pigment concentration is achieved by sympathetic stimulation and administra- tion of adrenaline, and that the chromatophores are innervated by the sympathetic nervous system (Parker, 1943, 1948). Tactile stimulation is known to cause chromatophore concentration and blanching in some fish and is possibly a significant factor in transient intensification of light emission in fish with bacterial light organs (Osborn, 1939). Chromatophores, overlying light organs, also control emission in cephalopods. In these animals, however, the chromatophores are actuated by muscle fibers, which expand the pigment cell when they contract, and allow it to contract when they relax. The appearance of light on concentration of chromatophores and occlusion on expan- sion of chromatophores have been described in Watosenio sclntiUans. As Harvey (1952) has pointed out, direct control of light production may also be involved in these screened photophores. Modes of Direct Control of Photogeny The basic problem in control is turning the light on and off. More- over, luminescence is a response with certain characteristics and parameters which, theoretically, can be varied in five ways, viz., in quality (spectral emission), intensity, duration, spatial distribution, and frequency or repetition. In actuality, we find that control of one or more of these variables is exercised by animals. Inhibition Perhaps the simplest and most direct form of control consists of inhibition of luminescence by illumination. This can be exercised in either of two ways: by direct inhibition of photogenic material, or by reflex inhibition in the sensori-neural system. It is probably significant that nearly all instances of direct inhibition of photogenic material or photocytes are confined to the Protozoa, Coelenterates, and Cteno- phores, animals either lacking a nervous system or provided with a nerve net. It is variously reported for Gomjaidax (personal observa- J. A. C. NICOL 309 tion of F. Haxo), and Noctihica (contradictory observations; see Harv^ey, 1952), pennatulids, possibly Pdagia and generally in cteno- phores (Peters, 1905; Moore, 1924; Heymans and Moore, 1924; Harvey, 1952). Primary inhibition of this kind effects conservation of photogenic material in dayhght, and the material is reserved for use in darkness. Su(?h a mechanism presents functional simplicity and advantages in organisms with restricted and common modes of regulation of diverse TIME Ml N Fig. 6. Course of reco\ery of luminescence in Renilla when transferred from liglit to darkness. Ordinate: intensity of response evoked by a burst of shocks. Abscissa, time in minutes after transferring animal from light to darkness. responses. In Renilla, for example, we find that luminescence attains full intensity only after 1 to 2 hours sojourn in the dark (Fig. 6). Now, tactile stimulation in this animal evokes both movement (con- traction) and luminescence, and it is highly probable that all these responses are controlled by the same nerve net. Nevertheless, the dependence of luminescence on previous dark exposure ensures that the luminescent response is reserved for times of darkness, even though photocytes and muscle fibers are reached by efferent impulses from tactile receptors at all times. Reflex inhibition of luminescence by daylight through photosensi- tive receptors might be expected, but no instances appear to have been recorded. Certainly, many animals, with differentiated central nervous systems, will luminesce when reflexly stimulated in daylight, e.g., polynoids, Chaetoptenis, Amphiura. There are some suggestive 310 PHYSIOLOGICAL CONTROL IN ANIMALS observations for the enteropneust Ptychodera bahamensis, in which luminescence is elicited with diflSculty by tactile stimulation after exposure to light. The photocytes themselves appear to be unaffected by illumination since they still respond normally to electrical stimula- tion (Crozier, 1917). Another form of inhibition is theoretically possible for intracellular luminescence. One can picture an animal in which photogeny is normally the active phase at the effector level and is continuous in the absence of external control. Inhibitory fibers would then act by suppressing light emission. Analogies are at hand in the transitory inhibition of continuous ciliary activity, as in veliger larvae (Carter, 1926). In fireflies, Buck (1948) has described four types of light emission, ranging from continuous glow to quick flash, and suggests that these may represent progressively more effective modes of con- trol. Fireflies showing the continuous glow are unable to control the response; intermittent glow and flash represent various levels of con- trol. It is still uncertain how control is achieved, but it can be argued that it involves some form of inhibition, either by direct nervous influence, or by restriction of oxygen supply. Records of induced luminescence in Lampyris noctiluca are shown in Fig. 7. B iMMimir Fig. 7. Luminescent responses of the European firefly Lampyris noctiluca to electrical stimulation. Electrodes were positioned over the ventral body vrall and stimuli consisted of condenser shocks. A, short burst at a low frequency (2-second burst of 11 pulses at 7 per second). B, 2-second burst at 18 per second (28 pulses). C, short burst (16 pulses), and D, a long burst (28 pulses), at 8 per second. E, long burst of 90 pulses at 8 per second. Heavy horizontal hne on bottom trace indicates position and duration of stimula- tion. Time scale above, 1 per second. J. A. C. NICOL 311 Excitation Let us consider now positive excitation of luminescent organs and the ways in which the response can be modulated and controlled. The luminescent response is often a triggered response, i.e., a single impulse will set it into operation. This is the case in Pelagia, Beroe, Pohpwe, Chaetopterus, and Vyrosoma, to mention some established ' ' * Fig. 8. Luminescent responses of Chaetopterus variopedatus, a polychaete which discharges a luminous secretion. A, response to a single electrical shock. Time scale 1 per second. B-F, responses to bursts of electrical shocks at frequencies shown. Time scale, 1 per 10 second (Nicol, 1952). examples (Figs. 2, 4, 5, 8, 9). In the nerve net of pennatulids {Renilla, Cavernularia ) on the contrary, facilitation is operative in the initiation of the response, as manifested by the several shocks necessary to produce the first flash (Fig. 1) (Harvey, 1917; Buck, 1953, 1955). There is also the observation that in Porichthys, a burst of impulses is necessary to bring out the response. 312 PHYSIOLOGICAL CONTROL IN ANIMALS Intensity of Response The intensity of response is influenced in two ways, either or both of which may be operative in the same animal: these pertain to summation and facihtation. Consider first Chaetopterus, an animal which discharges luminous matter (extracellular luminescence). The glow from this secretion lasts for as long as five minutes (Fig. 8). Now, facilitation is not operative peripherally in the luminescent responses of this animal, but with the protracted mode of response, summation occurs readily at low frequencies of stimulation (above 2 B Mill;;; ; ; 1 1 1 ) 1 1 1 Fig. 9. Luminescent responses of Pyrosoma, a colonial pelagic tunicate. A, elec- trical stimulation with 1 and 4 condenser shocks (the latter at 1 per second). B, 2-second burst of electrical stimuli at 5 per second. C, response to flash of light (pocket torch); approximate position of stimulation added to lower hue. D, responses to weak (left) and stronger (right) tactile stimulation (vibration). Amplification of A ten times that in B-D. Time scale 1 per second. per minute). Increased intensity of response results, therefore, from several or many stimuli, compared with one, and from short com- pared with long interval between impulses (Nicol, 1952b,c, 1954b). Similar response characteristics are encountered in Pyrosoma (Fig. 9). In this animal the response lasts some 4 seconds after a single electrical stimulus. Facilitation is not operative, at least not at slow frequencies of 1 per second. Prolonged stimulation at high frequencies, however, results in augmentation of light intensity as the result of summation of individual responses. Similarly, a strong mechanical J. A. C. NICOL 313 stimulus produces a brighter glow than a weak one, as the result of stronger tactile excitation producing more nerve impulses. The mode of luminescence in Tijrosoma is not known with certainty, and the problem is still obscured by Pierantoni's claim that light is due to intracellular symbiotic bacteria. In many animals, where the response is intracellular, luminescence takes the form of a short flash or series of flashes, and the intensity is often governed by facilitation, e.g., Renilla, Beroe, Polynoe. In 100 -\ > H 80 . U) I z 1 UJ \ t- 1 z 1 - 60 ■ 1 UJ 1 > 1 H 40 1 < -J Ul E 20 l A 4 0 TIME SEC. Fig. 10. Facilitation-decay cur\'es for Renilla (A) and Polijnde (B). In A, Renilla, the specimen was stimulated with paired shocks, separated by inter- vals ranging from 0.25 to 4 seconds. The first response appeared on tlie second shock, and ordinates represent the intensity of response at different stimulation intervals. B, Polynoe, shows increment of second o\'er first re- sponse to paired shocks, separated by intervals ranging from 1 to 120 seconds (from Nicol, 1954). Renilla and polynoids at least, facilitation occurs peripherally. It manifests itself by progressive increase in intensity of successive flashes and depends in some way upon arrival of successive impulses at the photocytes. It is generally assumed that augmentation of response is owing to a banking up of facilitator faster than it can be removed (Fig. 10). This effect is observed under normal modes of stimulation (tactile), as well as in response to electrical shocks, and, therefore, is probably a factor regulating the intensity of response to environmental agencies. 314 PHYSIOLOGICAL CONTROL IN ANIMALS Response duration Here we are concerned with two distinct phenomena, viz., duration of a single flash or glow and total duration of repetitive flashing. In those animals in which a response or glow is elicited by each pulse, lengthening of response duration is achieved by prolonged stimulation. Thus, in Chaetopterus, which secretes a luminous ma- terial, continued electrical stimulation, at frequencies high enough to produce fusion of separate responses, also results in a longer, more durable glow (Fig. 8). This is a consequence both of summation in a contractile mechanism, responsible for expressing secretion, and of an accumulation of photogenic material at a rate faster than it can be consumed. The second method concerns regulation of repetitive flashing. Continued flashing once stimulation has ceased, or the evocation of many flashes by a single stimulus, has been observed in Renilla, Beroe, and generally in luminescent polynoids (Figs. 1, 4, and 5). A strong tactile stimulus, or protracted electrical stimulation often sends Renilla into a hyperexcitatory state in which it continues to flash repeatedly, sometimes for long periods. This state of maintained flash- ing is an external manifestation of rhythmic discharge in the nerve net, and it would appear that nerve cells in Renilla can be charged up to high levels of excitability, when they pass into a rhythmic oscillatory state, expressing itself in periodic discharge across the nerve net. In polynoids repetitive flashing is the characteristic mode of response, and even a single electrical stimulus sets the elytram flashing. In these animals the response is regulated by a peripheral elytral ganglion, which passes into some form of oscillatory condition when excited and discharges with great regularity for many seconds. In these instances, which I have mentioned, the response can be greatly pro- longed beyond the original stimulus by maintained excitatory states engendered in the nervous system, be it nerve net or ganglion (Nicol, 1953, 1954a; Buck, 1953, 1955). Spatial Distribution In pennatulids, possessing a nerve net, each luminescent response sweeps over the entire surface of the animal. There are many instances J. A. C. NICOL 315 known, however, of animals in which the luminescent response can be quite localized, e.g., in teleosts where separate photophores or groups of photophores light up independently of others. Among lower animals I cite Cliaetopterus, in which the response can be caused to invade an increasingly greater number of segments by increasing frequency and number of stimuli. Moreover, the response spreads with greater facility posteriorly than anteriorly. Here we have instances in which the locus of response is controlled by the central nervous system, by functional polarization, facilitation, and possibly by central representa- tion of peripheral fields in higher forms (teleosts and cephalopods). Control of Spectral Emission There are a number of animals possessing several differently colored photophores, capable of emitting light of different colors, e.g., the beetle Phrixothrix and the squid Thaumatolampas. In some of these animals the differently colored photophores can respond indepen- dently of each other, thus providing light of different colors according to the stimulus. The possibility also exists, moreover, in certain animals in which the light is controlled by chromatophores, that these organelles may influence the color of the light emitted, according to their condition and pigment characteristics. Conclusions Regulation of luminescence must be related to neuro-effector con- trol in general, and in fact, involves control of four different mecha- nisms, viz., glandular secretion, muscular contraction, chromatophore movement, and intracellular photogeny. Most recent work on neuro- effector, and in particular, neuromuscular, control is being carried out at the molecular level, in an attempt to discover the details of trans- mitter action. Regulation at the interface between nerve fibers and light organs is liable to display great variation in different systems and in different animals. Accepting differences in transmitter action, we may, in the last analysis, discover much uniformity in patterns of energy release effecting the photogenic response. In those animals which hitherto have been studied, overall regula- tion of luminescence is achieved by transmission of an excitatory state, either across the surface of the cell (protozoa, eggs), or through 316 PHYSIOLOGICAL CONTROL IN ANIMALS nervous pathways ( metazoans ) . There are no unequivocal instances of hormonal regulation known, although this mode of control is not necessarily improbable. The interesting condition in the colonial tunicate, Fijrosoma, still awaits examination. In this animal a luminous wave progresses over the zooids making up the colony. Panceri ( 1873 ) found a common system of muscles extending between the zooids and suggested that associated nerve fibers might serve for the transmission of excitation affecting the photocytes. Since illumination evokes lumin- escence, it is also possible that the luminescent wave may be propa- gated by this means. In the study of any physiological process, one is assisted greatly in the design of suitable experiments by an appreciation of its role in the economy of the animal. It is indeed a curious fact that although certain aspects of luminescence have advanced greatly, notably the biochemistry of the process, we rarely have any clear appreciation of the significance of the luminescent response to the animal. As an example of how knowledge of the purpose of the response can in- fluence the design of experiments, I refer to the studies of Buck (1948) on the role of luminescence in the mating responses of the American firefly Photinus pyralis. Studies now available invite speculation on another aspect of luminescence, namely restricted modes of luminescent control. It is probable that neuro-effector control was established initially for mus- cular systems and extended secondarily to other effectors, including luminescent organs. It is not surprising, therefore, to find certain modes of regulation common to muscular and luminescent systems. Not only has luminescence arisen independently in several different groups of animals, but each of the several forms of luminescence has also appeared independently on several occasions, e.g., discharge of luminous secretion in worms, balanoglossids and many other forms; complex photophores in squid, euphausiids and teleosts; opaque, screening lids in squids and teleosts; and different processes of chro- matophore regulation in the same two groups. Within a few Hmited forms of structural patterns, combination and variation of detail provide multiplicity of response mechanisms. In the words of Sir Thomas Browne: "Studious observations may discover more analogies J. A. C. NICOL 317 in the orderly book of nature, and cannot escape the elegancy of her hand in other correspondencies." {Garden of Cyprus). ACKNOWLEDGMENTS I acknowledge with gratitude a fellowship grant from the John Simon Guggenheim Memorial Foundation, enabling me to visit the United States of America. Some part of the research mentioned in this paper has been car- ried out at the Scripps Institution of Oceanography, to which I am grateful for certain facilities. I should like to record my thanks to the National Institute of Oceanography (United Kingdom) for allowing me to take part in a research cruise of the R.R.S. Discovery II. The cost of part of the equipment used in my research was defrayed from a grant-in-aid of sci- entific research through the Royal Society. References Allman, G. 1872. Notes on Noctiluca. Quart. J. Microscop. Sci., N.S. 12, 326-32. Bonhomme, C. 1942. Recherches sur I'histologie de I'appareil lumineux des Polynoines. Bull Museum Oceanogr. Monaco, No. 823, 8 pp. Bozler, E. 1927. Untersuchungen iiber das Nervensystem der Coelenteraten. Z. Zellforsch., 5, 244. Buck, J. B. 1948. The anatomy and physiology of the light organ in fireflies. Ann. N. Y. Acad. Sci., 49, 397-482. Buck, J. B. 1953. Bioluminescence in the study of invertebrate nervous sys- tems (Abstr.). Anat. Record, 117, 594. Buck, J. B. 1955. Bioluminescence in Renillu, in relation to nerve net physi- ology. In press. Bullock, T. H. 1943. Neuromuscular facilitation in Scyphomedusae. /. Cellular and Comp. Physiol, 22, 251-72. Carter, G. S. 1926. On the nervous control of the velar beat of the nudi- branch veliger. /. Exptl. Biol, 4, 1-26. Crozier, W. J. 1917. The photic sensitivity of Bahnoglossus. J. Exptl. ZooL, 24, 211-17. Davenport, D., D. M. Wooton, and j. E. Gushing. 1952. The biology of the Sierra luminous millipede, Luminodrsmus sequoiae, Loomis and Daven- port. Biol. Bull, 102, 100-10. Enders, H. E. 1909. A study of the hfe history and habits of Chaetopterus variopedatus, Renier et Claparede. /. MorphoL, 20, 479-532. 318 PHYSIOLOGICAL CONTROL IN ANIMALS Gierse, A. 1904. Untersuchungen iiber das Gehirn und die Kopfnerven von Cyclothone acclinidens. Morph. Jahrb. Leipzig, 32, 602-88. Greene, C. W. 1899. The phosphorescent organs in the toadfish, Porichtliijs notatus Girard. /. MorphoL, 15, 667-96. Greene, G. W., and H. H. Greene. 1924. Phosphorescence of Porichthys notatus, the California singing fish. Am. J. Physiol., 70, 500-506. Handrick, K. 1901. Zur Kenntnis des Nervensystems und der Leuchtorgane von Argyropelecus hemigymntis. Inaug. Dissert. Erwin Nagele, Stuttgart. Also in ZooJogica, Stuttgart, Bd. 13, Heft. 32, 68 pp. Haneda, Y. 1950. Luminous organs of fish which emit light indirectly. Pacific Sci., 4, 214-27. Harvey, E. B. 1917. A physiological study of specific gravity and of lumi- nescence in Noctiluca, with special reference to anesthesia. Carnegie Inst. Wash. Piihl, No. 251, 235-53. Harvey, E. N. 1917. Studies on bioluminescence. VI. Light production by a Japanese pennatulid, Cavernularia haheri. Am. J. Physiol, 42, 349-58. Harvey, E. N. 1922. The production of light by the fishes Photohlepharon and Anomalops. Carnegie Inst. Wasli. Piihl., No, 312, 43-60. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Harvey, E. N. 1953. Bioluminescence: Evolution and comparative bio- chemistry. Federation Proc, 12, 597-606, Heymans, C., and A. R. Moore. 1924. Luminescence in Pelagia noctiluca. J. Gen. Physiol, 6, 273-80. Massart, J. 1893. Sur I'irritabilite des Noctiluques. Bull Sci. France et Belg., 25, 59-76. Moore, A. R. 1924. Luminescence in Mnemiopsis. J. Gen. Physiol, 6, 403-12. Moore, A. R. 1926. Galvanic stimulation of luminescence in Pelagia noc- tiluca. /. Gen. Physiol, 9, 375-79. Nicol, J. A. G. 1952a. Autonomic nervous systems in lower chordates. Biol Revs., 27, 1-49. Nicol, J. A. C. 1952b, Studies on Chaetopterus variopedatus (Renier) II. Nervous control of light production. /. Marine Biol Assoc. U. K., 30, 433-52. Nicol, J. A. G. 1952c. Studies on Chaetopterus variopedatus (Renier) III. Factors afl:ecting the light response. /. Marine Biol Assoc. U. K., 31, 113-44, Nicol, J. A. G. 1953. Luminescence in polynoid worms. /. Marine Biol Assoc. U. K., 32, 65-84. Nicol, J. A. G. 1954a. The nervous control of luminescent responses in polynoid worms. /. Marine Biol Assoc. U. K., 33, 225-55. Nicol, J. A. G. 1954b. Fatigue of the luminescent response of Chaetopterus. J. Marine Biol Assoc. U. K., 33, 177-86. J. A. C. NICOL 319 Osborn, C. M. 1939. The physiology of color change in flatfishes. /. Exptl. Zool, 81, 479-515. Panceri, P. 1872a. The luminous organs and light of the Pennatulae. Quaii. J. Microscop. Sci., N. S. 12, 248-60. Panceri, P. 1872b. Etudes sur la phosphorescence des animaux marins. Ann. sci. nat. Paris, zool. Ser 5. 16, Art. No. 8, 67 pp. Panceri, P. 1873. The luminous organs and light of Ptjwsoma. Quart. }. Microscop., N. S. 13, 45-51. Pantin, C. F. A. 1935. The nerve net of the Actinozoa. I. Facilitation. /. Exptl. Biol, 12, 119-38. Parker, G. H. 1919. The elementary nervous system. J. B. Lippincott Co., Philadelphia and London. Parker, G. H. 1920a. The phosphorescence of Renilla. Proc. Am. Phil. Soc, 59, 171-75. Parker, G. H. 1920b. Activities of colonial animals. II. Neuromuscular movements and phosphorescence in Renilla. J. Exptl. Zool, 31, 475-515. Parker, G. H. 1943. Animal color changes and their neurohumors. Quart. Rev. Biol, 18, 205-27. Parker, G. H. 1948. Animal colour changes and their neurohumours: a sur- vey of investigations 1910-1943. Cambridge Univ. Press. Peters, A. W. 1905. Phosphorescence in ctenophores. /. Exptl Zool, 2, 103-16. Ray, D. L. 1950. The peripheral nervous system of Lampamjctus leucop- sarus. J. Morphol, 87, 61-178. Robin, C, and C. Legros. 1866. De Taction exercee par I'electricite sur les Noctiluques miliaires. /. anat. Physiol Paris, 3, 558-59. Steche, O. 1909. Die Leuchtorgane von Anomalops katoptron und Photo- blephuron palpebratus, zwei Oberflachenfischen aus dem Malaiischen Archipel. Z. iviss. Zool, 93, 349-408. Yatsu, N. 1912. Observations and experiments on the ctenophore egg: I. The structure of the egg and experiments on cell-division. /. Coll Sci. Imp. Univ. Toktjo, 32, Art. 3, 21 pp. Young, J. Z. 1931. On the autonomic nervous system of the teleostean fish Uranoscopus scaher. Quart. J. Microscop. Sci., 74, 491-535. Discussion Dr. Harvey: For many years I have felt that records of the luminous response of a unicellular organism such as Noctiluca would be of great interest and value for comparison with single muscle fiber contractions or nerve cell potential changes. A graduate student, Mr. J. J. Chang, and I endeavored to obtain and culture Noctiluca for this purpose but without success, and therefore turned to the cteno- phore, Mnemiopsis leicleiji, common at the Marine Biological Labora- tory, Woods Hole. Luminescence appears from a group of cells within the radial canals. Last summer Mr. Chang made an exhaustive study of this luminous response to mechanical stimulation and electrical stimuli of various frequencies. The luminescence intensity was re- corded by photomultiplier tube, amplifier, cathode-ray oscillograph, and camera. The relationship between light emission and strength and duration of stimulus, repetitive stimuli, and fatigue at different tem- peratures has been particularly studied. The most interesting result is the almost exact parallel between a luminous response and a muscle contraction. At 21° to 23° C with single square-wave stimulation and a small piece of luminous tissue, a single flash appears having a latent period that varies considerably, a half-rise time of 35, a maximum-peak time of 60, a half-decay time of 48, and a 0.9-decay time of 114 milliseconds. Lowering the tempera- ture prolongs the time course of single flashes, especially the decay phase. Raising the temperature has the opposite effect. Light inten- sity increases as strength of stimulus increases, and repetitive stimuli elicit responses similar to summation of twitches, treppe, incomplete and complete tetanus of muscle. Fatigue appears soon and is a marked characteristic of the luminescent response. Repetitive flashes after a single stimulus have been observed in large pieces of tissue. The conduction rate of a luminous excitation along the canals averages 14 cm/sec. Bursts of action potentials, simultaneous with the lumines- cent responses, appear. An abstract of the work has appeared in Science (119, 581, 1954) and the complete paper in the Journal of Cellular and Comparative Physiology. Thus another organism in which the time relations of a luminous response are now well known can be added to those enumerated by Dr. Nicol. Dr. Mason: It is well known that the superficial colors of many organisms are under hormonal control. In one type of control of this 320 J. A. C. NICOL 321 pigmentation the pigment granules are either aggregated or dispersed within the cell. Is there any evidence that the photogenic granules within photocytes can be similarly aggregated or dispersed under extracellular influences? Dr. Nicol: There is no evidence at present available for hormonal control of movement of photogenic granules. There is, however, the interesting observation of the Greenes (1924) to the effect that injec- tion of pituitrin (extract of posterior lobe of the pituitary) induces luminescence in the midshipman (Porichthys). It is possible, although conjectural, that intermedin, the melanophore-expanding hormone, was the effective constituent in the extract. So far as I can determine, the Greenes made only a single test and did not continue the work. As I recall. Dr. Haneda has made an interesting observation con- cerning certain species of squid that possess a luminescent mantle organ partially bounded by the ink sac (e.g., Euprymna). It seems that movement of ink in the ink sac could be controlled in such a way as to blacken or expose the surface of the light organ and thus regulate light emission. Some Reflections on the Control of Bioluminescence John Buck National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Maryland From the information and data at hand it is evident that one cannot necessarily extrapolate from one luminescent system to another or from the behavior of one luminous organism to that of another. Nevertheless, the work review^ed by Dr. Nicol does point to certain tentative generahzations. For one thing, the simpler organisms such as bacteria, protozoa, coelenterates, ctenophores, and polychaetes seem either to luminesce continuously or to light up only in response to external stimuli, and it is only in groups with rather well-developed nervous systems that photogeny becomes subject to the precise sort of regulation so well exemplified in the mating signals of fireflies. Secondly, it is clear that the abihty to emit sharply delimited flashes of Hght does not depend on a highly developed nervous system but on intracellular Hght production. Tlius, for example, in Chaetopterus and Cypridina, animals with well-developed nervous systems, light production is extracellular and the luminosity decays slowly, whereas the primitively organized Noctiluca, ReniUa, and Mnemiopsis produce flashes not inferior in temporal control to those of many fireflies. Extracellular luminescences are characterized, in addition, by rapid "fatigue" due to exhaustion of the luminous excretion, whereas intra- cellular photogeny may sometimes persist through many hundred successive flashes. The main emphasis of Dr. Nicol's paper is on the properties of the nervous systems of various luminous organisms (and in this he pre- sents a strong case for photogeny being as accurate and critical an 323 324 CONTROL OF BIOLUMINESCENCE index of neural activity as is muscular contraction) rather than in the actual linking of the nerve impulse to the control of light production. However, I was happy to see that his records confirm the observation that facilitation occurs at the neurophotocyte junction in Renilla (Buck, 1953), because this suggests a parallel between photogenic control and events occurring at the motor end plate. In this connec- tion, and in apparent opposition to Parker's (1919) generalization that efiFectors have preceded neural regulation in evolution, Dr. Nicol makes the interesting suggestion that the photogenic system may have been able to hook into an already existent and widespread control mechanism, the neuromuscular system. This may make less puzzling at least the existence of diverse but highly developed photogenic control systems throughout the animal kingdom, if not the apparently haphazard development of the photogenic capacity itself (Harvey, 1952). The presence of two effector systems, muscles and photocytes, dependent on a single conduction system, has necessitated a notable degree of coordination even in so simple an animal as Renilla, in which hght production and zooid retraction may occur together or independently. At the same time, as Dr. Nicol has suggested, the inhibition of luminescence by light, as seen particularly in the more primitive luminous organisms, may have utility in sparing luminous substrate for the hours of darkness. In view of the teleological at- tractiveness of this speculation it would be of interest to know whether the photoinhibition involves, as in Mnemiopsis (Moore, 1924), the control system (and in this I suggest the possibility of direct action on the nerves as well as via light-sensitive receptors), or also, or instead, photochemical destruction of the photogenic substrate as sug- gested by Dr. Harvey's (1921) experiments with BoUna. In the latter instance it would be desirable to show that reversing the photoinac- tivation, or photodegradation, is less demanding, energetically speak- ing, than the synthetic replacement of substrate simply allowed to "burn" irreversibly. Dr. Nicol's figure for the response of Pelagia seems to indicate that duration of flash declines as intensity falls off in consecutive flashes. This would be contrary to the findings of Brown and King (1931) on fireflies and to what would be expected in luminous extracts, where JOHN BUCK 325 duration would not change with increasing total light (total luciferin) as long as the enzyme-substrate ratio was unchanged. Actually, of course, the shortening of the Pelogiu responses is an artifact of amplifi- cation and would disappear if the intensities were equalized, but it nevertheless justifies some discussion of the general relation between the kinetics of light emission by the organism as a whole and the kinetics of light production in the underlying photochemical system. The discussion will perhaps be useful also in reference to the striking similarity in form (though not in absolute rate), pointed out by Dr. McElroy, between the "pseudoflash" given by anaerobic firefly ex- tracts upon admission of oxygen and the normal flash of the intact firefly. Although twenty years ago it was still possible to think of the actual light-emitting reaction as involving primarily two reactants, luciferin and luciferase, and although the decay portion of the duration- intensity curve of some firefly flashes is logarithmic. Brown and King ( 1931 ) and Snell ( 1932 ) clearly recognized the difficulty of relating the control of luminescence directly to the underlying chemistry. At our more sophisticated level of knowledge today, with many enzyme cycles involved, there is much more difficulty in fixing upon the rate- limiting reaction except in highly purified, in vitro systems. In extracellular luminescence, as in CJmetopterus, obvious kinetic complications are caused by extraneous factors such as secretion delay and mixing delay ( Nicol, 1952 ) . However, even considering intracellular photogeny and making the simplifying assumptions that the response is triggered by a single nerve impulse and that the photogenic cell is a unit ( with all quanta being emitted simultaneously and with the accretion and decay phases clearly dependent on defined reactions), a formidable degree of kinetic indeterminism may enter purely at the biological level of organization as soon as the response of intact animals is considered. This is well illustrated in Renilla where the response is sufficiently leisurely for one to see that several ranks of siphonozooid clusters (each cluster being itself composed of several individual siphonozooids and each siphonozooid of several cells) are involved at a given instant in each luminous wave. One must sternly eschew, therefore, the temptation to guess that the beautifully reproducible and quite possibly logarithmic decay of 326 CONTROL OF BIOLUMINESCENCE H I- » OQ < I I .::'::'•" ^i::?::<-:v '''•:V;. V *••! ',•> . ;. .^ ■^- .•.■•';•'::';■.*%•.• . %*••; .M *•*,*,' •'«; ",; • .;. - .•*.' :•;:, •**»*•*" •,*,• .'•*•** ' ■ ••.••.•.■;:•.*•"* • *.;:,{; • •'** ••.*■ ; •' r,»^ ;. */. viiiiv ■rij) l3«g%iQQ) .•.'.'•.V'liV'. %.. J:- •J^y-'-'-'-'J:::- ••.:: #-#^ ".•:. .':, J::': • ••• ■•:V-".;.-.:.<< /- %0 ■«- •Vis':-* ^ /^ ■^^■^ ^ . ^ .****• •■•:•>;•:•?.•; .'•*• •'.**• Xv**-.. •"•." ''-'A •;.'i *- •'•* .•ri:-*- .V t '.'• f:V-> •■:•.•.•..•.;•;«• '•■• ;.*•:' •..•^^.j.- ■'•■••. ',•'«*•', ...«^-:-.-... !?:•■ ■'■it (NJ ro JOHN BUCK • -327 Plate I Schematized diagrams of the spread of luminescence (stippled areas) in light organs stimulated in various hypothetical ways and the corresponding intensity- duration curves which would be obtained by recording from the whole organ.* The three stages of spread of excitation (A, B, C) are indicated at the corre- sponding points on the graphs which form the last member of each series. The concentric spread of luminescence can be thought of as reflecting either a sort of nerve-net propagation of excitation from the initiating areas through succes- si\e ranks of effector units (as in Renilla) or as the pattern of direct effector innervation, the apparently progressive spread from each focus indicating con- duction delay from the underlying stimulation center (as would be more likely in the firefly). It has been assumed that only a single stimulus is given to each effector unit (repetiti\e stimulation would of course alter the cur\e forms dras- tically) and that a given area of organ lights only once. If the decay of lumi- nescence were long in comparison with the rate of spread of excitation, i.e., if the stimulated regions stayed lit at least until the whole organ was excited, all the patterns of excitation would gi\ e the same general type of intensity-duration curve. Fig. 1. Luminescence spread in an organ initially stimulated centrally. A possible instance of this sort of propagation is the circular prothoracic photophore of the elaterid beetle Pyrophorus in which the flash is so slow that the spread of luminescence is easily observed. In this organ, Heinemann ( 1886 ) observed that the light appeared first at the center of the organ and spread peripher- ally, and died out in re\erse order. (If the stimulation began peripherally, peak intensity would be gained early, giNing a cur\e skewed to the left.) Fig. 2. Luminescence spread in an organ in which the excited region is of con- stant area. The same form of intensity-duration curve is obtained if the excitation starts centrally and spreads laterally in both directions, and if there is only one excited region instead of two. An example of such propagation is seen in the long slender rachis of sea pens. Fig. 3. Luminescence spread in an organ with multiple excitation points. This type of propagation, which is the most likely type in lampyrid fireflies, gives a roughly symmetrical intensity-duration curve which is quite nonspecific. Fig. 4. Multifocal luminescence spread giving a bimodal type of intensity-duration curve reminiscent of that of the female of the firefly Photuris pennsylvanica (Brown and King, 1931). The initial phase of this curve is equivalent to Fig. 3A, but the subsequent spread of luminescence differs in that the excitation points are relatively closer together in relation to the total area of the organ so that after the initial peak is reached (at A), the die-away of luminescence at the 7 points of impingement of the 6 centers of spread exceeds, for a time (4B), the increase in luminous area due to peripheral spread of excitation. Later the peripheral area increase becomes dominant ( 4C ) , followed by a decay essentially similar to that in Fig. 3C. ' Thanks are due Dr. Margaret Keister for making the necessary calcula- tions and executing the figures. 328 CONTROL OF BIOLUMINESCENCE Renilla luminescence, as seen in Dr. Nicol's records, directly reflects the underlying chemical kinetics. There is indeed one possible deduc- tion which can be made about Renilla, though not from Dr. Nicol's present records (which integrate the light from the animal as a whole), which is that the time-intensity curve for the individual siphonozooid cluster must be skewed. This is shown by the fact, noted by Parker (1920), that the leading edge of the wave (first rank of clusters ) is the brightest. The firefly flash is even more difficult to analyze because its bril- liance and short duration prevent visual study of the course of excita- tion through the luminous tissue, and no recordings have been made of the time relations between light emission from different regions of the organ in the same flash. Since the organ itself occupies two abdominal segments in the males of most common American fireflies, asynchrony between the two segments is a distinct possibility, although the early firefly experiments of Briicke (1881), leniently considered, indicate that the innervation could conceivably be so arranged that the two segments are stimulated simultaneously some- what as in the unified response of the squid mantle (Young, 1938). Even so, one would be dealing with from 6000 to 15,000 effectors, depending on whether one considers the "cylinder" or the photocyte to be the smallest individually controlled unit (Buck, 1948), so that it is likely a fortiori that the time-intensity curve for the whole organ represents an integration of many separate asynchronous events rather than the simultaneous firing of all the photocytes in the organ. The strong resemblance of the time-intensity curves of some firefly flashes to the normal distribution curve may be significant in this connection, although I understand from Dr. Hastings that a symmetrical flash is the exception, rather than the rule. However, a gaussian type of curve might reflect the distribution of any one of several factors affecting the control of luminescence, for example the distribution of responsiveness among the individual effectors to a linear change in stimulation frequency (action potentials in nerve). Similarly, plateau- shaped, bimodal or skewed time-intensity curves might result rather simply from the mere architectonics of innervation in the light organ when the recording is made from the animal as a whole (Figs. 1-4), JOHN BUCK 329 or, assuming O2 to be the controlling factor, the tracheation of the organ (cf. Heinemann, 1886). Dr. Nicol's passing reference to the possibility of inhibiting lumi- nescence by limiting oxygen invites amplification. The case has in one sense been strengthened by the recent evidence that oxygen actually participates in the reaction in which the activated molecule emits the light — formerly it was possible to imagine a more subordinate role analogous to that in hexose resynthesis in muscle. As an essential reactant in light production in almost all organisms, oxygen is of course always a potential trigger at the chemical level, but the pos- sibility of its actually ever normally being limiting is strongly reduced by two circumstances. First, no organism, except possibly the firefly (see below), seems to have any anatomical arrangement even con- ceivably capable of rapidly affecting the oxygen supply to the lu- minous tissues. Second, in all systems thus far investigated (which includes the firefly) luminescence persists at oxygen partial pressures far lower than will support any significant amount of respiration. Dark periods, which are often of great length, would thus involve almost complete tissue anaerobiosis. Since oxygen control has often been postulated for the firefly, it may be worth while to examine the evidence thought to favor this view. Oxygen control has been strongly espoused by Snell (1932) and Alexander (1943), but in a critical review of the work (Buck, 1948) I concluded that no evidence had been presented which could not be interpreted as an indirect eflfect via the nervous system, rather than a direct limitation of oxygen in the photochemical reaction. Aside from the obvious dependence of light production on the pres- ence of oxygen, therefore, we have only the characteristically profuse tracheation of the light organ, which may of course be concerned with supplying some oxidative precursor or restoration reaction rather than with control, and the circumstantial but close correlation between the ability of certain species to produce sharp flashes and the pres- ence in those species (only) of "tracheal end cells" at the junction of the supply tracheae with the fine tracheal capillaries (tracheoles) which penetrate the photogenic tissue. From the standpoint of pure plumbing the end cells are indeed strategically situated to retard 330 CONTROL OF BIOLUMINESCENGE gaseous diffusion into the photogenic tissue. However, the volume of the tracheoles is neghgible compared with that of the tissues, and the final effect of oxygen limitation would in any case be seen in the cytoplasm and must involve diffusion of oxygen in solution. Hence, some reason must be given why dissolved oxygen will not diffuse into the photogenic tissue from the surroimding and presumably well- oxygenated blood, and from contiguous tissues. Furthermore, the slowness of diffusion of dissolved gases, particularly at the very low p02 and small gradients at which limitation would occur during the decay phase of luminescence, raises serious doubt that the observed time constants for the flash could be achieved. End-cell control was examined exhaustively in my 1948 review, and the conclusion was reached that control at the enzymatic level, probably via nervous tiiggering, was much more likely — a notable piece of clairvoyance, in view of Dr. McElroy's recent ingenious acetylcholine proposal. The effects of different oxygen tensions on the luminescence of in- tact fireflies (Snell, 1932; Alexander, 1943; Buck, 1948) bring up another facet of the control problem. Normally, captive fireflies do not luminesce visibly while at rest, but if the ambient p02 is gradu- ally reduced to about 4 mm, an "hypoxic glow" (not a flash) develops, which persists steadily for a long time. If the pOo is suddenly in- creased, the luminescence suddenly increases to a high level and then declines to zero ("pseudoflash"). If the p02 is decreased still further, the hypoxic glow dies out. These phenomena have been interpreted in terms of a control mechanism which is inactivated at low pOo, allowing oxygen to diffuse into the photogenic tissue unchecked but producing only a dim light because of the low pOo. The pseudoflash, on this interpretation, would represent a period of bright luminescence due to the increased pOo and terminated quickly by the recovery of the control mechanism. (The die-out with decrease in pOo below 4 mm would of course signify a straight oxygen limita- tion.) Snell (1932) interpreted the control mechanism directly in terms of the end cell, and I (1948) suggested the possibility of the pseudo- flash being due to burn-off of luciferin accumulating during hypoxia as in the bacterial "flash" (according to Dr. McElroy's present firefly scheme this would represent oxidation of accumulated "active inter- JOHN BUCK 331 mediate"). The Snell hypothesis is subject to the objections inherent in any hypothesis of control by oxygen hmitation. The hypothesis that formation of active intermediate is rate-hmiting in normal control has the attractive feature that the rate could be affected either by oxygen limitation ( the hypoxic glow-pseudoflash phenomenon ) or, in normally oxygenated animals, by making some other reactant limiting such as acetate, as proposed in Dr. McElroy's nerve-action control suggestion. There is, however, a possible obstacle in that I have not been able to demonstrate a relation between length of hypoxia and magnitude of pseudoflash. Possibly this obstacle can be resolved with additional work. The end-cell control hypothesis, though seriously defective in some respects, has in it the germ of a possibly important concept. As I pointed out in 1948, the mechanism which controls the hypoxic glow- pseudoflash phenomenon ought to be one which requires energy expenditure, that is, the animal has to do work to keep itself dark. This would accord well both with the effects of hypoxia and with the fact that fireflies develop a long-lasting glow during anesthesia and after death. Such an "active dark" mechanism need not be and ought not to be restricted to oxygen limitation. Although it is perhaps pre- mature to propose specific neuroeffector or chemical control mecha- nisms in either fireflies or other organisms, there is at least the suggestive analogy of skeletal muscle, in which work is done to main- tain the tissue normally in the relaxed state, and the well-established presence, in some Arthropods, of inhibitory nerves. Refere nces Alexander, R. E. 1943. Factors controlling firefly luminescence. /. Cellular and Comp. Physiol, 22, 51-71. Brown, D. E. S., and C. V. King. 1931. The nature of the photogenic re- sponse of Photiiris pennsylvanica. Physiol. Zool., 4, 287-93. Briicke, E. 1881. Vorlestingen iiber Physiologie. 3rd ed., 1, 59-61. Buck, J. B. 1948. The anatomy and physiology of the light organ in firefles. Ann. N. Y. Acad. Sci., 49, 397-482. Buck, J. 1953. Bioluminescence in the study of invertebrate nervous sys- tems. Anat. Record, 117, 594. 332 CONTROL OF BIOLUMINESCENCE Harvey, E. N. 1921. Studies on bioluminescence. XIII. Luminescence in the coelenterates. Biol Bull, 41, 280-87. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Heinemann, C. 1886. Zur Anatomic und Physiologie der Leuchtorgane mexikanischer Cucuyos. Arch, mikroscop. Anat., 27, 296-382. Moore, A. R. 1924. Luminescence in Mnemiopsis. J. Gen. Physiol, 6, 403-12. Nicol, J. A. C. 1952. Studies on Chaetopterus variopedatus (Renier). III. Factors affecting the Hght response. /. Marine Biol. Assoc. U. K., 31, 113-44. Parker, G. H. 1919. The Elementary Nervous System. Lippincott. Parker, G. H. 1920. Activities of colonial animals. II. Neuromuscular move- ments and phosphorescence of Renilla. J. Exptl Zool, 31, 475-515. Snell, P. A. 1932. The control of luminescence in the male lampyrid firefly, Photiiris pennsylvanica, with special reference to the effect of oxygen ten- sion on flashing. /. Cellular and Comp. Physiol, 1, 37-51. Young, J. Z. 1938. The functioning of the giant nerve fibers of the squid. /. Exptl. Biol, 15, 170-85. Discussion Dr. Davenport: To the interesting discussions presented by Dr. Nicol and by Dr. Buck I should Hke to add a few remarks in regard to an aspect of bioluminescence beyond that of the physiological mechan- isms of control. Greatest emphasis has been placed during the conference on the biochemistry of luminescence, for the obvious reason that the phenomenon has paramount importance as a tool of the investigator of the basic metabolism of cells. Some emphasis has been placed on mechanisms of control of luminescence, which in multicellular or- ganisms must of necessity be investigated over the years in each differing luminous species. Some emphasis, largely descriptive but of great interest, has been placed on the natural history of luminous organisms. But the subject of the experimental analysis of behavior in luminescent organisms has barely been touched upon. It is obvious that a careful analysis of the behavior of luminous forms should give us considerable information concerning the im- portance of luminescence in their evolution. Although the question of the adaptive significance of luminescence in particular continuously arises, practically no investigations resembling the interesting ones of Dr. Buck on Lampyrids have been undertaken. As an example of a case of luminescence which has been used, without comparative studies, to bolster important evolutionary theory we have that of the fungivoroid Diptera. Goldschmidt has used the New Zealand species to support the theory that by necessity there must have been sudden great evolutionary changes in morphology, physiology, and behavior for the New Zealand form to have been brought to its present level. It now appears as a result of Dr. Haneda's observations that there are luminous members of this single dipterous family as far separated as New Zealand, North Carolina, and Japan. Such discontinuous distribution in closely related luminous forms indicates that the habit of luminescence in these Diptera, although it is not the same in each species, may be of as great geological antiq- uity as that of the Lampyrids. Unquestionably, studies in comparative behavior in related lu- minous forms should give us a clearer picture of their course of evolution; let us hope that in the future more investigations will be devoted to this interesting facet of the whole subject of biolumines- cence. 333 Luminous Organisms of Japan and the Far East Y. Haneda Tokyo Jikeikai Medical College, Tokyo, and Yokosuka Museum, Yokosuka, Japan My investigation of luminous organisms started in 1934, with special interest in the problem of luminous symbiosis, bet\veen luminous bacteria and fish or squid. From 1937 to 1942 I had several opportu- nities to visit Micronesia, Tropical Asia, and New Guinea as a member of the staff of the Palao Tropical Biological Station and was able to collect luminous organisms and obser\-e their ecology. From 1942 to 1945 I was stationed in the Raffles Museum, Singapore. During my stay in Singapore I was able to visit the Malay Peninsula and the East Indies. During these trips I collected and observed luminous fishes, luminous fungi, fireflies, and other luminous organisms. Un- fortunately, most of my specimens, memoranda, and manuscripts were lost at the end of \\^orld War II while traveling in Middle Sumatra. Finally in 1946 I returned to Japan from Singapore, and in 1948 I again started to study deep sea luminous organisms of Suruga Bay, Japan. From 1951 to 1952 studies were made of luminous organisms of Hachijo Island, located 157 miles south of Tokyo, while I was a member of the Committee on Oceanographic and Biological Research of Hachijo Island. Since 1953 I have continued the work on luminous species of the Pacific coast of Japan, as a member of the Committee on Oceanographic and Biological Research for Marine Resources, sponsored by the Japanese National Commission for UNESCO. The present paper contains some of the results of my observations on the 335 336 LUMINOUS ORGANISMS OF FAR EAST various remarkable and interesting luminous organisms of Japan and the Far East. Bacteria In Japan very few taxonomic reports appear on luminous bacteria, but there are many observations on these organisms. Studies on morphology, immune reactions, cultivation, relation to pH and temperature, effect of salt, action of drugs, antibiotics, and others have been made by many workers, namely, Imamura (1904), Yasaki (1926), Ninomiya (1924), Majima (1931), Kishitani (1933), Takase (1938, 1939), Nakamura (1939, 1940, 1942a,b), Haga (1942), Yasaki and Kimura (1946), Yasaki and Kobayashi (1946), Kozukue (1952a, b), Odawara (1953a,b,c), and Shibata (1953a,b). Symbiotic Luminous Bacteria Studies on symbiotic luminous bacteria have been made by Yasaki (1928, 1929), Kishitani (1928a,b,c, 1930, 1932), Yasaki and Haneda (1935a,b), and Haneda ( 1938a,b, 1940, 1941, 1950, 1951). My own work has been chiefly concerned with fishes of the Acropomatidae, Leiognathidae, Gadidae, and Macrouridae, from which the following conclusions can be drawn. Two species of lu- minous bacteria exist in two species of Acropoma, namely, Acropoma jap07iicwn Giinther and Acropoma hanedai Matsubara; one group of luminous bacteria from fish of the Leiognathidae, and one group of luminous bacteria from fishes of the Gadidae and Macrouridae are all different and constitute new species. The details of the bacterio- logical work will be presented elsewhere. Parasitic Luminous Bacteria Studies on parasitic luminous bacteria were made by Yasaki ( 1927 ) , Majima (1931), and Haneda (1939). Nonluminous Crustacea and insects frequently become luminous when infected with luminous bacteria. Infection of fresh water shrimp has been investigated by Yasaki. The infected shrimp live only a few hours but they are brilliantly luminous. The luminous bacteria were isolated and called Microspira phosphoreum. Majima (1931), through his detailed study, later reidentified it as Vibrio yasakii. Y. HANEDA 337 Nonluminous Isopoda sometimes become luminous from infection by luminous bacteria. A specimen of Megaligia, Japanese name, Funomushi, among numerous nonluminous individuals, was observed at Tomioka Beach near Yokohama in the autumn of 1933. Again in 1939 in Palao, I observed a wood louse, Japanese name, Warajimushi, and a millipede infected by luminous bacteria moving on the ground behind the coral storehouse of Palao Tropical Station. The rather strong light from the wood louse was emitted from the whole body and lasted five days. The normally nonluminous millipede Trigoniulus rugosus is common in the Caroline and Marshall Islands of Micro- nesia, but occasionally luminous bacteria can be cultured from these animals. Saprophytic Luminous Bacteria There are many reports of luminous bacteria cultivated from marine fishes of Japan, and especially since World War II much progress has been made by Dr. Yasaki and his students in studies of luminous bacteria. Kozukue ( 1952b,c ) , Shibata ( 1953a,b,c ) , and Odawara (1953c) cultivated luminous bacteria from the digestive organs of some marine fishes, squid, and crab. The only observation on luminous bacteria from beef was made by Molisch (1926) at the city of Sendai, Japan. Several years before 1940 I tested the cultivation of luminous bacteria from beef, pork, and chicken in Japan. The beef, pork, and chicken, purchased from butcher shops in several cities, were cut into small pieces with ster- ilized knives, put into a dish, and sterilized salt water was poured on them. After 10 to 24 hours I observed these in the dark. Although there was some luminosity developed in experiments conducted during other seasons, in the winter season 65% of the beef, 46% of the pork, and 24% of the chicken became luminous. Strangely enough the luminous bacteria grew only on those purchased in the city of Ohgaki, Gifu Pref. I cultivated these luminous bacteria in culture media con- taining 0.5% salt and found that the bacteria emitted light strongly. Important research, using luminous bacteria as tests for antibiotics (Adawara, 1953a,b); for study of spectral distribution (Akaba, 1938; Takase, 1940; Haneda, Takase, and Kumagai, 1940); for observation with the fluoromicroscope (Shibata, 1953d), and for mutant investi- 338 LUMINOUS ORGANISMS OF FAR EAST gation (lizima and Kikkawa, unpublished) have been made, but space does not allow a description of the results. During World War II, luminous bacteria were grown for illumina- tion in blackouts, and recently an extract of luminous bacteria called "Florads" was made for sale by Dr. Takino (1953) of the Dainippon Zoki Institute for Medical Research, Osaka, Japan. Florads costs 500 yen per box of 5 ampules, each containing 2 cc for injection. Accord- ing to Dr. Takino, Florads has an anti-allergic and neurotropic action and acts effectively upon rheumatism of the joints and spontaneous gangrene without deleterious effects. Fungi It is a most remarkable fact that many species of luminous fungi appear at night in the rainy season in the forests or jungles of the tropics. The decayed wood that had grown luminous fungi was col- lected and brought back to Japan from tropical countries, where the fungi continued to glow in my laboratory in Tokyo during the summer season and I was able to observe in detail their ecology, the intensity of light, and color of their light. Eight new species of these luminous fungi from Micronesia and North Borneo were identified by the late Dr. S. Kawamura. The results of my observation of the luminosity were published by me (1939), and a taxonomical report was made by Kawamura (1940) in Japanese. Five more new species were also identified by Kawamura, but he died in 1943 before he was able to publish his final taxonomical report. Pleurotus lunaillustris and Mycena bambusa, named by Kawa- mura, are widely distributed in the tropics. In Singapore Dr. E. J. H. Corner and I collaborated in our collection and observations of lu- minous fungi and some of our findings were published by Dr. Corner (1950). Luminous fungi of Japan On the Pacific coast of South Japan and adjacent islands, some genera of luminous fungi are found which are common in Micronesia, East Indies, and the Malay Peninsula. In Japan the following four luminous species are well known: Lampteromyces japonicus (Kawamura) Singer, Armillaria mellea Y. HANEDA 339 (Vahl) Karsten, Dictyopamis pusillus (Lev), Singer, and Mycena cyanaphos from Bonin Island. Recently Dr. Kobayashi (1951) re- ported the following five luminous species from Miyazaki Pref. of South Kyushu, Japan: Poromyceiia Hanedai Kobayashi, Dictyopamis foliicohis Kobayashi, Mycena pseudostylobates Kobayashi, Mycena daisJiogunemis Kobayashi, and Mycena sp. Lanipteromyces japonicus is a large luminous fungus which grows on the dead trunks of beech trees in autumn. It is famous for the Japanese name Tsukiyo-ddke, meaning "moon night mushroom." It is a poisonous species and closely resembles the nonluminous and edible species Pleurotus ostreatus. According to Dr. S. Kawamura (1915), the luminescence comes only from the lamellae of the fruit- body, and not from the spores. However, according to my observa- tions, spores fallen upon moist blotting paper are slightly luminous. Polypoms Hanedai Kawamura (1940) is the synonym of Poromycena Hanedai. I found it in a jungle near Tawao, North Borneo, in March 1938. In August, 1953, Mr. Okuyama collected this fungus which had grown on decayed trunks of mulberry trees at Hachijo Island. Three other luminous species were found by us on Hachijo Island: Mycena chlorophos, Dictyopamis gloeocyst Corner, and Mycena lux- coeli Corner. The last mentioned species, illustrated in Fig. 1, is very common. Tropical Luminous Fungi Recently Dr. Corner has identified specimens of luminous fungi from the tropics and from Hachijo Island. According to him, seven Mycena, two Poromycena, two Dictyopamis and one Pleurotus were identified as Mycena chlorophos (B. & C.) (Palao, 1937; Celebes, 1943; Hachijo Island, 1952), Mycena rorida sp. nov. (Singapore, 1944), Mycena pruinvsoviscida sp. nov. var. rahaulensis var. nov. (Rabaul, 1942), Mycena lux-coeli sp. nov. (Hachijo Island, 1951), Mycena suhlucens sp. nov. (Amboina, 1942), Mycena noctilucens Kawam (Yap, 1937), Mycena illuminans Henn (Singapore, 1944), Poromycena Hanedai Kobayashi (Borneo, 1938; Rabaul, 1942; Miya- zaki, 1951; Hachijo Island, 1952), Poromycena manipularis sp. nov. (Ponape, 1940), Dictyopamis luminescens sp. nov. (Singapore, 1944), Dictyopanus gloeocyst sp. nov. (Hachijo Island, 1951), illustrated in 340 LUMINOUS ORGANISMS OF FAR EAST Fig. 1. The luminous fungus, Mycena lux-coeli, from Hachijo Island, Japan, photographed by daylight (top) and by its own light (bottom). Fig. 2, and Pleurotus noctilucens Lev, (Palao, 1937; Yap, Saipan, Rota, Truk, Ponape, 1940; North Borneo, 1938; Manukwari, New Guinea, 1942; Singapore, 1943; Java, 1944). Mycena hambusa Kawa- mura is a synonym of Mycena chlorophos, and Pleurotus lunaillustris is a synonym of Pleurotus noctilucens. Y. HANEDA 341 Fig. 2. Dictyopanus gloeocyst, a luminous fungus from Hachijo Island, Japan. Among these the most interesting fungi are Mijcena pruinosoviscida var. rabaulensis from Rabaul and Mycena rorida from Singapore. Both species are remarkable among many luminous fungi because only the fresh damp spores which have fallen out of the fruitbody are lu- minous. The spores which fall on a dry place are not luminous, but they become luminous when water is dropped on them. From Formosa Dr. K. Kominami (1930) reported a luminous species, Mycena photogena (Japanese name Hotaru-Dake, meaning 342 LUMINOUS ORGANISMS OF FAR EAST "firefly fungus"). According to him, the fruitbody of this fungus is green in color, and the color of the light is blue. Lamellae and spores are luminous. In India Dr. Bose (1926, 1930) reported that leaves, stalks, grass roots, and living roots from tlie forests of Bengal were luminous. As the result of pure cultivation from these luminous leaves, he de- cided that the light of leaves might be due to various species of Mycena. I have cultivated the luminous mycelium from spores of various luminous fungi and observed and measured the intensity of light, color of light, spectral distribution, and the relation to temperature. This work will be published in a separate paper. Other Miscellaneous Luminous Fungi Besides the above-mentioned luminous fungi, I collected on my expeditions to tropical countries the following luminous species identi- fied by Dr. Kawamura: Mycena phosphora, Mycena microiUumina, and Marasmius phosphorus from Palao, Mycena vapensis from Yap Island, and Mycena citrinella var. iUumina from Ponape Island. Un- fortimately, these specimens were lost during World War II in Tokyo and therefore have not been reidentified by Dr. Corner. The luminous fungi hitherto known all belong to Hymenomyceti- neae, which comprise the families of Agaricaceae and some Polypora- ceae. However, I collected in January, 1940, a minute luminous fungus which Dr. Kawamura identified as belonging to the Nidulariineae, shown in Fig. 3. These minute fungi were found on the bark of Rhizophora mucronata in a mangrove zone in Markyok village of Palao. Under the microscope they can be seen to emit a light, bluish in color. Dinoflagellata In Japan taxonomical studies of Dinoflagellata have been made by such workers as Nishikawa (1901), Okamura and Nishikawa (1904), Kofoid (1931), and T. H. Abe (1927, 1936, 1940). Many luminous species are known, but in the above reports the luminosity was not recorded. Y. HANEDA 343 Fig. 3. Minute luminous fungi of the family Nidulariineae from Palao, Micro- nesia ( X 54 ) . Noctiluca miliaris Noctiluca is one of the most remarkable luminous organisms, abun- dant in Japanese waters throughout the year. Especially from April to June large quantities of Noctiluca appear, and masses of them floating on the sea are sometimes blown to shore by the wind. A mass of Noctiluca can be several centimeters in thickness, making the sea about it gelatinous and coloring the water a reddish brick. Japanese 344 LUMINOUS ORGANISMS OF FAR EAST fishermen call it* "Akashio," meaning "red stream." Many fish are at times killed by such a stream, and the masses are strongly luminous at night, Noctiluca does not occur in the Pacific coral islands, such as Palao, Saipan, Truk, Yap, Ponape, and Marshall Islands, but appears abun- dantly near the waters of continental islands, such as New Guinea, Sumatra, Borneo, Java, Celebes, and also along the coast of the Malay Peninsula and Indo-China. However, I have collected it in the sea 300 miles north of the Celebes, where it developed along the coast and was carried far distances by an ocean current. Noctiluca from Japan is pale pink in color, but the tropical species from the East Indies and New Guinea are green. The green color is due to symbionts, which Weill ( 1929 ) has called Chloroflagellates, while Hada ( private communication ) considers them Chlamydomonas. Ostroumoff ( 1924 ) reported that green Noctiluca is not luminous, but I have observed brilliantly luminescent green Noctiluca on the north coast of West New Guinea and other islands of the East Indies, and also in Sandakan Bay of North Borneo. The luminosity of Sandakan Bay was the most beautiful. Dr. Ueno (1937) found Noctiluca and the luminous species, Cera- tium fusus, and also Peridlnium sp. in the swamp, Mokoto-Numa of Hokkaido, Japan. This swamp was connected with the sea, the upper surface being fresh water and the bottom sea water. According to him, Noctilucae were floating 2 to 3 meters deep at the surface. There were 44,000 per liter, and the color, due to green Chlamydomonas, was green as in all tropical species. Unfortunately the luminosity of these Noctilucae was not observed. Macroplankton Reports on luminosity of the macroplankton such as jellyfish, Pyro- soma, and Salpa are almost lacking in Japan. The only reference found was one on eggs of a Ctenophore, by Dr. Yo. K. Okada (1926). T. Komai has reported ctenophores (1918-21) in the neighborhood of the Misaki Marine Biological Station and also cubomedusae ( 1938 ) on the northern Pacific coast of Japan. Uchida ( 1928, 1929 ) described Japanese Hydromedusae around the coast of Japan, and T. Kawamura (1915) reported Siphonophores in the vicinity of Misaki. While Y. HANEDA 345 probably luminous, the light of these specimens was not actually observed. In my own observations, luminous jellyfish appear abundantly along the coast of Japan, especially from March to June. About the end of April, 1953, I collected many luminous Pelagia in a flying fish net near Hachijo Island, and in May of the same year observed luminous Pclagiu in great numbers in Yokosuka harbor and along the west coast of Izu Peninsula. In the flying fish nets at Hachijo Island, many luminous Ctenophora such as Beroe and Cesttim are caught. In June, 1953, along the coast of Namerikawa, Toyama Bay, numerous Beroe are often mixed with luminous squid, Wotasenia scintillans, the whole net becoming bril- liantly luminous. Several species of Siphonophora, such as Fraija, Diphes, and Abyla, which were caught in a plankton net at Aigae Bay, Hachijo Island, were also observed emitting light. In May, 1953, I collected Aequorea sp. by trawl net in Suruga Bay. This is a rather large Medusa, 200 mm in diameter, which emits a flash of light on irritation. In September, 1943, in the sea around the Thousand Islands, coral islands off Jakarta, Java, species of Ctenophora which emitted a flash of light when irritated were found. Observations by Undersea Observation Chamber Twice I have had the opportunity of observing luminous organisms in the sea at night while riding in an undersea observation chamber, called "Ktirosliw' (Inoue, Sasaki, and Oaki, 1952, 1953). The observation chamber was 3.15 meters in height, 1.48 meters in diameter, and weighed from 4440 to 5220 kg, with a buoyancy of from 4192 to 4297 kg. The chamber had a main observation window (150 mm diameter) with controllable reflector, 3 auxiliary observa- tion windows ( 100 mm diameter ) , telephone, teletalk, oxygen feeder, CO2 absorbing unit, projector lamps and miscellaneous gages. The Kuroshio observation chamber was hung down along the outside of the hull of the mother ship by means of a steel wire pulley on the end of a derrick. The lowering and raising was done by an electric motor winch, usually at a speed of about 20 meters per minute. In all, 130 diving observations were made in 1952 along the Japan 346 LUMINOUS ORGANISMS OF FAR EAST coast. The following summer observations were continued oflE the coast of Otaru, Hokkaido. My first experience was in December, 1952, in the sea off Ito, Izu Peninsula, and the second was in August, 1953, in the sea off Otaru, Hokkaido. In both descents, conducted at night, I saw the lumines- cence of Beroe and Cestum under natural conditions and at a depth of about 20 to 50 meters. I can never forget the beautiful luminous appearance of the transparent, long, bandlike Cestum through the window glass of the chamber, as the animal emitted a very beautiful flash of light. At the same time I saw numberless luminous Copepoda running across the window like a snowstorm. Below the surface, suspended materials like snowflakes could be seen with the naked eye, wherever the beam of the projector light fell. According to the laboratory experiments of Drs, Suzuki and Kato of Hokkaido University, these suspended flakes are assumed to be aggregates of distintegrating bodies of plankton organisms. These unidentified masses have been named "Marine snow" or "Sea snow" or "Plankton snow." One night in the summer of 1953, in the sea off Otaru, Hokkaido, I observed that some of these suspended flakes were luminous like the ctenophore Beroe. They were suspended in the water and, coming gradually to the window glass, they were crushed to minute luminous spots and then faded out. My own opinion is that luminous bacteria grow on the suspended flakes, causing them to be luminous. Pyrosoma and Salpa There is one report on a possible luminosity of Salpa in Japan. According to Tokioka (1937) CtjcJosalpa pin7mta var. polae (Sigl), 18 to 22 mm in length, appearing at Seto, Wakayama Pref., has one luminous stripe, but he never saw hght from these individuals in the dark. Pyrosoma and Salpa are common in Japanese waters. About the middle of April, great numbers occur in the waters of Hachijo Island. These Pyrosomae have been identified as Pyrosoma atlanticum Peron and two species of Salpa as Thetys vagina (Tilesius) and Pegea confoederata (Forskal) by Dr. T. Tokioka. Another small spe- cies of Pyrosoma, which was caught in a trawl net in Suruga Bay was identified as Pyrosoma verticillatum Neuman. Y. HANEDA 347 The two above-mentioned species of Pyrosoma are brilliantly lumi- nous when stimulated. The species of Salpa are not luminous under natural conditions. They emit light only on strong stimulation, such as cutting or tearing the body. As the luminous regions were small and the periods of luminosity very short, I could not determine the posi- tion of the luminous organs. Annelida Polychaeta There are many species of luminous Polychaeta in Japan. However, no reference to their luminosity has appeared except in Chaetopterus variopedatus and Mesochaetopterus japonicus Fujiwara ( 1935 ) . Chae- topterus variopedatus, which appears to be a species of worldwide distribution, is found on the sandy bottom of Japanese waters and also along the west coast of Korea. Fujiwara (1935) described somewhat difiFerent light regions occurring in the Mesochaetopterus japonicus of western Japan. During my stay in Palao, I collected some species of luminous Polychaeta. Among them Onuphis sp. and Stylarioides parmatus Grube (1878), both identified by Mr. K. Takahashi, are the most interesting species. Onuphis sp., belonging to the family Eunicidae, lives in coral in the neighborhood of the Palao Tropical Biological Station, Corror Island, Palao. This Polychaete, a beautiful pink in color, with brown stripes, is 234 mm long, 5 mm wide, and has 204 segments. The pairs of luminous spots are arranged from the first setigerous segment to the last on the abdominal lateral margin of each segment. The animal emits a bluish green hght when stimulated, but it does not discharge a luminous slime. It is closely related to the species of Onuphis investigatoris Fauvel from the Arabian Sea, the luminosity of which is unknown. A single specimen of Stylarioides parmatus was collected from the coral reef of the Corror Island, Palao. It lives in a tube 20 mm long and 2 mm in diameter, which is attached to a dead shell. Its cephalic plate becomes luminous when irritated, emitting a yellowish green light. No luminous slime is discharged. This species occurs in the Philippines, Ceylon, Madras, Madagascar, and New Zealand, but its luminosity has never been reported. 348 LUMINOUS ORGANISMS OF FAR EAST In April, 1953, I collected several pelagic species of luminous Tomopteris with a plankton net in Aigae Bay, Hachijo Island. At the pearl cultivating station of Usa, Kochi Pref., I recently col- lected luminous Tlielepus sp., the external appearance of which is similar to that of Pohjcimis. These organisms live in the living shell of pearl oysters and the tentacles become luminous when stimulated. Oligochaeta In Japan there appear two species of luminous earthworms. One is terrestrial (Microscolex phosplwreus) and the other is a sea earth- worm {Pondodrilus motsushimensis) . Microscolex phosplwreus is worldwide in distribution. In Japan it was first found in 1934 at Oiso, Kanagawa Pref., and later it was collected by many persons in every part of Japan. This worm is small, 40 mm in length and 1-1.5 mm in diameter, pale pink in color, and translucent. It is not luminous under normal conditions but after strong stimulation it discharges luminous mucus from mouth and anus. In November, 1939, Mr. Kuroki, a teacher of the Fukuoka Middle School in Kushu, observed many luminous earthworms crawling on the paved road near Fukuoka. On that cold rainy night they were trod under foot by pedestrians aiid crushed by bicycles and cars, and the soles of shoes, the tires of bicycles and cars became brilliantly luminous. I also collected some in my private garden at Zushi, Kanagawa Pref., in November, 1948. As these species are small and are not luminous under normal con- ditions, they are not known by most of the Japanese people. The sea earthworm is widely distributed along the coast of Japan. The body, which is pale pink and translucent, is about 100 mm in length and has 100 to 105 segments. The luminosity of this worm was discovered by the late Dr. Kanda and myself when we saw them in the wet sand at the tidal fine near Yokohama. This worm is also not luminous under normal conditions, but it will discharge a yellow luminous mucus from mouth and anus on strong stimulation or in- jury. I have collected luminous sea earthworms at the tidal line of tropical countries of Asia. According to Dr. Obuchi, these tropical species are not the same as the Japanese species. I have also endeavored to cultivate luminous bacteria from the luminous fluid of both species of worms and have obtained negative Y. HAXEDA 349 results. Microscopic examination did not reveal any luminous bac- teria. On blotting paper the luminosity of the luminous mucus faded out after a few minutes while the paper was drying, but dropping water on the blotting paper revived the luminosity. The color of the light of luminous mucus on the blotting paper was pale blue, which changed to yellowish when the paper was rubbed. The intensity of luminosity was also increased by rubbing. Mollusca Except Cephalopoda Luminous species of the great group of Mollusca, excluding squids, are comparatively rare. Seven genera of Gastropoda, Phillirrhoe, Tethys, Kaloplocamiis, Plocamophorns, Latia, Dijakia, and Tonrui, are known to be luminous. Two genera of Pelecypoda, Pholas and RoceUaria, are also known to be luminous. In Japan the luminous species of Gastropoda, Opisthobranchiata or sea slugs, in which the shell is absent, were studied by a number of workers. Dr. Okada and Dr. Baba observed luminous sea slugs, Plocamophorus tilesii, in Mutsu Bay. The mucous cells were scattered over the whole body, and the luminous slime came out of mucous cells. Beautifully colored plates of two luminous sea slugs, namely Ploca- mophorus tilesii and Plocamophorus imperialis, were reproduced in the book Opisfhohranchia of Sagami Bay collected by His Majesty, the Emperor of Japan (1949). The beautifully colored large sea slug, Calinga ornata, found in Japanese waters, is also known to be lumi- nous. Recently Kato (1949a and b) reported the luminosity of the small sea slug, Kaloplocamiis ramosum, which comes from the east coast of Japan. The animal has many luminous cells and no pores, but the luminous cells contain large granules. It emits intermittently bluish Hashes of light when irritated but it discharges no secretion into the water, although other luminous sea slugs are known to do so. I have observed it in the dark at the beach of Usa Marine Biological Station of Kochi University in November, 1953, and have tested for the luciferin-luciferase and ATP reactions, with negative results. The famous transparent pelagic nudibranch Pliyllirrhoe bucephala is occa- sionally collected in the neighborhood of the Misaki Marine Biological Station. 350 LUMINOUS ORGANISMS OF FAR EAST Among Pelecypoda, Pholas dactylus was at one time thought to be the only luminous species. However, in 1939 at Palao, I found that Rocellaria grandis, a coral boring shell, is also luminous. This animal has a thin white shell 45 mm long, 15 mm wide, and 15 mm in height. I have collected living Rocellaria from the living corals Favites virens, Goniastrea parvistella, and Pontes tenuis in Iwayama Bay of Corror Island, Palao. Rocellaria has a pair of luminous stripes on the mantle along the pallial lines and produces a luminous secretion which fills the mantle cavity and spouts from the siphon when irritated. The dried luminous tissue recovers its luminosity when moistened and the luciferin-luciferase reaction is positive. Among the Gastropoda Pulmonata it was thought that Latia neri- toides (Suter, 1890), a New Zealand fresh water limpet, was the only luminous species. It has been fully studied by Bowden (1950). How- ever, we found a luminous snail, Diakia striata of the Zonitidae, in Singapore. One night in September, 1943, when Mr. Kumazawa, entomologist, was collecting luminous larvae of fireflies on the lawn of the Good Wood Park Hotel, Scot Road, Singapore, he saw a weak Hght from a small land snail and informed me of the possibility of luminescence in land snails. The next evening we went to the place and were astonished to observe a true luminescence in this animal. The snail, about 10 to 15 mm in diameter, lives on grass or lawns in Singapore. The type specimen in Raffles Museum was collected from Gunung Pelai, Johore. I have also collected specimens in various regions of the Malay Peninsula and Kalimon Island of the Rhio Archipelago. The light appears inside the anterior region of the foot and cannot be seen when the animal is irritated and has withdrawn within its shell. When expanded the bluish white light passes through the translucent muscles of the head and of the foot and flickers like a firefly. The time relations of the flashing depend on the stage of development of the snail and various other conditions, such as tem- perature, humidity, and light. Young snails, immediately after hatch- ing, about 1 mm in shell diameter, emit a weak luminescence over the entire foot. The light appears continuous, but on close inspection the diflFused glow can be resolved into small flashes scattered over the area. As the snail grows, the flicker rate diminishes, and some full grown individuals do not luminesce at all. The normal duration of a Y. HANEDA 351 flash at 25° C is two to three seconds. The flashing is spontaneous; no kiminescence appears on stimulation. The intensity of each flash is the same and fairly bright, visible in electric light if shaded by the hand. The luminescence comes from luminous cells, not from luminous bacteria. The luminous organs, consisting of large luminous gland cells, lie below the mucous gland of the foot and surround its open- ing, but no luminous material is secreted to the outside. Hence this snail is self-luminous, and the light is intracellular. In sections, the luminous tissue can be easily distinguished. The luminous organ of a shell 15 mm in diameter is 2 mm long and 1.5 mm wide. The luminous cells are pale green, while the mucous cells of this region are yellowish orange when stained by safranin. Luminous Marine Snails Lu::.inous species of Gastropoda Prosobranchiata are also very rare. According to Turner (see Harvey, 1952) Tonna galea Linne, a marine Gastropoda, is luminescent. When this animal is moving about with its foot well extended, it emits a greenish white light. We have found two species of luminous Gastropoda on the beach of Borawazawa, Sueyoshi Village, Hachijo Island. During ebb tide on Apiil 23, 1953, my friend, Mr. H. Okuyama, saw some small marine snails emit light as they rolled in the sea water when he raised a stone. We were very much astonished and collected many specimens, examining them repeatedly to find out if these marine snails were really luminous, or were luminous only because of eating some lumi^ nous matter, or were infected luminous bacteria. As a result of these observations, I decided the snails are truly luminous, possessing a luminous organ on their mantles. These luminous marine snails, shown in Fig. 4, are very small. The shell, a beautiful pale pink with brown bands, is 10.5 mm high and has a diameter of 6 mm. The scientific name of this snail is Planaxis viratus Smith, identified by Dr. I. Taki of the National Science Mu- seum in Tokyo. I found another species, Planaxis perescelida Dall, in our collection and this also had the same luminous organ and emitted light. This species is rather more slender than the Planaxis viratus. Its height is 12 mm and its diameter 5 mm. The color is dark brown 352 LUMINOUS ORGANISMS OF FAR EAST with pale black or dark brown bands. Each species is widely dis- tributed in southern Japan. When the animal is under natural conditions on a rock with its foot well extended, no Hght can be seen. However, the light appears on strong stimulation. If many specimens are placed in a bottle and ^«t^ Fig. 4. The luminous marine snail, Planaxis viratus, from Hachijo Island, Japan. well shaken in the dark, some of them become luminous and twinkle. The light continues one or two minutes and then gradually disap- pears. If the body of the snail is irritated, the Ught reappears, and if placed in fresh water, the light continues for a longer time. When the animal is taken out of the shell by crushing the shell and is observed under a binocular or low power microscope in the dark, the luminous region is found to be on the dorsal part of the mantle in a limited area, as shown in Fig. 5. Under the microscope number- less luminous dots can be seen appearing or disappearing. The lumi- nescence of this animal is intracellular and the luminous organ con- Y. HANEDA 353 sists of many groups of luminous cells which run parallel to each other. No luminous cells are scattered over the whole body. When the dried materials of both species were ground in a mortar and moistened with water, luminescence reappeared and then disap- peared. The luciferin-luciferase reaction, tested by mixing hot water (70° C) and cold water extracts of crushed Planoxis, was negative. When the cold water extract is allowed to stand until the light disap- pears, it will not again emit light when ATP is added. Fig. 5. A Planaxis viratus removed from its shell to show the luminous organ in diagram at right. Cephalopoda There are many reports of luminous squid in Japan. According to Harvey (1952) luminous squid are divided into the following three groups, depending on their method of Hght production: (1) squid associated with luminous bacteria, (2) squid producing an abundant luminous secretion, and (3) squid with well-developed photophores and intracellular luminescence. In Japan the first and third groups have been studied by such workers as Watase (1905), Ishikawa 354 LUMINOUS ORGANISMS OF FAR EAST (1913), Harvey (1917), Shoji (1919), Y. K. Okada (1933), and Hasama ( 1941 ) . However, the second group has not been investigated until recently. Squid associated with luminous bacteria were studied by Dr. T. Kishitani. He reported that myopsid squid, such as Loligo edulis, Sepiola birostrata, and Euprymna morsel, are luminous species pos- sessing symbiotic luminous bacteria. All live in shallow water. Their luminous organs have openings to the exterior, and luminous bacteria live symbiotically in the ducts. In Singapore in 1944 I collected and observed another species of myopsid squid having luminous bacteria. Regarding the second group, I recently caught in a trawl net at Suruga Bay a luminous squid producing an abundant luminous secre- tion. The scientific name of this squid is Stoloteuthis leucoptera Ver- rill, identified by Dr. Y. Okada, and is a species closely related to Heteroteiithis. As shown in Fig. 6, its mantle is 20 to 30 mm long, with a large white band that has been mistaken by some observers for a luminous skin organ. The round white luminous organ is situated on the ink sac and is connected to the exterior by two pores. A sec- tion of the luminous organ is similar to that of Heteroteiithis, which also produces a luminous secretion. Stoloteuthis is not luminous when freely swimming, but, if touched, will spurt through the funnel a beautiful bright bluish secretion like that of the mollusc Pholas. This luminous secretion comes out of the pores of the luminous organ. I have endeavored to cultivate luminous bacteria from the luminous organ and the luminous secretion, but all the results were negative. As this luminous squid is comparatively easy to catch in trawl nets at Suruga Bay, I intend to study it in detail in the future. The third group of squid with well-developed photophores have been studied by many workers in Japan. Among many deep sea lumi- nous oegopsid squids, Watasenia scintiUans is most famous and re- markable. Watasenia scintillans (Japanese name, Hotaru-Ika, mean- ing firefly squid) comes to the surface in Toyama Bay, on the coast of the Nippon Sea, each year during late April to mid-June, to breed. This species is caught and dried on the beach in the sun for food and is an important commercial product of Uozu and Namerikawa, towns on Toyama Bay that are the best locations for collecting. Y. HANEDA 355 Fig. 6. Photograph (top) of luminous squid, Stoloteuthis leucoptera, which ejects a luminous secretion from its funnel when disturbed, with diagram of parts (bottom). Left, a dorsal view. Right, the body cavity has been opened to show the luminous organ ( L. O. ) with pores ( p ) . 356 LUMINOUS ORGANISMS OF FAR EAST Other interesting deep sea luminous squids, such as Abralia ja- ponica, Chiroteuthis, and CaUiteuthis, are obtained in shrimp nets of trawlers in Suruga Bay. Appearing in particularly great numbers are Chiroteuthis imperator (Japanese name, Yurei-Ika, meaning ghost squid), which fishermen throw away because it has no commercial value. This squid has many well-developed luminous organs on the eyeballs, tentacles, and abdomen. The tentacles of large specimens extend 2 meters and attached to them for some length are many lumi- nous organs, like many small lamps on one thread. Sasaki (1915) described two new luminous oegopsids from the Bay of Sagami, namely Meleagroteuthis separata and Symplectoteuthis hiininosa. The luminous organs of the former are uniform in appear- ance, thickly covering the ventral surface of the whole body, but are found in less number on the dorsal surface. Luminous organs are found also on the ventral and dorsal surfaces of the head and arms. The Symplectoteuthis huninosa were taken six miles off Misaki, at 700 fathoms in 1906. According to Sasaki, a macula is found on the ventral surface of the head. On the ventral surface of the mantle there occur a pair of longitudinal zones of the same character which run along the whole length of the mantle and are divided into three parts. He supposed the maculae and zones to be luminous organs, judging from their histological structures. Crustacea Many self-luminous species appear among the orders of Crustacea in Japan. The luminous species Cypridina, belonging to the Ostracoda and Heterocarpus sibogae of the Decapoda, which secrete a luminous liquid, while Sergestes prehensilis of the Decapoda, with photophores, and certain freshwater luminous shrimp are the most interesting and remarkable. Cypridina Cypridina hilgendorfii (Japanese name, Umihotaru, meaning sea firefly), shown in Fig. 7, is abundant along the coast of Japan from July to September. A well-known crustacean, it is often used for biochemical studies. It is 3 mm long and produces a strong lumines- cent secretion. It can be preserved in a dried condition, and the Y. HANEDA 357 Fig. 7. Photo of Cypridina Mgendorfii from Japan. luminescence can be restored with moisture. This Cypridina species Hves on the sandy bottom near the shore and comes out to feed at night. There are several methods of collecting Cypridina. During World War II, Japanese army officers used large earthenware pots baited with fish, which they lowered to the sandy bottom at night. This method was not too satisfactory because nonluminous Crustacea and marine snails were mixed with the Cypridina. A simple method for mass collection is to tie fish heads to long strings and suspend them in the sea at night. Cypridina gather on the fish heads and may be easily caught. The living specimens are then placed on blotters in sunhght to dry, or in a heater with a low temperature. If dried while still alive, the two hinged valves become transparent, and a bril- hant luminescence appears on moistening; if dried after death, the valve changes to a nontransparent white, and the luminosity is weak. The Mihtary Institution of Japan during World War II was plan- ning to utilize Cijpridina light for reading at night. I saw some sam- ples of Cypridina powder in Singapore which were sent from Japan, but in tropical countries the powder putrified very rapidly when moistened, and I do not know whether it was useful or not. Cypridina noctiluca, shown in Fig. 8, is a tropical species com- 358 LUMINOUS ORGANISMS OF FAR EAST Fig. 8. Photo of Cijpridina noctiluca from tropical Asia. monly seen in the waters of Palao, Java, Malay, and even Hachijo Island. It is slightly smaller than Cypridina hilgendorfii, being 2 mm long and pyramidal in form. It is a pelagic species and can be taken only by plankton net, since the animals are not attracted to fish bait. At Hachijo Island, as in tropical seas, Cypridina noctiluca is abundant, but no specimens of Cypridina hilgendorfii appeared. The light of Cypridina noctiluca and that of Cypridina hilgendorfii are quite sim- ilar in color. Luminous Shrimp In the winter season from October to May near the towns of Yui and Kambara in Suruga Bay, Shizuoka Pref., more than 100 trawl- fishing boats go out to catch deep sea shrimp. The scientific name of these shrimp is Sergestes prehensilis, and the Japanese name is Sakura-Ebi, meaning cherry shrimp. These are dried on the beaches in sunlight and are an important commercial product of the two towns. This shrimp possesses 157 photophores scattered over the body. The trawl nets which catch the shrimp are drawn from depths of from 50 to 100 fathoms by two fishing boats, and they bring up many other kinds of luminous animals, such as other species of shrimps, luminous squid, lantern fish, hatchet fish, Pyrosoma, jellyfish, and Y. HANEDA 359 other luminous deep sea animals. These forms live mostly at middle depths and migrate upward at night. Among them the most interest- ing and remarkable are the shrimp and squid which project luminous clouds into the sea water. Heterocarpus sibogae de Man. is 5 to 6 cm long and red in color. This species, like Heterocarpus alfonsi, has glands at the base of the antennae. If the material is dried rapidly while living by a heater with temperature not exceeding 40-50° C, its luminosity can be recovered upon moistening. When dead mate- rial is dried, the luminosity is weak. The luciferin-luciferase reaction is positive, and the species would be useful in biochemical studies. Pathogenic Luminous Shrimp The luminosity of the freshwater luminous shrimp Xiphocaridina compressa is much more interesting. These are found in Lake Suwa, 100 miles from the sea and 800 meters above the sea level, but are also widely distributed in freshwater in Japan. On hot summer nights the shrimp regularly become luminous be- cause of infection by pathogenic luminous bacteria, and the entire body, with the exception of the eye, but even tentacles and legs, will shine. The morphological aspects of the luminous bacteria (Microspira phosphoretim) are similar to the Cholera Vibrio or the Vibrio durnbar of Germany. The bacteria are easily cultivated from the shrimp, using a 0.5% NaCl culture media. Later, in 1928, the shrimp was found in a brook in the rice fields alongside the Tone River near the town of Sawara, Chiba Pref., about 30 miles from the mouth of the estuary. Because of the rare beauty of the shrimp as they mass and luminesce on hot summer nights, their destruction is prohibited by the govern- ment. Myriapoda Luminous Centipedes There appear to be no luminous Myriapoda in Japan, but small threadlike luminous centipedes {Orphaneus brevihbiatus) , 60-60 mm long, are distributed over Micronesia, the East Indies, Malay Penin- sula, and Indo-China, even Formosa and Okinawa Islands. Translu- 360 LUMINOUS ORGANISMS OF FAR EAST cent and pale brown in color, they live in the walls of native houses, in the furniture or beds made of pandanus leaves. Palao people call them Terai Was, or luminous paint, and think they are ear eaters, although they are not poisonous. The animal discharges a very strong greenish luminescent slime from both sides of the body segments when irritated. If put into chloroform vapor, they secrete a luminous mucus from every segment, which is very striking. One night in 1938 at Arumizu Village, Corror Island, Palao, I saw a centipede caught in the web of a spider which was very beautiful in the lumi- nosity of the discharged slime. Luminous Millipede Records of luminous diplopods or millipedes are few. I found a self-luminous millipede in the Truk Islands of Micronesia in 1939, from which no luminous bacteria could be grown. Takakuwa ( 1941 ) gave it the name of Spirobolellus phosphoretis, a new species. The whole body of this animal, with the exception of head and legs, emits a weak bluish light which becomes brighter upon irritation, but it excretes no luminous mucus as in the case of luminous centipedes or earthworms. It is very common in the Truk Islands, and I saw it many times during my stay of one week there. Many of them gather at the base of coconut trees. Although the light intensity is no stronger than that of mycelium of luminous fungi, the light can be recognized from afar. Dr. Y. Kobayashi, mycologist, observed this luminous millipede on Ponape Island, but I never saw it anywhere in tropical countries except Truk. Insecta Fireflies Regarding fireflies in Japan, Yo. K. Okada ( 1931 ) reported 33 spe- cies. However, as most of them live in Formosa, Okinawa, and Korea, we can scarcely count more than seven species in Japan proper. They are: Luciola cruckita Motschulsky, Luciola lateralis Motschulsky, Luciola parvula Kiesenwetter, Pyrocoelia fumosa Gorham, Pijrocoelia atripennis Lewis ( Amami-Oshima Island), Psilocladus variolosus Oliv- ier, and Lucidinu biplagiata Motschulsky. Among them are Luciola Y. HANEDA 361 cruciata and Luciola lateralis species, whose larvae live as aquatic glowworms; Luciola cruciata, distinguished by black wings and a red cross on the head, is the largest in Japan. Its larva develops in clear streams and has a pair of small abdominal luminous organs. The firefly appears from the end of June to July, and Japanese people go to various places where it swarms to view the splendor of its flashing light. As the glowworm tends to decrease in number, many famous places for fireflies have been under protection by the government as sanctuaries. The larva of Luciola lateralis appears in rather dirty water in rice fields or brooks. Luciola parmila is a mountainous spe- cies. Pijrocoelia fumosa is common near Tokyo; however, only its larva is luminous. Other species seem few in number because they are not easily recognized. In Micronesia fireflies live only in the coral Islands of Palao and Yap. The firefly Atyphella carolinae found in these islands is 7 mm in length and has black wings. After sunset swarms of the insects fly up for about an hour at a time and are a beautiful sight to see. Many species of fireflies live in the East Indies, Malay Peninsula, and New Guinea, and although I collected some specimens, they were burned in Tokyo during the war. Later specimens I collected were lost in Singapore, so unfortunately I am unable to report on them. Synchronous Flashing of Fireflies Although my specimens from these countries were lost, I can never forget the amazing spectacle of synchronous flashing of fireflies in New Guinea. I happened to see it in March, 1940, at the Rabaul Botanical Garden, Rabaul, New Britain. On the leaves of a big silk tree countless numbers of fireflies were alighting and flickering rhyth- mically, causing the whole tree to appear as if it were breathing. This species, with black wings, was 7 mm long. Its flicker is distinct, be- cause when the light disappears, it does so instantaneously and com- pletely. My detailed observations are as follows: 1. The silk tree was a big one, and the fireflies alighted forming three groups, one on the upper part, one on the middle part, and one on the lower part. The flicker was transmitted rhythmically, the upper group extinguishing its light first, followed by the middle 362 LUMINOUS ORGANISMS OF FAR EAST group, and last the lower group. Sometimes the rhythmical flashing was transmitted from the lower part to the upper. The flashes were repeated at the amazing speed of seventy per minute. The phenome- non continued every day for one week while I was there, lasting from sunset to dawn, notwithstanding the rain. It resembled the description reported by Smith (1935), except that he noted the phe- nomena occurring when the moon was half full. 2. When a strong electric light was directed on them for a few seconds the synchronous flashing became irregular. After thirty sec- onds new synchronisms arose from some other groups in the tree and extended over the whole tree. 3. Fireflies on the tree were male and female in equal number. This fact differs from the observations reported by Morrison (1929) and Smith (1935). I observed about 100 males and females, each in separate cages in the darkness. Only the males continued to flash synchronously. On the contrary the females showed irregular flashing. 4. Not only the luminous organs of the male and female, but the color also, differ. The difference is discernible with the naked eye. The color of the light of the males is yellow, while that of the females is bluish green. Their light looked like scatterings of yellow and bluish green powder when the tree. was shaken. 5. Even after dawn, with the sun shining brighdy, the fireflies re- mained on the leaves of the tree. 6. This species of firefly selected thin-leafed trees. Sometimes these fireflies flashed synchronously as they flew through the air. 7. Many copulating fireflies were found on the grass under the tree. At that time the males were not emitting light. In view of this ob- servation, it seems the synchronous flashing is a behavior pattern by which the males invite the females to a group. I do not believe there is any permanent leader in the group that acts as a continual pacemaker for the synchronous flashing. I think that when some individual or group emits light, it has a stimulating effect that causes the light to spread throughout the whole group as a wave. The larva of this firefly is terrestrial. I collected a luminous pupa in the Botanical garden at Rabaul. In March, 1943, I saw again this beautiful synchronous flashing at Y. HANEDA 363 Manukwari, Momi, and Walen in West New Guinea. This time the fireflies were different and larger than those seen in Rabaul. They were 7 to 8 mm in length with black wings and they differed in the following respects from the fireflies in Rabaul: (1) The color of the light is bluish in both male and female. (2) The fireflies have a migratory tendency, assembling one by one in a large group, then moving to another tree. Their whereabouts in the daytime is unknown; at any rate they were not on trees. I also saw synchronous flashing in Singapore on the Bukit timah Road and in the mangrove zone of Johore Baharu, as well as in Java, in certain rice fields in the Village of Provoringo, but these phe- nomena were on a much smaller scale. It is said that on the East coast of the Malay Peninsula fireflies that swarm in the mangrove zone used to be protected, since their light could be used for navigation (Watson). Their scientific name is not yet known, but Dr. R. Takahashi believes it to be the Vesta mene- triesi Motsch. Synchronous flashing is seldom seen in the Japanese firefly Lticioki cruciata. At Nagamori Village near Gifu City, Gifu Pref., which is noted as a gathering place, the fireflies swarm on the trees on the river banks and flicker synchronously. However, as the light of Lu- ciola cruciata does not extinguish entirely, the regular wave effect, as seen in the tropical species, is not produced. Instead, the flashing is such that the tree on which they are appears to be "breathing" tranquilly. Starworm A luminous insect called Urat intan or Urat bintang by Indonesians occurs over all tropical Asia. Urat intan means the diamond worm and Urat bintang the starworm. This animal has a pair of luminous dots on the second and twelfth segments, three luminous dots on each segment from the third to eleventh, and one dot on the last segment. The insect is very beautiful, emitting a bluish green light from each dot. The body is larval in form and wingless, similar to Phengodes or the railroad worm of South America (see Harvey, 1952). It is identical with the insect in Sumatra called Api-Api, or "fire." I collected it in North Borneo in March, 1938, and afterwards 364 LUMINOUS ORGANISMS OF FAR EAST many of them in Singapore and the Malay Peninsula. The maximum length is 15 mm, and the minimum 3 mm. On the night of December 15, 1945, at Jurong Village, Singapore, I was observing this starworm in a glass dish on a table under a rubber tree. Suddenly an insect resembling a firefly came flying down. It was about one-third the size of the starworm and had black wings, feather antennae, and no luminous organ. To my great surprise, it copulated with the starwonn, and I found that the insect was the male of this species and the full-grown starworm the female. The size of the starworm varies. Some are extremely tiny. Therefore the larva, pupa, and adult female of the starworm sometimes cannot be distin- guished. It is not yet known whether the larva of the male or its pupa emits light. Through the good offices of Dr. Harvey, a specimen was sent to the late Dr. Barber of the National Museum, Washington, who identi- fied it as Diploclodon Hosseltii. Dr. R. Takahashi observed that star- worms ate millipedes instead of snails. Luminous Fungus Gnat A most interesting luminous dipteron, belonging to the family Platyuridae, the luminous fungus gnat Ceroplatus occurs in Japan. On September 25, 1948, Mr. T. Shimizu observed the luminous larvae of a fungus gnat living in a web on the fungus Poria vaporaria, at Mt. Ryogami in Saitama Pref. The specimens were sent to Dr. T. Esaki (1949) for identification. Larvae hatched from eggs on the way. They were identified as Ceroplotus nipponicus Okada (1938). In September, 1950, Dr. Kato and Mr. Shimizu collected some speci- mens of these insects at the same place. There were two species, one Ceroplotus nipponicus and the other Ceroplatus testaceus Dalman. According to Dr. Kato (1952), these diptera have two kinds of fat tissue, one consisting of pure fat cells and the other luminous fat cells. The luminous fat tissue is found around the digestive organs. He suggested that the occurrence of luminous fat cells in the fungus gnat indicates a close relation between the luminous substance and the fat metabolism. One night in June, 1951, on Hachijo Island Mr. Okuyama and I collected luminous larvae of this insect living on a web on the under Y. HANEDA 365 surface of the fungus Ganodcrma oppJanatus, which grew in a hollow of a big root of the pasania tree {Sliiia Sieboldii) at Nakanogo Vil- lage. The whole body is luminous when it is in the larval and pupal stage, but not in the adult. The larva is 16 mm long and 2 mm wide, with pale brown stripes on a translucent body. The anterior and posterior parts are transparent. The pupa is enclosed in a pure white cocoon knitted by fine threads, cylindrical in form, 15 mm in length and 5 mm in diameter. The transformation from larva to pupa takes one day, from pupa to adult a week. Adults are nonluminous, but the larva and pupa emit a continuous weak, bluish white light. Stim- ulation does not increase the light intensity. The light of the pupa can be seen through the white cocoon, but the ovary or eggs are not luminous. I endeavored to culture luminous bacteria from the body and obtained negative results. The luminous larvae were placed in a des- sicator containing CaClo and were dried. This dried material became luminous when moistened with water in the dark, and its light in- tensity was stronger than that of the living larva or pupa. I tested for the luciferin-luciferase reaction as well as the ATP reaction and ob- tained negative results. Miscellaneous Small Groups Luminous Nemertean In the summer of 1936 the late Dr. Kanda and I stayed at the Asamushi Marine Biological Station on Aomori Bay. One night, in the aquarium of the station, I recognized luminescence on the surface of a common ascidian, Chelyosoma sihoja, when irritated. At the same time I collected a luminous nemertean and three species of luminous Polychaeta. These luminous animals were reported on by Kanda ( 1937 ) in Japanese. Among them the most remarkable, the luminous nemertean, was given the new species name of Emplectonema kandai by Kato (1939), and was also studied by Kanda in 1939. This animal is from 50 to 120 cm long and only 0.5 to 0.9 mm in diameter. Its body is unseg- mented and usually very thin and threadlike, sometimes extending to enormous lengths. 366 LUMINOUS ORGANISMS OF FAR EAST According to my observation the living animal is pale blue in color and glows only when stimulated, emitting a whitish blue light. The light may appear on all parts of the body except the head. No lumi- nous secretion is discharged. Luminous Enteropneusta Two species of Balanoglossus belonging to the family Ptychoderidae occur in Japan. However, the luminosity of these animals has not been observed. They are wormlike in form and about 50 to 60 cm long, live in the sand, and have a disagreeable odor. Recently I ob- served the luminosity of Balanoglossus carnosiis (Willey) on the sandy beach in front of the Usa Marine Biological Station of Kochi University, Kochi City. It is very difficult to collect perfect specimens because the body is very soft, especially the posterior part, which can be easily torn off by pulling. I was able to recognize the lumi- nosity of the body and posterior part, even with the head buried in the sand and unexposed. The pale bluish light is emitted from every portion of the body, but only upon stimulation, and luminous slime comes off on the fingers. Another species, Balanoglossus misakiensis Kuwano, was found in Misaki, Kanagawa Pref., but I have not yet verified their luminosity. I tested for the luciferin-luciferase reaction as well as the ATP reaction and obtained negative results. Luminous Snake Star In 1938 the late Dr. Kanda found a luminous snake star in the dredge net at the Mitsui Institute of Marine Biology near Shimoda, Izu Peninsula. This snake star was given the new species name of Amphiura kandai by Dr. S. Murakami (1942). The disk of this speci- men is 2.5 mm in diameter, and the anns, six in number, are 11 mm in length. According to Murakami, when a few drops of hydrogen peroxide were poured into a dish containing some specimens, the ani- mals emitted light very faintly in the dark. Kato (1947a) also studied the luminous cells of this animal and surmized that the light is intra- cellular. Bryozoa Kato (1950) reported on the luminous organ of Acanthodesia ser- rata (Syn. Memhranipora membranacea) , a common Japanese marine Y. HANEDA 367 bryozoan. He noted that the bryozoan colony emitted a bluish light for a few seconds when stimulated. A pair of light organs are situated on each anterolateral part of the ventral membranous area in each zoecium. Pisces Luminous fishes, depending on their method of light production and the type of luminous organs, are divided into three groups: (1) fish associated with luminous bacteria, (2) fish with well-developed luminous organs or simple luminous skin organs, ( 3 ) fish with indirect emission luminous organs, that is, the luminous gland lies inside the fish body and the light is reflected so as to pass through a translucent area of muscles. Fish Associated with Lmninous Bacteria In Japan the first group of luminous fishes have been studied in rather great numbers. In 1916 Dr. Harvey, during his stay in Japan, saw the light organs of the knightfish Monocentris japonicus, and he predicted that luminous bacteria would be found in the luminous organs. Later Yo. K. Okada (1926) studied the morphology and his- tology of the organs, and Yasaki (1928) cultivated luminous bacteria from the organs and confirmed Dr. Harvey's prediction. This fish is commonly found in shallow water along the coast of Japan. The fish is a beautiful golden yellow in color. Its luminous organs consist of two oval protuberances lying side by side at the tip of the lower jaw. In a fish 12 cm long the light organ is 4 mm long and 3 mm wide. The fishes of the families Gadidae and Macrouridae possess lumi- nous glands on their ventral surfaces. The Gadidae live in compara- tively shallow water and may be caught close to the coast of Japan. The Macrouridae are deep water fish and are always caught by trawlers as they inhabit a region over 100 fathoms deep. They are taken most abundantly during winter season along the Pacific coast of Japanese waters. Kishitani (1930) examined the luminous duct of Physicuhis japoni- cus of the Gadidae and discovered it was an open type of luminous gland containing symbiotic luminous bacteria, Micrococcus physiculus Kishitani. 368 LUMINOUS ORGANISMS OF FAR EAST Dr. Yasaki and I (1936) had reported that there were 10 species of the Macrouridae, which are closely related to the Gadidae and have a luminous organ of the same type. I ( 1938-1951 ) have added one species of the Gadidae and 4 species of the Macrouridae and I was able to obtain several strains of luminous bacteria from each species of the Gadidae and Macrouridae. These strains of bacteria were obtained from various species of fish caught at different times and in different localities. All the bacteria had the same general bio- logical characteristics but they varied in their optimum temperature, being higher in some cases but never varying greatly. I think all these luminous bacteria are of the same group. Fish with Indirect Emission Luminous Organ The fishes of the families Acropomatidae and Leiognathidae possess luminous glands containing symbiotic luminous bacteria. These fishes can also be classified under the third group since the luminous gland hes inside the body and the hght therefore passes through a trans- lucent area of muscles. Acropoma is a genus of fish of the family Acropomatidae found in the southern Sea of Japan. Acropoma japonicum Gunther is known as Hofani-jako in Japanese, meaning firefly small fish, and is consid- ered there to be a single species. However, I have observed that the luminous organ varies in shape and position in two types of fish. For this reason they may represent different species. Recently Matsubara (1953) reported another new species named Acropoma hanedai Matsubara, which I (1950) had mentioned as the second type of Acropoma. During the winter season they occur in southern Japan as a mid-water dweller, in depths ranging from about 80 to 200 fathoms. They are a beautiful pink in color and attain a length of 200 mm. There is no difficulty in obtaining specimens in the Mimase fish market near the city of Kochi, Shikoku, Japan. These fish differ from other luminous fish in possessing an unusually large luminous area. In fact, the lower part of the muscles of their entire body surface is utilized for this purpose. The diffused light comes from a luminous U-shaped filiform body in the muscle tissue not visible from the outside. It consists of the luminous gland, white reflector, lens, and an opening near the anus Y. HANEDA 369 with a long duct which connects with the luminous gland. The lumi- nous gland of Acropoma Jianedai is very long compared with that of Acropoma japonicum, and the muscles which form the lenses are comparatively poor. I obtained pure cultures of luminous bacteria from both species of Acropoma and found their general biological characteristics quite different. Other luminous fishes of the same general type as Acropoma belong to the Leiognathidae. These are true shallow-water forms and are abundant in southern Japanese waters and in tropical seas. In 1937 at Palao I observed that Gazza minufa and several other species of this family are luminescent when alive. Externally this fish does not present any unusual features. It was only by careful observation of the living fish by night that the luminosity of the lower half of the body was revealed. The source of the luminosity is a swollen ring of gland which en- circles the esophagus. The body cavity and the thoracic and ventral muscles are so modified as to increase the efficiency of the light- producing mechanism. The light is visible externally as a diffused greenish blue light, sometimes intermittent. The light control mech- anism is due to chromatophores which are scattered over the trans- parent membrane that covers the luminous gland. These fishes have a far more complex luminous organ than Acropoma. In Japan there are three species, L. argentum, L. rivulatum, and L. elongatum, which are usually dried for sale for food. In more tropical countries there are many species of Leiognathidae. I obtained and observed eleven species of Leiognathidae, distributed in three genera {Leio- gnathus, Secutor, and Gazza) in the southwestern Pacific. Gazza occurs in clear water in Palao at depths of about 30 meters, while Leiognathiis lives in turbid water in the mangrove zone at depths of 1 to 2 meters. In Palao, L. eqinihis ranged in length up to 70 mm, while in Sandakan Bay, North Borneo, it was 200 mm long. In Sandakan, Chinese fishermen catch these fish in nets and bring them for sale to the Sandakan fish market. Around Singapore and the Malay Peninsula these fish are easily obtained in all seasons from fishing traps known as "Kelong." In 1938 Fowler reported twelve species of this family in Malaya. In Java I collected many species of this family in the fish markets of Jakarta. 370 LUMINOUS ORGANISMS OF FAR EAST An interesting feature of the Japanese species L. rivulatum is that the luminous gland of the male is either very large or very small. I am of the opinion, though by no means certain, that this difference in size is due to age. The luminous gland of the female is smaller than that of the male and therefore gives off a weaker light. I was able to obtain pure cultures of luminous bacteria from vari- ous species of the fish caught at different times in Japanese and tropical waters. They had the same general biological characteristics, Fic. 9. The eye of the luminous iish Anomalops, showing the eUiptical white himinous organ with its opening (OP). with only shght variations, probably due to the effects of tempera- ture. All these luminous bacteria appear to be of the same group. The famous luminous fish Anomalops is found not only in the East Indies but also in the South Pacific Islands, the New Hebrides, Fiji, and Paumotus. This fish has a large half -moon shaped luminous organ below the eye, as shown in Fig. 9. The back of the organ is covered with a layer of black pigment. Anomalops is able to shut o£E its display of luminescence at will, by rotating the organ, so that the luminous part is turned away from sight and the black nonluminous back takes its place. Y. HANEDA 371 In June, 1942, I saw schools of Anomalops at the surface of the water in the harbor of Manukuwari on the northwest coast of New Guinea. At that time I collected three small specimens and was able to observe their luminescence. When this fish is swimming under natural conditions, the luminous surface appears and disappears in- termittently. If the fish is caught and put into a glass jar, its luminous display becomes irregular, if the water is in any way unsuitable. A fish in a dark place which is suddenly illuminated by switching on Fig. 10. A longitudinal section of the luminous organ of Anomalops. OP, open- ing; EP, outer epithelium; P, pores; PHOT, luminous ducts; REEL, reflec- tor; PG, pigment; ATR, artery. an electric light will cease to display luminosity in one or both organs. In daylight the fish will not display its luminosity, but if the place in which it is kept is suddenly darkened, its luminosity is immedi- ately displayed and appears as a bluish green light. It was very remarkable that 4 specimens of this fish were obtained near Japanese waters. The first specimen was caught in the sea near Kominato, Chiba Pref., Japan, and was recorded by Abe (1942). The second specimen was caught with hook and line near Kanminato Bay, Hachijo Island, and was also reported by Abe (1951). The third and 372 LUMINOUS ORGANISMS OF FAR EAST the fourth specimens I found among some preserved fish materials at the Hachijo Branch of Tokyo-to Fisheries Experimental Station, but the date of collection is unknown. The total length of the third specimen was 190 mm; the diameter of the eye was 20 mm, but the luminous organs had putrified. The total length of the fourth speci- men was 145 mm; the luminous organ was 14 mm long, 5.5 mm wide, and 1 mm thick. In a parallel section of the organ (Fig. 10) a large number of glandular tubes can be seen parallel to each other and extending completely across the organ from the back pigmented surface to the front transparent surface, with blood vessels running between them. A cross section of the organ is shown in Fig. 11. Harvey ( 1925 ) suggested that the luminescence is due to symbiotic luminous bacteria. He cultured bacteria from these luminous organs, but these cultured bacteria were not luminous. In a former report ( 1942 ) I noted that my culture test from the organs agreed with the result of Harvey's test, although I was doubt- ful of his bacterial theory. I had observed that the section of the organ was quite different from those in other symbiotic luminous fishes, Monocentris, Physiculus, Malacocephalus, Acropoma, and Lei- ognathus, and that the organ was closed, that is, it had no duct to the exterior. However, upon investigation of the large specimens of Hachijo Island, I was able to find the opening from the duct in the luminous organ, which I had overlooked in my former report. The opening to the exterior is considered to be the small depression in the dorso-anterior part of the luminous organ. There are also many pores scattered over the surface of the organ. Therefore the organ is of the open type and supports the bacterial theory of light emission, especially since the luminescence is continuous. Self-Luminous Fish The second group of luminous fish, namely fish with a well- developed luminous organ or simple skin photophores, were reported by such ichthyologists as Tanaka (1908, 1912), Ishikawa (1915), Nakasawa (1932), Kamohara (1936, 1940, 1952), Matsubara (1936, 1938, 1950), Abe (1942, 1951), Imai (1942), and Kuroda (1950). However, most of these are taxonomical reports. Recently Abe and Nakamura ( 1954 ) reported an adult female, with Y. HANEDA 373 Fig. 11. Transverse section of the luminous organ of Anomalops. Lettering same a'' in Fig. 10. 374 LUMINOUS ORGANISMS OF FAR EAST a supposedly parasitic male, of the deep sea angler fish Crytosparus couesi, from the Pacific coast of northern Japan which is supposed to be luminous. The structure of the light organs and the observation of luminous phenomena of self-luminous fish have been investigated by Ohshima (1911). He studied two species of deep sea luminous shark and some specimens of the Myctophidae and Sternoptychidae. In my observa- tion ( 1950 ) on living material of Yarrella and Pohjipmis from Suruga Bay, I recognized a peculiar color filter, hitherto considered a lens, situated between the luminous tissue and the lens. Recently I ob- served luminescence of the cheek organ of a deep sea luminous fish, Astronesthes ijimai Tanaka (1908), which was collected in a shrimp trawl net in Suruga Bay. This fish has a cheek organ and two rows of minute photophores along the ventral and lateral walls. The struc- ture of the cheek organ is very similar to that of Anomalops. Although luminescence is continuous, the luminous surface appears and disap- pears at will by rotating the cheek organ as in the case of Anomalops. Comparative studies of the structure and substance of both luminous organs should prove interesting. We may find luminous bacteria and a close relationship between the two organs. As deep sea luminous fish are caught with comparative ease by trawlers in many localities in Japan, it is to be expected that many unrecorded luminous species will be found in the future. Dr. A. Terao and others (1950) reported a luminous flying fish which was collected from Sagami Bay, Japan. They described the luminous organs as minute, weak, luminescent points scattered on the dorsal surface of the fish and rather strongly luminescent points on the head. The former resemble the skin organ of a luminous shark, and the latter are like the organs of Monocentris. However, I observed several living species of flying fish in the dark at Hachijo Island and Suruga Bay, but could not recognize any luminous points on the surface. ACKNOWLEDGMENTS I take pleasure in thanking the committee on the Luminescence Confer- ence at Pacific Grove, California, for inviting me to participate in the meet- ings of March 29-April 1, 1954, and express appreciation to Professor E. Newton Harvey for help in preparing this manuscript. Y. HANEDA 375 References' Abe, T. 1942. A record of Anomalops katoptron Bleeker from Japan. Annot. Zool. Japan, 21 (1), 55-57. Abe, T. 1951. A record of Anomalops katoptron from Hachijo Island. Japan. J.' Ichthyology, 1, 304-305. Abe, T. H. 1927. Report of the biological survey of Mutsu Bay. 3. Notes on the protozoan fauna of Mutsu Bay. 1. Peridiniales. Tohoku Imp. Univ. Sci. Repts. Biology, Sendai, Japan, 2 (4), 383-438. Abe, T. H. 1936. Report of the biological survey of Mutsu Bay. 29. Notes on the protozoan fauna of Mutsu Bay. II. Genus Peridinium; Subgenus Arclmeperidinium. Tohoku Imp. Univ. Sci. Repts. Ser. 4, Biology, 10 (4), 639-86. Abe, T. H. 1940. Studies on the protozoan fauna of Shimoda Bay genus Peridinium: Group Globula. Sci. Repts. Tokyo Bunrika Daigaku, B5 (81- 83), 21-51. Akaba, T. 1938. A supplement note on the nature of hght produced by luminous bacteria obtained from the deep-sea fish of the family, Macrou- ridae. Japan. J. Physiol, 3, 310-17. In Japanese. Annandale, N. 1900. Observations on the habits and natural surroundings of insects made on the "Skeat Expedition" to the Malay Peninsula, 1899- 1900. VI. Insect luminosity (an aquatic lampyrid larva). Proc. Zool. Sqc. London, 1900, 862-65. Arakawa, S. On a luminous centipede, Onja sp. from Formosa. /. Zool. Soc. Japan, 45 (532-33), 137-39. Baba, K. 1935. Report of the biological survey of Mutsu Bay. 27 Nudi- branchia. Tohoku Imp. Univ. Sci. Repts. Biology, 10 (4), 331-36. Baba, K. 1949. Opisthobranchia of Sagami Bay collected by His Majesty the Emperor of Japan. 43-44. PIXIV. Figs. 48, 49. Iwanami, Tokyo. Bose, S. R. 1926. Luminous leaves and stalks from Bengal. Nature, 117, 156-57. Bose, S. R. 1930. Relation of sunlight to the Ught of luminous wood. Natur- wissenshaften, 18, 787. Bose, S. R. 1935a. Photograph of a luminous fungus by its own light. Sci. Cult., 1, 57. Bose, S. R. 1935b. A luminous agaric (Pleurotus sp.) from South Burma. Trans. Brit. Mycol. Soc, 19, 97-101. Bowden, B. J. 1950. Some observations on a luminescent fresh water limpet from New Zealand, Biol. Bull, 99, 373-80. " This list contains references to papers by most of the authors mentioned in the text, and it represents the chief publications dealing with luminous organisms of Japan and the Far East, including Oceania. 376 LUMINOUS ORGANISMS OF FAR EAST Coiner, E. J. H. 1950. Descriptions of two luminous tropical agarics {Dic- tyopaniis and Mycena) Mycologia, 17 (3), 423-31. Esaki, T. 1949. Luminous larva of fungus gnat. Shin-Konchu, 2 (7-8), 11. Fujiwara, T. 1935. On the light production and luminous organs in a Japanese Chaetopterid, MesocJiaetopteriis japoniciis, Fujiwara. /. Sci. Hiroshima Univ., 3, 185-92. Hada, Y. 1939. Hydrographical observations and plankton studies of some brakish water lakes on the Ohotsuku Sea Coast of Hokkaido in winter. Trans. Sapporo Nat. Hist. Sac, 16, 147-74. Hamada, M. 1940. Physiologisch-morphologische Studien uber ArmiUoria mellea (Vahl) Quel., mit besonderer Rucksicht auf die Oxalsaure Bildung. Japan. }. Botany, 10, 387-463. Haneda, Y. 1938. On the luminescence of the deep-sea fish Malacocephaliis laevis (Lowe). Japan. }. Physiol, 3, 318-26; in Japanese. Also /. Med. Sci. Ill; Biophysics, 5, 355-66; in German. Haneda, Y. 1938. Luminous fishes from southern seas. Kagaku Nanyo {South Sea Science), 1, 21-27; also Zool. Mag. Tokyo, 51, 105. Haneda, Y. 1939a. A few observations on the luminescent fungi of Micro- nesia. Kagaku Nanyo, 1, 116-128. In Japanese. Haneda, Y. 1939b. Luminosity of Rocellaria grandis (Deshayes) (Lamelli- branchia). Kagaku Nanyo, 2, 36-39. In Japanese. Haneda, Y. 1939c. Luminescence of the Hving matters. Kagaku Nanyo, 9, 252-59. In Japanese. Haneda, Y. 1939d. The terrestrial luminescent animals and plants in Palau and Yap Island. Kagaku Nanyo, 2, 88-93. In Japanese. Haneda, Y. 1939e. Luminous bacteria symbiotic with fishes. Kagaku-Gaho, 28 (10), 49-56. In Japanese. Haneda, Y. 1939f. New type of luminous organ of the fishes. Reiko (cold hght) Ser. 1, 48-55, In Japanese. Haneda, Y. 1940a. Phenomena of bioluminescence. Seirigaku-Shidoshu, 5, 18-31. In Japanese. Haneda, Y. 1940b. On the luminescence of the fishes belonging to the family Leiognathidae of the tropical Pacific. Palao Trop. Biol. Station Studies, 2, 29-39. Haneda, Y. 1941a. Note on the light production in the Shan-Shan. Kagaku Nanyo, 3, 181-83. In Japanese. Haneda, Y. 1941b. Trip to Rabaul, New Britain (Observations on synchro- nous flashing of fireflies.) Kagaku Nanyo, 3 (3), 66-69. In Japanese. Haneda, Y. 1942. Further investigation upon luminescent fungi. Kagaku Nanyo, 4, 49-63. In Japanese. Haneda, Y. 1943. On the photogenous organs of Anomalops katoptron, a luminous fish. Kagaku Nanyo, 5, 81-88. In Japanese. Haneda, Y. 1946. A luminous land snail, Dyakia striata, found in Malaya. Seibutsu, 1 (5, 6), 294-98. Y. HANEDA 377 Haneda, Y. 1948. Luminous organisms. Kagaku Sekai (The world of Sci- ence), 23 (10), 34-38, Figs. 1-5. In Japanese. Haneda, Y. 1950a. Star-worm in Singapore. Shin Konchu (New Insect), 3, 2-5. In Japanese. Haneda, Y. 1950b. Harpodon nehereus, a non-luminous fish. Pacific ScL, 4 (2), 135-38. Haneda, Y. 1950c. Luminous organs of fish which emit light indirectly, y^^^ Pacific Sci., 4 (3), 214-27. ^ Haneda, Y. 1951. The luminescence of some deep-sea fishes of the families Gadidae and Macrouridae. Pacific Sci., 5 (4), 372-78. Haneda, Y. 1952. Some luminous fishes of the genera Yarrella and Polyipnus. Pacific Sci., 6 (1), 13-15. Haneda, Y. 1952. Observations on luminous organisms of the Miura Penin- sula, Japan. Yokositkashi-Shi., 5, 1-49. Haneda, Y. 1953. Observations on some marine luminous organisms of Hachijo Island, Japan. Records Oceanog. Works in Japan, 12 (2), n.s. 1 (1), 103-108." Haneda Y., and N. Kumagai. 1939. The luminous material of Pontodrilus matsiishimensis, Izuka. Japan. J. Physiol, 4, 328-33. In Japanese. Haneda Y., and M. Takase, and N. Kumagai, 1940. Spectroscopical study of the living light. Japan. J. Phy.siol, 5, 307-17. In Japanese. Harms, J. W. 1928. Bau und Entwicklung eines eigenartigen Leuchtorgans bei Eqiiula spec. Z. loiss. ZooL, 131, 157-79. Harvey, E. N. 1917. Studies on bioluminescence. IV. The Chemistry of light production in a Japanese ostracod crustacean, Cypridina hilgendorfii, Muller, Am. J. Physiol, 42, 318-41; V. By the fire-fly. Idem., 42, 342-48. VI. In a Japanese pennatulid, Cavernularia haheri. Idem., 42, 349-58. See also, Piih. Carnegie Inst. Washiiigton, No. 251, 171-234. Harvey, E. N. 1920. The Nature of Animal Light. Philadelphia, Harvey, E. N. 1923. The production of light by the fishes, Photoblepharon and Anomalops. Carnegie Inst. Wash. Piihl No. 312, 43-60; also in Science, 53, 314, 1921. Harvey, E. N. 1925. Luminous fishes of the Banda Sea. Nat. Hist., N.Y. 25, 353-56. Harvey, E. N. 1940. Living Light. Princeton University Press, Princeton, N.J. Harvey, E. N. 1952. Biohnninescence. Academic Press, New York. Hasama, B. 1941a. Uber die Bioluminescenz bei Chaetopterus variopedatiis Renier im bioelektrischen sowie histologischen Bild. Z. wiss. Zool, 154, 357-72. Hasama, B. 1941b. Uber die Bioluminescenz bei Watasenia scintillans im bioelektrische sowie histologischen Bild. Z. wiss. Zool, 155, 109-28. Hasama, B. 1942a. Uber die Bioluminescenz by Pyrocoeha rtifa im Aktion- strombild sowie im histologischen Bild. Ann. Zool Japan, 21, 59-77. 378 LUMINOUS ORGANISMS OF FAR EAST Hasama, B. 1942b. Uber die Biolumineszenz der Liiciola lateralis im zyto- logischen Bild sowie im Potentialbild ihres Leuchtorgans. Cytologia, 12, 366-77. Hasama, B. 1942c. Uber die Biolumineszenz der Larve von Luciola cruciata sowie von Pyrocoelia rufa im Aktionsstrombild und im histologischen Bild ihres Leuchtorgans. Cytologia, 12, 378-88. Hasama, B. 1942d. Zytologische Untersuchungen des Leuchtorgans von Luciola cruciata. Cytologia, 12, 389-96. Hasama, B. 1942e. Zytologische Untersuchungen des Leuchtorgans von zwei tropischen Leuchtkafern, Pyrocoelia analis und Luciola Gorhami. Cytologia, 12, 486-94. Hasama, B. 1943a. Lebensgeschichte der Luciola cruciata und der Luciola hteralis. Annot. Zool. Japon, 22, 23-47. Hasama, B. 1943b. Uber die Biolumineszenz des Plocamophorus tilesii Bergh sowie der Cavernularia habereri Moroff im Aktionsstrombild sowie im histologischen Bild. Cytologia, 13, 146-54. Hasama, B. 1944. Histologische Untersuchungen des Leuchtorgans der Luciola parvula. Cytologia, 13, 179-85. Hayashi, S. 1927. Studies on the luminescent organ of Watasenia scintillans (Berry) Folia Anat. Japon., 5, 417-27. Hayashi, K., and M. Okuyama. 1929. Bioluminescence, I, II, /. Okaijama Med. Soc, 41, 185-87, 270-72 (in Japanese). Biol. Ahstr., 8, 17565, 1934. Hennings, P. 1903. Ein stark phosphoreszierender javanischen Agaricus, Mycena illuminans P. Henn. n. sp. Hedwigia, 42, 309-10. Herre, A. W. C. T. 1939. On a collection of fishes from Nanyo, the Japanese mandated Islands. Ann. Zool. Japon, 18 (4), 298-307. Hirasaka, K. 1922. On a case of discoloured sea water. Annot. Zool. Japan., 10. Honjo, I. 1944. Supplemental knowledge of the neural physiology of Caver- nularia obesa Valenciennes. Physiol. Ecolog. Contr. from Otsu JJydrobiol. Expt. Sta., Zool. Lab. Kyoto Univ., No. 9, 13 pp. In Japanese, lida, T. T. 1934. Osmotic pressure of Noctiluca scintillans. J. Faculty Sci. Imp. Univ. Tokyo, 3, 495-97. Imai, H. 1942. Studies on the symbiotic luminous bacteria. Sei-i-kwai Med. J., 61, 454-67 (in Japanese). Imamura, T. 1904. Ueber ein neues Photobacterium. /. Hyg. Chem. Soc. Japan, 1, 48. In Japanese. Abst. in Zbl. Bakt., 38, 37 (1906). Inoue, N., T. Sasaki, and R. Oaki. 1952. /. Sci. Research Inst., 46 (1284); also 1953. Records Oceanog. Works in Japan, n.s. 1(1). Ishikawa, C. 1891. Vorlaufige Mitteilung uber die Conjugationserscheinungen bei der Noctilucen. Zool. Anz., 14, 12-14. Ishikawa, C. 1894. Ueber die Kernteilung bei Noctiluca miliaris. Ber. natur- forsch. Ges. Freiburg im Breisgau, B8, 54-67. Y. HANEDA 379 Ishikawa, C. 1894. Studies on reproductive elements. II. Noctihica miliaris Sur: its division and spore formation. /. Coll. Sci. Imp. Univ. Tokyo, 6, 297-334. Ishikawa, C. 1899. Further observations on the nuclear division of Noctihica. J. Coll. Sci. Imp. Univ. Tokyo, 12, 243-62. Ishikawa, C. 1913. Einige Bemerkungen iiber den leuchtenden Tintenfish, Watasenia nov. gen. scintillans. Berry, aus Japan. Zool. Anz., 43, 162-72. Ishikawa, C. 1915. On a new species of Maiirolicus japoniciis. }. Coll. Agr. Imp. Univ. Tokyo, 6 (2), 183-91. Josserand, M. 1953. Sur la luminescence de "Mycena rorida" en Europe occidental. Bull. mens. soc. Linn, de Lyons, 22 (4), 99-102. Kajiyama, E. 1912-1913. Study on Ostracoda of Misaki (1-3). Zool. Mag. Tokyo, 24 (287), 488-92; (289), 609-19; 25 (291), 1-16. Kamohara, T. 1952. Revised description of the offshore bottom fishes of Prov. Tosa, Shikoku, Japan. Kept. Kochi Univ. Nat. Sci., 3, 1-22. Kanda, S. 1920. Physico-chemical studies on bioluminescence. I. Am. J. Phijsiol, 50, 544-60; II. ibid., 50, 561-73; III. ibid., 53, 137-49. Kanda, S. 1921. IV. Am. J. Phijsiol, 55, 1-12. Kanda, S. 1924. V. Am. J. Physiol, 68, 435-44. Kanda, S. 1928. VI. Sci. Papers Inst. Phys. Chem. Research (Tokyo), 9, 265-69. Kanda, S. 1929. VII. Sci. Papers Insi. Phys. Chem. Research (Tokyo), 10, 91-98. Kanda, S. 1930. The Chemical nature of Cypridina luciferin. Science, 71, 444; also in Sci. Papers Inst. Phys. Chem. Research (Tokyo), 13, 246-47. Kanda, S. 1932. Crystalline luciferin. Sci. Papers Inst. Phys. Chem. Re- search (Tokyo) Supp., 18, 1. Kanda, S. 1935. Hotarii (Firefly). Tokyo. In Japanese. Kanda, S. 1937. Some newly found luminous organisms. Rigaku-Kai, 35 (2), 109-15. Kanda, S. 1938. The luminescence of Pontodrilus matsushimensis. Riga- ku-Kai, 36 (3), 1-7. Kanda, S. 1939. The luminescence of a Nemertean, Emplectonema kandai, Kato. Biol Bull, 77, 166-73. Kato, K. 1939. A new luminous species of the Nemertea, Emplectonema kandai sp. nov. Japan. J. Zool, 8, 251-54. Kato, K. 1947a. Luminous Ophiuroidae. Zool Mag. Tokyo, 57, 10. In Japanese. Kato, K. 1947b. A new type of luminous organ of fishes. Zool. Mag. Tokyo, 57, 195-98. In Japanese. Kato, K. 1949a. Luminous organ of Kaloplocamus ramosum. Zool. Mag. Tokyo, 58, 163-64. In Japanese. Kato, K. 1949b. On a luminous Hydroid, Clytia linearis. Zool Mag. Tokyo, 58 (11), 17-18. 380 LUMINOUS ORGANISMS OF FAR EAST Kato, K. 1950. On a luminous Bryozoan, Acanthodesia serrata. Zool. Mag. Tokyo, 59 (1), 9-10. Kato, K. 1953. On the luminous fungus gnats in Japan. Sci. Rept. Saitama Univ. Bl (2), 59-63. Kawamura, S. 1915. Studies on the luminous fungus, Pleurotus japonicus, sp. nov. /. Coll. Sci. Imp. Univ. Tokyo, 35, art. 3, 29 pp. Kawamura, T. 1915. Siphonophores. Zool. Mag. Tokyo, 27 (318), 191-98; (320), 317-24; (322), 428-40; (325), 577-88. Kawanaka, T., and H. Matsubara. 1941. Study of lanternflies. I. Chemical components of Liiciola cniciata. Sci. and Ind. Tokyo, 16, 441-44; 17, 480-86. In Japanese. Review in Chem. Ahstr., 36, 169. Kiesenwetter, H. 1874. Die Malacodermen Japans nach den Ergebnisse der Sammlungen des Herrn Lewis wahrend der Jahr 1869-1871. Berlin. Ent. Z., 18. Kishinouye, K. 1910. Some medusae of Japanese waters. /. Coll. Sci. Imp. Univ. Tokyo, 27 (9), 1-35, pi. 1-5. Kishintani, T. 1928a. On the luminous organs of Watasenia scintillans. Annot. Zool. Japan., 11, 353-61. Kishitani, T. 1928b. Drei neue arten von Leuchtbakterien. Pwc. Imp. Acad. Japan, 4, 69-75. Kishitani, T. 1928c. Uber das Leuchtorgan von Eiipnjmna morsei Verill und die symbiontische Leuchtbakterien. Pwc. Imp. Acad. Japan, 4, 306-09. Kishitani, T. 1928d. Preliminary report on the luminous symbiosis in Sepiola birostrata Sasaki. Proc. Imp. Acad. Japan, 4, 393-96. Kishitani, T. 1930. Studien liber die Leuchtsymbiose in Physiculus japonicus Hilgendorf, mit der Beilage der zwei neuen Arten der Leuchtbakterien. Sci. Rept. Tohokii Univ. 5, 801-23. Kishitani, T. 1932. Studien iiber Leuchtsymbiose von japaninishen Sepien. Folia Anat. Japon. 10, 315-418; Stud. Tokitgawa Inst., 2, 315-418. Kishitani, T. 1933. Zur morphologic und Biologic einer Leuchtbakterienart. /. Sci. Hiroshima Univ., Bl, 183-96. Kobayashi, Y. 1937. Several luminous Mycomycetes from the Bonin Islands. Bull. Biogeogr. Soc. Japan, 7, 1-7. Kobayashi, Y. 1949. A luminous fungus newly found in Japan. Bull. Nat. Sci. Mus. No. 26, 13-19. In Japanese with English summary. Kobayashi, Y. 1951. Contributions to the luminous fungi from Japan. /. Hattori Botan. Lab., 5, 1-6. Kofoid, C. A. 1931. Report of die biological survey of Mutsu Bay. 18. Protozoan fauna of Mutsu Bay. Subclass DinoHagellata. Tribe Gym- modinioidae. Sci. Repts. Tolioku Imp. Univ., 4 Ser. Biology 6(1), 1-43. Y. HANEDA 381 Komai, T. 1918. On Ctenophores of the neighbourhood of Misaki. Annot. Zool. Japan., 9 (4), 451-74. Komai, T. 1921. Notes on the two Japanese Ctenophores, Lampetia pan- cerina Chun and Beroe ramosa n. sp. Annot. Zool. Japon., 10 (2), 15-18. Kominami, K. 1930. Fungus (Luminous Fungi) Iwanami Koza (Botany): 53-63. Kozukue, H. 1952a. Studies on the influence of uramin on hght of luminous bacteria. /. Tokyo Jikei-Kai Med. Coll., 67, 1-4. Kozukue, H. 1952b. Isolation of luminous bacteria from alimentary canal of symbiotic luminous fish. /. Tokyo Jikei-Kai Med. Coll., 67 (3), 22-25. Kozukue, H. 1952c. Luminous bacteria of the intestine of some marine fishes. /. Tokyo Jikei-Kai Med. Coll., 67 (3), 18-21. Kuroda, N. 1950. On the lanternfish of Suruga Bay. Zool. Mag. Tokyo, 59 (2-3), 7. Kuroki, S. 1940. Luminous earthworm and its luminosity. Shizen to Shiimi 9(1), 1-4. Majima, R. 1931a. Studies on luminous bacteria (Kept. I). Further studies on pathogenic luminous bacteria, Microspira phosphoreum Yasaki. Sei-i- kai /., 50 (2), 1-23. Majima, R. 1931b. (Rept. II). A study on photogenic bacteria establishing their saprophytic abode in flatfish procured in the market. Sei-i-kai J., 50 (3), 33-54. Majima, R. 1931c. (Rept. III). On the fermentation of carbohydrates by various luminous bacteria and the immune-agglutination of the organ- isms. Sei-i-kai J., 50 (4), 41-67. Matsubara, K. 1936. On two new species of fishes found in Japan. Annot. Zool. Japon. 15 (3), 355-60. Matsubara, K. 1938. Studies on the deep sea fishes of Japan. VI-VIII. /. Imp. Fish Inst., 33 (1), 37-66. Matsubara, K. 1953. Revision of Japanese serranid fish, referable to the genus Acropoma. Mem. Coll. Agr., Kyoto Univ. 66, 21-29. Matsumura, M. 1918. Fireflies of Japan. Kyoiku-Gaho, 6 (3). In Japanese. Molisch, H. 1925. Botanische Beobachtungen in Japan. III. Uber das leuchten des schlacht-viehfleishes in Sendai, Japan. Set. Repts. Tohoku Imp. Univ. Biology, 1, (4) 97-103; also Planzenbiol., in Japanese, Jena, 1926. Morrison, T. F. 1929. Observations on the synchronous flashing of fireflies in Siam. Science, 69, 400-401. Motschulsky, 1854. Ent. p. 53. Shrenck. Reis, p. 144. 1860. Siberia. BttU. Soc. Imp. Nat. Mosc, 39 (1866), 163. Murakami, S. 1942. Ophiurans of Izu. Japan. J. Dept. Agr. Kyushu Imp. Univ., 7(1), 11-12. 382 LUMINOUS ORGANISMS OF FAR EAST Nakamura, H. 1939. Hydroperoxid und bakterielles Leuchten. Acta Phyto- chim. (Tokyo), 11, 159-66. Nakamura, H. 1940. Light production by bacteria. Botan. Zool. 8, 154, 701- 10. In Japanese. Nakamura, H. 1941. Luminosity of earthworm (Microscolex phosphoreus) . Kagaku, 2 (10), 413. Nakamura, H. 1941. On a luminous earthworm. Kagahifin, 2 (1), 44-47. Nakamura, H. 1942a. Zur Kenntnis der leuchtenden Dinoflagellaten in Japan. Botan. Mag. Tokyo, 56, 553-54. In Japanese with German sum- mary. Nakamura, H. 1942b. Co-luciferase in luminous bacteria and fungi. Med. Biol, 2 (6), 318-20. In Japanese. Nakamura, H. 1942c. Cold Light. Kobundo, Tokyo. In Japanese. Nakamura, H. 1944. Luminescent Microorganisms. Tokyo. In Japanese. Nakamura, H. 1947. Ueber die Bedeutung von Flavin bei der Biolumines- zenz. Botan. Mag. Tokyo, 60, 703-14. In Japanese with German summary. Nakazawa, K. 1915. Report of Sakura-Ebi (Sergestes prehensilis) of Suruga Bay. Kept. Siiisan Koshu-sho, 11 (2), 1-21. Ninomiya, R. 1924. Der Einfluss von Antikorpern mid Kompliment auf biologische Functionen von Bakterien. Z. Itnmiin. Forsch. 29, 498-512. Nishikawa, T. 1901. Gonyaulax and the discolored water in the Bay of Agu. Annot. Zool. Japon., 4, 31-34. Odawara, X. 1953a. Studies on the effects of antibiotics for the growth and luminescence of luminous bacteria. Rept. 1. Effects of penicillin. /. Tokyo Jikeikai Med. Coll. 68 (3), 171-74. Odawara, X. 1953b. Studies on the effects of antibiotics for the growth and luminescence of luminous bacteria. Rept. 2. Effects of streptomycin, Chloromycetin, aureomycin, terramycin and colistin. 7- Tokyo Jikeikai Med. Coll., 68 (3), 175-82. Odawara, X. 1953c. Luminous bacteria of the intestine of marine crab, Portiinus trituberculatus. }. Tokyo Jikeikai Med. Coll., 68 (3), 183-86. Ogihara, Y. 1944. Luminous earthworm. Collecting and Breeding, 5 (6), 180. Ohshima, H. 1911. Some observations on the luminous organs of fishes. /. Coll. Sci. Imp. Univ. Tokyo, 27, art. 15, 1-25. Okada, Yo. K. 1925. Luminescence in sponges. Science, 62, 566-67. Okada, Yo. K. 1926. On the photogenic organ of the knightfish, Monocentris japonicus (Houttuyn). Biol. Bull, 50, 365-73. Okada, Yo. K. 1926. Light locahzation in ctenophores. Science, 63, 262. Okada, Yo. K. 1927. Luminescence et organe photogene des ostracodes. Bull. soc. zool. France, 51, 478-86. Okada, Yo. K. 1928a. Notes on the tail organs of Ascetes. Ann. Mag. Nat. Hist., 1 (10), 308. Okada, Yo. K. 1928b. Two Japanese aquatic glowworms. Tram. Roy. Ento- molog. Soc. London, 76, 101-08. Y. HANEDA 383 Okada, Yo. K. 1931. Fireflies of Japan. Zool. Mag. Tokyo, 43 (508-910). Okada, Yo. K. 1935. Origin and development of the photogenic organs of lampyrids with special references to those of Luciola cruciata and Pijw- coelia rufa. Mem. Coll. Sci. Kyoto, 108, 209-28. Okada, Yo., and K. Baba. 1938. On the luminous organs of a nudibranch PlocamopJwnis tilesii, Berge. Annot. Zool. Japon., 17, 276-81. Okada, Yo., and K. Kato. 1949. Studies on luminous animals in Japan. III. Preliminary report on the life history of Cypridina hilgendorjii. Bull Bio- geogr. Soc. Japan, 14 (3), 21-25. A translation from Kagaku, 16, 64-66, 1946. Okada, Yo. K., S. Takagi, and H. Sugino, 1933. Microchemical studies on the so-called photogenic granules of Watasenia scintillans. Proc. Imp. Acad. Japan, 10, 431-34. Okada, I. 1938. Beitrage zur Kenntnis der Ceroplatinen-Fauna Japans (Dipt., Fungivoridae). Insecta Matsumurana, 13 (1), 21. Okamura, K., and Nishikawa, T. 1904. A list of Ceratium in Japan. Annot. Zool. Japon., 5. Ostroumoff, A. 1924. Noctiluca miliaris in Symbiosis mit griinen Algen. Zool. Anz., 58, 162. Sasaki, M. 1915. On three interesting new oegopsids from the Bay of Sagami /. Coll. Agr. Tohoku Imp. Univ. Sapporo, Japan, 6 (6), 131-50. Shibata, S. 1953a. Luminous bacteria of the intestine of some marine fishes. /. Jikeikai Med. Coll., 68 (3), 203-15. Shibata, S. 1953b. Studies on high temperature luminous bacteria. /. Jikeikai Med. Coll., 68 (3), 187-202. Shibata, S. 1953c. Luminous bacteria obtained from the intesting of the squid, Loligo bleekeri Keferstein. /. Jikeikai Med. Coll., 68 (3), 116-17. Shibata, S. 1953d. Methods of observation of luminous bacteria by a fluoro- microscope. 7- Jikeikai Med. Coll., 68 (3), 218-24. Shima, G. 1927. Preliminary note on the nature of the luminous bodies of Watasenia scintillam Berry. Proc. Imp. Acad. Tokyo, 3, 461-64. Shoji, R. 1919. A physiological study on the luminescence of Watasenia scintillans- (Berry) Am. J. Physiol, 47, 534-57. Smith, H. M. 1935. Synchronous flashing of fireflies. Science, 82, 151-52. Suter, H. 1890. Miscellaneous communications on New Zealand land and fresh water molluscs. Trans. New Zealand Inst., 23, 93-96. Takagi, S. 1933. Mitochondoria in the luminous organs of Watasenia scintil- lans (Berry). Proc. Imp. Acad. Japan, 9 (10), 651-54. Takakuwa, Y. 1941. Fine neue leuchtende Spirobolellus-art (Diplopoda) und eine neue Lamyctes-art Trans. Nat. Hist. Soc. Formosa, 31, 84-87. In Japanese with German summary. Takase, M. 1938. Undersuchungen uber den Einfluss von Rohrzucken auf die Leuchtproduktion der Leucht-bakterien. Sei-i-kai Med. J., 57, 220-36. In Japanese with German summary. 384 LUMINOUS ORGANISMS OF FAR EAST Takase, M. 1939. Untersuchungen uber den Einfluss von verschiedenen Salzen auf die Lichtproduktion der Leucht-bacterien. Sei-i-kai Med. ]., 58, 54-80. In Japanese with German summary. Takase, M. 1940. Studies on the action of salts on the spectrum of the luminous bacteria. Sei-i-kai Med. J., 59, 124-42. In Japanese. Takino, M. 1953. Effective component taken from fluorescent bacteria for rheumatism and spontaneous gangrene. Nippon Tour. Angio-Cardiology, 17 (1), 1-7. Tanaka, S. 1908. /. Coll Sci., Imp. Univ. Tokyo, 23, art. 13, 9, Pi. 1, Fig. 1. Tanaka, S. 1912. Figures and descriptions of the fish of Japan. 6, 88-91 (Etmoptenis), PL 29, Fig. 117. (Astronesthes) 7, 116-119 {AstronestJies ijimai). Terao, A. 1917. Notes on photophores of Sergestes prehensilis. Bate Annot. Zool. Japon., 9, 299-316. Tokioka, T. 1937. Notes on Salpa and Doliohim occurring on the Pacific coast of middle Japan. 219-232. Uchida, T. 1925. Some Hydromedusae from Northern Japan. Japan. J. Zool, I, 77-100. Uchida, T. 1926. Report of the biological survey of Mutsu Bay. 2. Medusae of Mutsu Bay. Sci. Repts. Tohoku Imp. Univ., Biology, 2 (3), 215-38. Uchida, T. 1928. Studies of Japanese Hydromedusae. 2. Trachomedusae and Narcomedusae. Japan. J. Zool, 2 (1), 73-97. Uchida, T. 1929. Studies on the Stauromedusae and Cubomedusae, with special reference to their metamorpohsis. Japan. J. Zool, 2 (2), 103-193. Uchida, T. 1938. Report of the biological survey of Mutsu Bay. 33. Actiniaria of Mutsu Bay. Sci. Repts. Tohoku Imp. Univ., Biology, 13 (3), 287. Ueno, M. 1938. Stratification of Noctiluca in a brackish water lake of Hok- kaido, Japan. Proc. Imp. Acad. Japan, 14, 231-32. Usami, S., and M. Yokoseki. 1948. On the effect of certain substances upon hght production of a luminous bacterium. Seibutu, 3, 210-17. In Japanese with English summary. Wassink, E. C. 1948. Observations on the luminescence in fungi. I. Including a critical review of the species mentioned as luminescent in literature. Rec. trav. hot. neerland., 41, 150-212. Watanabe, H. 1897. The phosphorescence of Cypridina hilgendorfii. Annot. Zool. Japon., 1, 67-70. Watase, S. 1895. On the physical basis of animal phosphorescence. Biological lectures delivered at the Marine Biological Laboratory, Woods Hole, 177-192. Watase, S. 1905. On the luminous organs of Ahraliopsis. Zool. Mag. Tokyo, 17, 119-22. In Japanese. Weill, R. 1929. Observation sur la symbiose entre Noctiluques et Alges vertes. Arch. zool. exptl gen., 69, Notes 13-21. Yamada, T. 1937. On the spawning of the squid, Watasenia scintillans in Y. HAXEDA 385 the waters off the east coast of Tyosen. Bull. Japan. Soc. Sci. Fish, 6, 75-78. Yamaguchi, E. 1935. Occurrence of luminous OHgochaetae, Microscolex phosphoreiis (Dug) in Japan. Annot. Zoo/. Japon., 15 (2), 200-202. Yasaki, Y. 1927. Bacteriologic studies on biokiminescence. 1. On the cause of luminescence in the fresh water shrimp. Xiphocardina compressa (De Haan). /. Infect. Diseases, 40, 404-407. Yasaki, Y. 1928. On the nature of the luminescence of the knight fish, Monocentris Japoniciis (Houttuyu). /. Exptl. ZooL, 50, 495-505. Yasaki, Y. 1929. On the so-called symbiotic bioluminescence. Zool. Mag. (Tokyo) , 41, 490-91. Yasaki, Y. 1934. On luminous bacteria. Nihon-Iji-Shtiho, 1959, 1-10. In Japanese. Yasaki, Y. 1943. Studies on bioluminescence /. Physiol. Jap. 253-68. In Japanese. Yasaki, Y. 1949. Observation on tubercle bacilH b>' fluoromicroscope. Sei-i- kai,]., 64 (1), 1-8. Yasaki, Y. 1953. Fluoromicroscope and recent progress of its application. Igaku no Ayumi, 16 (5), 279-90. Yasaki, Y., and Y. Haneda. 1935a. Bioluminescence of the deep-sea fish, Macrouridae. /. Appl. Zool. Tokyo, 7, 165-76 (in Japanese). Yasaki, Y. and Y. Haneda. 1935b. Uber einen neuen Typus von Leucht- organ in Fische (Acropoma japonicum) Proc. Imp. Acad. Japan, 12 (2), 55-57. Yasaki, Y., M. Nishio, A. Ichikawa, R. Majima, and O. Ishikawa. 1926. Bacteriological studies on bioluminescence II. On the nature of the new luminous bacteria, Microspira phosphorctim Yasaki. Sei-i-kai Med. J., 45 (3). In Japanese. Yasaki, Y., and Y. Sinoto. 1952. Studies on nuclei and chromosomes by fluoromicroscopy I. Introductory remarks. Cytologia International J. Cytology, 17 (4), 3S6-44. Yatsu, N. 1912. Observations and experiments on the ctenophore egg: 1. The structure of the egg and experiments on cell-division. /. Coll. Sci. Tokyo, 32, art. 3, 21 pp. Yatsu, N. 1917. Note on the structure of the maxillary gland of Cypridina hilgendorfii. J. Morphol. 29, 435-40. Yoshida, K., T. Nakamura, and T. Okada. 1935. Uber die Lichtproduktion des Leuchtkafers. 1. Mitt. med. Akad. Kioto, 15, 1149 (Abstract in Ger- man). Ecology of Autotrophic Marine Dinoflagellates with Reference to Red Water Conditions* John H. Ryther Woods Hole Oceanographic Institution, Woods Hole, Massachusetts Although dinoflagellate luminescence was at one time believed to be restricted to a few highly specialized forms (i.e., Noctihtca, Pyro- cijstis), as early as 1830 Michaelis described 9 luminous species of the common genera Ceratium, Peridinium, and Prorocentrum. Kofoid and Swezy ( 1921 ) attribute this property to "many if not most of the Perindiniales and Gymnodiniales." The extent to which biolumines- cence occurs in the group as a whole has not been systematically investigated, but the opinion now appears to be widespread among workers in the field that most dinoflagellates, at least under certain conditions, exhibit luminescence. Kofoid, in a personal communica- tion to Harvey (1952), went so far as to suggest that perhaps all dinoflagellates have this property. The dinoflagellates are found in all the oceans of the world and at least a few are usually present at every time of the year. Among all forms of marine life, they are probably second in abundance only to the diatoms and, in many cases, may greatly outnumber them. Together with the diatoms they constitute the bulk of the so-called phytoplankton, the community of unicellular, autotrophic organisms that are the basis for all life in the sea. In view of the widespread occurrence of dinoflagellates in the ocean and the apparent preponderance of luminous forms, this group must be considered as one of the principal producers of biolumines- " Contribution No. 712 from the Woods Hole Oceanographic Institution. 387 388 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS cence in the sea. Allen (1939) stated categorically that they are the most common cause of this phenomenon. It follows that any consid- eration of the factors contributing to the occurrence of marine bio- luminescence must include a consideration of the ecological factors that influence and control the growth and distribution of its princi- pal causitive agents, the dinoflagellates. The following discussion will be concerned with the general ecology of these organisms. In addi- tion, special attention will be devoted to a possible explanation for the frequently reported occurrences of dense "blooms" or "swarms" of dinoflagellates in many parts of the world. No attempt will be made in the following discussion to differen- tiate between luminous and nonluminous dinoflagellates, but it will be assumed that the presence or absence of this property will not appreciably affect the interrelationships between the organisms and their environment. A consideration of their ecology as a group is com- plicated, however, by the fact that individual species may employ almost any type of nutritional habit, including autotrophic, holozoic, saprophytic, and parasitic. Although the number of species which employ the second mode of nutrition, either obligatorily or faculta- tively, is apparently quite large (Kofoid and Swezy, 1921), in total abundance they are probably insignificant in comparison with the photosynthetic forms. For this reason we will consider here primarily the autotrophic dinoflagellates. General Ecology of Dinoflagellates Temperature Dinoflagellates are found in all parts of the ocean and are usually present in some quantity at all times of the year. It is therefore im- possible to make broad generalizations concerning tHe temperature relationships of so large and widespread a group. Some species are obviously arctic forms and may be observed living in subzero tem- peratures associated with the polar seas (Gran, 1924; Braarud, 1935). However, there does appear to be some justification for classifying the great majority of the dinoflagellates as warm-water organisms. This is particularly true if they are compared as a group with the diatoms, which, in contrast, are often considered cold-water forms. JOHN H. RYTHER 389 In the temperate regions of the ocean, the relatively unproductive winter season is normally followed by a spring "flowering" or "bloom" of phytoplankton, which consists predominantly of diatoms. This spring maximum, which usually develops into the largest population of the year, becomes limited by the supply of available nutrients and ma)' pass through its entire cycle of growth and decline in a few weeks. As the nutrient level falls and the temperature rises, the num- bers of diatoms decrease and many species may disappear entirely from the plankton. By late spring, the dinoflagellates appear in significant numbers, and while the\' seldom attain the abundance of the spring diatom bloom, they often persist as the dominant member of the plankton community throughout the summer months. The sequence of events described above occurs in temperate regions in the open ocean (Herdman, 1922), in coastal and slope waters (Gran and Braarud, 1935), and in estuarine situations (Gaarder and Gran, 1927; Marshall, 1947; Braarud, 1945). While northern waters support larger populations of dinoflagellates, the number of temperate species is extremely small compared with the number of tropical forms ( Sverdrup, Johnson, and Fleming, 1942 ) . This is particularly true of the liighly specialized Dinophysiales, which are characteristic of tropical seas (Fritsch, 1935) but also ap- plies to the ubiquitous genus Ceratium of which Peters ( 1932 ) found 33 of 55 South Atlantic species confined to warm water. The dense populations of dinoflagellates which create "red water" conditions are known only in the tropics or in temperate water dur- ing the warmer (and usually the warmest) time of the year. This subject will be discussed in more detail in another section. Dino- flagellate luminescence shows a similar seasonal periodicity in the temperate parts of the ocean. Allen (1939) reported that luminous displays by these organisms are common in the La Jolla region during the summer months, but never occur between October and May. Such fragmentary physiological evidence as is available concerning the temperature relations of dinoflagellates appears to support the view that they are predominantly a warm-water group. Barker ( 1935), who is one of the pioneers in developing successful culture methods for dinoflagellates, observed optimal temperatures for the growth of 390 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS some 14 species between 18° and 25° C. Braarud and Pappas (1951) noted a temperature optimum for Peridinium triquctrum at 18° C, while Nordli (1953) found that Ceratium fiisiis and C. fiirca grew most rapidly at temperatures of 15° and 20° C respectively. Provasoli (personal communication) finds temperatures of 20-25° C most suitable for growing Gyrodinium californicum. Although the preceding experimental evidence is obviously insufficient to form a basis for generalization, it is perhaps significant that all the above species, despite their relatively high temperature optima for growth, were isolated from temperate regions of the ocean. Salinity Dinoflagellates occur in fresh, brackish, and full sea water, and it is again impossible to define optimal or limiting salinity conditions for the group as a whole. The unarmored dinoflagellates appear to be most abundant in the open ocean plankton, while the armored forms are more typical of coastal and estuarine regions ( Fritsch, 1935 ) . This suggests that the optimal growth conditions for the latter may occur at somewhat lower salinities than are observed in the open ocean. Several plankton investigations have revealed that dinoflagellate maxima may often be correlated with the seasonal or geographical occurrence of relatively low salinity water (Gran, 1924; Marshall, 1947; Gaarder and Gran, 1927; Marshall and Orr, 1927; Gran and Braarud, 1935). It is noteworthy that the majority of such studies have been concerned principally with the armored species and have largely neglected the smaller, naked dinoflagellates owing to the difficulties involved in collecting and preserving them. Nordli (1953) found optimum salinities for the growth of three species of Ceratium (C. furca, C. fusiis, and C. tripos) at 20-25 o/oo- He was able to correlate the relatively low salinity and high-tempera- ture optima of these fonns with similar conditions in the regions where they occur along the Norwegian coast. NordH suggests that the so- called Tripos-plankton region of Gran (1902) off the northern coast of Norway may be a biogeographical area limited by high-tempera- ture and low-salinity borders. - Table I gives a summary of the optimum and maximum range of JOHN H. RYTHER 391 TABLE I Optimum and Maximum Range of Salinity for Growth of Some Neritic Dinoflagellates Reference Nordli (1953) Braarud (1951) Species 0 ptimum, "/oo Range, "/oo Ceratiurn furca 25 10-40 Ceratium tripos 20 10-35 Ceratiurn fusus 20 10-40 Amphidinium sp. 15 5-45 Exuviella balitca 18 5-35 Peridinium trochoideum 20 5-60 Braarud and Rossavik (1951) Prorocentrum micans 15-20 10-45 Braarud and Pappas (1951) Peridinium triquetrum 15-20 10-40 salinity for the growth of several species of common neritic dino- flagellates as determined by Braarud and his co-workers. While all the species have salinity optima well below that of full sea water, it is perhaps of even greater significance that they are also able to grow within an extremely wide range of salinities. This high degree of adaptability is a definite advantage to life in the variable environment of coastal and estuarine waters, and it is perhaps one of the means by which the neritic dinoflagellates are able to compete successfully with other organisms, such as diatoms, which in general have a more narrow range of salinity tolerance. Nutrient Requirements The autotrophic dinoflagellates, as other members of the phyto- plankton, are dependent upon dissolved mineral salts for their nutri- tion. According to Vinogradov (1935) the peridinians contain ap- proximately the same relative concentrations of nitrogen, phosphorus, calcium, and iron as does sea water. The dinoflagellates have often been credited with the ability to utilize and flourish in extremely low concentrations of nitrogen and phosphorus (Gran, 1926-27; Gilson, 1937). This concept has stemmed largely from observations that dinoflagellate maxima, in temperate 392 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS waters, follow the decline of the spring diatom flowerings and rela- tively large populations often persist throughout the summer months when the supplies of these nutrients are almost undetectable. Gran (1926-27) has proposed that the dinoflagellates require less nutrition for growth than the diatoms on account of their relatively low rate of metabolism. There is some experimental evidence that dinoflagellates can utilize nitrogen and phosphorus at rather low concentrations. Thus Barker (1935) observed that increasing the nitrogen content of aged sea water by 1000 to 10,000 times did not increase the growth rates of Prorocentrum micans, P. gracile, or Peridiniwn sp. Similarly P. micans grew equally well in K0HPO4 concentrations ranging from 5 X 10~^ to 5 X \0~^%. King (1950) found that increasing nitrogen and phos- phorus 1 to 200 times their concentrations in aged sea water did not increase the growth rate of Gymnodinium simplex. There is no indication, however, that such low nutrient concentra- tions are necessary for the optimum growth of these organisms. Braarud ( 1945 ) found that maximum populations of several dino- flagellates occurred in the regions of heaviest pollution in the Oslo- fjord. Marshall (1947) observed a dense growth of P. triquetrum in Loch Craiglin immediately following fertilization of the loch. In the following section, a factor other than the nutritional physiology of the dinoflagellates will be discussed as a possible explanation for their occurrence and growth in nutrient-poor waters. As mentioned earlier, dinoflagellates as a group show a continuous variation in their modes of nutrition from autotrophic to holozoic, while many species are facultative, obtaining their food by either or both methods (Kofoid and Swezy, 1921). We are concerned here principally with the former type of nutrition. However, it is perhaps questionable whether any of the dinoflagellates are completely auto- trophic, in the literal sense of the term. For many years they have remained among the most difficult of marine organisms to grow and maintain in culture. The author knows of no case in which dinoflagellates have been grown in a completely inorganic medium, either of the completely artificial or the enriched sea water type. The few media which have been developed and used successfully for growing these organisms include, almost without JOHN H. RYTHER 393 exception, soil extract as their common ingredient (Barker, 1935; Gross, 1937; Braarud, 1951; Sweeney, 1951). Sweeney found that this substance best supported the growth of Gymnodinium splendens when aged for 4-6 weeks, and it was rela- tively ineffective when freshly prepared or aged for more than two months. Neither Barker nor Sweeney was able to replace soil extract with trace element mixtures. The latter author concluded that the active ingredient is probably organic in nature, and later (Sweeney, in press) succeeded in replacing soil extract with vitamin Bio. King (1950) cultured Gymnodinium simplex in a medium consisting of McClendon's artificial sea water, Hoagland's trace element mixture, and yeast extract. Again, she was unable to obtain growth with the inorganic constituents alone, but was able to replace yeast extract with a mixture of amino acids added as pure chemicals in the same ratio as they occur in the yeast extract. Provasoli and Pintner (1953) synthesized an artificial medium for the growth of Gyrodinium californicum which contained, in addition to the common inorganic nutrients and trace metals, a chelating agent (EDTA), NaH glutamate, dl-\ycine, JMeucine, and vitamin B12. Although the growth-promoting ingredients of soil extract have not been identified, it would appear likely, in view of the preceding evi- dence, that they include growth factors such as vitamin Bio. The organisms may also derive some benefit from humic acid and other ingredients which may act as chelating agents, reducing the concen- trations of some one or more of the trace metals to a nontoxic or noninhibitory concentration. In inshore waters dinoflagellates are often abundant, and in many parts of the world their populations may develop bloom proportions. Here the close association of the plankton with land masses and the contribution of runoff water to their enviromnent provide what may be considered as a natural "soil extract," and, in the vicinity of heavily populated areas, frequently a source of organic pollutants. As men- tioned earlier, Braarud (1945) observed a heavy growth of dino- flagellates in the highly polluted regions of the Oslofjord. Braarud and Pappas (1951) later found that the addition of small amounts of raw sewage to the medium stimulated the growth of Peridinium trique- trum,. 394 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS In the open ocean, the presence of free organic compounds is more difficult to account for. In this case it is perhaps the metabohtes or decomposition products of other plankton organisms which create the necessary growth conditions for the dinoflagellates. It has been pointed out that many of the dinoflagellates found in temperate waters have relatively high-temperature optima, a fact which may explain their paucity in the winter plankton community and the spring blooms. However, there are certainly many species which are able to show some growth in temperatures of 0° C and below (Braarud, 1935). Yet even these are conspicuously absent from the usual spring flowering of diatoms. Thus temperature alone may not be sufficient to account for the seasonal periodicity of the dino- flagellates. It would appear that, in many cases, the growth of these organisms may be also dependent upon, or at least benefited by, the previous existence of a flowering of diatoms. The underlying cause for this type of succession is probably nu- tritional, but its exact mechanism is obscure. The diatoms may reduce the concentrations of one or more of the nutrients or trace metals to a level favorable for the growth of the dinoflagellates. This type of relationship was proposed as an explanation of the succession of plankton elements in freshwater by Pearsall (1932) and Hutchinson ( 1944 ) . A somewhat different type of relationship has been suggested by Lucas ( 1947, 1949 ) , who has proposed the production of external metabolites or "ectocrines" by one group of plankton organisms which may have a beneficial effect upon the succeeding population and an inhibitory effect upon other competing organisms. This has been demonstrated in fresh water by Rice ( 1954 ) . Ecological Significance of Motility in Dinoflagellates The presence of motility in the dinoflagellates, and its absence in the diatoms, may have an important bearing in the relationships of these organisms with such environmental conditions as temperature, salinity, and the nutrient concentration of the water. This is the hypothesis of Gran ( 1926-27 ) and is supported by Braarud ( 1935 ) . The diatoms are dependent upon vertical mixing and their natural buoyancy for remaining in the upper, photosynthetic zone of the ocean. They may derive some benefit from the synthesis of fats and JOHN H. RYTHER 395 its inclusion as oil droplets in their protoplasm and may possess structural adaptations of various types which increase their surface area and hence retard sinking (Gran, 1912; Russell, 1927; Sverdrup, Johnson, and Fleming, 1942). Gross and Zeuthen (1948), on the other hand, attribute the buoyancy of diatoms to their ability to maintain extremely low concentrations of divalent ions in their cell sap. Accord- ing to these authors, under suitable physical conditions plankton diatoms have a specific gravity equal to that of sea water and do not sink, but at temperatures of 20° C or above, this equilibrium does not exist and the cells settle to the bottom. In general diatoms may be said to thrive in cold, fully saline ocean water at times of the year when there is considerable mixing of the surface water. In temperate or northern summer conditions, the density of the water decreases as its temperature rises. For example, an increase in temperature from 0-25° C decreases the viscosity of sea water by one-half ( Sverdrup, Johnson, and Fleming, 1942 ) . In addition, vertical mixing is usually at a minimum in summer. Under these conditions, the diatoms find it increasingly difficult to remain in the upper water layers, and in Gran's opinion, their populations are unable to maintain themselves in the euphotic zone. In the tropics, this situation usually prevails throughout the year and provides a definite disadvantage to the diatoms. For the same reason, they may encounter suboptimal conditions in the low-salinity (and hence low-density) waters of many coastal and estuarine situa- tions. The dinoflagellates, on the other hand, possessing the advantage of motility, are able to maintain themselves in water of low density with comparative ease and are relatively independent of vertical mixing. Furthermore, they, too, are often assisted by morphological adapta- tions which increase their surface area and thereby retard sinking. In the tropical forms, these structures far surpass similar features of the diatoms. Thus Ceratium is reputed to have the ability of prolonging or shedding its horns as it moves into warmer or colder water masses. (Fig. 1). Others, like Ornithocercus, grow wide, wing-shaped mem- branes (Fig. 2). Some species, such as Triposolenio, have asymetri- cally arranged appendages so that, as soon as they stop swimming and start to sink, they are quickly oriented horizontally in the water, 396 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS Fig. 1. Ceratium trichocews, showing progressive and proportionate reduction of the horns in autotomy. (After Murray and Hjort.) Fig. 2. (a) Ornithocercus splendidus; (b) Ornithocercus steinii. (After Murray and Hjort. ) providing a maximum surface as resistance to sinking (Kofoid, 1906). Thus the dinoflagellates, in contrast to the diatoms, appear to be particularly well adapted for life in relatively calm, low-density water. In addition, and of perhaps greater significance, the same factor of JOHN H. RYTHER 397 motility provides the dinoflagellates with an advantage in waters of low nutrient content. The nonmotile diatoms are dependent upon the dissolved nutrients contained in the water which immediately surrounds them and through which they sink (see Munk and Riley, 1952). In contrast, the dinoflagellates, though not strong swimmers, can move about for considerable distances and localize In the most advantageous depth for photosynthesis. If the nutrient level of the water is low, they may, by their vertical migrations, utilize all the nutrients available within the entire photic zone. According to Peters ( 1929 ) Cerotium can move through 5 to 10 meters in 12 hours or less. Thus motility may be equally as important as the basic physiological characteristics of dinoflagellates in providing the means for their existence in waters of high temperature, low salinity, and reduced nutrient concentration. "Red Water" Phenomena Background In certain parts of the world, usually in coastal or estuarine regions, dinoflagellates either grow or aggregate to such an extent that they impart a distinct coloration to the water. Such manifestations are known popularly as "red tide" or "red water," since the predominant color is of a reddish hue, although yellow or brown dinoflagellate blooms are not uncommon. According to Allen (1946) a cell concentration of one-half to one million organisms per liter may give a chocolate-brown color to the water, while a doubling or trebling of this number is sufficient to produce red water. Concentrations of over 50 million cells per liter are not uncommon in patches of red water (Woodcock, 1948; Davis, 1948). Frequently these aggregations of dinoflagellates are accompanied by a mass mortality of marine organisms (Whitelegge, 1891; Aiyar, 1936; Gunter et al., 1948; Connell and Cross, 1950). However, many red water reports make no reference to such calamities. In many in- stances these outbursts are also ac-companied by brilliant displays of luminescence. The reported occurrences of red water are too numerous to review 398 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS 2 1^ m +i Si 03 PU CO D _o *c > ^ c o C! 0/ c T3 P PQ ^ • " o c 3 o CD s o m c3 3 to G O c o O c .2 o O a 3 ii 3 i 3 3 o 3 c , O ^ -i '^ 3 ^ rt " .2 3 5 "" ji "^ 4;S 3 °J^ ~~^ 3 3 01' .3 CO -3 ^ CD _l: 3 O "O 3 ^ a; "S ^ .2 ^ c3 5 ^ E ^ " -1^ w .X >, ■XL -f^ SC 3 O O CO o ,—1 ^^ ^j) S o i: ?H "c -*^ ^ 3r^'o S .2 O 3 3 "^ 00 o3 3 O c3 to .i; -^^ a; 05 00 T— 1 3 to 1— 1 1—1 a; a) bfi 03 p:3 _2 "3 .■S ^ 1—1 a; > X o ^ fe c3 Dh fa J 03 o to 3 ^ 3 . > -n §1 2 "^ JS O o -Q CO li P'Sfi.^ S: S 55, 5 s d H ■g s 03 -2 •« -3 e "o Si § 5 s s S E "tj O Si c O i;:? ^ 0 C^ 0 5^ I— 1 1 ^ 1 p, 0 a;> n 3 3 c3 0 '$. c? c5 1-5 ►^ >; bD 3 m < 0 0 1 '> r^ c3 '0 M 0 05 o3 o3 O C5 3 -3 CO 3 O K o3 O o 3 O m 05 a; s - s ^ 3-^ .S 3 CO 5 -Q -fi 3 *^ --^ o i; — ' O ^ -r ■^. ° c3 fe a; 9 -^ ■^ 2 SI 03 • s s O a; -3 aj o a; ^ O v.- 1.1 ? "C fe .3 CD § -►^ -3 3 03 . 3 ;§ 03 3 03 aj 3 O o S = 03 ^ 2 r° n 03 3 O .- 3 oj o3 aj -►J O O lq 3 -3 ec a; a bh 3 3 <5 2 -3 CO O 3 3 ;>> 03 eo .'^ "S d h~ Q 02 e s g •S S S ■^ g^ '?* 0 ^ Si 0 1 ;-i CD ■ -1-2 +3 a. 0 a; tJ . Tfi --1 £ |-^ 05 ^ o3 o3 1—1 V aj 3 3 >i 3 3 3 -1-J 0 >-i < 0 .2 o3 "3 S Si 0 t— ( 1— 1 ^ -ij" 0 CO CO s 03 o3 0 0 3 u u L. o3 =3 CO T3 3 o3 ;a -2 ^ o3 w -^ T3 _> c -^ 03 1-1 03 "TS CD 3 CO o3 ^^ 03 1— ) — CO "3 K < 05 3 O 3 aj -rj O u — ' aj I ^ £ aj JOHN H. RYTHER 399 CO c3 "^-i O) =- ° . 73 w: ■C ^ C ^ S ^ "~C_S?:rto.«^SK ^:^P-^^r":S ^;n-3S S.^--ferJ"^^ -3 .S - -^ '--^^ « .S; = CO - <^' ^.S-^.So^1- -_ ^ -2 t^ ^ ^ bC2 IS -^ -^- c^ £ .5 - 3 H -^ E b S-^ ^^i.cc-3- _$=t:; ■p g ".5— DH-- E '-""-^ ~ i-cC-T3~ 3 O O ~ ^ , — I '^ Oj •?•.—. — '^ 3^ C G O bl t^ e « «3 e -^ e e e ■§ -5 3 •§ -5 .g .= .^ i o CD (M 2vX~"'^00 05 o P-53^doi C3 C3 3-3 't ^ "~^ "- § "f -f 9 " " 00 .^ 03 c3 3 l> CO _fl i^ 0 OJ 1—* i^ S -1^ CO ^ Is c3 0 « W oj >> E 0" -3 (h 0 0) ^ 3 CL, .— « 0 3 c3 § 03 CC 73 "3 50 E -7-03 o5 ^o3 o3 b»^0 CO [C j"™ 0) o o3 Bay, Scotl 03 -*^ A 0 -H3 ►H 0) a; ^ CO > S '^^ -3'" ^ 4^ ;-4 ^H^«2 0 ^ ^^ ■o tH ;3 ^ ^ '^ c ^ -' :3t-33 o""^o 3 _G (h a 10 00 2 J 0 -^ c; z 10 -3 ^ -3 s^ ■3 QC 35 3 02 C3 1—1 "a3 "B ^"^ 3 3i ■^ — S2 S 3 35 -3. -t^ 05 "3 £2 ^ r-> T-H r-H rf c3 0 0 3 0 yA, § § « 400 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS here completely. Table II gives a partial review of the red water literature, including the time and place of the occurrence, the causa- tive organism, and a brief statement of the ecological conditions preceding or accompanying the outbreak, in so far as these have been observed. Prerequisites for Red Water Conditions The basic requirements for an outbreak of red water may be summarized under the following three headings: (1) a seed popula- tion of dinoHagellates, (2) the existence of favorable conditions for the growth of one or more of the species present, and (3) either the concentration of sufficient nutrients to permit the dense growth of the organisms, or the concentration of the organisms tliemselves, to the degree to which they are found in "red water." The presence of a seed population of dinoflagellates is a condition which is probabh' always met, for there are at least a few of these organisms in almost every part of the ocean at all times. However, this is undoubtedly an important factor, together with the environmental conditions, in determining the particular species of dinoflagellate dominating a red water outbreak. Thus, it is somewhat more difficult to account for the origin of Gijmnodinitim hrevis, the causative agent of the Florida "red tide," which appears to be present in the region only during its periods of blooming. Slobodkin ( personal communica- tion) is of the opinion that seed populations of this organism are main- tained in the brackish to freshwater regions of the Florida Everglades. In maximum developments of "red water," the optimal growth conditions for the species involved are perhaps met or closely ap- proached. However, such ideal situations are probably less common than the existence of an environment in which the growth of some one species of dinoflagellates is favored over that of all other phyto- plankton forms. In most parts of the ocean the diatoms are the principal competitors of the dinoflagellates, though other forms, such as Chrysophceae, Chlorophyceae, and Euglenineae may be important in estuarine conditions. In the preceding sections of this report we have seen that the dinoflagellates can compete successfully with the diatoms under conditions of high temperature, low sahnity, and reduced nutrient JOHN H. RYTHER 401 concentrations. They appear to require the presence of some one or more unknown organic substances, and there is some indication that they benefit from the previous existence of a large popuhition of diatoms. It is doubtful if all these conditions are usually satisfied at any one time, and unlikely that they are all necessary for the domi- nance of dinoflagellates. An examination of Table II reveals one factor which is almost universal in red water outbreaks, the occurrence of a high water temperature. In temperate or boreal regions of the ocean, red water appears to be restricted to the summer months, and the notation is frequently made that it is preceded by periods of unusually hot, calm weather. Along the Indian coast, red water occurs during the clear, hot periods between the southwest and the northeast monsoons (Hornell and Nayudu, 1923; Menon, 1945; Bhimachar and George, 1950). Off the Peruvian and Southwest African coasts it appears during the southern summer when the upwelling of cold water is at a minimum and water temperatures are the highest of the year (Bron- gersma-Sanders, 1948 ) . Allen ( 1946 ) described a number of occur- rences of red or yellow water along the California coast which have always occurred in mid-summer during periods of hot, calm weather and smooth seas. There is less evidence that low salinity is an important factor in red water outbreaks. Slobodkin (1953) has shown a close correlation of red tides along the west coast of Florida with previous periods of exceptionally heavy rainfall. He has proposed that the organisms develop in small, discrete masses of relatively low-salinity water which result from the increased land drainage during heavy rains and are maintained in the ocean by density gradients. Table II gives several other instances in which red water has been preceded by heavy rainfall ( Whitelegge, 1891; Hornell and Nayudu, 1923; Lund, 1936; Menon, 1945; Connell and Cross, 1950). However, this situation is by no means universal. On the South African, Peruvian, and California coasts, red water occurs during periods when precipitation is at a minimum. Furthermore, where such measurements have been made, the salinity in patches of red water does not appear to be significantly lower than that of the surrounding, clear ocean water (Ketchum and Keen, 1948; Chew, 1953; Torrey, 1902). 402 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS It is quite likely, however, that land drainage is important to the dinoflagellates from another aspect, that of providing their necessary organic matter. Whatever the function of these substances, it is possi- ble that they are required in such minute concentrations as to satisfy the requirements of the organisms, although the entraining fresh- water may not appreciably dilute the sea water. The concentration of dissolved nutrient salts in a given water mass prior to the onset of red water has not, to the author's knowledge, been determined. There are, to be sure, small, local occurrences which appear to be related to the introduction of domestic pollutants (Braarud, 1945; Connell and Cross, 1950). In these instances there are perhaps sufficient nutrients, as well as the necessary organic growth factors, to support populations approaching red water proportions. However, in the majority of red water occurrences there is no indica- tion of an unusually high enrichment of the water prior to the out- break. Brongersma-Sanders ( 1948 ) emphasized that red water is associated with those regions of the world where upwelling brings deep, cold, nutrient-rich water to the surface (i.e.. Southwest Africa, Peru, Cali- fornia). She also pointed out, however, that such upwelled water normally supports a luxuriant growth of diatoms, and that red water occurs in these regions only during the summer months when upwel- ling, and presumably the nutrient level, is greatly reduced. On the Indian coast, the southwest monsoon is associated with high enrich- ment and maximum productivity of the waters. Under these condi- tions it is again the diatoms which flourish, and red water appears only when the monsoon has ended. Ketchum and Keen (1948) measured the total phosphorus content of sea water collected in and outside of patches of red water which occurred off the west coast of Florida in 1946-47. They found con- centrations of total phosphorus in the red water (including that contained in the organisms) up to ten times as high as that of the clear water outside of the local patches. These authors emphasize the necessity for a mechanism either for accumulating these concen- trations of phosphorus in the water prior to the growth of the organ- isms, or for concentrating the organisms after they have grown. There was no indication that the first of these alternatives was true. JOHN H. RYTHER 403 The maximum concentration of phosphorus in the deep offshore waters of the Caribbean ( 2 microgram atoms per hter ) , as determined by Rakestraw and Smith ( 1937 ) , does not approach the maximum observed in the red water patches (20 microgram atoms per hter), so upwelhng could not have provided sufficient enrichment to be an important factor. It is also unlikely that land drainage could account for this phosphorus. Since the red water region contained only about 10% freshwater, runoff would have to carry tremendous concentrations of this element to account for the amount present in the organisms (estimated at 17,000 lb of pure phosphorus per square mile of red water ) . Chew ( 1953 ) , who investigated a minor outbreak of red water in the same region in 1952, found that the lower salinity, coastal water, which contained the river drainage, had lower con- centrations of phosphorus than the open waters of the Gulf of Mexico. Thus it would appear that there may be insufficient nutrients normally present in sea water, and no known mechanism for concen- trating them to a sufficiently high level, to support the development of a typcial red tide. The remaining possibility, that the organisms themselves become concentrated after growth, will be discussed below. In every case in which the vertical distribution of dinoflagellates in red water has been examined, it was found that the organisms were concentrated in relatively narrow bands usually at the surface of the water ( Whitelegge, 1891; Nishikawa, 1901; Hirasaka, 1922; Martin and Nelson, 1929). In three instances of small, local outbreaks of red water, cell counts were made in a vertical column, revealing in each case that the maximum concentration of dinoflagellates occurred at the surface (Table III). It is very doubtful that the organisms could have grown in such a pattern of distribution. As pointed out above, there would appear to be insufficient nutrients available to support such populations. Fur- thermore the high light intensities at the surface of the ocean are normally inhibitory to photosynthesis (Stanbury, 1931; Jenkin, 1937), particularly in the case of dinoflagellates, many of which appear to prefer reduced illumination (Barker, 1935; King, 1950). Hence, dense aggregations of dinoflagellates at the surface of the water must be 404 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS Vertical Distributioi (Figures iiid TABLE III 1 of Dinoflagellates in Red Water icate organisms per liter) Depth (meters) Peridinium triquetrum Gonyaulax monilata Braarud (1945) Marshall (1927) Slobodkin (1953) 0 0.3 2.0 2.6 4.0 6.0 361,000 45,000 11,000 32-56,000 800 8,200,000 5,700,000 450,000 explained by concentrations of the organisms themselves, either by means of their active motility, or by passive floatation, should they become less dense than sea water. Gran and Braarud (1935) observed that Peridinium triquetrum was often found in maximum concentrations at the surface, but that this organism could occur at any depth from 0 to 25 meters. They con- cluded that it could seek out the level at which the general conditions for growth were most favorable. Halse ( 1950 ) demonstrated marked vertical diurnal migrations in several dinoflagellates, the character- istics of which differed considerably from species to species. Thus Ceratium fusus and C. tripos rose to the surface at night and sank to the lower depths during the day, while Gomjaulax polyedra and Pro- rocentrum m^icans showed the opposite response to light, rising in the daytime and sinking at night. Concentration of dinoflagellates by means of such migrations may, at times, create red water conditions. Hirasaka (1922) observed red water in Gokasho Bay, Japan, in which the entire bay was apparently affected, in contrast to the usual situation in which red water is re- stricted to patches or streaks. According to this author, the organisms were confined to a band no more than 3 to 4 ft thick which appeared to migrate diurnally, the red color being most conspicuous in the late afternoon. In many cases, however, the red water organisms remain at the surface of the water at all times. Woodcock ( personal communication ) collected samples of Gyinnodinium brevis during the 1946-47 red tide off Florida and observed that the organisms remained at the JOHN H. RYTHER 405 surface of the vessel at all times of the day and night. Ketchum (personal communication) obsewed the same behavior in Noctiluca miliaris collected during a bloom at Friday Harbor. Homell and Nayu'du (1923) and Nishikawa (1901) found that the red water organisms remained floating at the surface of a bottle apparently until they encysted; then they sank to the bottom. Harvey (1917) and Ketchum (unpubhshed) studied the specific gravity of Noctiluca collected from patches of red w^ater. In each case the organisms remained floating at the surface of the sea water if the latter was diluted with fresh water until the dilution approached 50%, beyond which they became suspended or sank. It must be em- phasized, however, that these observations were made upon cells collected during typical red water conditions. It must not be assumed that such low densities are typical of Noctiluca at all times, and it cannot be so assumed if one subscribes to the theory that they accumulate at the surface after their growth. In this connection, the present author has observed a healthy culture of Noctiluca grown by Dr. L. Provasoli in which the organisms were evenly distributed throughout the medium. Pratje (1921) has suggested that Noctiluca undergoes physiological changes in response to its environment, and that senescent, nondivid- ing cells lose density and become buoyant in sea water. He subscribes to the theory suggested above that the concentration of these organ- isms at the surface does not result from active migration, but from floatation. Spoehr and Milner (1949) and others have demonstrated that Chlorella ceases to divide and stores fats when the nitrogen in its medium becomes limiting. Fat storage in the autotrophic dino- flagellates is common. Many species accumulate bright yellow or red oil droplets (Graham, 1951). Noctiluca apparently does not store fats, but its buoyancy is explained on the basis of the low specific gravity of its cell sap possibly through the accumulation of NH4+ ions (see Krogh, 1939), Such mechanisms as these are adequate to account for the floatation of dinoflagellates. It remains to be demon- strated that changes occur in their metabolism or osmo-regulation such as to make them buoyant only at certain stages of their development. The accumulation of buoyant organisms at the water surface would, of course, be enhanced in the absence of vertical mixing of the water, 406 MARINE DIXOFLAGELLATES AND RED WATER CONDITIONS which would tend to carry them down again into deeper layers. Harvey ( 1917 ) mentions that Noctiluca could not be observed at the surface during windy days. A re-examination of Table II will reveal that most of the occurrences of red water throughout the world were accompanied or preceded by periods of calm weather and smooth seas. Such conditions, together with the high temperatures which usually accompany them, may further stabilize the water through thennal stratification, providing additional resistance to the vertical mixing of the organisms. PREVAILING WIND < Fig. 3. Accumulation of floating material along shore by prevailing onshore wind. Once the organisms have accumulated at the surface of the water, there are several means by which they may become further con- centrated. Three such mechanisms will be discussed briefly below. (1) Prevailing onshore winds: Surface water driven shoreward by prevailing onshore winds establishes a circular pattern, sinking at the waters edge and returning seaward at lower depths. Buoyant organ- isms will accumulate in windrows along shore or at the region of descent (Fig. 3). (2) Where brackish coastal water, particularly in the vicinity of river mouths, meets open ocean water, there is a mixing and sinking of the two water masses along a line of convergence. Both types of water flow toward this line, and buoyant organisms will accumulate at or near the convergence line, producing streaks of floating material. (Fig. 4). (3) Convection cells: Wind-driven vertical convection- cells may be established which rotate alternately clockwise and counterclockwise JOHN H. RYTHER 407 with their vertical axes perpendicular to the direction of the prevailing wind. Floating objects will accumulate in the region between the descending components of two such adjoining cells. Under these con- ditions parallel streaks of floating matter are produced ( Fig. 5 ) . ( For a more detailed description of this process, see Langmuir, 1938; Stommel, 1949.) BAND OF aCCUMUL/XTION Fig. 4. Accumulation of floating material at con\ergence of water masses of dif- ferent density. STREAKS OF ACCUMULATED F LOAT I r\JG MATTER Fig. 5. Accumulation of floating material by wind-driven convection cells. These and other similar mechanisms are undoubtedly important means of concentrating dinoflagellates to the extent to which they are observed in red water conditions. The existence of such processes is indicated by the repeated mention of patches or streaks of organisms in many descriptions of red water (see Table II). Bary (1953) has indicated that convection cells are responsible for the streaky distribu- tion of floating masses of the ciliate Cyclotrichium meunieri in Wel- lington Harbor, New Zealand. 408 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS Thus through the combined processes of floatation and surface con- centration, a method is possible by means of which red water condi- tions may be produced without the excessive enrichment of the water, and, indeed, without an unusually heavy growth of dinoflagellates. This may be illustrated by the following hypothetical situation. According to Riley (1937, 1938) phosphate-phosphorus values for the surface waters of the Gulf of Mexico range from 0.02 to 0.5 micro- gram atom per liter. Since red water often occurs when nutrient levels are low, let us propose a situation in which the phosphate-phosphorus concentration is 0.05 microgram atom or 1.55 X 10^® gram per liter. Let us further assume that other mineral nutrients, organic growth factors, etc., are present in the same or higher concentrations as phosphorus relative to the requirements of a given species of dino- flagellate. No data are available concerning the dry weight and elementary analysis of red water organisms. However, these may be roughly estimated from known figures for other plankton organisms, and values may be obtained which are probably reliable within an order of magnitude. Gymnodinium brevis, the dinoflagellate responsible for the Florida red tide of 1946-47, measures approximately 28 by 28 by 13 mi- crons (Davis, 1948), and its volume may be roughly estimated at 7.5 X 10-^ cc. If it is assumed to have about the same densit}^ as sea water (at about 34<'/oo and 25° C in the coastal waters of the Gulf of Mexico), its wet weight will approximate 7.7 X lO"'' gram per cell. Ketchum and Redfield (1949) found that the dry weight of a va- riety of planktonic algae was approximately 25% of their wet weight, and that their phosphorus content was rather constant at about 2.5% of the dry weight. Using these figures for Gymnodinium (which are probably somewhat high for dinoflagellates in general) a value of 0.48 X 10^^^ gram of phosphorus per cell is obtained. A dissolved phosphorus concentration of 1.55 X 10"^ gram per liter will then support a population of 1.55 X 10-« 0.48 X 10-10 = .3.2 X 10" cells per liter JOHN H. RYTHER 409 if all of the phosphorus were utilized and no other nutrient was limit- ing. If the dinoflagellates in this hypothetical situation are also able to utilize all the nutrients in a water column 10 meters deep, and then, due to changes in their specific gravit}% accumulate in the upper meter, their concentration at the surface will be in the range of 3.2 X 10^ cells per liter. The action of winds, convergence, or convection may then easily concentrate this surface layer by another factor of 10 or 20, producing concentrations of organisms typical of red water con- ditions, which range anywhere from 10^' to lO'^ cells per liter. Thus there is no necessit\' to postulate obscure factors which would account for a prodigious growth of dinoflagellates to explain red water. It is necessary only to have conditions favoring the growth and dominance of a moderately large population of a given species, and the proper hydrographic and meteorological conditions to permit the accumulation of organisms at the surface and to effect their further concentrations in localized areas. References Aiyar, R. G. 1936. Mortality of fish of the Madras Coast in June, 1935. Current Sci., 4, 488-89. Allen, W. E. 1939. Phosphorescence in the sea. Naut. Gaz., Aug. 26, 1939. Allen, W. E. 1946. "Red water" in La Jolla Ray in 1945. Trans. Amer. Microscop. Soc, 65, 149-53. Barker, H. A. 1935. The culture and physiology of marine dinoflagellates. Arch. Mikrobiol 6, 157-81. Rar\', B. M. 1953. Sea-water discoloration by living organisms. New Zealand J. Sci. Technol. B34, 393-407. Rhimachar, B. S., and P. C. George. 1950. Abrupt set backs in the fisheries of the Malabar and Kanara Coasts and "red water" phenomenon as their probable cause. Proc. Indian Acad. Sci., B31, 339-50. Bonnot, P., and J. B. Phillips. 1938. Red water, its cause and occurrences. Calif. Fish and Game, 24, 55-59. Braarud, T. 1935. The 0st expedition to the Denmark Strait, 1929. II. The phytoplankton and its condition of growth. Hvalradets Shifter, Norske Videnskaps — Acad. Oslo, No. 10. 410 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS Braarud, T. 1945. A phytoplankton survey of the polluted waters of inner Oslofjord. Hvalmdets Skrifter, Norske Videnskaps-Akad. Oslo, No. 28. Braarud, T. 1951. Salinity as an ecological factor in marine phytoplankton. Physiol Plant., 4, 28-34. Braarud, T., and I. Pappas. 1951. Experimental studies on the dinoflagellate Peridinium triquetrum (Ehrb.) Lebour. Avhandl. Norske Videmkaps- Acad. Oslo Mat-Naturv. Klasse, 1951, No. 2. Braarud, T., and E. Rossavik. 1951. Observations on the marine dinoflagel- late Prorocentrurn micans Ehrenb. in culture. Avhandl. Norske Videnskaps- Akad. Oslo Mat. Natiirv. Klasse, 1951, No. 1. Brongersma-Sanders, M. 1948. The importance of upwelling water to verte- brate paleontology and oil geology. Verhandl. Konin. Nederland. Akad. Wetenschap. Natiiur., Tweede Sectie, 45, No. 4, 112 pp. Chew, F. 1953. Results of hydrographic and chemical investigations in the region of the "red tide" bloom on the West Coast of Florida in Nov., 1952. Bull. Marine ScL, Gulf and Carib., 2, 610-25. Connell, C. H., and J. B. Cross. 1950. Mass mortality of fish associated with the protozoan Gonijaulax in the Gulf of Mexico. Science, 112, 359-63. Davis, C. C. 1948. Gijmnodinium brevis sp. nov., a cause of discolored water and animal mortality in the Gulf of Mexico. Baton. Gaz., 109, 358-60. Fritsch, F. E. 1935. The structure and reproduction of the algae. Vol. I. The University Press, Cambridge, Mass. Gaarder, T., and H. H. Gran. 1927. Production of plankton in the Oslofjord. Cons. Perm. Int. Explor. Mer; Rapport et Proc. Verb. 42, 1-48. Gilchrist, J. D. F. 1914. An enquiry into fluctuations in fish supply on the South African Coast. Marine Biol. Kept., Union Sotith Africa, 2 (2), 8-35. Gilson, H. C. 1937. The Nitrogen Cycle. John Murray Exped., 2, 21-81. Graham, H. W. 1951. Pyrrophyta. In Manual of Phycology: An Introduction to the Algae and Their Biology. The Chronica Botanica Co., Waltham, Mass. Gran, H. H. 1902. Das plankton des norwegischen Nordmeers von bio- logischen und hydrographischen Gesichtspunkten behandelt. Bergen. Kept, on Norwegian Marine — und Fishery — Investigations, 2. No. 5. Gran, H. H. 1912. Pelagic Plant Life. In J. Murray and J. Hjort, The Depths of the Ocean. MacMillan and Co., London. Gran, H. H. 1924. Quantitative plankton investigations carried out during the expedition with "Michael Sars" July-Sept., 1924. Cons. Perm. Exp. Mer; Rapport et Proc. Verb., 56. Gran, H. H. 1926-27. Investigation of the production of plankton outside the Romsdalfjord. Cons. Int. Exp. Mer; Rapport et Proc. Verb., 56. Gran, H. H., and T. Braarud. 1935. A quantitative study of the phyto- plankton in the Bay of Fundy and the Gulf of Maine. /. Biol. Board Can- ada, 1, 279-467. JOHN H. RYTHER 411 Gross, F. 1937. Notes on the culture of some marine plankton organisms. /. Marine Biol. Assoc, 21, 753-68. Gross, F., and E. Zeuthen. 1948. The buoyancy of plankton diatoms: a problem of cell physiology. Proc. Roy. Soc. (London), B135, 382-89. Gunter, G., R. H. WiUiams, C. C. Davis, and F. G. Walton Smith. 1948. Catastropic mass mortality of marine animals and coincident phytoplank- ton bloom on the west coast of Florida, November 1946 to August 1947. Ecol. Monographs, 18, 309-24. Halse, G. R. 1950. Phototactic vertical migration in marine dinoflagellates. Oikos, 2, 162-75. Harvey, E. B. 1917. A physiological study of specific gravity and of lumi- nescence in Noctiliica with special reference to anaesthesis. Carnegie Inst. Wash., Papers Dept. Marine Biol, 11, 235-53. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Herdman, W. A. 1922. Spolia Runiana V. Summary of results of continuous investigation of the plankton of the Irish Sea during 15 years. /. Linnean Soc. London Botany, 46, 141-70. Hirasaka, K. 1922, On a case of discolored sea-water. Annot. Zool. Japan, 10, 161-64. Hornell, J., and M. R. Nayudu. 1923. A contribution to the Life history of the Indian Sardine. Madras Fish. Bidl., 17, 129-97. Hutchinson, G. E. 1944. Limnological studies in Connecticut, VII. A critical examination of the supposed relationship between phytoplankton peri- odicity and chemical changes in lake waters. Ecology, 25, 3-26. Jenkin, P. M. 1937. Oxygen production by the diatom, Coscinodiscus excen- triciis Ehr. in relation to submarine illumination in the English Channel. /. Marine Biol. Assoc. U. K., 22, 301-43. Ketchum, B, H., and J, Keen, 1948, Unusual phosphorus concentrations in the Florida "red tide" sea water, /. Marine Research, 7, 17-21. Ketchum, B. H., and A. C, Redfield, 1949. Some physical and chemical char- acteristics of algae grown in mass culture. /. Cellular and Comp. Physiol., 33, 281-300, King, G, S, 1950. Production of red tide in the laboratory. Proc. Gulf and Carib. Fish. Inst. 2nd Ann. Sess., Nov., 1949, 107-109. Kofoid, C. A. 1906, On the significance of the asymmetry in Triposolenia. Univ. Calif. Puhl. Zool, 3, 127-33, Kofoid, C, A,, and O, Swezy, 1921, The free living unarmored Dinoflagellata, Mem. Univ. Calif. ,5, 1-562, Krogh, A, 1939. Osmotic Regulation in Aquatic Animals. The University Press, Cambridge. Langmuir, I. 1938, Surface motion of water induced by wind. Science, 87, 119-23. 412 MARINE DINOFLAGELLATES AND RED WATER CONDITIONS Lucas, C. E. 1947. The ecological effects of external metabolites. Biol. Revs., 22, 270-95. Lucas, C. E. 1949. External metabolites and ecological adaptation. Symp. Soc. Exptl. Biol. III. Selective toxicity and antibiotics. Academic Press, New York. Lund, E. J. 1936. Some facts relating to the occurrence of dead and dying fish on the Texas Coast during June, July, and August, 1935. Cited from: G. Gunter et al. 1948. Catastrophic mass mortality of marine animals and coincident phytoplankton bloom on the west coast of Florida, November, 1946 to August, 1947. Ecol. Monographs, 18, 309-24. Marshall, S. M. 1947. An experiment in marine fish cultivation III. The plankton of a fertihzed loch. Proc. Roy. Soc. Edinburgh, B63, 21-33. Marshall, S. M., and A. P. Orr. 1927. The relation of the plankton to some chemical and physical factors in the Clyde Sea area. /. Marine Biol. Assoc., 14, 837-68. Martin, G. W., and T. C. Nelson. 1929. Swarming of dinoflagellates in Dela- ware Bay, New Jersey. Botan. Gaz., 88, 218-24. Mead, A. D. 1898. Peridinium and the "red water" in Narragansett Bay. Science, 8, 707-709. Menon, M. A. S. 1945. Observations on the seasonal distribution of the plankton of the Trivandrum Coast. Proc. Ind. Acad. Sci., B22, 31-62. Michaelis, G. A. 1830. Ueber das Leuchten der Ostsee nach eigenen Beb- bachtungen. Cited from: C. A. Kofoid and O. Swezy. 1921. The free- living unarmored Dinoflagellata. Mem. Univ. Calif., 5, 1-562. Munk, W. H., and G. A. Riley. 1952. Absorption of nutrients by aquatic plants. J. Marine Research, 11, 215-40. Nishikawa, T. 1901. Gonyaulax and the discolored water in the Bay of Agu. Annot. Zool. Japon, 4, 31-34. Nordli, E. 1953. Salinity and temperature as controlling factors for distribu- tion and mass occurrence of Ceratia. Saertrykk av. Blyttia, 11, 16-18. Pearsall, W. H. 1932. Phytoplankton in the English lakes. II. The composi- tion of the phytoplankton in relation to dissolved substances. /. Ecol, 20, 241-62. Peters, N. 1929. Uber Orts- und Geisselbewegund bei marinen Dinoflagel- laten. Archiv. f. Protistenk. 67: 291-321. Peters, N. 1932. Wissenschaftliche Ergebnisse der deutschen atantischen Expedition auf dem Forschungs-und Vermessungsschiff "Meteor" 1925- 1927. Bd 12, Lief 1: Die Bevolkerung des Siidatlentischen Ozeans mit Ceratien. Cited from: Bot. Centr. [N. F.] 21, 412-13. Pratje, A. 1921. Noctiluca miliaris Suriray. Beitrage zur Morphologic, Physi- ologic, und Cytologic. I. Morphologie und Physiologic, Arch. Protistenk., 42, 1-98. Provasoli, L., and I. J. Pintner. 1953. Ecological implications of in vitro JOHN H. RYTHER 413 nutritional requirements of algae flagellates. Ann. N. Y. Acad. Sci., 56, 839-51. Rakestraw, N. W., and H. P. Smith. 1937. A contribution to the chemistry of the Caribbean and Cayman Seas. Bull. Bingham Oceanog. Coll., 6 (1), 1-41. Rice, T. R. 1954. Biotic influences affecting population growth of planktonic algae. U. S. Fish Wildlife Service Fishery Bull, 54, 227-45. Riley, G. A. 1937. The significance of the Mississippi River drainage for biological conditions in the northern Gulf of Mexico. /. Marine Research, 1, 60-74. Riley, G. A. 1938. Plankton studies. I. A preliminary investigation of the plankton of the Tortugas region. /. Marine Research, 1, 335-52. Ritchie, G. S. 1952. H. M. S. Challangers investigations in the Pacific Ocean. ;. Inst. Nav., 5, 251-61. Russell, F. S. 1927. The vertical distribution of plankton in the sea. Biol. Revs. 2, 213-62. Slobodkin, L. B. 1953. A possible initial condition for red tides on the coast of Florida. /. Marine Research, 12, 148-55. Spoehr, H. A., and H. W. Milner. 1949. The chemical composition of Chlo- rella; effect of environmental conditions. Plant Physiol., 24, 120-49. Stommel, H. 1949. Trajectories of small bodies sinking slowly through con- vection cells. /. Marine Research, 8, 24-29. Stanbury, F. A. 1931. The effect of light of different intensities, reduced selectively and non-selectively, upon the rate of growth of Nitzschia closteriiim. J. Marine Biol. Assoc, 17, 633-53. Sverdrup, H. U., M. W. Johnson, and R. H. Fleming. 1942. The Oceans, Their Physics, Chemistry, and General Biology. Prentice-Hall, New York. Sweeney, B. M. 1951. Culture of the marine dinoflagellate, Gymnodinium with soil extract. Am. J. Bot., 38, 669-77. Sweeney, B. M. Gymnodinium splendens, a marine dinoflagellate requiring vitamin B12. In press. Torrey, H. B. 1902. An unusual occurrence of dinoflagellates on the Cali- fornia Coast. Am. Naturalist, 36, 187-92. Vinogradov, A. P. 1935. Elementary chemical composition of marine organ- isms. Part I. Cited from: H. U. Sverdrup, M. W. Johnson, and R. H. Flem- ing. 1942. The Oceans, Their Physics, Chemistry, and General Biology. Prentice-Hall, New York. Whitelegge, T. 1891. On the recent discoloration of the waters of Port Jack- son. Rec. Aust. Mus., Sydney. 1, 179-92. Woodcock, A. H. 1948. Note concerning human respiratory irritation asso- ciated with high concentration of plankton and mass mortality of marine organisms. /. Marine Research, 7, 56-62. Discussion Dr. Sweeney: Dr. Ryther has made clear in his presentation, an understanding of the nutritional requirements of the dinoflagellates is prerequisite to any explanation of the occurrence of "red tides." These organisms have been found to require for growth, substances not present in aged autoclaved sea water and usually supplied by the addition to the culture medium of small amounts of soil extract. In our laboratory, an analysis of the activity of soil extract with respect to the growth of bacteria-free Gymnodinium splendens has shown that soil extract may be replaced by vitamin B12, the optimum concentra- tion being at 0.01 microgram vitamin B12 per liter. Provasoli has shown that the dinoflagellate Gyrodinium sp. also requires vitamin B12. The organic requirements of Prorocentrum micans and Gonyau- lax polyedra appear to be more complex and cannot be met by vitamin B12 alone. Gonyaulax polyedra may be grown successfully in mineral-enriched aged sea water with soil extract. Maximum populations reached are of the order of 40,000 cells per milliliter and the generation time is 40 to 60 hours. This organism is bioluminescent in culture. (Prorocen- trum micans, Gymnodinium splendens, and Gyrodinium in our cultures are not luminescent.) Because it is both luminescent and photosyn- thetic, Gonyaulax provides interesting material for the study of biolu- minescence. 4J4 Bioluminescence in Gonyaulax polyedra* F. T. Haxo and Beatrice M. Sweeney Scripps Institution of Oceanography, University of California, La Jolla, California Gonyatilax polyedra is a photosynthetic marine dinoflagellate which is occasionally responsible for striking luminescent displays in the coastal waters of Southern California. The individual cells are ar- mored; they are polyhedral in shape, 45/a long and 41)U. wide and con- tain a heavily pigmented brown protoplast, in which the chlorophyll is masked by an abundance of carotenoid pigments. The unialgal culture of Gomjaulax polyedra investigated was started from net samples collected off Scripps Pier on September 22, 1952. Since that time, the organism has been maintained in aged sea water, diluted to 75% of full strength with twice-distilled water, and supplemented with 2 X IQ-^M KNO3, 2 X IQ-'^M K2HPO4, 6 X 10-^ M FeCls, 6 X 10-"M MnClo, ethylene diamine tetracetic acid (EDTA), 10 mg per liter, and 2% soil extract. Liquid flask cultures were grown either at room temperature in a window with a northern exposure, or at 20° C under continuous illumination from white fluorescent lamps, at an intensity of about 700 foot-candles. Growth was about equally good in full-strength and half-strength sea water, or in media in which the supplementary salts were reduced to one- half the above values. Soil extract was required for growth in aged sea water and could not be replaced by vitamin B]2 alone. No growth, however, was ob- tained in artificial sea water supplemented with soil extract. At 20° C, Scripps Contribution No. 744. 415 416 BIOLUMINESCENCE IN GONYAULAX POLYEDRA maximum rate of growth was obtained at 500-800 foot-candles. The generation time was 56 hours (0.43 division in 24 hours), and the maximum cell densities were 30,000-50,000 cells per milliliter. Photosynthesis of Gonyaulax cell suspensions was measured in arti- ficial sea water (formula of Emerson and Green, 1934, modified to contain one-fourth the normal amount of Ca and Mg) by the direct manometric method. In saturating hght (800 ft-c) at 20° C and non- limiting COo concentration, as provided by 0.0234 M bicarbonate - 0.0016 M carbonate buffer, the rate of photosynthetic oxygen production was as follows (corrected for respiration): 350 mm^ Oa/hr/lO^ cells; 14,800 mm^ Oo/hr/g sohds; 5000 mm^ Os/hr/mg chlorophyll a plus c, estimated by the equations of Richards (1952). Under these conditions, the rate of oxygen production was about 5 times that of respiratory oxygen consumption. The temperature opti- mum for photosynthesis in Gonyaulax is about 30° C. Gonyaulax has retained the ability to luminesce in culture for 1^4 years. Light is emitted on mechanical stimulation as a bright bluish- green flash of short duration. Stimuli repeated at intervals of one minute elicit progressively weaker responses. Since preliminary ob- servation with dark-adapted eyes indicated that luminescence of cell suspensions exposed to light was considerably weaker than those darkened for a few hours, a study was undertaken of the effect of Hght on the luminescence of this organism. The apparatus employed for measuring and recording light emission included a photomultiplier tube (931A) and appropriate ampHfying circuit, with variable sensitivity, coupled to a Speedomax recorder. Insertion of a con- denser in the circuit served to slow down the instrumental response and to provide smooth curves for convenience in planimetric calcula- tions of total light emission.* Aliquot cell suspensions of 3 ml were taken from mature cultures (20,000-40,000 cells per milliliter) grown at 20° C in constant light of 700 ft-c. The aliquots were held in the lighted incubator for 4 hours and then placed in darkness. Measure- ments of light emission by different aliquots were made at intervals during the dark adaptation period. Luminescence, stimulated by pass- ing a stream of air through the cell suspension at a constant rate of " This apparatus was designed by Mr. James Snodgrass, Division of Special Developments, Scripps Institution of Oceanography. F. T. HAXO AND BEATRICE M. SWEENEY 417 flow for one minute, rose sharply to a maximum in less than 7 seconds (less than a second when measured with a faster recording device) and decreased to a low steady state. Cell suspensions taken directly from the lighted incubator lumi- nesced only slightly on stimulation. When such cells were placed in f \\ O 2-6-S4 SERIES X 2-IS-64 SERIES 20 30 40 50 DARK PERIOD IN HOURS Fig. 1. Variation in luminescence of Gonyaulax polyedra cell suspensions with time in darkness, after gro\\th in white hght at 700 ft-c. Each measurement represents the integrated light emission during mechanical stimulation of one minute duration by different aliquots, not otherwise agitated until the indicated time. darkness, the capacity for luminescence increased with time in dark- ness and reached a peak after 6-8 hours, as shown in Fig. 1. With the time in darkness further increased, luminescence decreased pro- gressively, reaching a low value after 15-18 hours. This decline was followed by a second increase in light emission, which reached a peak after 28-30 hours in darkness. The second peak was lower than the first and was followed bv a second decrease in luminescence. This 418 BIOLUMINESCENCE IN GONYAULAX POLYEDRA fluctuation pattern in luminescence has been observed a number of times and appears to be independent of the time of day at which cultures are placed in darkness. Instrumental fluctuations and varia- tions between duplicates were of the order of 2-4% and could not account for the changes observed. Cell suspensions placed in white light after an optimum dark adap- tation time (5-6 hr) showed a progressive decrease in luminescence z o t- < v> m z o a. «n UJ a: UJ o CO UJ 20 40 60 MINUTES IN WHITE LIOHT 100 Fig. 2. Decrease in luminescence of Gonyaulax polyedra suspensions upon ex- posure of dark-adapted cells to 700 ft-c white light. The cells were grown in the light and placed in darkness for 6 hours prior to irradiation. Relative light emission prior to dark adaptation was 0.75. F. T. HAXO AND BEATRICE M. SWEENEY 419 with time of irradiation ( Fig. 2 ) . When exposed to fluorescent hght of 700 ft-c intensity, maximum inhibition of luminescence was obtained in 90 minutes, the final level of light emission being about equal to that of cultures maintained continuously in the light. Preliminary experiments with broad band pass filters intercepting the fluorescent hght indicate that the blue-violet end of the spectrum is the most effecti\e in suppressing luminescence. There is a suggestion of a small effect of red hght, and a more detailed determination of the effective- ness spectrum is planned. The inhibition of luminescence by light in Gomjaulax suggests a photooxidation of one of the components of the luminescent system, as in luminous bacteria and Cypridina extracts. The observation that the maximum light emission of dark-adapted cultures increases with the duration of the previous photosynthetic period implies that light exerts a second effect on luminescence, probably through the building up of photosynthetic products or interaiediates necessary for the maintenance of the luminescent system. The decline in luminescence capacity in the dark observed during the period 8-18 hours (Fig. 1) may be interpreted as a progressive depletion of these substances. It has been suggested that, as starvation proceeds, degenerative proc- esses within the cell result in the release of additional reserves and lead to the second peak observed in darkened cells. The final decline in luminescence after 30 hours is associated with loss in cell viability and death. This undoubtedly results from starvation, since Gomjaulax is nutritionally dependent on photosynthesis. The above findings recall earlier reports (cf. Harvey, 1952) based on visual observations that natural populations of dinoflagellates show fluctuations in luminescence when brought into the laboratory. In unialgal cultures of Gonyaulax polyedra maintained in the labora- tory, luminescence is partially suppressed by exposure to bright light. The ability to luminesce on stimulation is enhanced in darkened cultures, and subsequently falls and rises before the cells die of starvation. 420 BIOLUMINESCENCE IN GONYAULAX POLYEDRA References Emerson, R., and L. Green. 1934. Manometric measurements of photosyn- thesis in the marine alga Gigartina. J. Gen. Phijsiol, 17, 817-42. Harvey, E. N. 1952. Bioluminescence. Academic Press, New York. Richards, F. A., and T. G. Thompson. 1952. The estimation and characteri- zation of plankton populations by pigment analyses. II. A spectrophoto- metric method for the estimation of plankton pigments. /. Marine Research, 11, 156-72. Author Index Abe, T. H., 342, 371, 372, 375 Adawara, 337 Aivar, R. G., 397, 398, 409 Akaba, T., 337, 375 Albrecht, H. O., 77, 97 Alexander, R. E., 329-331 Allen, W. E., 388, 389, 397, 399, 401, 409 Allman, G., 300, 317 Anderson, R. S., 9, 128-133, 137, 138, 140, 153 Annandale, N., 375 Arakawa, S., 375 Arnold, W. A., 3, 24, 47, 50 Audubert, R., 246 Raba, K., 375 Backovasky, J. M., 8 Baker, R. S., 283, 293 Ballentine, R., 114, 123, 130, 154 Barker, H. A., 389, 392, 393, 403, 409 Bar\% B. M., 407. 409 Bavlor, E. R., 266, 267, 293 Baylor, M. B., 275, 294 Becker, R. S., 25, 43, 44 Beijerinck, M. W., 10, 266, 293 Bellis, D. R., 74 Bemanose, A., 78, 85, 97 Bessey, O. A., 140, 155 Bhimachar, B. S., 398, 401, 409 Blinks, L. R., 68, 74 Bonhomme, C., 305, 317 Bonnot, P., 399, 409 Bose, S. R., 342, 375 Bowden, B. J., 6, 350, 375 Boyle, R., 275, 293 Bozler, E., 302, 317 Braarud, T., 388-394, 399, 402, 404, 409, 410 Branch, G. E. K., 95, 98 Brealey, G. J., 31, 41 Bremer, T., 77-80, 85, 91, 97 Brigham, E. H., 23, 116, 123, 138, 154, 159 Brongersma-Sanders, M., 399, 401, 402, 410 Bronk, T. R., 23, 130, 154 Brown, D. E. S., 5, 239, 267, 272- 274, 286, 287, 293, 294, 324, 325, 327, 331 Briicke, E., 328, 331 Buchner, P., 2, 23 Buck, J. B., 12, 14, 23, 300, 302, 310, 311, 316, 317, 323, 324, 3'^ 8— 331 Bullock, T. H., 302, 317 Calvin, M., 35, 37, 41 Campbell, D. H., 275, 282, 294 Carter, G. S., 310, 317 Chakravorty, P. N., 114, 123, 130, 154 Chambers, L. A., 97 Chance, B., 81, 97, 224, 238 Chang, J. J., 5, 7, 12, 23, 24, 131, 156, 240, 244, 258, 262, 264, 295 Chaplin, H., 239, 294 Chase, A. M., 6, 9, 23, 116, 123, 127-132, 138-140, 154, 159, 264 Chew, F., 399, 401, 403, 410 Coblentz, W. W, 110, 123 ConneU, C. H, 397, 399, 401, 402, 410 421 422 Cook, R. M., 74 Cormier, M. J., 7, 24, 131, 156, 211, 213, 216, 221, 223, 224, 226, 233, 238-240, 244, 258, 262, 264, 269, 284, 293, 295 Corner, E. J. H., 338, 376 Coulombre, J., 7, 24, 127, 155, 163, 168, 173, 176, 178, 182, 183, 197, 198, 239, 258, 261, 262, 264, 295 Coulson, C. A., 27, 31, 42 Cross, T. B., 397, 399, 401, 402, 410 Crozier, W. J., 310, 317 Cushing, J. E., 4, 14, 23, 317 Davenport, D., 4, 14, 23, 300, 317, 333 Davidson, J. B., 47, 50 Davis, C.C, 397, 408, 410, 411 Davis, E. P., 131-133, 155 Dorough, G. D., 7, 34, 35, 37, 41, 42 Doudoroff, M., 266, 293 Dougherty, G., 77, 92-94, 96, 98 Drew, H. D. K., 77, 92, 94, 97, 235, 239 Dubois, R., 13, 128, 154, 164, 197, 263, 264 Duvsens, L. N. M., 43, 51, 52, 66, 71, 73, 103, 123 Emerson, R., 416, 422 Enders, H. E., 301, 317 Esaki, T., 364, 376 Evans, M. G., 246 Evstigneev, V. B., 33, 42 Evmers, J. G., 108, 110, 123, 210, 239, 240 Eyring, H., 24, 75, 130, 155, 229, 239, 244-246, 269-274, 276, 277, 279-282, 286, 290, 293-295 Fair, W. R., 74 Farghaly, A. H., 258, 264, 266, 268, 293, 295 Farr, A. L., 155 Fehr, K. L. A., 150, 155 Fisher, K. C, 280, 281, 293 Flagler, E. A., 278, 294 Fleming, R. H., 389, 395, 413 AUTHOR INDEX Flosdorf, E. W., 77, 97 Forster, Th., 102, 123 Foster, R. A. C, 275, 283, 293 Franck, J., 42 Fraser, D., 274, 275, 283, 293, 294 French, C. S., 51, 52, 63, 69-71, 74 Friedman, S., 260, 295 Fritseh, F. E., 389, 390, 410 Fujiwara, T., 347, 376 Gaarder, T., 389, 390, 410 Garwood, R. F., 92, 94, 97 Gavrilova, V., 33, 42 George, P. C., 398, 401, 409 Gerretsen, F. C, 210, 239 Gherardi, G., 239, 294 Gierse, A., 306, 318 Giese, A. C, 128, 131, 132, 140, 154, 266, 293 Gilchrist, T. D. F., 399, 410 Gilson, H. C., 391, 410 Goldfinger, P., 97 Goodwin, R. H., 72, 74 Graham, H. W., 405, 410 Gran, H. H., 388-392, 394, 395, 404, 410 Green, A., 259, 261, 264 Green, L., 416, 420 Greene, C. W., 306, 318. 321 Greene, H. H., 306, 318, 321 Gregg, J. H., 131, 154 Grimpe, G., 20 Gross, F., 393, 395, 411 Grube, 347 Gunter, G., 397, 399, 411 Hada, Y., 376 Haga, 336 Halse, G. R., 404, 411 Hamada, M., 376 Handrick, K., 306, 318 Haneda, Y., 6, 20, 23, 196, 197, 308, 318, 335-338, 348, 368, 372, 374, 376, 377, 385 Harms, J. W., 377 Harris, D. G., 51, 57, 61, 74 Harris, L., 85, 97 AUTHOR INDEX Harvey, E. B., 300, 318, 405, 406, 411 Harvey, E. N., 1, 2, 4-10, 12, 20, 21, 23, 24, 77, 79, 97, 127, 128, 130, 131, 138, 147, 154-156, 161, 195-198, 209-211, 223, 238-241, 243, 244, 258, 262-264, 268, 294, 295, 299, 300, 302, 307-309, 311, 318, 320, 324, 332, 351, 353, 354, 372, 377, 387, 405, 406, 411, 419, 420 Hasama, B., 5, 354, 377, 378 Hastings, J. W., 7, 9, 23, 24, 127, 130, 155, 161, 182, 183, 197, 198, 214, 239, 257, 258, 261, 262, 264, 295 Hatt, H. H., 97 Haxo, F. T., 68, 74, 415, 417 Hayashi, K., 378 Hayashi, S., 378 Hays, R., 176, 178, 197 Heinemann, C, 327, 329, 332 Hendricks, Sterling, 42 Hennings, P., 378 Herdman, W. A., 389, 411 Herre, A. W. C. T., 378 Heymans, C, 20, 302, 309, 318 Hickling, C. F., 20 Hirasaka, K., 378, 398, 403, 404, 411 Hobart, F. A., 97 Hoffman, H., 20 Hollaender, A., 74 Holt, W. W., 74 Honjo, I., 378 Hopkins, F. G., 281, 294 Hornell,]., 398, 401, 405,411 Huber, C, 239, 294 Huennekens, F., 42 Hughes, C. W., 123 Hutchinson, G. E,, 394, 411 Ichikawa, A., 385 lida, T. T., 378 Imai, H., 372, 378 Imamura, T., 336, 378 Inoue, N., 345, 378 423 Ishikawa, C., 353, 372, 378, 379 Ishikawa, O., 385 Jacobsen, C. F., 295, 297 Jenkin, P. M., 403, 411 Jockusch, H., 236, 239 Johnson, F. H., 23, 24, 130, 154, 155, 210, 224, 229, 238, 239, 240, 244-246, 265-287, 290, 293-295, 297 Johnson, M. W., 389, 395, 413 Josserand, M., 379 Jowett, M., 281, 294 Kaiser, K. H., 94, 97, 123 Kajiyama, E., 379 Kamohara, T., 372, 379 Kanda, S., 130, 155, 365, 379 Kasha, M., 25, 26, 27, 28, 29, 30, 31, 34, 36, 37, 39, 41, 42, 43, 44, 45 Kato, Kojiro, 3, 14, 24, 349, 364- 366, 379, 380, 383 Kautsky, Hans, 94, 97, 102, 103, 106, 123 Kauzmann, W., 158, 251, 295, 297 Kawamura, S., 338, 339, 380 Kawamura, T., 344, 380 Kawanaka, T., 380 Keen, J., 401, 402, 411 Ketchum, B. H., 401, 402, 408, 411 Kiesenwetter, H., 380 Kimura, 336 Kincaid, J. F., 246 King, G. S., 5, 324, 325, 327, 331, 393,403,411 Kipnis, D. M., 210, 239 Kishinouye, K., 380 Kishitani, T., 336, 367, 380 Kluyver, A. J., Ill, 123, 130, 155 Kobayashi, Y., 336, 339, 380 Koffler, H., 278, 294 Kofoid, C. A., 342, 381, 387, 388, 392, 396, 411, 412 Komai, T., 344, 381 Komarek, J., 8 Kominami, K., 341, 381 424 AUTHOR INDEX Korr, I. M., 9, 130, 132, 155, 210, 211, 239 Koski, V. M., 63, 72, 74 Kozukue, H., 336, 337, 381 Krasnovski, A. A., 33, 42 Krogh, A., 405, 411 Kubista, V., 8 Kumagai, N., 337, 377 Kunkel, H. G., 137, 155 Kuroda, S., 372, 381 Kuroki, S., 348, 381 Langmuir, L., 407, 411 Lauffer, M. A., 281, 295 Legros, C, 300, 319 Lewin, L, 271, 294 Lewis, G. N., 27-29, 42, 43 Linderstr0m-Lang, K. U., 295, 297 Livingston, R., 32, 33, 35, 38, 42, 66,74 Loomis, W. E., 42 Lorenz, P. B., 129, 154 Lowry, O. H., 140, 150, 152, 155 Lucas, C. E., 394, 412 Lumry, Rufus, 75 Lund, E. J., 399, 401, 412 Macbeth, A. K., 115, 123 MacdowaU, F. D. H., 51, 74 Macrae, R., 13 McArdle, J., 32, 42 McConneU, H., 31, 42 McDowell, C. A., 236, 239 McElroy, W. D., 6, 7, 20, 24, 116, 123, 127, 146, 155, 156, 161-163, 166-168, 170, 172-180, 182, 183, 186-188, 196-199, 207, 208, 210, 214, 235, 239, 257-259, 261-264, 266, 284, 295 McGeer, K., 278, 294 McLaren, A. D., 273, 296 McMurry, H. L., 30, 42 Magee, J. L., 274, 293 Majima, R., 336, 381, 385 MalisofiF, W. M., 97 Marshall, S. M., 389, 390, 392, 399, 412 Marsland, D. A., 239, 267, 272, 274, 286, 287, 293, 294 Martin, G. W., 399, 403, 412 Mason, H. S., 44, 125, 131-133, 149, 155, 157, 159, 297, 320 Massart, J., 300, 318 Matsubara, H., 380 Matsubara, K., 368, 372, 381 Matsumura, M., 381 Mayer, J. E., 248 Mead, A. D., 399, 412 Menon, M. A. S., 398, 401, 412 Michaelis, G. A., 412 MiUer, H., 266, 295 Miller, J. R., 42 Miller, V. K., 275, 283, 293 Millikan, G., 238 Milner, H. W., 405 Molisch, H., 337, 381 Moore, A. R., 10, 20, 302, 304, 309, 318, 324, 332 Moore, S., 135, 150-153, 155, 156 Morrison, T. F., 79, 97, 362, 381 Motschulsky, 381 Mott, N. F., 253, 255 Mulliken, R. S., 30, 39, 42 Munk, W. H., 397, 412 Murakami, S., 366, 381 Nakamura, H., 336, 372, 382, 385 Nakasawa, 372 Nakazawa, K., 382 Nayudu, M. R., 398, 401, 405, 411 Neitzke, O., 123 Nelson, T. C, 399, 403, 412 Nicol, J. A. C, 5, 13, 24, 299, 305, 306, 311-314, 318, 321, 325, 332 Ninomiya, R., 336, 382 Nishikawa, T., 342, 382, 383, 398, 403, 405, 412 Nishio, M., 385 Nordli, E., 390, 391, 412 Oaki, R., 345, 378 Odawara, X., 336, 337, 382 Ogihara, Y., 382 _ Ohshima, H., 374, 382 AUTHOR INDEX 425 Okada, I., 382, 383 Okada, Yo. K., 344, 354, 360, 364, 367, 382, 383, 385 Okamura, K., 342, 383 Okuyama, M., 378 Oppenheimer, J. R., 50 Orr, A. P., 390, 399, 412 Osborn, C. M., 308, 319 Ostroumoff, A., 344, 383 Owens, O. v. H., 72, 74 Panceri, P., 301, 302, 304, 316, 319 Pantin, C. F. A., 302, 319 Pappas, I., 389, 391, 393, 410 Parker, A. S., 85, 97 Parker, G. H., 12, 300, 301, 308, 319. 324, 328, 332 Pearman, F. H., 92, 97 Pearsall, W. H., 394, 412 Peters, A. W., 301, 304, 309. 319 Peters, N., 389. 397, 412 Phillips. J. B., 399, 409 Pieper, G., 236, 240 Pierantoni, U., 2 Pintner, I. J., 393. 412 Piatt. J. R., 27, 30, 33, 42 Polissar, M. J., 24, 239, 244-246, 269, 270, 272-274, 276, 277, 279-282, 286, 290, 294 Potts, W. J., Jr., 39, 42 Pratje, A., 405, 412 Price, J. R., 123 Provasoli, L., 393, 412 Prudhomme, R. B., 77, 97 Rabinowitch, E. I., 51, 56, 66, 74 Rainwater, C. S., 163, 198 Rakestraw. N. W., 403, 413 Randall, R. J., 155 Ray, D. L., 306, 319 Redfield, A. C., 408, 411 Reman, G. G., 51, 58, 74 Rexford, D. R., 130, 155 Rice, T. R., 394, 413 Richards, F. A., 416, 420 Riley, G. A., 397, 408, 412, 413 Ritchie, G. S., 398, 413 Robin, C., 300, 319 Robinson, F. A., 150, 155 Roothaan, C. G. J., 39, 42 Rosebrough, N. J., 155 Rossavik, E., 391, 410 Russell, F. S., 395, 413 Rvdon, H. N., 149, 155 Ryther, J. H., 387 Sasaki, M., 345, 356, 378, 383 Schlegel, F. McK., 275, 295 Schuiling. A. L., 117, 123, 210, 240 Shibata, S., 336, 337, 383 Shima, G., 383 Shoji, R., 354, 383 Shoup, C. S., 210, 211, 235, 239, 240 Simpson, R., 278, 294, 295, 297 Sinoto, Y., 385 Sites, J. R., 199, 208 Slobodkin, L. B., 401, 413 Smith, F. G. Walton, 411 Smith, H. M., 362, 383 Smith, H. P., 403, 413 Smith, J. H. G., 63, 74 Smith, P. W. G., 149, 155 Sneddon, I. N., 255 Snell, P. A., 5, 325, 329, 330, 332 Sonnenfeld, V., 7, 24, 198, 239, 258, 261. 262, 264, 295 Spoehr, H. A., 405, 413 Spruit, C. J. P., 77, 85, 93, 99, HI. 115-117, 123, 130, 155, 210, 240 Spruit-van der Burg. A.. 10, 11, 98, 99, 210, 240, 252, 255 Stanbury, F. A., 403, 413 Stearn, A. E., 271, 295 Stearn, J. R., 280, 281, 293 Steblav, R., 239, 294 Steche, O., 307, 319 Stein, W. H., 135, 150, 152, 153, 155, 156 Stommel, H., 413 Strehler, B. L., 3, 7, 20, 24, 47, 50, 116, 123, 131, 146, 156, 166, 167, 170. 172, 198, 199, 207-211, 213, 216, 221, 223, 224, 226, 229, 233, 235, 238-241, 244, 246, 258, 262, 426 264, 269, 283-285, 287, 290, 293, 295 Stress, F. H., 95, 98 Sugino, H., 383 Sumner, J. B., 131, 156 Suter, H., 350, 383 Sverdrup, H. U., 389, 395, 413 Sveshnikov, B. J., 78, 98 Sweeney, B. M., 393, 413-415 Svvezy, O., 387, 388, 392, 411, 412 Takagi, S., 383 Takakuvva, Y., 360, 383 Takase, M., 336, 337, 377, 383, 384 Takino, M., 338, 384 Tanaka, S., 372, 374, 384 Terao, A., 374, 384 Thomas, J. J., 236, 239 Thompson, T. G., 420 Tisehus, A., 137, 155 Tokioka, T., 346, 384 Torrey, H. B., 399, 401, 413 Totter, J. R., 172, 198 Towner, G. H., 74 Tsuji, F. I., 7, 23, 127, 147, 156, 158 Uchida, T., 344, 384 Ueno, M., 344, 384 Usami, S., 384 van der Burg, A., 104, 109, 123, 124, 130, 155 van der Kerk, G. J. M., Ill, 123, 124, 130, 155, 156 Van Norman, R., 51, 74 van Schouwenburg, K. L., 108, 110, 113, 114, 116, 123, 124, 210, 239, 240 AUTHOR INDEX Vermeulen, D., 51, 58, 74 Vinogradov, A. P., 391, 413 Virgin, Hemming, 63, 74 Warren, G. H., 268, 296 Wassink, E. C, 51, 58, 74, 384 Watanabe, H., 384 Watase, S., 354, 384, 385 Watson, W. F., 32, 42, 66, 74 Weber, G., 148, 156 Weill, R., 344, 384 Wenig, K., 8, 20 Werbin, H., 273, 296 Whitelegge, T., 397, 398, 401, 403, 413 Wilhelmsen, P. C, 75 Williams, R.H., 411 Williams, R. W., 270 Wilson, P. W., 278, 294 Winzor, F. L., 123 Wittig, G., 236, 240 Woodcock, A. H., 397, 413 Wootton, D. M., 4, 14, 23, 317 Yamada, T., 385 Yamaguchi, E., 385 Yasaki, Y., 336, 367, 368, 385 Yatsu, N., 301, 319, 385 Yokoseki, M., 384 Yoshida, K., 385 Young, J. Z., 306, 319 Young, V. K., 52, 66, 69-71, 74, 328, 332 Zellner, C. N., 77, 92-94, 96, 98 Zeuthen, E., 395, 411 ZoBell, C. E., 275, 283, 294 Zscheile, F. P., 51, 57, 61, 74 Genera and Species Index Abralia japonica, 356 Abraliopsis, 384 Abyla, 345 Acanthodesia, 3; serrata, 366, 380 acclinidens (Cyclothone) , 318 Acholoe, 17; astericola, 305 Achromohacter, 16; fischeri, 24, 78, 216, 218, 225, 229, 258, 268, 270- 273, 284-286, 287, 291, 295; har- veyi, 269, 272, 273 Acropoma, 19, 307, 336, 368, 369, 372, 381; hanedai, 336, 368, 369; iaponicum, 336, 368, 369, 385 Aequarea, 9, 16, 345 Agalma, 16 Aglaophenia, 16 Agwtis, 18 alfonsi (HeteTocarpus), 359 Amphidinium, 391; fusiforme, 399 Amphiura, 19, 309; ^-anJat, 366 a;w/w (Pyrocoelia) , 378 Afioma/op5, 1, 2, 19, 307, 318, 370, 371, 373, 374, 377; katoptron, 319, 375, 376 Apogon, 19 Appendicularia, 19 applanatus (Ganoderma), 365 Arachnocampa, 18 arbutifolia (Photinia), 63 Archaeperidinium, 375 Arctic, 3 argentiim ( Leio gnathus) , S69 Argyropelecus, 4, 20, 306; hemigym- nus, 318 Armillaria mellea, 109, 110, 338, 376 A.scefe5, 383 astericola (Acholoe), SOS Asteroscopus, 18 Astronesthes ijimai, 374 atlanticitm (Pyrosoma) , S46 atripennis (Pyrocoelia), 360 Atyphella carolinae, 361 Bacillus subtilis, 275, 283, 294 B. co^i, 158 bahamemsis (Ptychodera) , SIO Balanoglossus, 19, 317, 366; carno- sus, 366; misakiensis, 366 balitca (Exiwiella), 391 bambxisa (Mycena), 338, 340 Bassozeius, 3 Beroe, 9, 16, 304, 311, 313, 314, 345, 346 birostrata (Sepiola) , 354, 380 bleekeri ( Loligo ) , 383 BoZinfl, 324 Bolinopsis, 16 brevilabiatiis (Orphaneus) , 359 brevis (Gxjmnodinium) , 399, 400, 404, 408, 410 Brisingia, 3 bucephala (Phyllirrhoe) , 349 Caenis, 18 californicum (Gyrodinium) , 393 Calinga ornata, 349 Calizonella, 17 Callitetithis, 17, 356 Campanularia, 16 Camptonotits, 19 carnosus (Balanoglossus) , 366 carolinae (Atyphella), 361 casei (Lactobacillus) , 259 390, 427 428 catonella (Gonyatilax) , S99 Cavernularia, 5, 16, 301, 311; habe- reri, 378; haheri, 318, 377; ohesa, 378 Cavolinia, 3 CentrosyUium, 19 Ceratia, 19, 412 Ceratium, 16, 383, 387, 389, 390, 395, 397; furca, 390; fusus, 344, 390, 391, 404; trichoceros, 396; tripos, 390, 391, 404 Ceratoisis, 16 Ceroplatus, 18, 364; nipponicus, 364; testaceus, 364 Cestiim, 345, 346 Chaetoptems, 5, 13, 17, 300, 309, 311, 312, 314, 315, 318, 323, 325; variopedatus, 24, 311, 317, 318, 332, 347, 377 Chauliodus, 20 Chehjosoma siboja, 365 Chimaera, 3 Chironomus, 18 Chiroteuthis, 17, 356; mperator, 356 ChJamydomonas, 344 Chlarella, 47, 50, 51, 52, 405, 413; pyrenoidosa, 47 chlorophos (Mycena), 339, 340 Chromatium, 51 Ciona, 3 Cirratulus, 17 Cirrothauma, 3 citrineUa (Mycena) , S42 Clio, 3 closterium, minutissima (Nitzschia) , 67 closterium ( Nitzschia ) , 4 13 C/y/ia linearis, 380 Coelorhynchus, 19 co/i (B.)> 158 co^i (Escherichia), 271, 294 Collosphaera, 16 Collozoum, 9, 16 Colossendeis, 3 compressa (Xiphocaridina) , 359, 385 Conchoecia, 17 Corynocephalus, 17 GENERA AND SPECIES INDEX Coscinodiscus excentricus, 411 couesi (Cnjtosparus),S74 Cranchia, 17 Creseis, 3 cruentum (Porphyridium) , 68, 69 Crytosparus couesi, 374 cyanophos (Mycena), 339 Cyclosalpa pinnata var. polae, 346 Cyclothone, 306; acclinidens, 318 Cyclotrichium meunieri, 407 Cypridina, 6, 8, 9, 11, 14, 17, 23, 110, 111, 114, 116, 123, 125, 127- 133, 135, 137-140, 142, 146, 147, 153-158, 164, 196, 197, 224, 239, 243, 261, 263, 264, 323, 356, 357, 419; hilgendorfii, 24, 127, 154, 356, 358, 377, 383, 384, 385; noc- fi7uca, 357, 358 J^cfyh/s (P;ioZfl5),128,350 daishogunensis (Mycena), 339 diadema (Lycoteuthis) , 20 Diakia striata, 350 Diaphus, 20 Dictyopanus, 339, 376; folucolus, 339; gloeocyst, 339, 341; lumines- cens sp. nov., 339; pusillus, 339 Dioptoma, 19 Diphes, 345 Diphyes, 16 Diplocladon, 19; Hasseltii, 364 Doliolum, 384 Dolopichthys, 19 Drosera tvhittakeri, 123 d»m[?flr (ViZ;no), 359 Dyo^w, 17, 349; ^friafa, 376 Echiosioma, 4, 20 eJu/w (Loligo), 354 Eisenia, 13, 17; foetida, 8; submon- tana, 8 Electra, 3 Eledonella, 3 elongatum (Leiognathus) , 369 Emplectonema, 17; kandai, 365, 379 enopla (OdontosylUs) , 6 equulus (Leiognathus) , 369 Escherichia coli, 271, 294 GENERA AND SPECIES INDEX 429 Etmopterus, 19 Eucharis, 9, 16 Euphausia, 18 Euprymna, 321; morsei, 354, 380 excentricus (Coscinodiscus) , 411 Exocetus, 3 Exuviella balitca, 391 Favites virens, 350 fischeri ( Achromobacter) , 24, 78, 216, 218, 225, 229, 258, 268, 270-273, 284-287, 291, 295 fiscJieri (Photobacteriiim) , 109, 110, 113, 114 favida (Omphalia) , 110 Flustra, 3 foetida (Eisenia), 8 fohicohis (Dictyopanus), 339 Fulgora, 14, 18 fumosa (Pyrocoelia) , 360, 361 /j/rca (Ceratium) , 390 fusiforme ( Amphidinium) , 399 /usu5 (Ceratium), 344, 390, 391, 404 gc/ea (Tonna), 351 Ganoderma applanatus, 365 Gastrosaccus, 18 Gazza, 19, 308, 369; minuta, 369 Geophilus, 18 Glenodinium rtibrum, 398 gloeocyst (Dictyopantis) , 339, 341 Gnathophausia, 18 Goniastrea parvistella, 350 Gonostama, 20 Gonyaulax, 16, 308, 382, 399, 410, 412, 418, 421; catonella, 399; wo- njtoa, 404; polyedra, 399, 404, 414-419; polygramma, 398 Gorhami { Luciola), S78 gracile (Prorocentrum) , S92 grandis (Rocellaria) , 350, 376 Grantia, 16 Gryllotalpa, 18 Gymnodinium, 16, 398, 399, 408, 413; fcreut^, 399, 400, 404, 408, 410; sangiiineum, 398; simplex, 392, 393; splendens, 413, 414 Gyrodinium, 415; californicum, 390, 393 habereri (Cavernularia) , 378 Imberi {Cavernularia) , 318, 377 Halistaura, 16 Halosaurus, 3 hanedai (Acropoma), 336, 368, 369 Hanedai (Polyporus), 339 Hanedai (Poromycena) , 339 Harmothoe, 17 Harpodon, 21; neheretis, 377 Hasseltii (Diplocadon) , 364 hemigymnus (Argyropelecus) , 318 Heterocarpus, 6, 18; alfotisi, 359; sibogae, 356, 359 Heterochaeta, 18 Heterocirrus, 17 Heteroteuthis, 17, 354 hilgendorfii (Cypridina), 24, 127, 154, 356, 358, 377, 383-385 Hippopodius, 16 Histioteuthis, 17 Hoplophorus, 18 t'/injat (Astronesthes) , 374 Ileodictyon, 3 illuminans (Mycena), 339, 378 imperator (Chiroteuthis) , S56 imperialis { Plocam^phorus) , S49 Inioteuthis, 17 investigatores (Onuphis) , S47 Ipnops, 3 Iridomyrex, 19 japonicum (Acropoma), 336, 368, 369, 385 japonicus (Lampteromyces) , 338, 339 japonicus (Maurolicus) , 379 japonicus (Mesoclmetopterus) , 347, 376 japonicus (Monocentris), 367, 382, 385 japonicus (Physiculus) , 367, 380 japonicus (Pleurotus) , 380 430 Kalchbrennera, 3 Kaloplocamus, 17, 20, 349; ramosum, 349, 379 kandai (Amphiura), 366 kandai (Emplectonema), 365, 379 katoptron (Anomahps) , 319, 375, 376 Kollikeri (Renilla), 20 Krijptophaneron, 19 lacineata (Porphyra) , 52 Lactobacillus casei, 259 Laemargus, 19 laevis (Malacocephahis) , 376 Lampadena, 20 Lampanyctus, 306, 307; leucopsarus, 319 Lampetia pancerina, 381 Lamprogrammus, 3 Lampteromyces japonicus, 338, 339 Lampyris, 19; noctiluca, 310 Lamyctes, 383 lateralis (Luciola), 360, 361, 378 Lfli/a, 17, 349; neritoides, 6, 350 leideyi (M nemiapsis), S2l Leiognathus, 19, 307, 369, 372; argentum, 369; elongatum, 369; equulus, 369; rivitlatiim, 369, 370 Leiicicorus, 3 Leuckartia, 18 leucopsarus (Lampanyctus) , 319 leucoptera (Stoloteuthis) , 354, 355 linearis (Clytia), 380 Linophryne, 19 Loligo bleekeri, 383; edulis, 354 Lotella, 19 Lucidina biplagiata, 360 Luciola, 19; cruciata, 360, 361, 363, 378, 380, 383; elongatum, 369; Gorhami, 378; lateralis, 360, 361, 378; parvula, 360, 361, 378 luminescens (Dictyopanus) , 339 Luminodesmus, 4, 14, 18, 299; se- quoiae, 23, 317 luminosa {Sijmplectoteuthis) , 356 lunaillustris (Pletirotus) , 338, 340 lux-coeli (Mycena), 339, 340 GENERA AND SPECIES INDEX Lycoteuthis, 17; diadema, 20 Macrurus, 3 Malacocephahis, 2, 19, 372; ^eow, 376 Malthopsis, 3 Mammestra, 18 manipularis (Poromycena), 339 Marasmius phosphorus, 342 matsushimemsis (Pontodrilus) , 20, 348, 377, 379 Maurolicus, 4, 20; japonicus, 379 Megaligia, 18, 337 Meleagroteuthis separata, 356 me/Zea {Armillaria), 109, 110, 338, 376 membranacea (Membranipora) , 366 Membranipora, 3; membranacea, 366 menetriesi {Vesta), 363 Mesochaetopterus, 17; japonicus, 347, 376 meunieri (Cyclotrichium) , 407 micans (Prorocentrum) , 391, 392, 399, 410, 414 Micrococcus physiculus, 367 microillumina (Mycena), 342 Microscolex, 17; phosphoreus, 348, 382, 385 Microspira phosphoreum, 336, 359, 381, 385 mt7jam (Noctiluca), 343, 378, 379, 383, 398, 405, 412 minuta (Gazza), S69 misakiensis (Balanoglossus) , 366 A/ixonMS, 3 Mnemiopsis, 9, 10, 16, 23, 318, 323, 324, 332; ?d£Zeyi, 321 monilata (Gonyaulax) , 404 Monocelis, 3 Monocentris, 19, 372, 374; japoni- cus, 367, 382, 385 morsei (Euprymna), 354, 380 mucronata (Rhizophora) , S42 Mycena, 339, 342, 376; bambusa, 338, 340; chlorophos, 339, 340; citrinella var. illumina, 342; cy- anophos, 339; daishogunensis, 339; illuminans, 339, 378; lux-coeli. GENERA AND SPECIES INDEX 431 339, 340; microillnmina, 342; noc- tilucens, 339; phosphora, 342; photogena, 341; pohjgramma, 110; pruinnosoviscida var. rabaulensis, 339, 341; pseudostylobates, 339; rorida, 339, 341, 379; subhtcens, 339; vapensis, 342 Myctophum, 20 My sis, 18 Mxjxosphaera, 16 Neanura, 18 nehereus (Harpodon), 377 Nematoscelis, 18 Neoscopelus, 20 Nereis, 3 neritoides (Latia), 6 nipponicus (Ceroplatus) , 364 Nitzschia, 67, 68; closterium, 413; closterium, minutissima, 67 Noctiluca, 11, 12, 16, 300, 309, 317, 318, 323, 343, 344, 379, 384, 387, 398, 399, 405, 406, 411; miliaris, 343, 378, 379, 383, 398, 405, 412; scintillans, 378 noctiluca (Cypridina), 357, 358 noctiluca (Lampyris), 310 noctiluca (Pelagia), 302, 303, 318 noctilucens {Mycena),SS9 noctilucens (Pleurotiis) , 340 notatus (Porichthys) , 318 Nyctiphanes, 18 Obe//a, 16 obesa {Cavernularia),S78 Octochaetus, 17 Odontosyllis, 13, 17, 21; enopla, 6 Omphalia, 16; fiavida, 110 Oncaea, 18, 20 Onuphis, 17, 347; investigator is, 347 Onychiurus, 18 Ophioscolex, 19 Opiopsila, 19 Orchestia, 18 ornata (Calinga), 349 Ornithocercus, 395; splendidus, 396; steinii, 396 Orplianeiis brcvilabiatus, 359 Orj/fl, 18, 375 ostrcatus (PJeurotus) , SS9 palpebratus (PliotoblepJiaron) , 319 pancerina (Lanipetia), ^81 Paniis, 16; stipticus, 13 Parapronoe, 18, 20 Parehippus, 3 parmatus (Stylarioides) , 347 parvistella (Goniastrea) , 350 parvula (Luciola), 360, 361, 378 Pcgea confoederata, 346 Pe/flg/a, 9, 16, 309, 311, 324, 325, 345; noctiluca, 302, 303, 318 Pennatula, 16, 301 pennsylvanica (Photuris), 327, 331, 332 perescelida { Planaxis) , S51 Peridinium, 344, 375, 387, 392, 399, 412; triquetrum, 390, 391, 393, 399, 404, 410; trochoideum, 391 Phengodes, 4, 19, 299, 363 Phialidum, 16 Phillirrhoe, 349 P/io/a5, 13, 17, 263, 349, 354; dacft/- /t/s, 128, 350 phosphora (Mycena), S42 phosphoreum (Microspira) , 336 359, 381, 385 phosphoreum (Photobacterium) , 78, 109-114, 116, 123, 269, 272-274, 276, 277, 279, 280, 282, 286, 287 phosphoreus (Microscolex) , 348, 382, 385 phosphoreus (Spirobolellus) , 360 phosphorus (M arasmius) , S42 Photinia arbutifolia, 63 Photinus, 19; pyralis, 7, 111, 116, 162, 316 Photobacterium, 16; jxscheri, 109, 110, 113, 114; phosphoreum, 78, 109-114, 116, 123, 269, 272-274, 276, 277, 279, 280, 282, 286, 287; splendidum, 109, 110, 113, 114, 266, 269; splendidum harveyi, 269 Photohlepharon, 1, 2, 19, 307, 318, 377; palpebratus, 319 432 photogetia (Mycena), 341 Photuris, 19 Photiiris pennsylvanica, 327, 331, 332 Phrixothrix, 9, 15, 19, 315 PhyUirhoe, 17; bticephala, 349 Physiculiis, 19, 372; iaponicus, 367, 380 physiculus (Micrococcus) , 367 pinnata (C y closalpa) , S46 Planaxis, 17, 353; perescelida, 351; uirafus, 351-353 Platyura, 18 Plesiopenaeus, 18 Pleurobranchia, 16 Pleuromma, 18 Pleiirotus, 16, 339; japonictis, 380; lunaillustris, 338, 340; noctilucens, 340; ostreatus, 339 Plocatnophorus, 5, 17, 349; imperi- alis, 349; fi/esn, 349, 378, 383 Pohjcirrus, 17, 348 pohjedra {Gonyaulax) , 399, 404, 414-419 polygramma (Gonyaulax) , 398 poly gramma (Mycena), 110 Pohjipnus, 20, 23, 374, 377 Pohjnoe, 17,311,313 Polyopthalmus, 3 Pohjporm, 16; Hanedai, 339 Pontodriliis, 17; jnatsushimensis, 20, 348, 377, 379 Porcellio, 18 Porj'a vaporaria, 364 Porichthys, 4, 20, 306, 307, 311, 321; notatus, 318 Porites tenuis, 350 Poromycena, 339; Hanedai, 339; ma- nipidaris, 339 Porphyra lacineata, 52 Porphyridium, 69, 70; cruentum, 68, 69 Portunus tritubercidattis, 382 Praya, 345 prehensilis (Sergestes), 356, 358, 382, 384 Primnoisis, 16 Prorocentrum, 387; gracile, 392; mi- GENERA AND SPECIES INDEX carw, 391, 392, 399, 404, 410, 414; triquetnim, 392 pruinosoviscida (Mycena), 339, 341 pseudostylobates (Mycena), 339 Psilocladus variolosus, 360 Pteroeides, 16 Pterotrachea, 3 Ptychodera, 19, 20; bahamensis, 310 pusillus (Dictyopanus) , SS9 pyralis (Photinus), 7, 11, 116, 162, 316 pyrenoidosa (Chlorella) , 47 Pyrocoelia analis, 378; atripennis, 360; fitmosa, 360, 361; m/c, 377, 378, 383 Pyrocypris, 14, 17 Pyrocystis, 16, 387 Pyrophorus, 19, 161, 327 Py/womc, 19, 311, 313, 316, 319, 344, 346, 347, 358; atlanticum, 346; verticillatum, 346 Pyrrophyta, 410 ramosum (Kaloplocamus) , 349, 379 reniformis (Renilla), 20 Ren/7/a, 16, 20, 301, 302, 309, 311, 313, 314, 317, 319, 323-325, 327, 328; Kollikeri, 20; reniformis, 20 Rhizobium, 295; fn/o//j, 278 Rhizaphora tnucronata, 342 rividatitm (Leiognathus) , 369, 370 Rocellaria, 17, 349, 350; grandis, 350, 376 Rondeletia, 17 roric/a (Mycena), 339, 341, 379 nibrmn (Glenodinium) , 398 n//fl (Pyrocoelia), 377, 378, 383 rugosus (Trigoniulus),SS7 Riwettus, 3 Saccopharanx, 20 Sagitta, 3 Sa/pa, 19, 344, 346, 347, 384 sanguineum (Gymnodinium) , 398 scaber (Uranoscopus), 319 scintillans (Noctiluca), 378 scintillans (Watasenia), 13, 308, 345, 354, 377-380, 383, 385 GENERA AND SPECIES INDEX 433 Scohoplanes, 18 Scypholanceola, 18 Secutor, 19, 308, 369 separata (Meleagroteuthis) , 356 Sepiola, 17; birostrata, 354, 380 sequoiae {Ltiminodesmus) , 23, 317 Sergestes, 18; prehensilis, 356, 358, 382, 384 serrata (Acanthodesia), 366 Shiia Sieboldii, 365 sibogae (Heterocarpiis), 356, 359 siboja (Chehjosoma) , 365 Sieboldii (Shiia), S65 simplex (Gytnnodinium) , 392, 393 Siriella, 18 Sphaerozoum, 16 Spirobolellus, 18, 383; phosphoreus, 360 Sp/n//a, 17, 20, 299 splendens (Gymnodinitim) , 413, 414 splendidum (Photobacterium) , 109, 110, 113, 114,266,269 splendidum harveyi (Photobacte- rium), 269 splendidus (Ornithocercus), S96 steinii (Ornithocercus) , 396 Stigmatogaster, 18 stipticus (Panus), 13 Stoloteuthis, 17, 20, 354; leucoptera, 354, 355 Stomias, 20 Streetsia, 18 striata (Diakia), 350 striata (Dyakia), S76 Styiola, 3 Stylarioides parmatus, 347 sublucens ( Mycena ) , 339 submontana (Eisenia),8 subtilis (Bacillus), 275, 283, 294 Symplectoteuthis luminosa, 356 Systellaspis, 6, 18, 21 Talitrus, 18 Teloganodes, 18 tenuis (Porites), 350 testaceus (Ceroplatus) , 364 ref/i|/5, 349 Thalassicola, 9, 16 Thaumatolampas, 315 Thelepus, 17, 348 Thetys vagina, 346 fi/esu' (Plocamophorus) , 349, 378, 383 Tomopteris, 17, 348 Tonfw, 3, 349; ga/ea, 351 trichoceros (Ceratium), 396 f rj/o/// ( Rhizobium ) , 278 Trigoniulus, 18; rugosiis, 337 m'pos (CeraiiJim), 390, 391, 404 Triposolenia, 395, 411 triquetrum (Peridinium) , 390, 391, 393, 399, 404, 410 triquetrum ( Prorocentrum ) , 392 trituberculatus (Portunus) , 382 trochoideum (Peridinium) , 391 L/Zufl, 61 Uranoscopus scaber, 319 vagina (Thetys), 346 Vampyroteuthis, 17 vapensis (Mycena) , ^42 vaporaria (Poria), 364 variolosus (Psilocladus) , 360 variopedatus (Clxaetoptenis) , 24, 311,317,318,332,347,377 verticillatum (Pyrosoma) , S46 Vesta menetriesi, 363 Viforfo, 16; dumbar, 359; j/flw/cii, 336 Vicifl, 72, 74 viratus (Planaxis), 351-353 virens (Favites), 350 Wafa^enifl, 17; scintillans, 13, 308, 345, 354, 377-380, 383, 385 whittakeri (Drosera), 123 Xenodermichthys, 3 Xiphocaridina compressa, 359, 385 Xylaria, 3 Xyphocaridina, 18 Yare/fe, 20, 23 Yarrella, 374, 377 yasakii (Vibrio), 336 Subject Index Absorption mirror image of fluorescence, 100 Absorption spectra of anthraquinones, 118-119 bacterial luciferin, 113, 114 chlorophyll, magnesium role, 33 Cypridina luciferin, 130, 132, 138-146 effect of pH on, 140-146 effect of oxygen on, 140-145 1 ,4-dihydroxynapthyl-2-hydroxy- methylketone, 112, 116, 118 DPD's, 107 firefly luciferin, 166, 200-202 methyl acridone, 102 Photiniis pyralis, 116 phycoerythrin, 56 riboflavin, 202 Absorption spectrum relation to solvent, 31 relation to structure, 118-121, 125-126 Absorption spectrum, UV, of Cypridina luciferin, 138-146, 151 firefly luciferin, 200-201 naphthoquinones, 118-121 Acetylcholine, 330 in firefly luminescence, 194, 330 Acropomatidae, 19, 336, 368 Action potentials, 320, 328 Action spectrum, for fluorescence in Nitzschia, 68 quenching of bacterial light, 111- 114 Activation energy, of bact. luminescence in vitro, 220- 221, 223 firefly luminescence, 167 luminol luminescence, 85 oxvgen use in chemiluminescence, 90-91 ADP, effect on bact. luminescence in vitro, 212 fireflv luminescence, 164, 171, 'l72, 175, 176 Adrenalin, 306-308 Africa, 402 Aigae Bay, 348 Akashio, 344 Alciopidae, 17 Alcohol, effect on bacterial luminescence, 279 bact. luminescence in vitro, 291 globulin denaturation, 282 Aldehyde, dodecyl, in bact. luminescence in vitro, 262 Aldehydes, effects on bact. luminescence in vitro, 216- 218, 223, 224, 234-238, 259, 284, 285, 288, 291-292 respiration of bact. extracts, 224, 225 Algae, 43 blue green, 52 red, 51, 52, 68-72 chlorophylls in, 68-72 phycobflins in, 68-72 Alpha luciferin, 149 Amami-Oshima Island, 360 Amboina, 339 L-Amino oxidase, 242 Amphipoda, 18 Annelida, 347-349 Anomalopidae, 2, 19 435 436 Anthracene, singlet-triplet split, 34 Antibiotics, effects on luminous bacteria, 336, 337 Aomori Bay, 365 Api-Api, 363 Arabian Sea, 347 Araboflavin, 242 Aranea, 3 Ariophantidae, 17 Arumizu, 360 Asamushi Marine Station, 365 Ascidiacea, 3 Ascomycetes, 3 Asteroidea, 3 A tube, 78-79 ATP assay by firefly extracts, 172 ATP, eflFect on bact. luminescence in vitro, 212 firefly luminescence, 162, 164, 'l65, 168, 170-173, 175-182, 186-197, 263 luminescence, various organisms, 16-20 ATP reaction, 6, 7, 349, 353, 365, 366 Australia, 398 Autotomy, of dinoflagellates, 396 Azide, eflFect on bact. luminescence in vitro, 223 firefly luminescence, 168 Bacteria, luminous mutants, 10 pathogenic, 1, 359 saprophytic, 1, 16, 337 symbiotic, 1, 2, 299, 313, 335, 336, 353, 354, 367, 368, 372 Bacteria, photosynthetic, 60 Bacterial luciferase, see Luciferase, bacterial Bacterial luciferin, see Luciferin, bacterial Bacterial luminescence, see also Luciferin, bacterial; lucifer- ase, bacterial; emission spec- tra, etc. SUBJECT INDEX alcohol inhibition of, 279 change between steady states, 286 during growth, 266-269 extraction of, 210-211 hydrostatic pressure on, 272-279 inactivation by UV, 210 irreversible denaturation of, 275, 276 photochemical inactivation, 112- 114 pressure-temperature eflFects on, 210 recovery after UV, 113 relation to oxygen, 210 relation to respiration, 210 reversible denaturation, 269-279 sulfanilamide inhibition, 281-282 temp. -pressure relations, 272-276 urethan inhibition of, 278-283 volume change of activation, 272- 274 Bacterial luminescence in vitro activation energy, 220-221 alcohol inhibition of, 291 change between steady states, 230-232, 284-292 comparative biochemistry, 223- 224 diffusible components of, 210- 218 DPN in, 212-216, 284, 285, 288- 290 effects of aldehydes, 216-218, 259, 261, 262, 264, 284, 285, 288, 291, 292 effects of inhibitors, 221-223, 260, 262, 291 effects of KCF (pal. aid.), 228, 230 effects of malate, 226 effects of malic acid, 212-213 effects of naphthoquinones, 212, 223 effects of riboflavin, 223, 241-242, 260 effects of various substances on, 212-213 effects of UV, 221-222 SUBJECT INDEX 437 emission spectrum, 218-219 energy relations in, 235-237 hydrogen peroxide in, 211, 234- 238, 244, 246, 259 hydrostatic pressure effects, 284- 292 limiting components, 226 mechanism of reactions, 233-238, 241-248, 251-255 pH dependence, 220, 259 pressure-temperature relations, 284-292 relation to oxygen tension, 225 reversible denaturation, 287 ")2 rise time," 224-229 role of aldehydes, 234-238 scheme of reactions, 235, 290 temperature dependence, 221 volume changes of activation, 288 Bacteriochlorophyll, 5 1 fluorescence spectrum, 58—59 green pigment accompanying, 59- 60 Banana squash, 64, 65 Banda Islands, 2 Barley, 61, 63, 64, 66 Basidiomycetes, 16 Batrachoididae, 20 Bay of Aug, 398 Beef, 337 Behavior, comparative, 333 Bengal, 342 Benzimidazole, 169 Benzothiazole, 169 Benzotriazole, 169 Benzoxazole, 169 Bermuda, 6, 20 Beryllium, 178 Beta lactoglobuHn, 297 Bioluminescence colors of, 16-21 diurnal rhythm, 10 extracellular, of various organisms, 16-20 in cytolysis, 12 inhibition by light, in various organisms, 16—20 oxygen requirement, 9, 12, 16-21, 127, 130, 210 without oxygen, 9, 12 Bioluminescence spectra, sec Emis- sion spectra Bivalvia, 17 Blue shift, 31, 33 Bonin Island, 339 Borawazawa, 351 Borneo, 339, 344 Br\'ozoa, 366 Calcium, effect on firefly lumi- nescence, 167, 168, 177, 186, 187 California, 399, 402 Carohne Islands, 337 Caterpillars, 1 Celebes, 339, 344 Cephalopoda, 353-356 Ceratioidea, 19 Ceylon, 347 Chaetognatha, 3 Chaetopteridae, 17 Chemiluminescence after photosynthesis, 3 mechanisms of excitation, 102-104 rate equations of, 76 relation to fluorescence, 100-108 transfer of excitation energy, 103- 104 Chemiluminescence of 3-aminophthalic hydrazide, 10 dimethvldiacridinium nitrate, 10 DPD's' activation energy, 85 catalysts of, 77 emission maximum, 91 end products of reactions, 93- 96 energy of, 91 entropy of activation, 87 flow method measurement, 81- 82 influence of Oo cone, 80-81 influence of substituents, 92- 93, 106-108 438 intensity vs. oxygen cone, 87- 91 intermediate products, 93-96 mechanism of, 91-97 quantum yield, 85 rate of oxygen use, 87-91 role of oxygen, 77-81 role of peroxide, 94-96 structure of emitting molecule, 108 temperature dependence, SC- SI temperature dependence of oxy- gen use, 90-91 yield, compared to fluorescence, 93 hydrazine and ferricyanide, 79- 80, 91 luminol hydrogen peroxide in, 244 mechanism of, 244 metal porphyrins, 9, 10 pyrogallol, 10 riboflavin, 235, 246 Chemiluminescence spectra, see also Emission spectra effect of temperature on, 103-105 relation to fluorescence, 122 Chemiluminescence spectra of dimethylbisacridinium nitrate, 103, 105 DPD's, 107 methylacridone, 104 organic substances, 10 Chiba Pref., 359, 371 Chickens, 337 Chiloprda, 9, 18 Chimaerae, 3 Chloroflagellates, 344 p-Chloromercuribenzoate, effect on bacterial lucif erase, 219 bact, luminescence in vitro, 223, 260, 262 firefly luminescence, 168 Chlorophyceae, 400 Chlorophylls dissipation of excitation energy, 40-41 SUBJECT INDEX electronic transitions, 31 energy transfer between, 43, 71 ethylenic potential functions, 26 fluorescence activation, 32 fluorescence and phosphorescence, 40 in red algae, 68-72 intercombinations, 26 luminescence emission properties, 33-36 role of triplet states in energy transfer, 38 n,7r-transitions, 26, 30 Chlorophylls, a and b energy transfer between, 38-39, 43, 66 fluorescences, 35 ff. fluorescence spectra, 57-58 interpretation of spectra, 33 lowest triplet state,- 33-36 phosphorescences, 35 ff. singlet-triplet splits, 35 Chlorophyll a, 30 fluorescence peaks, 61 fluorescence spectra, 62, 67-70 fluorescence spectrum in hve Ulva, 61 in red algae, 68-72 Chlorophyll b, 30 fluorescence spectrum, 66 triplet-singlet emission, 34 Chlorophyll c, 67 Chlorophyll d, in red algae, 68 Cholera vibrio, 359 Choline, 194 Chromatophores, 1, 307, 308, 315, 369 Chrysophyceae, 400 Chymotrypsin activity, volume change of activation, 272- 273 Cirratulidae, 17 Co-dehydrogenase, see DPN Coenzyme A, 169, 194, 195, 212 Coenzyme I, see DPN Coenzyme II, see TPN Coleoptera, 14, 19 Collembola, 14, 18 SUBJECT INDEX Control, physiological, of lumines- cence absence of, 299-301 acetylcholine, 194, 330 action potentials, 320, 328 adrenalin, 306, 307, 308 facilitation, 301-304, 311-313 fatigue, 303, 320 firefly flash, 194-195, 310, 325, 328-332 hormonal, 4, 14, 306, 307, 316, 320 hyperexcitatory state, 314 inhibition of luminescence, 308- 310 intermedin, 321 latent period, 306, 320 oxygen role, 325, 329-331 pituitrin, 306, 321 repetitive flashes, 302, 305, 310, 314 response duration, 314 rhythmic flashes, 304, 305 screening devices, 1, 2, 299, 307- 308, 316, 370, 374 spectral emission, 315 summation, 304, 305, 312, 314 temperature effects, 320 tetanus, 320 treppe, 320 Control, physiological, of lumines- cence in Acholoe, 305 Acropoma, 307 Amphiura, 309 Anomalops, 307 Argijropelecus, 306 balanoglossids, 316 beetles, 299, 315 Beroe, 304, 311, 313. 314 Bolina, 325 Cavernularia, 311 cephalopods, 307 Chaetopterus, 309, 311, 312, 315, 323, 325 coelenterates, 301, 308, 323 ctenophores, 300, 301, 303, 304, 308, 309, 323 439 Cyclothone, 306 Cypridina, 323 eggs, 300, 316 elaterid beetles, 327 enteropneusts, 310 euphausiids, 316 Euprymna, 321 fireflies, 194-195, 310, 316, 325, 328-332 fishes, 299, 306-307 Gazza, 308 Gonyatdax, 416-419 Lampanyctus, 306, 307 lampyrids, 300, 310 Lampyris, 310 Leiognathns, 307 medusae, 302, 303 midshipman, 306 millipedes, 299 Mneiruopsis, 320, 323, 324 mvctophids, 306 Noctiluca, 11, 12, 300, 309, 320, 323 Pelagia, 303, 309, 311, 325 Pennatula, 301 pennatulids, 301, 309, 311, 314 Photinus, 316 Photuris, 327 Photoblepharon, 307 Phrixothrix, 315 polvchaetes, 300, 323 Polynoe, 311,313 polynoid worms, 304-305, 309, 314 Porichthys, 306, 307, 312, 321 protozoa, 300, 308, 309, 315, 316, 323 Ptychodera, 310 Pyrophorus, 327 Pyrosoma, 311, 312, 313, 316 Rendla, 301, 302, 309, 312-314, 323-325, 328 Scyphomedusae, 302-303 sea pansy, 302, 309 sea pens, 301 Secutor, 308 squid, 299, 316 squid mantle, 327 440 SUBJECT INDEX teleosts, 306-308, 316 Thatimatolampas, 315 tunicates, 316 veliger larvae, 310 Watasenia, 308 worms, 304, 305, 311, 316 Copepoda, 9, 18, 346 Copper, and porphyrin phospho- rescence, 37 Copper, efiFect on firefly lumines- cence, 178 Copper phthallocyanines, fluores- cence and phosphorescence, 37 Corn, 61 Corror Island, 347, 350, 360 Crustacea, 356 Ctenophora, 16, 345 Cultivation, of dinoflagellates, 392- 393, 414-419 Cvanide, combination with luciferin, 131 Cyanide, effect on bacterial luminescence, 210 bact. luminescence in vitro, 223, 260 firefly luminescence, 168 Cyanines, singlet-triplet split, 34 Cypridina luminescence, see under Absorption spectra; Emission spectra; Lucif erase, Cypri- dina; Luciferin, Cypridina; Luminous organs; Total light, etc. Decapoda, 356 Delaware Bay, 399 Delayed light production measurement of, 47-48 reaction kinetics, 49-50 reaction scheme, 50 saturation of intensity, 47—50 Denaturation, irreversible catalysis by narcotics, 281-283 in bacterial luminescence, 275, 276 volume change of activation, 275, 276 Denaturation, reversible, in bacterial luminescence, 269-279 bact. luminescence in vitro, 287 growth of bacteria, 271 promotion by narcotics, 275-277 volume change of reaction, 273- 274 Diatoms, 67, 68, 387, 389, 392, 394-397, 400 2,2'-Dicarboxylbenzil, in DPD chemiluminescence, 94, 96 l,4-dihydroxynaphthyl-2-hydroxy- methylketone, 112, 130 absorption spectrum, 116-118 structure, 116 Dimethylbisacridinium nitrate emission spectrum, 103, 105 fluorescence spectrum, 105 Dinoflagellata, 16, 342-344 Dinoflagellates autotomy in, 396 autotrophic, 387, 388, 405 cultivation of, 392-393, 414, 415 diurnal migrations, 404 holozoic, 388 in red water conditions, 400-409, 414 luminous, 16, 387, 414-419 nutrient requirements, 391-394, 414, 415 photosynthetic, 388, 415-419 physical properties, 408 salinity relations, 388-391 saprophytic, 388 temperature relations, 388-391 vertical distribution, 404 fran^-Diphenylethylene, 39 Diplopoda, 18 Diptera, 14, 18, 333 Dipyrimidopyrazines and firefly luciferin, 207 DPD, see also Luminol; Chemilu- minescence DPD addition product with Oo, 96 ionization constants, 95 DPD's keto and enol forms, 95 SUBJECT INDEX 441 structure of, 92, 106 DPN (coenzyme I), effect on bact. luminescence in vitro, 210- 216, 220-229, 231, 232, 234- 237, 257-259, 262, 284, 285, 288-290 Drilidae, 19 Drugs, effects on luminous bacteria, 278-283, 336 East Indies, 335, 338, 344, 359, 361 "Ectocrines," 394 EDTA, 393, 415 EflFector systems, 299 Efficiency, of light emitting reactions, 81-86, 244-255 4,, 76, 81, 83-86 temperature dependence, 81- 86 Eggs, luminous, 4, 300, 316, 344 Elasmobranchii, 19 Elateridae, 19 Electrolumigrams, 5 Electronic transitions, influence of solvents, 32 TT-Electronic states, 27 and fluorescence, 27 Elytra, 305 Elytrophore, 305 Emission spectra control of, physiological, 315 relation to fluorescence, 108 Emission spectra, effect of resorption 10-11, 62, 106, 109, 110 temperature, 103—105 Emission spectra, of A. mellea, 109-110 bacteria and fungi, 109-110 bact. luminescence in vitro, 218- 219 Cypridina luminescence, 11, 110 dimethylbisacridinium nitrate, 103-105 DPD's, 107 firefly extracts, 163 luminous bacteria, 11, 109 luminous fungi, 109, 110, 342 Mycena poly gramma, 110 Omphalia flavida, 110 organic substances, 10 Pli. phosphoreiim, 109 Ph. spendidum, 109 Photinus pyralis, 116 various organisms, 10 Emission spectrum, mirror image of inactivation spectrum, 114 Energy, of bacterial luminescence, 233 bact. luminescence in vitro, 235- 237, 244-255 bioluminescence, 161-162 DPD luminescence, 91 Energy relations, in photochemical reactions, 244-255 Energv states, of a two-atom mole- 'cule, 100 Energy transfer, between photosyn- thetic pigments, 51-52 Enteropneusta, 19, 366 Entropy of activation, in luminol luminescence, 87 Ephemerida, 18 Epinephrin, 4 Ethylene, singlet and triplet states, 39 Ethylenic molecules, potential en- ergy, 40 Euglenineae, 400 Eunicidae, 17, 347 Excitation energy dissipation, in cells and solution, 40-41 transfer between dyes, 103 transfer between pigments, 103 transfer in chemiluminescence, 103, 104 Excited states internal conversion, 26 hfetime, 28-30 radiationless combination between, 26 n,7r-Excited states, singlet, 37 TTjTT-Excited states, 31, 37 442 SUBJECT INDEX Facilitation, neural, 301-304, 311- 313 FAD, effect on, bact. luminescence in vitro, 212-213, 223 Fatigue, of luminescence, 303, 320, 323 Firefly flash, 5, 20, 180-182, 184, 194-195, 310, 325, 328-331, 361-363 Firefly flashing, synchronous, 20, 361-363 Firefly luciferin, 130; see also Lucif- erin, firefly Firefly luminescence acetylcholine in, 194, 330 activation energy, 167 ATP relations in, 162, 164, 165, 168, 170-173, 175-181, 186- 197, 263 control of, 194-196, 310, 316, 325, 328-332 eflect of, azide, 168 beryllium, 178 calcium, 168, 178 p-chloromercuribenzoic acid, 168 copper, 178 cyanide, 168 emission spectrum, 163 fluoride, 168, 186 oxygen tension, 182-186, 195- 196, 310, 325, 329-331 temperature, 167 flash, in vitro, 180-182, 184 intermediary reactions, 181, 194- 197 mechanism of, 196-197 pH optimum, 166, 167 reaction scheme, 194 relation to pyrophosphate, 186- 195 relation to triphosphate, 168, 172, 177, 179, 180-182, 186, 191- 195 role of Mg, 162, 170-175, 178, 180, 181, 186, 188, 193, 194 Firefly, luminescent system, 127. 162 Flash of luminescence, in Beroe, 304, 305 Gonyaulax, 416 snails, 351 Flavins, in bioluminescence, 11 in bacterial luminescence, 130, 210-216, 223-232, 234-237, 241-243, 258-262, 284, 288- 290 Flicker, luminescent, 350 Florads, 338 Florida, 399, 400, 404, 408 Flow method, in luminol reaction, 81-82 Fluorescence action spectrum for excitation, 67- 68 distinction from phosphorescence, 44-45 effect of incident intensity, 71 intrinsic quantum yield, 29 lifetime, 44-45 mirror image of absorption, 100 relation to chemiluminescence, 100-108 relation to electronic states, 26, 27 Fluorescence excitation, wave length effectiveness, 70-71 Fluorescence of bacterial luciferin, 8 chlorophyll magnesium role, 33 relation to solvent, 32 ctenophore organs, 8 Cypridina luciferin, 8, 134-136, 138, 146-149, 153 DPD's, 93, 107 earthworm lymph, 8 firefly luciferin, 8, 165-166, 200- 201 hematoporphyrin, 9 luminol, 85 luminous organisms, 16-21 luminous organs, 8, 16-21 lymph, earthworm, 8 Mg phthallocyanines, 37 SUBJECT INDEX Fluorescence spectra distortion by scattering, 63 measurement of, 52-55 relation to chemiluminescence, 122 significance of resorption, 62, 106 Fluorescence spectra of new leaf pigment, 72 bacteriochlorophyll, 59-60 chlorophyll a, 62, 67-72 in hve Ulva, 61 chlorophyll h, 66 chlorophylls a and h, 57-58 dimethylbisacridinium nitrate, 105 DPD's, 107 ivy leaf, 67 methyl acridone, 102, 104 Nitzschia, 67 partially greened leaf, 66 phycocyanine, 68-72 phycoerythrin, 56, 68-72 Porphyridium, 68-72 protochlorophyll, 63-66 uroporphyrin, 72-73 squash seed coat, 65 Vicia epidermis, 72-73 Fluoride, effect on fireflv lumines- cence, 168, 186 Formosa, 341, 359, 360 Friday Harbor, 405 Frog larvae, 278-279 Fruiting bodies, luminous, 342 Fucoxanthin, 67-68 Fukuoka, 348 Funamushi, 337 Gadidae, 19, 336, 367, 368 Gastropoda, 349, 350, 351 Gifu City, 363 Gifu Pref., 337, 363 Glucoisoalloxazine, 242 Glutathione, 168, 212 Gokasho Bay, 398, 404 Gorgonacea, 16 Gramicidin, 132 Grammaridea, 18 Grape leaf, 61, 62 443 Gulf of Mexico, 403, 408 Gunung Pelai, 350 Gymnodiniales, 387 Habitats, of luminous organisms. 16- 20 Hachijo Island, 335, 339, 340, 341, 345, 346, 348, 351, 352, 358, 364, 371, 372, 374 Hematoporphyrin, fluorescence of, 9 Hemiptera, 18 Hexokinase, 164, 171 Histology, of luminous tissues, 16-20 Hokkaido, 346 Honshu, 398 Hormonal control, 306, 307, 316, 320 Hotaru-Dake, 341 Hotaru-Ika, 354 Hotaru-jako, 368 Hydrazine, luminescence with fer- ricyanide, 79-80, 91, 93 Hydrogen peroxide, in bact. luminescence in vitro, 211, 234-238, 244, 246, 259 DPD chemiluminescence, 94-96 luminol luminescence, 244 snake star luminescence, 366 Hydroidea, 16 Hydrostatic pressure, effect on bact. luminescence, 272-276 bact. luminescence in vitro, 229- 232, 284-292 bacteriophage inactivation, 275, 283 Cypridina luminescence, 130 denaturation with urea, 297 disinfection of spores, 275, 283 globulin denturation, 275 methemoglobin denaturation, 275 narcosis of larvae, 278-279 temperature optima, 272-274 tobacco mosaic virus denaturation, 275, 283 Hymenomycetineae, 342 Hymenoptera, 19 Hyperexcitatory state, 314 Hyperiidea, 18 iS^^CAi 444 Immune reactions, of luminous bac- teria, 336 Illumination, by luminous bact., 338 Cypridina luminescence, 357 India, 342, 398 Indo-China, 344, 359 Indole, 169 Infrared absorption, of Brefly luciferin, 200-202 riboflavin, 202 Inhibition, of luminescence by light, 9, 10, 16- 21, 111-114, 309-310, 325, 371, 416-419 luminescent responses, 308-310 Inhibitors, of bacterial luciferase, 219, 278-284 bacterial luminescence, 278-284 bact. luminescence in vitro, 221- 223, 260, 291 firefly luminescence, 168-169, 186-188 Inhibitory fibers, 310 Intermedin, 321 Intrinsic lifetime, 30 lodoacetate, effect on bact. luminescence in vitro, 223 Iron porphyrins, nonfluorescence, 37 Isopoda, 18, 337 Isoptera, 3 Ivy leaf, 61, 62 Iwayama Bay, 350 Izu Peninsula, 345, 366 Jakarta, 345, 369 Japan, 13, 333, 335-338, 344. 345, 348, 349, 353, 359, 360. 361, 369, 374, 398, 404 Java, 344, 345, 358, 363, 369 Johore, 350 Johore Baharu, 363 Jurong, 364 Kalimon Island, 350 Kanagawa Pref., 366 Kanminato Bay, 371 Kautsky effect, 71-72 SUBJECT INDEX KCF (palmitic aldehyde), effects on bact. luminescence in vitro, 216— 218, 228, 230, 231 Kelong, 369 Kochi, 349, 366, 368 Kochi Pref., 348 Kominato, 371 Korea, 347, 360 Kuroshio, 345 Kushu, 348 Kyushu, 339 La JoUa, 389 Lampyridae, 19 Lampyrids, 9, 333 Latent period, of luminescent re- sponses, 306, 320 Latiidae, 17 Leiognathidae, 19, 336, 369 Lepidoptera, 3, 18 Loch Craighn, 392 Loch Striven, 399 Luciferase, bacterial electrophoresis, 258 general properties, 219 inhibitors of, 219, 223 molecular weight, 257 purification, 219, 257-258 Luciferase, firefly complex with ATP, 181-182 complex with phosphate, 181-196 concentration vs. light, 174-176 immobilization, 181-182 purification of, 162-166 Luciferase, specificity, 13-14 Luciferin acetyl, 131 definition, 7, 241-243, 263 Luciferin, bacterial absorption spectrum, 113, 114 definition, 241-243, 263 dual function, 113 fluorescence of, 8 identity of, 261 possible nature of, 122, 130 reversible oxidation, 113, 114 Luciferin, Cypridina absorption spectra, 116, 138-146 SUBJECT INDEX 445 alpha and beta, 133, 149, 150 amino acids in, 132, 149-153, 158 chromatography of, 132, 133-138, 150-153 combination with CN, 131 combining weight, 132 definition, 243, 263 dual oxidation, 129, 196-197 effects of oxygen on spectra of, 140-145 electrophoresis, 137-138, 151-152 ferricyanide oxidation of, 132 fluorescence of, 8, 134-136, 138, 146-149, 153 hydrolysis of, 132, 149-153 infrared absorption, 132, 150, 157-159 ion exchange fractions, 135-136 possible nature of, 130 purification, 128, 132-138 redox potential, 132 reversible oxidation, 114-115 Rf values, 134, 135 stabihty vs. pH, 140 UV absorption, 138-146, 151 yellow pigment from, 132 Luciferin, firefly, 130 absorption maxima, 166 chromatography, 163-166, 199 concentration vs. light, 173 electrophoresis, 202 fluorescence of, 8, 165-166, 200- 201 infrared absorption, 200-202 mass spectrocopy, 205-207 pK„ 200-201 polarography, 203 purification of, 162-166 relation to dipyrimidopyrazine, 207 relation to riboflavin, 200-202 role in flash, 330-331 stability of, 166, 196 structure proposed, 206 UV absorption, 200 Luciferin-luciferase reaction specificity, 13-14 various organisms, 6, 7, 16-21, 161, 211, 349, 353, 359, 365, 366 Luciferin, Pholas, 130, 263 Luciferin, relation to naphthoqui- nones, 115—127 Lumichrome, 8, 242, 249 Lumiflavin, 229, 242 Luminescence in "red water," 397 intrinsic lifetime, 29 relation to respiration, 223 Luminescence efficiency, ^, 76, 81, 83-86 Luminol, see also DPD; Chemilumi- nescence anaerobic reaction of, 80 fluorescence yields, 85 ionization constants, 95 quantum yield, 85 rate of oxidation, 87-91 role of oxygen in luminescence of, 77-81 structure of, 92 total luminescence, temperature dependence, 82-85 Luminor nerves, 5, 12 Luminous bacteria, see also Bacte- rial luminescence; Luciferin, bacterial; etc. Luminous bacteria detectors of oxygen, 78-79 illumination by, 338 parasitic, 1, 336-337 pathogenic, 1, 359 saprophytic, 1, 16, 337 symbiotic, 1, 2, 299, 335, 336, 353, 354, 367, 368, 372 Luminous bacteria, extracts, see also Bact. luminescence in vitro Luminous bacteria, extracts anti-allergic action, 338 anti-gangrenous action, 338 anti-rheumatic action, 338 neurotropism, 338 Luminous organisms (common names ) amphipods, 18 446 SUBJECT INDEX annelids, 13, 347-349 ascidia, 365 balanoglossids, 15 bacteria, 1-6, 8, 10, 11, 15, 16, 113, 114, 116, 130-131, 209, 223, 265, 336, 337 beetles, 4, 5, 19, 161 bryozoans, 366, 367 bugs, 18 centipedes, 14, 359-360 cephalopods, 307 chilopods, 14 coelenterates, 12 copepods, 13, 18 crabs, 337 Crustacea, 6, 13, 14, 17, 127 ctenophores, 9, 10, 12, 16, 344- 346 cubomedusae, 344 decapods, 13, 17, 18 dinoflagellates, 10-12, 342-344, 387, 388, 414-419 diplopods, 4, 14, 18, 360 diptera, 364, 365 dipterons, 364, 365 earthworms, 8, 14, 348 marine, 348 eggs, 4, 300, 316, 344 elaterid beetles, 327 enteropneusts, 19, 310, 366 euphausiids, 316 fireflies, 4, 5, 8-11, 14, 21, 146, 161, 162, 310, 316, 328-332, 335, 361, 363 fishes, 1, 2, 4, 6, 9, 10, 15, 19, 20, 335-337, 367-374 flies, 18 flying fish, 374 fungi, 3-5, 10, 13, 16, 109, 110, 122, 335, 338-340 glow worm, 5, 161, 360, 364 hydroids, 16 hydromedusae, 16, 344 isopods, 18 jellyfish, 344, 345 lampyrids, 300, 310 larvae, 364, 365 hmpets, 6, 350 macroplankton, 344—347 medusae, 9, 12, 16, 302, 303, 345 midges, 1 midshipman, 306 millipedes, 4, 337, 360, 364 molluscs, 5, 13, 128 myctophids, 306 myriapods, 14 mysids, 13 nemerteans, 365 nudibranchs, 349 ohgochaetes, 348-349 ophiuroids, 15 ostracods, 13, 21, 127 pennatulids, 5, 12, 13, 16 peridinians, 398, 399 polychaetes, 300, 323, 347-348, 365 protozoa, 4, 10, 11, 300, 308, 309, 315, 316, 323 pupae, 4, 300, 362 radiolarians, 9 railroad worm, 9, 15 sand fleas, 1 scyphomedusae, 302-303 sea pansy, 302, 309 sea pens, 301 sea slugs, 349 shrimp, 6, 18, 356, 358, 359 siphonophores, 16, 344 snails, 350-353 snake stars, 366 sponges, 16 spores, 339, 341 squid, 1, 2, 13, 15, 316, 337, 345, 353-356 squid mantle, 327 starworms, 363, 364 teleosts, 306-308, 316 tunicates, 15, 19, 316 veliger larvae, 310 wood louse, 337 worms, 5, 13, 17, 316 Luminous organisms, list of, 3, 16- 20 Luminous organs, of annelids, 347 SUBJECT INDEX bryozoans, 366-367 Cypridina, 127 fireflies, 362 fishes, 367-374 fungus gnat, 364, 365 snails, 351, 352 squid, 353-356 starworms, 363 Luminous varieties, crosses between, 13 Macroplankton, 344-347 Macrouridae, 19, 336, 367, 368 Madagascar, 347 Madras, 347, 398 Magnesium, effect on bact. luminescence in vitro, 212- 213 firefly luminescence, 162, 167, 170-175, 178, 180, 181, 186, 188, 193, 194 porphyrin phosphorescence, 37 Magnesium phthallocyanines, fluor- escence, 37 Malabar, 398 Malay, 358 Malay Peninsula, 335, 338, 344, 350, 359, 361, 363, 364, 369 Malaya, 369 Malic acid, effect on bact. lumines- cence in vitro, 212-213, 226 Manganese, effect on bact. luminescence in vitro, 212- 213 in firefly luminescence, 167, 177, 186, 187 Manukvvari, 363, 371 Markyok, 342 "Marine snow," 346 Marshall Islands, 337, 344 Mass spectroscopy, of firefly luciferin, 205-207 riboflavin, 203-205 Mercuric ion, effect on bacterial luciferase, 219 bact. luminescence in vitro, 223 Metal porphoryrins, 9 447 Methylacridone absorption spectrum, 102 fluorescence spectrum, 102, 104 luminescence spectrum, 104 Methylethylenes, 39 2-Methylnapthoquinone, photolabil- ity of, 115 Michaelis constant, 223, 224 Micronesia, 335, 337, 338, 343, 359, 360, 361 Midges, 1 Millipedes, 364 Misaki, 344, 349, 356, 366 Mitsui Institute, 366 Miyazaki, 339 Miyazaki Pref., 339 Mollusca, 349-356 Momi, 363 Monocentridae, 19 Monterey Bay, 399 Mt. Ryogami, 364 Mucus, luminous, 348, 349, 360 Muscle tension temp. -pressure relation, 273 Mutants, of luminous bacteria, 10, 265-266, 337-338 Mutsu Bay, 349 Mycelium, luminous, 342, 360 Myctophidae, 374 Myctophoidea, 20 Myokinase, 171, 172 Myopsida, 17 Myriopoda, 359 Mysidacea, 18 Nagamori, 363 Nakanogo, 365 Namerikawa, 345, 354 Naphthalene, singlet-triplet split, 34 Naphthoquinones, 130 relation to luciferin, 115-122 Naphthoquinones, effect on bacterial luminescence, 210 bact. luminescence in vitro, 212, 223 Narcosis, reversal by pressure, 278- 279 448 Narragansett Bay, 399 Nemertinae, 17 Neuro-effector, 315 Neurophotocyte, 324 New Britain, 361 New Guinea, 335, 344, 361, 371 New Jersey, 399 New Zealand, 6, 333, 347, 407 Nidulariineae, 342, 343 Nippon Sea, 354 N-methylalloxazine, 242 North Borneo, 338-340, 344, 369 North CaroUna, 333 Norway, 390, 399 Nudibranchia, 9, 17 Oak Ridge, 199, 200, 207, 238 Octopoda, 3 Oegopsida, 17 Offats Bayou, 399 Ohgaki, 337 Oiso, 348 Okinawa, 359, 360 Ohgochaeta, 17, 348, 349 Ophiuroidea, 19 Opisthobranchiata, 349 Orthoptera, 18 Osaka, 338 Oslofjord, 392, 399 Ostracoda, 17, 356 Otaro, 346 Oxygen detection bv luminous bacteria, 78-79 ' effect on absorption spectra, 140-145 chemiluminescence intensity, 87-91 luciferin fluorescence, 147-149 in chemiluminescence, 77-81, 90- 91 measurement by luminol, 97 relation to firefly flash, 182-186, 195-196, 310, 325, 329-331 Oxygen requirement, in biolumines- cence, 9, 12, 16-21, 127, 130, 210 Oxygen tension, effect on SUBJECT INDEX bact. luminescence in vitro, 225 firefly luminescence, 182-186 respiration of bact. extracts, 225 Oxynonylhydroperoxide, 236 Ovsters, 348 Palao, 335, 337, 339, 340-344, 347, 350, 358, 360, 361, 369 Palmitic aldehyde (KCF), 269 effect on bact. luminescence in vitro, 216-218 Paramagnetic molecules, and quench- ing, 37 Pasania tree, 365 Pelecypoda, 349, 350 Pennatulacea, 16 Peridiniales, 387 Peroxyaldehydes, in bact. lumines- cence in vitro, 234-238 Peru, 402 pH, effect on bacterial luminescence in vitro, 220, 259 Cijpridina luciferin, 140-146 firefly luciferin, 200-202 firefly luciferin fluorescence, 201 firefly luminescence, 166 luminous bacteria, 336 Phallales, 3 Phengodidae, 19 Pheophytin, 36 Phihppines, 347 Phosphorescence distinction from fluorescence, 44— 45 heavy atom effects, 36-37 intrinsic quantum yield, 29 lifetime, 44-45 low temperature spectrum, 28 paramagnetic atom effects, 36-37 relation to viscosity, 28 Photochemical inactivation of bacterial luminescence, 112- 114 of luminescence, 9, 10, 16-21, 111-114, 309-310, 325, 371, 416-419 SUBJECT INDEX Photochemical inactivation spectra, mirror image of emission spectra, 114 Photochemical reactions energy relations in, 244-255 maximum eflBciency of, 244-255 Photocytes, 299, 304, 308-310, 313, 321, 324, 328 Photophores, 4, 14, 306, 307, 315, 316, 327, 353, 356, 370-374 Photosynthesis delayed light production, 47-50 induction period, 71 quantum efficiency, 38 Phthallocyanines, singlet-triplet split, 35 Phycobilins, in red algae, 68-72 Phycocyanine, fluorescence spectra, 68-72 Phycoerythrin absorption spectrum, 56 fluorescence spectra, 56, 68—72 Pigments, photosynthetic, energy transfer between, 51-52 Pisces, 367-374 Pituitrin, 306, 321 "Plankton snow," 346 Platyuridae, 364 Polarography, of firefly luciferin, 203 Polychaeta, 3, 9, 17, 347-348, 365 Polymixin, 132 Polynoinae, 17 Polysporaceae, 342 Polyzoa, 3 Ponape, 340, 342, 344, 360 Porifera, 16 Pork, 337 Porphyrin, chemiluminescence of, 9 Porphyrin, singlet-triplet split, 34 Port Jackson, Australia, 398 Potential energy, of ethylenic mole- cules, 40 Pressure optima, 278-279 Pressure-temperature relations, of bact. luminescence in vitro, 284-292; muscle tension, 273 Prosobranchia, 17 Prosobranchiata, 3, 351 449 Protochlorophyll, 61 absorption maximum, 64 fluorescence spectrum, 63-66 in squash seed coat, 64, 65 Protozoa, 11, 12, 300 Provoringo, 363 Pseudoflash, 330, 331 Pteridins, 131, 200 Pteropoda, 3 Ptychoderidae, 366 Pulmonata, 17 Purple bacteria, 51, 52 Pyrophosphatase, 175-177, 186-195 Pyrophosphate, role in firefly lumi- nescence, 186-195 Quantum yield in luminol luminescence, 85 intrinsic, 29 Quenching by paramagnetic molecules, 37 collisional bimolecular, 30 of chemiluminescence, 76 of luciferin fluorescence, 148 of luminescence, by light, 9, 10, 16-21, 111-114, 309-310, 325, 371, 416-419 temperature dependence, 83-86 Rabaul, 339, 341, 361-363 Radiolaria, 12, 16 Raffles Museum, 335, 350 Rat brain, 281 Receptors, for luminescence, 306 Redox potential, of FMNH2, 241 "Red Water" prerequisites, 400-409, 414 survey of occurrences, 398-399 Regulation of luminescence, see Con- trol, physiological Repetitive flashes, 302, 305, 310, 314, 320, 323 Respiration of bacterial extracts, 224, 225 effects of aldehydes, 224, 225 relation to luminescence, 223 Rhagophthalmidae, 19 Rhio Archipelago, 350 450 SUBJECT INDEX Rhode Island, 399 Rhythmic flashing, of Beroe, 304, 305 Riboflavin, and firefly luciferin, 200-202 chemiluminescence of, 235, 246 effect on bact. luminescence in vitro, 223, 241-242, 260 infrared absorption, 202 mass spectroscopy, 203-205 relation to bioluminescence, 8, 111 relation to firefly luciferin, 200- 202 "Rise time, ^2," of bacterial lumines- cence in vitro, 224-229 Rota, 340 Saccopharyngidae, 20 Sagami Bay, 356 Saipan, 340, 344 Saitama Pref., 364 Salamander larvae, 278-279 Salinity, relation to red water con- ditions, 398, 399, 401, 402 Salt, effect on luminous bacteria, 336 Sandakan Bay, 344, 369 San Pedro, Calif., 399 Sawara, 359 Schizopoda, 18 Scotland, 399 Screening devices, in luminescence control, 1, 2, 299, 307-308, 316, 370, 374 Scyphomedusae, 16 Sea lettuce, 61 Sea of Japan, 368 "Sea snow," 346 Sendai, 337 Serranidae, 19 Seto, 346 Shikoku, 368 Shimoda, 366 Shizopoda, 18 Silver sulfate, eflfect on bact. lumi- nescence in vitro, 223 Singapore, 335, 338, 339, 341, 350, 354, 361, 363, 364, 369 Singlet-singlet transitions, 27 Singlet-triplet split, definition, 34 Singlet-triplet transition, 27 significance of atomic numbers, 36, 37 Siphonophora, 16 Slime, luminous, 366 Snake venom, 242 Soret band, 73 South Africa, 399 South Kyushu, 339 Southwest Africa, 402 Spores, luminescent, 339, 341 Squash seed coat, fluorescence spec- trum, 65 Sternoptychidae, 374 fran5-Stilbene, 39 Stomiatoidea, 20 Sueyoshi, 351 Sugami Bay, 374 Sulfanilamide, effect on bact. lumi- nescence in vitro, 291 Sumatra, 335, 344 Summation, of flashes, 304, 305, 312, 314, 320 Suruga Bay, 335, 345, 346, 354, 356, 374 Syllidae, 17 Symbiosis, of luminous bacteria, 1, 2, 299, 313, 335, 336, 353, 354, 367, 368, 372 Synchronous flashing, 20, 361-363 Tawao, 339 Teleostomi, 3, 19 Temperature, effect on bacterial luminescence, 269-277 bact. luminescence in vitro, 221, 285-289 chemiluminescence intensity, 80- 81 chemiluminescence spectrum, 103- 105 eflSciency of luminescence, 81-86 firefly luminescence, 167 growth and luminescence, 266- 269 luminescent response, 320 luminous bacteria, 370 SUBJECT INDEX luminous fungi, 342 oxygen use in chemiluminescence, 77-81, 90-91 quenching reaction, 83-86 total light of DPD, 82-85 Temperature optimum, variation with pressure, 272-274, 277 Temperature optimum, for dinoflagellates, 390 firefly luminescence, 166-167 photosynthesis in Gomjaulax, 416 Temperature-pressure relations, of bacterial luminescence, 272-276 muscle tension, 273 Temperature, relation to red water conditions, 398, 399, 401 Temperature relations, of dinoflagel- lates, 388-391, 394, 395, 397 Terai Was, 360 Terebellidae, 9, 17 Tetanus, 320 Texas, 399 Thiamin pyrophosphate, effect on bact. luminescence in vitro, 212-213 Thousand Islands, 345 Tokyo, 335, 338, 342, 351, 361 Tokyo-to Fisheries Station, 372 Tomioka Beach, 337 Tomopteridae, 9, 17 Tone River, 359 Torsional potential, effects on spec- tra, 39 Torsional vibration, influence of phenyl groups on, 40 Total light Cypridina luminescence, 128-129 of DPD, 82-85 of Pelagia luminescence, 303 Toyama Bay, 13, 345, 354 TPN (coenzyme II), in bact. lumi- nescence in vitro, 212, 257 Transfer of excitation energy between chlorophylls, 43, 71 between dyes, 103 between pigments, 103 n,7r-Transitions, 26, 30-32 TTjTT-Transitions, 32 451 Transitions, radiationless, 101 Treppe, 320 l,3,4-Trihydroxynapthyl-2-methylke- tone, photosensitivity, 115 Triphosphate, in firefly lumines- cence, 168, 172, 177, 179- 182, 186, 191-195 Triplet state and energy transfer between chlo- rophylls, 43 in photochemical reactions, 28-30 lifetime, 28 lowest, nature of, in chlorophyll, 37-38 of normal molecules, 27 role in energy transfer, 38 Trivandrum coast, 398 Tropical Asia, 335, 348 Truk, 340, 344, 360 Trypsin activity, volume change of activation, 272-273 Tsukiijo-dake, 339 Tunicata, 19 Turbellaria, 3 Ultraviolet effect on bact. luminescence in vitro, 259-261 stimulation of bacterial lumines- cence, 221-222 Ultraviolet absorption, of Cypridina luciferin, 138-146, 151 firefly luciferin, 200-201 naphthoquinones, 118-121 Ultraviolet inactivation, of bact. luminescence in vitro, 222 Umihotaru, 356 UNESCO, 335 Uozu, 354 Urat-bintang, 363 Urat intan, 363 Urea, denaturation of ovalbumin by, 297 Urethan dual action of, 281-283 effects on bacterial luminescence, 278- 283 452 bact. luminescence in vitro, 291 TMV denaturation, 283 multiple reactions, 283 Uroporphyrin I octamethyl ester, fluorescence spectrum, 72-73 Usa, 348, 349, 366 Utrecht, 99, 104 Vampyromorpha, 17 Veliger larvae, 310 Versene, effect on bact. luminescence in vitro, 260 Vinyl group, as dissipation of ex- citation energy, 40 Vitamin Bi„ 393, 414, 415 Volume change of activation, in bacterial luminescence, 272-274 bact. luminescence in vitro, 232, 288 chymotrypsin activity, 272-273 denaturation of proteins, 275, 283 trypsin activity, 272-273 Volume change of reaction, in re- SUBJECT INDEX versible denaturation, 273- 274 Wakayama Pref., 346 Walen, 363 Walvis Bay, 399 Washington, 364 Waves of luminescence, 301, 302, 326-328 Wellington, 407 West New Guinea, 363 Yap, 339, 340, 342, 344, 361 Yeast, 280 Yokohama, 337, 348 Yokosuka, 334, 345 Yurei-Ika, 356 Zinc, and porphyrin phosphores- cence, 37 Zinc tetraphenylporphyrin, singlet- triplet split, 34 Zonitidae, 350 Zushi, 348 m: M th iSa:. y- •■•iM bM, i