• • " 'I'l,,. , . . . . ' > f - wrrv*> ru nj cr =o a nj CD m a CAROTINOIDS AND RELATED PIGMENTS THE CHROMOLIPOIDS BY LEROY S. PALMER, PH.D. PROFESSOR OF AGRICULTURAL BIOCHEMISTRY, UNIVERSITY OF MINNESOTA American Chemical Society Monograph Series BOOK DEPARTMENT The CHEMICAL CATALOG COMPANY, Inc. 19 EAST 24TH STREET, NEW YORK, U. S. A. 1922 COPYRIGHT, 1922, BY The CHEMICAL CATALOG COMPANY, Inc. All Rights Reserved Press of J. J. Little & Ives Company New York, U. S. A. GENERAL INTRODUCTION American Chemical Society Series of Scientific and Technologic Monographs By arrangement with the Interallied Conference of Pure and Ap- plied Chemistry, which met in London and Brussels in July, 1919, the American Chemical Society was to undertake the production and publication of Scientific and Technologic Monographs on chemical subjects. At the same time it was agreed that the National Research Council, in cooperation with the American Chemical Society and the American Physical Society, should undertake the production and pub- lication of Critical Tables of Chemical and Physical Constants. The American Chemical Society and the National Research Council mu- tually agreed to care for these two fields of chemical development. The American Chemical Society named as Trustees, to make the nec- essary arrangements for the publication of the monographs, Charles L. Parsons, Secretary of the American Chemical Society, Washington, D. C.; John E. Teeple, Treasurer of the American Chemical Society, New York City; and Professor Gellert Alleman of Swarthmore Col- lege. The Trustees have arranged for the publication of the American Chemical Society series of (a) Scientific and (b) Technologic Mono- graphs by the Chemical Catalog Company of New York City. The Council, acting through the Committee on National Policy of the American Chemical Society, appointed the editors, named at the close of this introduction, to have charge of securing authors, and of considering critically the manuscripts prepared. The editors of each series will endeavor to select topics which are of current interest and authors who are recognized as authorities in their respective fields. The list of monographs thus far secured appears in the publisher's own announcement elsewhere in this volume. The development of knowledge in all branches of science, and espe- cially in chemistry, has been so rapid during the last fifty years and the fields covered by this development have been so varied that it is difficult for any individual to keep in touch with the progress in 3 4 GENERAL INTRODUCTION branches of science outside his own specialty. In spite of the facilities for the examination of the literature given by Chemical Abstracts and such compendia as Beilstein's Handbueh der Organischen Chcmic, Richter's Lexikon, Ostwald's Lehrbuch der Allgemeinen Chemie, Abegg's and Gmelin-Kraut's Handbueh der Anorganischen Chemie and the English and French Dictionaries of Chemistry, it often takes a great deal of time to coordinate the knowledge available upon a single topic. Consequently when men who have spent years in the study of important subjects are willing to coordinate their knowledge and present it in concise, readable form, they perform a service of the highest value to their fellow chemists. It was with a clear recognition of the usefulness of reviews of this character that a Committee of the American Chemical Society recommended the publication of the two series of monographs under the auspices of the Society. Two rather distinct purposes are to be served by these monographs. The first purpose, whose fulfilment will probably render to chemists in general the most important service, is to present the knowledge available upon the chosen topic in a readable form, intelligible to those whose activities may be along a wholly different line. Many chemists fail to realize how closely their investigations may be con- nected with other work which on the surface appears far afield from their own. These monographs will enable such men to form closer contact with the work of chemists in other lines of research. The second purpose is to promote research in the branch of science covered by the monograph, by furnishing a well digested survey of the prog- ress already made in that field and by pointing out directions in which investigation needs to be extended. To facilitate the attainment of this purpose, it is intended to include extended references to the litera- ture, which will enable anyone interested to follow up the subject in more detail. If the literature is so voluminous that a complete bibli- ography is impracticable, a critical selection will be made of those papers which are most important. The publication of these books marks a distinct departure in the policy of the American Chemical Society inasmuch as it is a serious attempt to found an American chemical literature without primary regard to commercial considerations. The success of the venture will depend in large part upon the measure of cooperation which can be secured in the preparation of books dealing adequately with topics of GENERAL INTRODUCTION 5 general interest ; it is earnestly hoped, therefore, that every member of the various organizations in the chemical and allied industries will recognize the importance of the enterprise and take sufficient interest to justify it. AMERICAN CHEMICAL SOCIETY BOARD OF EDITORS Scientific Series:- Technologic Series:- WILLIAM A. NOYES, Editor, JOHN JOHNSTON, Editor, GILBERT N. LEWIS, C. G. DERICK, LAFAYETTE B. MENDEL, WILLIAM HOSKINS, ARTHUR A. NOYES, F. A. LIDBURY, JULIUS STIEGLITZ. ARTHUR D. LITTLE, C. L. REESE, C. P. TOWNSEND. American Chemical Society MONOGRAPH SERIES Other monographs in the series of which this book is a part now ready or in process of being printed or written. Organic Compounds of Mercury. By Frank C. Whitmore. 397 pages. Price $4.50. Industrial Hydrogen. By Hugh S. Taylor. 210 pages. Price $3.50. The Chemistry oj Enzyme Actions. By K. George Falk. 140 pages. Price $2.50. The Vitamins. By H. C. Sherman and S. L. Smith. 273 pages. Price $4.00. The Chctnn-nl Ktjtct* oj Alpha Particles and Electrons. By Samuel C. Lind. 180 page.*. Price $3.00. Zirconium and Its Compounds. By F. P. Venal. lc. Price $2.50. Tin ProjH rtti .s- dj /'„'/< ctricallt/ Conducting Systems. By Charles A. Kraus. Price $4.50. The Analysis of I\nhl>,r. By John B. Tut tic. Price $2.50. Tin Orii/ht. of Spectra. By Paul D. Foote and F. L. Mohlt.-r. Price $4.50. Thyroxin. By E. C. Kendall. The Properties of SHim unit the Silicates. By Robert B. Sosman. Coal Carbonization, By Horace C. Porter. The Corrosion of Alloys. By C. G. Fink. Piezo-Chemistry. By L. H. Adams. Cyanamidc. By Joseph M. Braham. Liquid Ammonia as a Solr< nt. By E. C. Franklin. Wood Distillation. By L. F. Hawley. Solubility. By Joel H. Hildebrand. Glue and Gelatin. By Jerome Alexander. Organic Arsenical Compounds. By George W. Raiziss. Jos- eph L. Gavron. Yah nc( , and the Structure of Atoms and Molecules. By Gil- bert N. Lewis. Shale Oil. By Ralph H. McKee. Aluminothermic Reduction oj Metals. By B. D. Saklatwalla. The Chemistry of Leather Manufacture. By John A. Wilson. Absorptive Carbon. By N. K. Chaney. Refining Petroleum. By George A. Burrell, et al. Extraction of Gasoline jrom Natural Gas. By George A. Burrell. The Animal as a Converter. By H. P. Armsby and C. Robert Moulton. Chemistry of Cellulose. By Harold Hibbert. The CHEMICAL CATALOG COMPANY, Inc. 19 EAST 24TH STREET, NEW YORK, U. S. A. PREFACE Color in nature may properly be divided into two group?, namely, color due to structure, which is caused by light reflections from col- loidal particles of air or water, and color due to pigments, which is caused by substances having remarkable powers of absorption of light rays of certain wave length and reflection of others. The reflected rays, of course, give the pigment its color. The present monograph treats of pigmented substances having a yellow, yellow-orange, orange, red-orange and red color. So far as the author is aware no authentic instances of structural colors of these hues have been reported. In fact, the wave length of light in these regions of the spectrum is probably too great for such a phenomenon to occur for colloidal emulsions having the refractive index of air or water. The particular pigments to be considered are widely dis- tributed in every stage of living matter, and are perhaps more fre- quently encountered than any other class of natural pigments. They have attracted the attention of the biologists for at least 100 years. Among the earliest inquiries were those of Caventau (1817) and Goebel (1823). The former was interested in the yellow pigment of the daffodil, the latter in the pigments of the crab and in the feet of doves and geese. Active investigation of these pigments in plants and animals has been confined to the past fifty years. It has only been within the past fifteen years, however, that the chemical composition of any of these pigments has been definitely established. Their constitution still offers a fascinating problem for the organic chemist. The writer favors Tswett's terminology of carotinoids for these pig- ments. From the standpoint of phytochemistry there is definite evidence for the existence of five carotinoids, with indications that several others also occur. When it was discovered that certain of the carotinoids occur in animals it was believed that both plants and ani- mals synthesize these pigments. It soon became apparent, however, that the chromolipoids found in the higher animals, at least those 7 8 PREFACE which have been identified as carotinoids, are in reality merely derived from the food. The assumption therefore seems justified that a similar biological relationship exists between all the chromolipoids of plant and animal life; in other words, that all animal chromolipoids are derived pigments and are either true or modified carotinoids. The writer has had three main ideas in mind in preparing this mono- graph. First, he has attempted to compile a thorough history of the development of the chemistry of the plant and animal chromolipoids. This has not been attempted before in this particular field. Second, he has tried to present such information regarding the pigments as would be useful to workers who desire to attack the many interesting problems in this branch of plant and animal chromatology. Third, he has made an effort to point out lines of research which might prove attractive to those interested in this subject. The author hopes that he has had a reasonable measure of success in his efforts. For the convenience of readers who have not been trained in sys- tematic nomenclature the scientific name of the individual species of plants and animals in which carotinoids occur has been supplemented wherever possible by the common name. For the plants, this informa- tion has been drawn largely from Bailey's Cyclopedia of Horticulture. It may be of interest to the reader to know that the carotinoid pig- ments in plants and animals have proved to be of some practical importance. The uses to which their occurrence in animals have been put are reserved for discussion in Chapter XI of the monograph. The occurrence of carotinoids in plants, particularly green plants, formed the basis for the construction of the light filters used by the American Army during the late \var for the detection of camouflaged foliage. Natural green foliage reflects both green and red light, due to the fact that the chlorophylls and carotinoids are present together in the chloro- plastids. The visibility of the rays reflected from the carotinoids is so low in the presence of the chlorophylls which are present in five to six times the concentration of the orange and red pigments, that green color only appears to be reflected. However, it was found possible to construct a light filter which absorbed practically all light rays except a wide band in the red at about 700|i|i, and a narrow band, with low transmission in the green at about 500|4i, so that natural green foliage viewed through this filter appeared red, while camouflaged foliage on which green paint only was used, appeared green. The writer has encountered so much difference of opinion regarding the correct pronunciation of certain words which are used very fre- 9 qucntly in this monograph that he hcu-. (,, sii«_:.iic.-l , for the sake of uniformity, the following pronunciations which arc believed to be in keeping with the best modern Kn.dish usage. Carotin :K fir' 6- tin. Carotinoid = Kar" 6 - tin - oid'. Lipochrome = L I p' 6 - k r 6 m. Chromolipoid = Kr6m-6-lip'6id. Chromatogram - : K r 5 m" - a t o - g r a m'. In conclusion the author wishes to acknowledge his indebtedness to his colleague, Dr. R. A. Gortner, for many helpful criticisms in the preparation of the manuscript; to Dr. Josephine E. Tilden of the Department of Botany, University of Minnesota, for classifying the algae in which carotinoids occur, and to Mr. Lloyd A. Jones, of the Eastman Kodak Co., for information regarding the light filters devised in the Eastman Research laboratories during the war. ST. PAUL, MINN. July 1, 1922. CONTENTS PAGE CHAPTER I. General Distribution of Carotinoids. The Pigments Defined 13 Luteins. - - Lipochromes. - - Lipoxanthins. - - Chromolipoids. —Carotinoids. — Non-carotinoid plant pigments. — Non-curo- tinoid animal pigments. CHAPTER II. Carotinoids in the Phanerograms 25 The pigments of the carrot. — Carotinoids in other roots.— Carotinoids in the chloroplastids. — Separation of yellow pig- ments from chlorophyll. — Crystalline carotinoids from chlo- roplastids.— Plurality of yellow pigments in chloroplastids. —Carotinoids in etiolated leaves. — Carotinoids in naturally yellow leaves. — Carotinoids in yellow autumn leaves. — Caro- tinoids in autumn and winter reddening. — Carotinoids in flowers. — Carotinoids in fruits. — Carotinoids in seeds and grains. CHAPTER III. Carotinoids in the Cryptogams 92 Carotinoids in the algae. — The Phaeophyceae. — Fucoxanthin. -The Rhodophyceae. — The Charales. — The Chlorophyceae. -The Baccilariae (Diatomaceae). — The Peridinieae. — The Flagellata. — The Myxophyceae (Cyanophyceae). — Carotin- oids in the fungi. — The Basidiomycetes. — The Ascomycetes. -The Phycomycetes. — The Myxomycetes. — The Imperfects. —Carotinoids in Bacteria. CHAPTER IV. Carotinoids in the Vertebrates 125 Carotinoids in mammals. — Corpus lutcum. — Blood serum.— Milk fat. — Adipose tissue. — Internal organs. — Nerve<. Skin. —Carotinoids in birds. — Egg yolk. — Body tissues. — Retina. — Feathers. — Carotinoids in fishes. — Carotinoids in amphib- ians.— Carotinoids in reptiles. CHAPTER V. Carotinoids in Invertebrates 154 Carotinoids in insects. Lepidoptera. — Rhynchota. — Coleop- tera. — ( hllmptera. — Acerata. — ( 'arotinoids in ( Vustacea.- Carotinoids in Echinoderms. — Carotinoids in molluscs. —Carotinoids in worms. — Carotinoids in sponges. 11 12 CONTENTS PAGE CHAPTER VI. Clu-mical Relations between Plant and Animal Carotinoids 173 Egg yolk xunthophyll. — Corpus luteum carotin. — Crustacean carotin. CHAPTER VII. Biological Relations between Plant and Animal Carotinoids 182 Earlier views for and against a biological relationship. — -Iso- lated facts supporting a biological relationship. — Experi- ments proving a biological relationship. — Insects. — Cattle.— Fowls. — Man. — Distributions of carotinoids among different species. CHAPTER VIII. Methods of Isolation of Carotinoids . . . 199 Isolation of carotin. — From carrot?. — From green leaves. — From animal fat. — From blood serum. — Isolation of xan- thophylls. — From green leaves. — From egg yolk. — From blood serum. — Isolation of lycopin. — Isolation of fucoxan- thin. — Isolation of rhodoxanthin. CHAPTER IX. General Properties and Methods of Identification of Carotinoids 218 Properties of carotinoid solutions. — Carotin. — Lycopin.— Xanthophylls. - - Rhodoxanthin. - - Fucoxanthin. - -Proper- ties of crystalline carotinoids. — Carotin. — Other pigmented hydrocarbons. - - Xanthophyll. - Lycopin. - - Fucoxanthin. —Methods of identification in biological products. — Plant tissues. — Animal tissues. CHAPTER X. Quantitative Estimation of Carotinoids . . . 248 Estimation of carotin and xanthophyll. — Method of Arnaud. — Arnaud's results. — Method of Monteverde and Lubimenko. — Monteverde's results.— Methods of Willstatter and Stoll. — Results by Willstatter and Stoll's method. — Estimation of fucoxanthin. — Application to other biological materials. CHAPTER XL Function of Carotinoids. in Plants and Animals . 262 Possible function in plants. — Possible function in animals.— Possible relations to vitamins. — Relation between yellow pig- mentation and fowls and egg laying. — Possible relation be- tween yellow pigmentation of cattle and milk secretion. Bibliography 279 Index to Authors 296 Index to Subjects 300 CAROTINOIDS AND RELATED PIGMENTS Chapter I General Distribution of Carotinoids. The Pigments Defined Red, orange and yellow pigments which can be extracted from tis- sues by fat solvents are found abundantly in all forms of living mat- ter. In the plant world they are present in nearly all species ranging from bacteria, the lowest forms of cryptogams, to the dicotyledons, the highest forms of phanerogams. Similarly in the animal kingdom we find yellow to reddish pigments in all forms of both invertebrates and vertebrates, from protozoa to man. The earliest workers in both the plant and animal fields naturally based the classification of the pigments on simple properties, so that it is not surprising to find that many names have been proposed for what is obviously the same pigment. This diversity in nomenclature is found to be especially true among the yellow animal pigments, and can be traced in most instances to slight variations in certain of the simple properties which were re- garded as specific for various types of pigment. In some cases these variations were due to the fact that the method employed for the iso- lation of the pigment did not insure its freedom from other pigments of similar but not identical properties. In other cases the variations were due to the examination of the pigment in amorphous condition or in solution, without reference to the possible effect which these states might exert upon the particular properties being studied. Again, there was frequently an abundant contamination with lipoid impurities, which are invariably separated with the pigments from animal tis- sues. Another, still more important cause for these variations was the failure to protect the pigments from oxidation. The true caro- tinoids, which unquestionably comprise the great majority of yellow 13 14 CAROTINOIDS AND RELATED PIGMENTS and red tinted animal pigments included under the older term lipo- chrome, are characterized by the ease with which they oxidize when in solution or in the solid state. The earlier workers did not recognize, however, that some of the most characteristic properties of these pig- ments are subject to modification even in the earliest stages of oxida- tion. This is particularly true of the color reactions with various reagents, and the spectroscopic properties, which have been used so widely, and many times exclusively, as the basis for the classification of the animal chromolipoids. Confusion in terminology, however, has not been confined to the animal pigments. The chief difficulty regarding the plant carotinoids has been the proposal of names already in use for pigments of obvi- ously different composition and properties. For example, the name xanthophyll, as used by various workers in the field of plant pigments, has been the cause of so much confusion in the nomenclature as to be very disconcerting to many students of this subject. It is not surprising, therefore, that certain investigators have attempted to bring some semblance of order to the confusion by pro- posing one name to cover all the names previously proposed for pig- ments of like or similar properties. A brief history of these attempts with their resulting influence on the nomenclature of plant and animal chromatology may prove of interest at this point. Luteins The first attempt to bring various yellow pigments together under one name is found in Thudichum's (1869) classic paper, in which the yellow pigments found in many tissues of both vegetable and animal origin are grouped under the name "luteine," or luteins. The name was obviously suggested by the fact that the characteristic yellow pig- ment of the corpus luteum on the ovaries of mammals, especially that of the cow, is one of the representatives of the "luteine" pigments. It is doubtful whether Thudichum was familiar with the work of Pic- colo and Lieben (1866), who had crystallized the corpus luteum pig- ment a few years previously and named it luteohamatoidin or hamo- lutein. However, Thudichum mentioned the work of Holm (1867), who isolated the corpus luteum pigment and called it hamatoidin. Thudichum's luteins included, besides the corpus luteum pigment, the yellow pigment of blood serum, adipose tissue and butter, and the yellow pigment of egg yolk. The vegetable pigments in the lutein GENERAL DISTRIBUTION OF CAROTINOIDS 15 group included the pigment of yellow maize, and annatto seeds, the pi.umrnt of the carrot root and of yellow leaves, such as those of the Coh us, and the pigments which characterize the stamens and petals of many flowers. The basis for the classification of these pigments into one group was: (1) their common solubility in alcohol, ether and chloroform, and in albuminous liquids like blood serum; (2) the fact that their solutions showed three absorption bands in the blue, indigo and violet region of the spectrum; (3) the fact that they could be crystallized in the form of rhombic plates; (4) certain common chemical reactions, such as their precipitation by mercuric acetate and mercuric nitrate, and their blue color reaction with nitric acid, when the pigments were in the solid state or in solution in acetic acid; (6) and their affinity for albumin, as in blood serum and the fluid of ovarian cysts, from which the pigments are extracted with difficulty. Thudichum's classification never received wide adoption. In fact, the luteins, as defined by Thudichum, comprise a number of different pigments. Moreover, our present knowledge regarding practically all the pigments which were included in this classification shows that cer- tain of the characteristic lutein properties 1 are specific only for cer- tain individuals of the group. The final abandonment of this classifi- cation appears in the recent application of the name lutein by Will- st litter and Escher (1912) to the specific crystalline pigment isolated by them from the yolks of hen's eggs. This use of the name appears to the author to be illogical both from the standpoint of function and anatomy as well as on other biological grounds. The name lutein obviously suggests the body from which the name was derived, namely, the corpus luteum. The yolk of the egg of the oviparous animal is certainly not related to the corpus luteum either functionally or an- atomically. Moreover, the egg yolk pigment has been demonstrated by Palmer (1915) to be physiologically as well as chemically identical with at least one, and probably a group of the plant pigments which are known as xanthophylls. Egg yolk xanthophyll is, in fact, a true carotinoid, or mixture of carotinoids, and no further designation appears necessary. 'The heretofore inexplicable property of being precipitated by mercury salts, ascribed to the luteins by Thudichum, becomes clear only in the light of Palmer's (1914 c) obser- vations that the albumin with which carotin is sometimes associated in the blood serum of animals is precipitated by mercury salts. It is also possible that Thudichum observed the phenomenon of the adsorption of carotin by mercury salts described by Tswett (1906 b). 16 CAROTINOIDS AND RELATED PIGMENTS Lipochromes Krukenberg (1882k, 1886), a number of years after Thudichum, proposed the name lipochrome to cover all the animal and plant pig- ments which had previously been known as luteins, carotins, zoonery- thrin, tetronenrthrin, chlorophan, xanthophan and rhodophan. The name lipochrome has been widely adopted and due to the very broad basis upon which the name was founded it has been applied to numer- ous plant and animal pigments not mentioned by Krukenberg, or unknown to him. Krukenberg believed that all the pigments which he proposed to designate as lipochromes were associated with fat in their natural state, and the name suggests this supposition as well as their capability of existing in association with fats and oils. It was obviously the intention of the originator of the name lipo- chrome to limit it to pigments of yellow or reddish tints, but the name itself is applicable to pigments of many other colors, such as chloro- phyll and many vegetable dyes of various colors, which have a marked affinity for fat. Numerous workers object to the use of the name lipo- chrome on this account. Kohl (1902a), for example, in his extensive monograph on carotin, objects to designating this pigment as a lipo- chrome because of the numerous cases in which it is known to occur free from fat, and also because he believes that where carotin is actually found associated with fat it is in combination with the fat and not merely in solution. The particular properties by which Krukenberg (1886) proposed to judge whether a pigment should be classified as a lipochrome are, in general, as follows: They are soluble in alcohols (methyl, ethyl and amyl), ether, chloroform, benzene, carbon disulfide, petroleum ether and acetone; in the solid state they are colored blue-green to blue by concentrated sulphuric and nitric acids and generally blue-green with iodine in potassium iodide ; they show two and sometimes three absorp- tion bands in the blue and violet region of the spectrum; they are not destroyed on boiling with alcoholic caustic alkalies; in the solid state they are greenish-yellow, yellow, orange or red, and their solutions are yellow; they are very sensitive to light and readily bleach, the bleached pigments being similar to cholesterol. Subsequent investigations of the lipochromes, using the class char- acteristics defined by Krukenberg, have added very little to our knowl- edge of the properties of these pigments considered as a group, but have served merely to define more closely certain of the criteria enu- GENERAL DISTRIBUTION OF CAROTINOIDS 17 merated. KrukenlxTg believed that all lipochromes should be regarded as composed of carbon, hydrogen and oxygen, and free from nitrogen. At the present time hydrocarbons, such as carotin and its isomers, as well as the oxyhydrocarbons, fulfill all the characteristics of lipo- chromes. Probably wider use has been made of the color reactions with concentrated sulphuric and nitric acids and with iodine in potas- sium iodide than any of the other class characteristics for identifying pigments as lipochromes although many studies have also included spectroscopic observations. Unfortunately the color reactions and spectroscopic properties are subject to greater variation than any of the others upon which the classification is based. The result of the color tests as well as the quality of the color is often influenced strongly by admixture with foreign substances, and this is apparently especially true for the reaction with iodine in potassium iodide. Simi- larly the spectroscopic absorption properties are subject to wide varia- tion as to the position of the bands as well as their definiteness by reason of admixture with impurities, concentration of pigment, and the solvent employed. Lipoxanthins A more recent attempt than Krukenberg's to bring all the known plant and animal pigments with like properties under one name is that of Schrotter-Kristelli (1895a), who proposed to group together all the various plant and animal coloring matters which had previously been known as etiolin, chlorophyll yellow, xanthin, anthoxanthin, lutein, xanthophyll, chrysophyll, carotin, phylloxanthin, phycoxanthin, erythrophyll, solanorubin, lipoxanthin, haematochrom, ehlororufin, bacteriopurpurin, haemolutein, vitellorubin and tetronerythrin. He regarded these pigments as at least an homologous group, if not com- pletely identical, and chose the name lipoxanthin as the most suitable for a general designation. The chief characteristics of the lipoxan- thins, according to Schrotter-Kristelli, arc their affinity for fats, their insolubility in water, their blue color reaction with concentrated sul- phuric acid, their absorption of the violet end of the spectrum, their lack of fluorescence when in solution, and their ease of destruction by light and oxygen. Schrotter-Kristelli believed that the slight dif- ferences in the spectroscopic properties of the various pigments were due to their ease of destruction. According to this author lipoxanthins have been demonstrated to occur in all green leaves, in autumn leaves, in many flowers and fruits, 18 CAROTINOIDS AND RELATED PIGMENTS in arils and roots, in algse lichens, fungi and bacteria; among ani- mals they have been demonstrated in the egg yolk of the sea-spider, in the retina of bird's eyes, in insects, such as Chrysomelidce and Coc- cinellidcB, and in the secretions of various Crustacea-, such as various kinds of Diaptoma, and Maia squinado as well as in still lower forms of animal life. The lipoxanthins are thus seen to be a more or less indefinite group of pigments, whose classification together under one head is secured just as well by the older term lipochrome, which no doubt explains why the proposed term never received wide recognition. Chromolipoids As our knowledge of the so-called lipochromes and lipoxanthins has been extended by exhaustive researches regarding the various indi- vidual representatives from both plant and animal sources the objec- tions which have been raised by various workers to terms such as lipochrome and lipoxanthin seem to be more and more valid. The botanists have been the first to definitely break away from the old terminology as exemplified by the citation from Kohl's monograph. Czapek (1913a) proposes to meet the objections to the name lipo- chrome by calling the pigments chromolipoids. His point of view is that the lipochromes, at least in plants, are to be classed with the lipoids by reason of their many fat-like properties, especially solu- bility, and also because of their widespread occurrence in cells in which lipoids are known to exist. Moreover, the lipochromes, in com- mon with phosphatides and sterols, absorb oxygen very readily. Czapek's terminology has much in its favor, in the opinion of the author. It is at least preferable from many standpoints to the more or less misleading term lipochrome. Carotinoids Attempts have not been wanting to secure uniformity in the termi- nology of the yellow plant pigments. The first yellow plant pigment to be isolated in crystalline form was carotin, the pigment of the root of the cultivated carrot, Daucus carota. At one time the name caro- tin was used to cover all the plant chromolipoids. When it became known that differences existed between many of the so-called caro- tins, the name was changed to carotinen, or investigators spoke of the GENERAL DISTRIBUTION OF CAROTINOIDS 19 "Carotin group." The discovery that carotin itself is a hydrocarbon li-d to the adoption of the name "carotene," as proposed by Arnaud 1 1886). The London Chemical Society favors the spelling "carrotene" for the hydrocarbon. Xopf (1893a, 1895) proposed to distinguish between two groups of carotins, namely, eucarotins (true carotins) which were hydrocarbon in nature and carotiriins, which contained oxygen as well, and formed compounds with the alkali and alkaline earth metals. It should be stated, however, that Zopf used the term carotin synonymously with lipochrome in most of his extensive studies of the pigments of the lower forms of plants and animals. His eucarotins, which were some- times called yellow carotins, unquestionably contained representatives of our present group of xanthophylls whose chemical relation to caro- tin was not discovered until several years later. The carotinins of Zopf were red in color. The belief that they contained oxygen was based on the fact that they appeared to form alkali and alkali earth compounds. Obviously the carotinins are not related to the oxygen- containing xanthophylls, as known at the present time. None of the true carotinoids so far isolated in pure, crystalline state show acid properties like the so-called carotinins. The nature of the compounds which the latter are stated to form with sodium, calcium and barium remains to be determined, as well as their true relation to the caro- tinoids. The carotinins appear to be constituents of both plants and animals, as will appear from a fuller account of them given in Chap- ters III and V. Tswett (1911a), to whose ingenuity we owe much of our knowledge regarding the physico-chemical properties of the chromolipoids, has proposed the term "carotinoide" for the various chromolipoids which are chemically and generically related to carotin. He would desig- nate as carotins all those chromolipoids whose constitution and prop- erties show themselves to be hydrocarbons, and as xanthophylls all those whose constitution and properties show themselves to be oxy- hydrocarbons and which are chemically, as well as generically, related to carotin. Tswett's terminology has been widely adopted. The author has also used it consistently in his own writings. The term carotinoid has the objection, however, that the -oid ending is derived from the Greek ei5i>$, shape, so that strictly speaking the carotinoids are pigments which resemble carotin in form or structure only. As yet nothing definite is known regarding the structure of the carotinoids. The 20 CAROTIN VIDS AND RELATED PIGMENTS word form cannot be restricted to crystalline form, inasmuch as the crystalline form of the carotinoids varies widely depending upon the. solvent from which they separate. As will be pointed out later, how- ever, the carotinoids must of necessity be closely related structurally. Their close chemical relations and the fact that they are invariably found together in chlorophyllous organs support this view. Tswett's terminology has given promise of presenting a very simple solution of the difficulties of nomenclature in connection with the vari- ous red and yellow tinted pigments which conform to the properties of the so-called lipochromes so widely distributed in all forms of plant life. Unfortunately, however, Lubimenko (1914, 1915, 1916) has greatly complicated the system on very inadequate evidence by using the ending -oid for a group of pigments which he believes to corre- spond to each of the definitely known carotinoids. Thus, Lubimenko speaks not only of carotin, xautliophyll, lycopin, etc., but of caro- tinoids, xanthophylloids, lycopinoids, etc., as well. One cannot but express the opinion that our knowledge of the carotinoids in the sense used by Tswett, and followed in this monograph, is not sufficiently extensive to warrant a belief in the existence of numberless interme- diate products. As a matter of fact, the chemistry of the specific indi- viduals of Lubimenko's terminology, namely carotin, lycopin, xantho- phyll and rhodoxanthin, argues against the existence of many plant chromolipoids of the nature of those mentioned. Certainly in view of the fact that there is every evidence to believe that all the xan- thophylls bear the simple relation to carotin that is expressed in their respective formulae, C40H56 and C40H56O2, it seems little short of pre- posterous to assume the existence of a group of "carotinoids" which are oxidation products of carotin and another group of "xanthophyl- loids" which are reduction products of xanthophylL Tswett's terminology, therefore, seems entirely adequate for our present knowledge of the chromolipoids of plant origin. If the chemi- cal and physiological relation of the carotinoids to the yellow animal chromolipoids of the tissues and fluids of the higher mammals and man, and of the egg yolk and bodies of oviparous animals, is a cri- terion of similar relations throughout the entire realm of the animal kingdom, then Tswett's terminology is equally applicable to the yellow and red tinted chromolipoids so widely distributed in all forms of animal life. The probability of such a relationship is, in fact, the basis of the present monograph. GENERAL DISTRIBUTION OF CAROTINOIDS 21 Non-car otinoid Plant Pigments Carotinoids, however, are not the only yellow and orange colored pigments found in the plant and animal kingdoms, which fact must not be lost sight of in the examination of plant and animal products for the presence of carotinoid pigments. Although all plant pigments have a hydrocarbon nucleus there are only a few yellow to red hydrocarbons known which are not carotin- oids. They are the acenaphtylene of Behr and van Dorp (1873), Blumenthal (1874) and Graebe (1893), the di-biphenylenathene of de la Harpe and van Dorp (1875) and Graebe (1892), the fulvenes of Thiele (1900a), the cinnamylidenindene of Thiele (1900b), and the rubicene of Pummerer (1912). Each of these is discussed more fully in Chapter IX, in connection with the probable constitution of carotin. Among the naturally occurring yellow vegetable pigments which contain carbon, hydrogen and oxygen, but which have no relation to the xanthophyll group of carotinoids, two especially well defined groups are known, namely, the xanthones and the flavones. Five xanthones are known, (1) Cotoin, C13H1204, (2) Euxanthone, C13H6(OH)202, (3) Maclaurin, C13H5(OH)50, (4) Datiscetin, or di-methyl-tetraoxyxanthone, C15H1206, and (5) Gentisein, C13H0- (OH)302. The structural constitution of each of these pigments is known. A much larger number of flavones are known, all of which are characterized by the common nucleus, (3-phenyl-benzo-y-pyrone. Many of the natural pigments occur as glucosides and are regarded as the chromogens from which anthocyanins are derived (Wheldale, 1916). Some of the more interesting members with a yellow color are (1) Luteolin, which is not to be confused with the carotinoid, Eteolin, but which is 1,2, 3, 4-tetra-oxyflavon ; and (2) Gossyptin, the yellow dye in the yellow flowers of the Indian cotton (Gossypium herbaceum) , occurring there as a glucoside. No doubt the yellow color of cottonseed meal is due in part to Gossyptin, which can be extracted from cotton flowers with hot alcohol. The pure pigment exists as glistening yellow needles. Besides the xanthones and flavones other yellow pigments are found in plant parts, among which may be mentioned chrysophanic acid, a methyl di-hydroxy anthracene whose solution in alcohol, ether, ace- tone, benzene, chloroform or petroleum ether will dye animal tissues a deep yellow color. Of special interest in this connection is the yellow pigment of the seeds of annatto (Bixa orellana), called bixin or 22 CAROTINOIDS AND RELATED PIGMENTS annatto, which is widely used for the artificial coloring of butter and cheese, and which' has been commonly regarded as a lipochrome and was included by Thudichum among the luteins. The annatto pig- ment does, in fact, correspond in almost every particular with the class characteristics of the lipochromes. It is not entirely unattacked by alkalies, however, and furthermore is decomposable into a number of well known substances, such as m-xylene, m-ethyl toluene, and even palmitic acid. It reduces Fehling's solution even in the cold. Its constitution is unknown, as yet, but its elementary composition cor- responds to the formula C28H3405, according to Etti (1878), or C29H3405, according to van Hasselt (1909). It is thus seen that bixin, while corresponding well to the lipochrome classification, is in no sense a carotinoid. Palmer and Kempster (1919c) have shown that the annatto pigment has no effect on the coloration of the tissues when fed to fowls. Other vegetable coloring matters of a yellow color giving reactions in some cases similar to carotinoids, but of entirely different com- position, are Crocin, and Crocitin, flavones which are found in the petals and pollen grains of the Indian crocus (Crocus sativus), which dissolve in concentrated sulphuric and nitric acids, with a deep blue color, which passes, however, into a brown shade. Another yellow vegetable dye showing a like reaction, although the after shade with the acid reagents is yellow, is the nyctanthin which Hill and Sikar (1907) described a few years ago. The empirical formula for this dye C20H2704, has an interesting resemblance to that of the carotinoids, at least when one doubles the above formula. The yellow, orange, and red colors seen frequently among the fungi of the lichen and mushroom types appear to be due in many cases to pigments of a nature quite different from the carotinoids. Chryso- phanic acid, mentioned above, sometimes occurs among these plants, as well as many other like coloring matters which have been named of Zopf (1889b, 1892b, 1893b) and which other workers have found occurring among the Basidiomycetes. In color and in some of their solubility properties these pigments resemble the carotinoids, and cer- tain of them give a color reaction with concentrated sulphuric acid which is not unlike that regarded as characteristic of the lipochromes. Non-carotinoid Animal Pigments Several yellow pigments are present in animal tissues and fluids which are not to be mistaken for carotinoids. One of these, whose GENERAL DISTRIBUTION OF CAROTINOIDS constitution is unknown, but which is thought to be derived from the blood corpuscles, is hiimatoidin, a yellow crystallizable pigment found in old blood c-xudates, in mummified embryos, and sometimes in the urine and other excreta. First described and named by Virchow (1847) and later by others, its origin as well as chemical properties and crystalline form have been recently studied anew by Neumann (1904, 1905). Holm (1807) thought the corpus luteum pigment was hiimatoidin in his early study of this pigment which has since been shown to be carotin. Another much more widespread yellow animal pigment with cer- tain lipochrome properties is the bile pigment bilirubin. There is less danger of confusing it with carotinoids, however, save with respect to its color, inasmuch as it is a nitrogen containing substance which readily forms salts with the alkali and alkaline earth metals, and has many other properties at variance with those of the carotinoids. Other non-carotinoid pigments exist in animal tissues, but which resemble the carotinoids in color and in solubility in fat solvents. Palmer and Kempster (1919a) have recently encountered such a pig- ment in the carotinoid-free egg yolks of hens raised from hatching on rations devoid of carotinoids, the eggs being produced likewise on xanthophyll-free rations. The small amount of pigment found in the yolks could be extracted by acetone, but hardly at all by ether, was almost entirely saponifiable and failed to respond to characteristic xanthophyll tests. The author finds that a similar pigment can be extracted from the carotinoid-free and apparently colorless "corpus luteum" of the sow, if a sufficient number of these organs are mac- erated and extracted with fat solvent. These cases are cited in order to point out the danger in assuming that all animal pigments of a yellow color are carotinoid in nature. Such a sweeping conclusion cannot be justified. The same statement can also be made, although with less assur- ance, for certain red pigments which appear among the lower animals and birds. These pigments, as indicated, are red in the solid condition but their dilute solutions are usually yellow. They have been studied by certain of the older investigators, such as Kiihne, Maly, Kruken- berg, MacMunn and Zopf and others, and have received various names at the hands of these authors, such as rhodophan, vitellorubin, crustaceorubin, tetronerythrin, lina-carotin (from the Lina species of beetles in which they occur) and diaptomin. The pigments are strik- ingly similar in many respects to the carotinoids, but differ from them 24 CAROTINOIDS AND RELATED PIGMENTS in showing only one wide absorption band at F, and in forming, according to the statement of certain of their investigators, true com- pounds with sodium, calcium and barium. These points of divergence from the carotinoids should be examined in the light of our present knowledge of carotin and xanthophylls before it can be stated with assurance that these pigments are distinct substances. They all cor- respond completely to the class characteristics of the older termi- nology of lipochromes. Summary Red, orange and yellow pigments which have certain simple prop- erties in common arc found in almost all forms of plants and animals. These pigments have been variously classified as luteins, lipo- chromes, lipoxanthins and chromolipoids. These classifications have been based on general properties rather than on composition and are accordingly subject to both error and criticism. As a general class term the name chromolipoid seems to conform more nearly to present conceptions of these pigments as well as to more common usage in connections with substances with fat-like properties. Investigations regarding the composition of the chromolipoids show that a large majority of them are apparently chemically and gener- ically related to carotin, a specific chromolipoid widely distributed in plant and animal tissues. It seems reasonable to believe, therefore, that a great many chromo- lipoids can be classified more specifically as carotinoids, a name pro- posed for them by Tswett (1911a). Two classes of carotinoids are recognized in Tswett 's definition; carotins, whose constitution and properties show them to be hydro- carbons identical or isomeric with carotin; and xanthophylls, whose constitution and properties show them to be oxy-hydrocarbons and which are chemically, as well as generically, related to carotin. Carotinoids are not the only yellow to red colored pigments occur- ring in plants and animals. Many of these non-carotinoids resemble the true carotinoids in one or more properties and some even in com- position. The reader is referred to the text for the detailed discussion of the non-carotinoids and the properties which they have in common with the carotinoids as well as their distinguishing characteristics. Chapter II Carotinoids in the Phanerogams There is no special reason, either physiological or genetical, for considering the carotinoids in the phanerogams and cryptogams sepa- rately, as is done in this and the subsequent chapter. In fact, there appears to be no logical reason for subdividing the plants into groups in connection with the distribution of carotinoids, inasmuch as the pigments appear to be widely distributed in all forms, both chloro- phyllous and non-chlorophyllous, from bacteria to dicotyledons. The subdivision, then, is merely one of convenience. The Pigments of the Carrot The pigment of the carrot root (Daucus carota) was first described by Wachenroder (1826), nearly 100 years ago, and called carotin by him. This serves as the starting point of our knowledge of the prop- erties, as well as the nomenclature of the carotinoids, and this pigment today represents our most typical chromolipoid. For this reason the carrot pigment will be considered first. Wachenroder made an ether extract of the dried macerated roots, or the coagulum obtained on heating the carrot juice. The golden yellow salve-like residue left on evaporation of the solvent was shaken repeatedly with ammonia to separate admixed fatty material, dis- solved again in ether and the ether allowed to evaporate slowly with addition of small amounts of alcohol. Ruby-red quadratic crystals, imbedded in fatty material, were obtained. Wachenroder described the crystals as tasteless and odorless, soluble in alcohol and ether, readily soluble in fats and ethereal oils, but insoluble in acetic acid and alkalis. Vauquelin -and Bouchardat (1830) are credited with the next study of the carrot pigment, but it was a number of years before Zeise (1847) isolated carotin from c/arrot roots in quantity sufficient for analysis. Zeise discovered the ready solubility of the pigment in carbon disul- fide with its characteristic blood red color, as well as the fact that 25 26 CAROTINOIDS AND RELATED PIGMENTS alcohol when added to the concentrated carbon disulfide solution will throw down the carotin in crystalline form. The beautiful, glistening, copper colored crystals were described by Zeise, who also mentioned their insolubility in alcohol and their difficult solubility in ether and acetone. The crystals melted at 168° (+)C. Zeise made the first analysis of carotin and ascribed to it the formula C5H8. He was thus the first to show the hydrocarbon nature of the pigment, but due to the authority of the next investigator (Husemann (1861)), this fact was not proved until Arnaud (1886) made his careful analyses of the carrot pigment. Husemann (1861) pressed the juice from finely grated carrots and then added weak sulphuric acid to the diluted juice, following Zeise's method, throwing down a coagulum which was partially dried and then extracted with hot 80 per cent methyl-alcohol. The residue was then dried completely and extracted with carbon disulfide. Carotin crystals were thrown out of the concentrated carbon disulfide solution by addition of absolute alcohol. Husemann purified the crystals merely by repeated washing on a filter with hot 80 per cent alcohol and finally with absolute alcohol. Husemann described the ruby-red color and velvety appearance of the carotin crystals, and their violet- like odor, which he found was especially noticeable on warming. He noticed the bleaching of the crystals in the air with the complete re- versal of solubility, the bleached crystals being very difficultly soluble in carbon disulfide and benzine, but easily soluble in alcohol and ether. Husemann found that carotin was not precipitated by metallic salts but he observed the green color produced by adding ferric chloride to an alcoholic solution of the pigment. Palmer and Thrun (1916) have recently studied the reaction of ferric chloride on the carotinoids and have found it a most useful test for confirming the presence of these pigments in oils and fats and in various extracts of plant and animal tissues. Husemann was the first to show the unsaturated nature of the carotin molecule, although he regarded the chlorine and iodine deriva- tives which he made as substitution products. Husemann's analyses led him to propose the formula C18H.,40 for carotin and his figures were accepted over those of Zeise. Arnaud (1886) was the next investigator of the carrot chromolipoid. He isolated the pigment from 600 kilograms of carrots by pressing the juice from the grated roots, adding lead acetate to the juice, drying the precipitate in vacuum and extracting it with carbon disulfide. The IX THE PHANEROGAMS 27 dried press cake was extracted similarly and the carbon disulfidc dis- tilled off of the combined extracts. The residue was washed with cold petroleum ether and the pigment purified by crystallization from car- bon disulfide with absolute alcohol and then allowing it to crystallize spontaneously from cold petroleum ether. About 3 grams of crystals per 100 kilograms of carrots were obtained in this way. Arnaud found the bleaching of carotin noticed by Husemann to be an oxidation, analyses which he made of the bleached product show- ing an addition of 21 per cent oxygen. The rapid bleaching of caro- tin solutions was also noticed; and Arnaud pointed out the influence of this fact on the securing of pure preparations for analysis. Arnaud's analyses of freshly prepared crystals showed an average composition of 88.67 per cent carbon and 10.69 per cent hydrogen, definitely proving the correctness of Zeise's assertion regarding the hydrocarbon nature of the pigment. This investigator was also the first to prepare the crystalline iodine derivative of carotin by adding iodine crystals a little at a time to a solution of carotin in anhydrous petroleum ether, maintaining the while an excess of carotin in the solution. It was the elementary composition of this product, con- sidered together with the composition of the pure carotin, that led Arnaud to ascribe to carotin the formula C26H38, and to the iodine derivative the formula C26H38I2. Kohl (1902b) has given us one of the most detailed descriptions of the chemical and physical properties of carotin. His analyses of the crystalline pigment, however, gave unsatisfactory results, as did also his molecular weight determinations, using the cryoscopic method. He therefore accepted Arnaud's formula as representing the correct composition of carotin. Certain of Kohl's detailed descriptions of carotin will be summarized in Chapter IX, where the physical and chemical properties of the carotinoids are considered. Willstatter and Mieg (1907) definitely settled the composition of the carrot carotin at the time they proved its identity with the carotin of the chloroplastid. Their data show a mean ratio of C:H of 1:1.406 for the carrot carotin, for which the simplest formula is C5H7. Molecu- lar weight determinations in CHC13 and CS2, using the ebulloscopic method, show an average of 536, which corresponds exactly with the formula (C5H7)9. or C40HGO, which thus appears to be the correct formula for carotin. Escher (1909) and Willstatter and Escher (1910) have confirmed these results completely. Escher furthermore attempted to ascertain 28 CAROTINOIDS AND RELATED PIGMENTS the structure of carotin using 150 grams of the pigment isolated from carrot meal by extraction with petroleum ether. His efforts led only to the production of amorphous products, all of high molecular weight. The constitution of the pigment thus remains to be determined. Euler and Nordcnson (1908) also isolated carotin from carrots in quantity sufficient for analysis. Their results confirm the Willstiitter formula. The purified crystals from 25 kilos of fresh carrots were found to contain xanthophyll, which was identified by the color of the crystals and their solubility properties. Palmer and Eckles (1914g) have also shown the presence of xanthophyll carotinoids in the carrot root by the Tswett (1906c) chromatographic method of analysis, but van Wisselingh (1915), using microchemical crystalliza- tion methods, did not observe any. xanthophyll crystals. It appears that anthocyanins, also, may accompany carotinoids in the carrot root. Wittmack (1904) has described a red variety of carrots (Daitcus carota, var. Boissieri Schweinfurth) which contains both carotinoids and anthocyanin. Many other investigators have isolated carotin crystals from car- rots without, however, submitting them to chemical examination. According to Schimper (1885) and Courchet (1888) carotin exists in the carrot tissue in crystalline form. Van Wisselingh (1915), how- ever, has shown that the little tubules which Schimper and Courchet observed are not true crystalline forms. The author, also, has never observed any but granular deposits of carotin in sections of the fresh carrot tissue. Carotinoids in Other Roots Very few other roots have been examined for carotinoids although several which are widely used as food are characterized by their yellow color, e.g., the yellow parsnip root (Pastinaca sativa), and the sweet potato (Ipomoea batatas) , especially the highly colored varieties grown in the southern part of the United States, popularly called Yams. The pigments of these' roots should be examined. Formanek (1900) has studied the pigment of the red beet (Beta vulgaris) , and believes that the red pigment changes into a yellow one under certain conditions. The absorption bands of the latter are iden- tical in -position with those of carotin. Formanek's red pigment showed only one absorption band in the yellow part of the spectrum and is undoubtedly an anthocyanin. Its apparent transformation into carotin cannot at present be explained. CAROTINOIDS IN THE PHANEROGAMS 29 Lubimcnko (1914a) has examined the pigment of the yellow turnip root (Brassica Rapa L.) and finds that it contains a yellow pigment soluble in 95 per cent alcohol, but which he was not able to crystal- lize, and also a pigment which closely resembles lycopin, the red pig- ment of the tomato. Spectroscopically the pigment appears to be identical with lycopin but because of a difference in the relative inten- sity of the bands as compared with lycopin, and a greater ease of solubility in alcohol and concentrated acetic acid, Lubimcnko pre- ferred to call the pigment a lycopinoid, a term which the author regards as very unfortunate in view of the more generally accepted use of the terminology -oid as applied to the carotins and xanthophylls. It seems possible that the pigment of the related variety of turnips, namely, rutabaga (Brassica campestris L.) is of the same character. The question of this type of carotinoid in roots deserves confirmation and further study. Carotinoids in the Chloroplastids The tissues of all chlorophyllous plants are characterized by certain specialized bodies, probably protein in nature, of microscopic size, called plastids. In early stages of the plant's development and often in the subterranean parts of the plant after maturity the plastids are colorless. They are then called leucoplastids. More commonly they develop green pigments, chlorophylls, when the plastids are called chloroplastids, or chlorophyll granules. The chlorophylls in the chloroplastids are always accompanied by carotinoids of both types, namely, carotin and xanthophylls. Investigations regarding these yellow chromolipoids in the chloro- plastids apparently did not begin until the observation of Fremy (1860) that a yellow pigment can be obtained from green leaves by allowing strong HC1 and ether to act upon the residue from the alco- holic extract, or by similar treatment of the precipitate thrown down • from the alcoholic leaf extract by A1(OH)3. In this procedure the ether took on a yellow color, the pigment of which Fremy called phylloxanthine, leaving a blue pigment, which he called phyllocyanine, in the aqueous acid layer. Fremy believed that his phylloxanthin pre-existed in the leaves. It is now quite certain that Fremy's phylloxanthine was a mixture of some of the natural carotinoids of the leaf with an acid decompo- sition product of chlorophyll, a view which was expressed first by Stokes (1864). The name phylloxanthin is, in fact, at present re- 30 CAROTINOIDS AND RELATED PIGMENTS served for a product of rather indefinite composition which results from the action of acid on chlorophyll b (Tswett, 1907, 1908b) . In a later study of methods of isolation of phylloxanthine, however, Fremy (1865) undoubtedly obtained much more valid proof of the existence of yellow pigments associated with chlorophyll although he regarded the pigment which he isolated as the same yellow phylloxan- thine isolated by the ether-HCl method. He found that a careful addition of Mg(OH)2, or A1(OH)3 to alcoholic chlorophyll solutions carried down the green pigment only, leaving the yellow pigment in solution. Ca(OH)2 and Ba(OH)2 gave similar results, but the best procedure with the last named reagent was to add an excess, which threw down all the pigments, from which the phylloxanthine (caroti- noids) could be extracted with alcohol. Especially interesting was Fremy's observation that when his chlorophyll was saponified with strong bases, alcohol took up the yellow "phylloxanthin" from the residue, and yellow plate-formed or reddish colored prismatic crystals, soluble in alcohol and ether, could be obtained from this solution. The red crystals were described as being very much like crystals of 'potassium dichromate, and having a strong coloring power. It would appear as though Fremy succeeded in obtaining for the first time crystals of carotin, and possibly xanthophyll also, from green plants. Fremy's observations precipitated a lively interest in the subject of yellow pigments in the chloroplastids which resulted in a number of investigations during the succeeding years, some quite independent of the others. These investigations seem to fall quite naturally into several groups. The first of these was a series of studies confirming the presence of yellow substances accompanying chlorophyll through the development of suitable methods for their separation. Separation of Yellow Pigments from Chlorophyll Stokes (1864a) is to be credited with first discovering a method for separating the actual yellow pigments accompanying chlorophyll and for recognizing the existence of distinct green and yellow constituents in the plastids. This investigator states, "I find the chlorophyll of land-plants to be a mixture of four substances, two green and two yellow, all possessing distinctive properties"; and later referring to phylloxanthine he states, "When prepared by removing the green bodies by A1(OH)3 and a little water, it (phylloxanthin) is mainly one of the yellow bodies, but when prepared by HC1 and ether, it is a CAROTINOIDS IN THE PHANEROGAMS 31 mixture of the same yellow body (partly, it may be, decomposed) with the product of decomposition by acids of the second green body." Stokes never published his method of separation in detail but he gives a hint of its character in a paper in another publication (1864b), in which he states in a discussion of the advantages of a partition between solvents for the separation of various substances, "Bisulphide of car- bon in conjunction with alcohol enabled the lecturer to disentangle the colored substances which are mixed together in the green coloring matter of leaves." Stokes was not the only one of the earlier investigators to express the belief that Fremy's pigments were decomposition products. Filhol (1865) also reached this conclusion, as did Askenasy (1867). Filhol (1868) a little later noticed that it is possible to remove the green constituent of crude alcoholic chlorophyll solutions by treating them with animal charcoal insufficient to completely decolorize the solution. A yellow colored solution remained on filtering off the bone-black, the color of which Filhol believed to be due to a pre-existing pigment or pigments associated with the green one. C. A. Schunck (1901), many years later, employed this method of obtaining his xanthophyll group of pigments free from chlorophyll. Schunck's "xanthophylls" included carotin also, so that Filhol's observation was in reality of much more importance than he realized. Timiriazeff (1871), studying Fremy's phylloxanthin, also found that alcohol alone would extract the yellow pigment from the barium com- pound thrown down from alcoholic leaf extracts by an excess of Ba(OH)2. He preferred to call the yellow pigment xanthophyll, the name previously employed by Berzelius (1837a) — from xavSo?, yellow and (f>v\\ov, leaf — for the yellow pigment which he extracted from the yellow autumn leaves of the pear tree (Pyrus communis). Sorby (1871b) employed the same term for a group of yellow and orange pig- ments which, with chlorophyll, he believed caused the green color of leaves, and were represented as types by pigments which could be extracted from carrots by CS2. Notwithstanding the previous observations of Fremy, Stokes, Filhol, Timiriazeff and Sorby, credit is given in most quarters to Gregor Kraus (1872a) for making the first actual separation of the pigments of leaf extracts from one another. Kraus' method is frequently re- ferred to as an "ausschiittlungs" method, for he shook the green alcoholic leaf extracts with benzene and observed that the benzene had extracted the green pigment leaving the alcohol layer a beautiful 32 CAROTINOIDS AND RELATED PIGMENTS yellow. Kraus named the green pigment cyanophyll and the yellow pigment xanthophyll. From our present knowledge of the relative solubility properties of the carotinoids which have been built very largely upon Kraus' observations, it is obvious that his xanthophyll pigment was composed almost entirely of the xanthophyll carotinoids, although traces of carotin may have been present also. Kraus found that the absorption bands of his xanthophyll fraction in the blue part of the spectrum corresponded with Bands V and VI of the leaf extract. The residue from Kraus' xanthophyll fraction gave a deep blue colora- tion with concentrated sulphuric acid, and the xanthophyll solution itself bleached very quickly in the sunlight. Several studies of the leaf pigment, using the Kraus procedure, soon followeH. Konrad (1872) observed that the separation of xanthophyll from chlorophyll by benzene was effective only when an alcohol of about 70 per cent (by volume) strength was employed. Treub (1874) substantiated the necessity of using weaker alcohol for the benzene separation, and found that CS, was effective when the chlorophyll extract was in strong alcohol. Ccmpert (1872) found that linseed oil could be used in place of benzene and Wiesner (1874a, b) found that a number of vegetable oils (linseed, walnut, poppyseed, olive) and ethereal oils (turpentine oil, rosemary oil, oil of wintergreen) and also carbon disulfidc could be employed. Wiesner used 85 per cent (by volume) alcohol and benzene, boiling at 92-94° C., for the regular Kraus separation. Equally good results were obtained with toluene and xylene, or mixtures of these with benzene. Wiesner even suc- ceeded in shaking the green chlorophyll out of alcoholic leaf extracts with dilute egg albumin. Especially important was his observation that dilute ammonia or caustic alkali solutions acting on the residues from alcoholic leaf extracts would take up most of the yellow color leaving behind the chlorophyll, mixed with a little xanthophyll. This observation, pointing to the resistance of the yellow chromolipoids to alkalis and the attacking;; of chlorophyll by the same reagents, was later developed by Hansen (1884a) and is still widely used for the separation of carotinoids from chlorophyll. Hansen boiled young wheat plants in water for one-half hour, dried the product, extracted with cold 96 per cent alcohol in the dark, concentrated the extract to one-eighth volume, saponified with NaOH, diluted the soap solution with water, and salted out the soap with NaCl. The green soap thus obtained was extracted with petroleum ether, which gave a yellow extract which Hansen called "chlorophyll gelb," the residue being CAROTINOIDS IN THE PHANEROGAMS 33 called "chlorophyll grim." Carl Kraus (1875) observed the same facts when he found that benzene extracts a yellow color from alco- holic leaf extracts made strongly alkaline with KOH. He called the yellow pigment xanthin and regarded it as a decomposition product of G. Kraus' xanthophyll. Possibly following the hint given by Stokes (1864b), Sorby (1873) developed a separation method for the yellow and green constituents of a number of tj^pes of plants using alcohol and carbon disulfide. Sorby named five members of a "xanthophyll" group of yellow pig- ments as well as two chlorophylls, but pure pigments could not have been obtained in most cases, since the methods which he employed will not give a true separation of the various carotinoids of the chloro- plastids. Of the various pigments named by Sorby the "yellow chlorophyll" is obviously a xanthophyll mixed with some chlorophyll, and the "orange xanthophyll" is for the most part carotin. Sorby's "xanthophyll" and "yellow xanthophyll" are the only true xantho- phylls, the former being in all probability a mixture of what are now called a and a' xanthophyll, while the "yellow xanthophyll" appar- ently consisted almost wholly of our present (3 xanthophyll, which is characterized, as Sorby found for it, by the development of a blue color when its alcoholic solution is treated with HC1. Crystalline Carotinoids from Chloroplastids The second group of studies relating to the yellow pigment in the chloroplastids deals with isolation of crystals of carotinoids, and ter- minated with the isolation and analysis of one of the pigments and the discovery of its identity with the carotin of carrots. The various studies were, for the most part, independent of each other and accord- ingly resulted in the proposal of several different names for the chloroplastid chromolipoids. Fremy's (1865) observations regarding crystalline chromolipoids have already been mentioned. He apparently regarded the crystals as related to his phylloxanthin, as no special name was proposed by him for the crystalline pigment. Hartsen (1873a), however, several years later, observed golden-red crystals in the deposit from the spon- taneous evaporation of an alcoholic extract from green leaves. He called the pigment chrysophyll, a name previously applied by Sorby (1871b) to a group of water-soluble pigments from autumn foliage, and later (1875) described a method for purifying the crystals by 34 CAROTINOIDS AND RELATED PIGMENTS washing away the fat and chlorophyll with petroleum ether, taking up the pigment in alcohol and recrystallizing. Hartsen regarded his chrysophyll as probably identical with xanthophyll (G. Kraus) and as existing together with chlorophyll in the leaf. According to Will- statter and Mieg (1907) Hartsen's chrysophyll was probably a xan- thophyll in the sense that this name is u^'d at the present time, but most writers have regarded it as identical with carotin. Bougarel (1877), a little later, isolated red crystals from the alcoholic extract of peach and sycamore-tree leaves. He described the insolubility of the red, green reflecting crystals in alcohol and ether, and their solu- bility in chloroform, benzene and carbon disulfide, as well as the rose color of the solution in the last named solvent. Notwithstanding his familiarity with Hartsen's chrysophyll, which he mentions, Bougarel regarded his pigment crystals as a new substance and unfortunately proposed the name erythrophyll for it, which had already been given by Berzelius (1837b) many years before for the red, alcohol soluble pigment which he isolated from red cherries (Prunus cerasus), black Johannis berries (Ribus nigmm) and the red autumn leaves of vari- ous plants, and which was also used by Sorby (1871, 1873) for a group of water-soluble pigments. The erythrophyll of Bougarel is unques- tionably to be regarded as carotin. Dippel (1878) made a careful study of the absorption spectra of G. Kraus' xanthophyll and cyanophyll and the products of the action of KOH and acid on the pigments prepared according to the Kraus method. He found that yellow pigments could be prepared in each case but that the absorption spectrum of the yellow pigment from the acid treatment of cyanophyll was entirely different from the spectra of the yellow pigment from the alkali treatment of both xanthophyll and cyanophyll. Dippel proposed the name xanthin (compare C. Kraus (1875)) for the yellow pigment obtained from Kraus' xanthophyll and cyanophyll on treatment with alkali and extracting with alcohol, and regarded it as the true yellow constituent of chlorophyll. The absorption bands of DippePs xanthin obtained by alkali treatment of Kraus' benzene-cyanophyll layer lay at 490- 456|4i and 455-435(.iu, while the bands of Kraus' xanthophyll, as measured by Dippel, lay at 483-460j.iu and 446-433^1. These meas- urements correspond almost exactly with those of carotin and xan- thophyll, respectively, as known at the present time. Dippel's xanthin is to be regarded, therefore, as composed of carotin for the most part. Borodin (1883) made one of the most striking contributions to our CAROTINOIDS IN THE PHANEROGAM* 35 knowledge of crystalline carotinoids accompanying chlorophyll in the chloroplastids. Two groups of pigments were described by him, one characterized by their slight solubility in alcohol and great solubility in benzine, corresponding with one of the properties of carotin, the other group being characterized by their ease of solubility in alcohol and their slight solubility, in benzine, which corresponds with one of the most distinguishing properties of the xanthophylls. No names were proposed by Borodin for his crystalline pigments but lie described in detail the crystal forms and certain properties of two carotins and two xanthophylls. One of the carotins formed orange-red rhombic crystals (he obtained these from the alcoholic extract of Spirogyra) and the other bright yellow needles with a strong violet or rose-red nuance. Of the two xanthophylls one formed straw yellow, ribbon- like scales or dark brown, crooked, branching rods, and the other golden-yellow "navikeln," an English synonym for which the author has not been able to find. The latter were observed especially clearly by Borodin in extracts from parsley (Petroselinum sativum). Borodin regarded the red and violet tinted, benzine-soluble group as widely distributed in all chlorophyllous plants, the red forms being identical with Bougarel's erythrophyll. The alcohol soluble forms were not regarded by Borodin as being so widely distributed, especially the pigment forming the golden-yellow "navikeln." With the exception of the red rhombic-formed crystals in the benzine-soluble group Borodin's crystal forms do not correspond with the carotins and xanthophylls which have since been isolated in pure form by various investigators so it is not known whether they represent forms which were modified by the solvents employed or isomeric carotin and xanthophyll carotinoids not yet isolated in quantity. The latter pos- sibility is not to be disregarded in view of the various yellow chromo- lipoids which are revealed in chloroplastids using Tswett's (1906c) chromotographic analytical procedure. Guignet (1885) observed orange crystalline material from extracts obtained by a method similar to that used by Dippel for xanthin. Alcoholic leaf extract in strong alcohol was agitated with one-tenth its volume of petroleum ether (Sachsse (1877) introduced the use of petroleum ether instead of benzene in the Kraus separation) and the green petroleum ether extract agitated with a solution of NaOH in 95 per cent alcohol, leaving a yellow solution from which the crys- talline material was obtained. The pigment was no doubt carotin although no name was proposed for it by Guignet. 36 CAROTINOIDS AND RELATED PIGMENTS It remained for Arnaud (1885), however, to first recognize and prove the identity of the orange-red crystals apparently first observed by Fremy (1865) and later called chrysophyll, erythrophyll, xanthin, etc., with the carotin from carrots, which had been known and studied for 60 years. Arnaud (1885) first observed the identity in form and properties of carotin which he isolated from carrots and the red rhombic crystals which he isolated from spinach leaves (Spinachia oleracea and glabra) , mulberry leaves (Moms alba), the leaves of peach (Persica vulgaris) and sycamore (Acer pseudoplatanus) trees, and the leaves of the English ivy vine (Hedera helix) , as well as from pumpkins (Cucurbita pepo). In a succeeding paper Arnaud (1886) proved this identity by his analyses of the crystals obtained from carrots, to which reference has already been made, the results leading to the proposal of the formula C2GH38 for the pigment. Arnaud did not make any analyses of the apparently identical crystals which he obtained from leaves, so that strictly speaking the final proof of the identity of the crystals was not furnished until Willstatter and Mieg performed their comparative analysis many years later (1907). How- ever, following Arnaud, investigators with few exceptions adopted his terminology and called the red crystalline pigment carotin which could be isolated so generally from chlorophyll forming plants, as well as many fruits and seeds, and from cryptogamic forms. Immendorff (1889), in fact, soon after Arnaud's work, isolated carotin from barley and rye leaves and submitted the crystalline pigment to analysis. His data corresponded best with Zeise's older formula, C5H8, but he pre- ferred to accept the Arnaud formula because it appeared to be sub- stantiated by Arnaud's analysis of the iodine derivative of carotin. Willstatter and Mieg, however, starting with 100 kilos of dried nettle (Urtica) leaves, isolated carotin in sufficient quantity to estab- lish for it the correct formula, C40H56. Their analyses gave the average composition of 89.29 per cent carbon and 10.53 per cent hydro- gen as compared with the theoretical values 89.48 and 10.52 per cent carbon and hydrogen, respectively. Plurality of Yellow Pigments in the C hloroplastids The next group of investigations dealing with the yellow pigments of the chloroplastids had to do with the question whether more than one yellow pigment is a constant accompaniment of the chlorophyll. This question brings us up to the present time for notwithstanding the CAROTINOIDS IN THE PHANEROGAMS 37 fact that two crystalline carotinoids in the xanthophyll group have now been isolated from plant forms and their composition determined, the isolation of other known members of this group of pigments still remains to be carried out and their composition and relation to the known members determined. Stokes (1864a) is to be credited with the first suggestion of the presence of more than one yellow pigment in chloroplastids, but in spite of the various yellow pigments isolated by Sorby (1873), using Stokes' carbon disulfide procedure, this method could not have led to a true isolation of the various members of the carotinoid pigments which are recognized today. The observation of Dippel (1878) that besides Kraus' xanthophyll a yellow pigment accompanied the cyano- phyll in the petroleum ether layer, could have led to the discovery of the actual existence of two groups of carotinoids. Borodin (1883), however, first demonstrated the existence of more than one yellow chromolipoid when he obtained various forms of crystals from green plants. As already pointed out, these crystals naturally fell into two groups according to their solubility properties, one group, to which Borodin recognized the erythrophyll (carotin) of Bougarel belonged, being very soluble in benzine (petroleum ether) and dif- ficultly soluble in alcohol, and the other group being easily soluble in alcohol but dissolving with difficulty in benzine. These observations of Borodin's are the basis of the classification of the carotinoids which prevail at the present time, considerably extended and supported, of course, by other chemical and physical properties; they also furnish the basis for the separation of the carotinoids into the two groups now recognized, namely, the carotin and xanthophyll groups. This sys- tem of classification is the only logical one, as has been pointed out by Tswett and proven by the chemical analyses of members of each group by Willstatter and his co-workers. Before reviewing the history and evidence in favor of this classification and the proof of the existence of individuals in the group, it should be stated that the plurality of yellow chromolipoids in chloroplastids has been recognized by other investigators who have proposed other systems of classification. These latter studies will be reviewed first. Tschirch found proof of the existence of more than one yellow chromolipoid in his well-known series of spectroscopic studies of chlorophyll. In his first papers Tschirch (1884, 1885, 1887) con-, sidered the yellow constituents of the chloroplastid to be erythrophyll (adopting Bougarel's terminology) and a. group of five xanthophylls, 38 CAROTINOIDS AND RELATED PIGMENTS which he called a, p, y, 8 and e xanthophyll, respectively, although he concluded from the fact that each pigment showed two absorption bands showing no significant spectro-analytical differences that they were probably identical substances. Kohl (1902c) later considered all these xanthophylls to be carotin, but a comparison of the meas- urements of the spectrum bands of y and 8 xanthophyll as given by Tschirch (1885) would indicate that the former may have been due to carotin, but that the latter was undoubtedly one of the xantho- phylls as now recognized. In his next paper Tschirch (1896) regarded all the yellow chromo- lipoids as xanthophylls and distinguished between two, one of which was obtained in metallic glistening crystals, which he called xantho- carotin. This pigment showed three beautiful absorption bands, the measurements of which correspond with those now recognized for xanthophyll. The other xanthophyll could not be obtained in crystal- line form and its solutions were characterized by showing no absorp- tion bands, only end absorption of the violet and ultra-violet. Tschirch used fresh grass as the source of his material for this study. In his most recent paper T.schirch (1904) turned hi* attention to a comparison of his xanthocarotin and xanthophyll with the carotin from carrots. Spectroscopic absorption properties only were considered. There has always been a question in the author's mind as to which group of carotinoids Tschirch'* xanthocarotin belongs. Tschirch him- self considered that it might be identical with the carotin from car- rots, inasmuch as the absorption spectra of the crystalline pigment which he isolated from carrots and that of his xanthocarotin from grass were identical. Kohl (1902d) believed that Tschirch's xantho- carotin was carotin contaminated with phytosterin, and Tswett (1911a) apparently also regarded the pigment as a carotin although he recognized the absorption spectra of Tschirch's carrot carotin did not correspond exactly with the measurements given by other inves- tigators. Willstatter and Mieg (1907) also regard the xanthocarotin as carotin, but with these views the author is not in agreement on the following grounds. The author believes that Tschirch's carotin crys- tals from carrots were xanthophyll, not carotin, for he obtained them merely by spontaneous evaporation of an ether extract of sugar-free carrots, which would be more likely to yield xanthophyll crystals than carotin. Moreover, the crystals had the reddish yellow color and steel blue reflection described by Willstatter and Mieg for crystals of xantho- phyll. In addition the absorption spectra of these crystals correspond CAROTINOIDS IN THE PHANEROGAMS 39 exactly with (hose of xanthophyll, not carotin, as the following table shows: Carotin in alcohol Xanlhojih t/ll ni alcohol (Willstatter :md Stoll (1913n) (WilUiiiicr and Micj? (1907) 1—492-476 nn 1—488-471 mi 11—459-445 " 11—454-440 " 111—4307419 " . Ill— 425-420 " Carrot carotin in alcohol Xanthocarotin in alcolml (Tschirch) (Tschirch) I— 487H170 mi 1—485-468 nn 11—457-439 " 11—455-438 " III— 429-417 " III— 430-418 " Much discussion has also resulted from the statement made by T-chireh in the paper under consideration that he was able to observe the transformation of xanthocarotin into xanthophyll. As a matter of fact Tschirch observed merely that certain impure xanthocarotin solutions lost their absorption bands without losing their color appre- ciably. In view of the fact that the so-called xanthophyll of Tschirch showed no absorption bands but merely end absorption, he conclude* 1 that xanthocarotin readily changes over into xanthophyll, a most sweeping conclusion from such indefinite evidence. The author has observed many times that impure solutions of carotinoids lose their spectroscopic absorption bands in the earliest stages of decomposition with little or no loss in color of the solutions. A somewhat different system of yellow chromolipoids was proposed by Schunck (1899, 1901, 1903) in his series of papers. He depended largely upon the spectroscopic absorption properties of the pigments for their differentiation, as did Tschirch, and in his later studies upon the action of certain chemical agents upon the absorption bands. It may be stated of Schunck's work, faulty as it was in certain respects, particularly in his adoption of Sorby's method for separating the various yellow coloring matters by carbon disulfide, that he has given us some of the most beautiful spectro-photographs of the carotinoids that exist in the literature. Schunck accepted from the outset that more than one yellow chromolipoid was present in the chloroplastids. Inasmuch, however, as he modified his views somewhat regarding the number and nomenclature of these pigments during the course of his studies his final views only will be discussed. Schunck proposed to call all the yellow pigments accompanying chlorophyll xanthophylls, the chief member of the group being chryso- phyll, thus adopting Hartsen's terminology for carotin in spite of the fact that Schunck not only referred to Arnaud's work but confirmed 40 CAROTINOIDS AND RELATED PIGMENTS it from a spcetroscopic standpoint. Besides chrysophyll, the only member of the "xanthophylls" which he was able to obtain in crystal- line form, Schunck separated two other xanthophylls from green leaves by shaking the alcoholic chlorophyll-free1 solution with successive equal portions of carbon disulfide, each volume of carbon disulfide being equal to about one-lialf the volume of the crude solution experi- mented upon. This was continued until no more color was extracted, three or four extractions being sufficient as a rule to accomplish this result. With the exception of the first carbon disulfide extract, which con- tained the crystallizable chrysophyll as well as one of the xantho- phylls, Schunck erroneously believed that the various carbon disul- fide fractions represented more or less pure solutions of individual xanthophylls with varying degrees of relative solubility in alcohol and carbon disulfide. The various carbon disulfide fractions were now allowed to evapo- rate spontaneously, the residue was taken up again in alcohol and the spectroscopic alt-orption bands photographed. The effect on these bands of adding rmuviitrated HC1, HNO:!, H2S04, H202 and nascent hydrogen was studied, as well as the effect of these reagents on the color of the alcoholic solution. Certain marked differences were observed with the various fractions. The first fraction from green leaves contained, besides ehrysophyll, a pigment which Schunck called L. xanthophyll, whose spectroscopic absorption bands differed from those of chrysophyll by being shifted only slightly towards the ultra-violet and whose solution, like chryso- phyll, changed to a green tint before fading on addition of HC1 or HN03, the absorption bands disappearing. The subsequent rarbon disulfide extracts contained a second xan- thophyll, called B. xanthophyll, which differed from the first in two respects, (1) the absorption bands (Schunck observed three distinct bands for all his "xanthophylls") were shifted slightly more towards the ultra-violet, (2) the effect of acids on the alcoholic solution was to produce a brilliant green color which gradually changed to a beau- tiful peacock blue, then purple, and gradually bleached entirely. Especially striking was the observation that the addition of ammonia 1 This was obtained by one of two methods, either by adsorbing the chlorophyll on animal charcoal, which does not remove the "xanthophylls," according to Schunck, or by saponifying the alcoholic leaf extract and extracting the soap with ether, the latter taking out the yellow pigments. After evaporation of the ether the pigments were taken up in alcohol for the "xanthophyll" separations. CAROTINOIDS IN THE PHANEROGAMS 41 to the blue solution restored the original yellow color of the solution, although less intense, the blue color reappearing on acidifying again. Sorby (1873) mentioned this reaction for his "yellow xanthophyll." The author 2 has observed that the change from yellow to blue and vice versa can be repeated apparently indefinitely with one of the xanthophylls obtained from plants by Tswett's chromotographic method. In his last paper Sclmnck found evidence of the existence in flowers of still another xanthophyll, called Y. xanthophyll, with properties similar to B. xanthophyll, except that it was much less readily extracted from alcohol by carbon disulfide and was accordingly found in the alcohol after the carbon disulfide extractions. Schunck found no evidence of the existence of Y. xanthophyll in his leaf extracts. Kohl (1902e) attempted to harmonize the views of Tschirch and Schunck as well as his own belief that carotin is the principal pigment in the chloroplastids. He recognized the difference between carotin and the xanthophyll proper of Schunck, but apparently did not recog- nize the existence of several of these xanthophylls, as proposed by Schunck. Kohl recognized also the existence of the xanthophyll of Tschirch, which showed no absorption bands, and believed, like Schunck, who proposed no name for the pigment, that it could be extracted from the chloroplastids by hot water, as well as by alcohol. Kohl, therefore, proposed to call Schunck's xanthophyll a xanthophyll and the xanthophyll of Tschirch /? xanthophyll, and expressed the belief that carotin and these two xanthophylls comprised the yellow pigments in the chloroplastids. We will now return to a consideration of the investigations leading up to the classification of the carotinoids which prevails at the pres- ent time. Following Borodin, Monteverde (1893) found that the yellow pigments accompanying chlorophyll can be divided into two groups according to their relative solubility in alcohol and petroleum ether, and he was the first to show that this fact offers a very simple means of separating the pigments from each other. Using the pro- cedure of Fremy and Timiriazeff, Monteverde precipitated the chloro- phyll from an alcoholic leaf extract with an excess of Ba(OH)2, which carries down with it the carotinoids also, and extracted the yellow pigments from the precipitate with alcohol. Petroleum ether and a few drops of water were added to this yellow solution, and the mix- ture shaken. The liquids soon separated into two layers, each con- J Unpublished observation. 42 CAROTINOIDS AND RELATED PIGMENTS taining a yellow pigment with distinguishing characteristics. Monte- vcrde found the pigment in the upper petroleum ether layer to be spcctroscopically as well as in other respects identical with carotin and accordingly called it carotin. The pigment remaining in the alcohol layer, on the other hand, was found to be different in many respects and was called xanthophyll, following Gregor Kraus' termi- nology. Monteverde regarded it as not unlikely that this "xantho- phyll" itself consisted of two yellow pigments. In order to separate completely the carotin and xanthophyll the petroleum ether and alcohol layers after separation were shaken with fresh quantities of alcohol and petroleum ether, respectively. On spontaneous evapora- tion of tin- alcoholic xanthophyll solutions Monteverde obtained crys- tals which corresponded exactly in form with the "strohgelben Krys- tallen" described by Borodin. There is some doubt, however, whether the pale yellow crystals observed by Monteverde, and the similar ones observed by Borodin, were actually xanthophyll. Reinke (1885) sev- eral years previously obtained yellow platelets on evaporation of alco- holic solutions of the yellow chloroplastid pigments and found them to be merely phytosterol or a mixture of sterols colored with pigment. It is likely that Monteverde was misled by the same phenomenon, as the great solubility of xanthophyll in alcohol undoubtedly prevents the formation of crystals when one is dealing with the very small quantities of pigment present in Monteverde's solutions. Monteverde, however, described very clearly the difference between the absorption spectra of carotin and xanthophyll, as did Schunck, some years later, between chrysophyll (carotin) and the L. B. and Y. xanthophylls which he separated. Monteverde also described the green coloration, chang- ing to a blue on addition of concentrated HC1 to the alcoholic xan- thophyll solution, a reaction which also characterized the B. and Y. xanthophylls of Schunck, as mentioned in an earlier paragraph. Tswett was very quick to recognize the importance of Monteverde's work and the significance of the Kraus method of separation in indi- cating the existence of alcohol-soluble xanthophylls in contrast with benzine-soluble carotin. This investigator's keen appreciation of the significant properties of carotin and xanthophylls is what makes pos- sible today the extension of our knowledge of the distribution of these pigments in all forms of plant and animal matter. Tswett's important observations are accessible to us in a series of papers (1906a, b, c, 1911a) from 1906 to 1911. The last paper is more of the nature of a summary but by reason of its clear-cut statements it CAROTINOIDS IN THE PHANEROGAMS 43 may well serve today as our best laboratory outline for working with the class of pigments with which this monograph deals. It was in this paper that Tswett proposed the nomenclature for the carotinoids which has been adopted in this monograph. Tswctt's most important contribution to the subject, from an inves- tigational standpoint, was on certain physico-chemical properties of the pigments. He showed (1906b) that the various colored constitu- ents of the chloroplastids, when carefully obtained in certain solvents by methods which avoid the action of plant acids, exhibit very char- acteristic adsorption coefficients towards finely divided materials, such as CaCO:!, inulin and sucrose, as well as many other inert materials which are insoluble in the solvent employed and which can be obtained in a finely divided state. This exceedingly interesting phenomenon is no doubt due to the fact that the various green and yellow chromo- lipoid constituents of the chloroplastids exist in organic solvents in colloidal aggregates of various sizes, the larger colloidal particles being the more strongly adsorbed, and some, like carotin, which is not adsorbed at all, existing in true solution. Tswett found petroleum ether, the carotin solvent, to serve best for the study of these prop- erties, although carbon disulfide was also very useful because of the brilliant color which all the chloroplastid pigments show in this sol- vent, and also because the xanthophylls are especially well differen- tiated in this solvent. This latter fact is no doubt closely related to Schunck's (1903) observations regarding the relative solubility of xanthophylls in carbon disulfide by which he believed he was able to separate them from one another by a shaking-out method. Schunck's observations were near the truth but can not be compared in accuracy with the method of separation which Tswett was able to develop from the colloidal properties of xanthophylls. Tswett hit upon a very ingenious method indeed of applying the results of his study. He filtered the moisture-free petroleum ether solution of the mixed chloroplastid pigments for carbon disulfide solu- tion) through a column of perfectly dry CaC03, packed as tightly and evenly as possible in a glass tube, and found that the various pig- ments differentiated themselves according to their adsorption affinity (colloidal aggregation) for the CaC03. The resulting chromatogram (as Tswett proposed to call it) presented a most surprising picture of the chloroplastid pigments, which is strikingly similar in effect, if not in principle, to the well-known Liesegang phenomena. By applying this chromatographic method of analysis to petroleum 44 CAROTINOIDS AND RELATED PIGMENTS ether and carbon disulfide solutions of the chloroplastid pigments from plantain (Plantago) and dead nettle (Lainium album) leaves Tswctt has shown that at least three and possibly four xanthophylls accom- pany carotin. He has provisionally designated these a, u', a", and (3 xanthophylls, respectively. Tswett has characterized these pigments further, as follows: Xanthophyll a. This pigment is least adsorbed by the CaC03 and is closest to carotin in this respect, which is not adsorbed at all. Its adsorption zone is the lowest in the column of the xanthophyll zones and has an orange-yellow color when carbon disulfide is the solvent. It is hypophasic in the Kraus separation, i.e., remains in the alcohol layer. It shows three well marked absorption bands, the first two of which, in alcohol or petroleum ether solution, lie at 485-470|i(j, and 455-440^1. Its alcoholic solutions are merely bleached on addition of con. HC1. Xanthophylls a' and a". These pigments lie very close together in the column but above the zone of xanthophyll a. In CS2 their zones are yellow. They are similar in properties to xanthophyll a, i.e., in the Kraus separation and spcctroscopically, but their absorp- tion bands are shifted slightly towards the violet. The effect of HC1 on the alcoholic solutions is not mentioned but the author (1914g) has found that for xanthophyll a , at least, no color reaction is produced. Xanthophyll /?. This pigment shows the greatest adsorption affinity for CaC03 (exists in the largest colloidal aggregates) and comprises the highest yellow zone in the column. This pigment is hypophasic in the Kraus separation like the other xanthophylls, but may be dif- ferentiated from them by the fact that its alcoholic solution gives a blue color on addition of con. HC1, and also by the fact that its absorption bands are shifted perceptibly towards the violet from those of xanthophylls a, a', and a", the first two bands lying at 475-462|j|i and 445-430(i(.i, when in alcoholic solution. The xanthophyll (5 of Tswett appears to be identical with th.e "yellow xanthophyll" of Sorby and the Y. xanthophyll of C. A. Schunck, but bears no relation what- ever to the xanthophyll (3 of Kohl. According to Tswett (1908b) the latter is not a xanthophyll at all, in fact does not exist in the plant but is merely a post-mortem decomposition product derived from colorless chromogens whose alkali salts are yellow and which assume a dark color on oxidation. The relative solubility properties of carotin and xanthophylls as CAROTINOIDS IN THE PHANEROGAMS 45 exhibited in the Kraus separation indicated to Tswctt (1906a> a fun- dament ;il chemical difference between the two groups of carotinoids. The proof of this theory as well as the nature of the difference was soon brought to light by Willstattcr and Mieg (1907) when they iso- lated (lie first crystalline xanthophyll and submitted it to analysis. Working on the same elaborate scale, which has characterized all the researches on carotinoids in Willst Jitter's laboratory, a crystalline xanthophyll was isolated from 100 kilos of dried nettle (Urtica) leaves. The average of five ultimate analyses of crystals prepared both by recrystallization from methyl alcohol and from chloroform (by addi- tion of petroleum ether) showed 84.22 per cent carbon and 9.92 per cent hydrogen, which corresponds very closely with the theoretical values of 84.44 per cent carbon and 9.93 per cent hydrogen for the formula C4,,H-,.,0.,. This was confirmed fairly well by a molecular weight determination (found 512, theory 564), and better by an analysis of the iodine content of the theoretically simplest iodine addi- tion product, C40H5(;02I2 (found 31.68 per cent, theory 30.86 per cent). The chemical properties of the crystalline xanthophyll isolated by Willstatter and Mieg will be considered in detail elsewhere. Several points, however, may profitably be considered at this point. The crystalline product showed the greatest solubility difference from caro- tin in alcohol and low boiling petroleum ether, being practically in- soluble in the latter, but readily soluble in the former, which is just the reverse of carotin in these solvents. The Kraus method of separa- tion of the pigments was further confirmed by Willstatter and Mieg by applying the test in several ways to solutions of the purified pig- ments. The difference between the position of the absorption bands of carotin and the xanthophylls, first pointed out by Monteverde, was confirmed, the first two bands as measured by Willstatter and Mieg lying at 480-470i.in and 453-437^. Willstatter and Mieg expressed their belief in the existence of a group of xanthophylls in the paper under consideration although they were apparently not familiar with Tswett's demonstration of this fact a year before their paper appeared. The question naturally arises as to which xanthophyll was obtained in crystalline form by these in- vestigators. Tswett (1910a) has expressed the opinion that the xanthophyll crystallized by Willstatter and Mieg was a mixture of two or three xanthophylls in which xanthophyll a predominated, a possibility which was later acknowledged by Willstatter and Stoll (1913b). The evi- 46 CAROTINOIDS AND RELATED PIGMENTS dence available on this question indicates, however, that xanthophyll (3 may have formed a considerable proportion of the crystalline prep- aration. Willstatter and Mieg mention the fact that their prepara- tion dissolved in strongly alcoholic HC1 with a blue color, a reaction which is apparently characteristic of xanthophyll (3 only. In the Sorby and C. A. Schunck separation, however, the pure pigment differen- tiated itself almost equally between the alcohol and carbon disulfide layers, a reaction which obviously characterizes the a group of xan- thophylls because of their lesser adsorption from this solvent by CaC03. Still further evidence of a mixture of xanthophylls in the Willstatter and Mieg preparation is the fact that its spectroscopic absorption bands apparently lie in an intermediary position between the bands of xanthophylls a and (5 as recorded by Tswett. The isolation of the various members of the xanthophyll group in crystalline form seems greatly to be desired in order that the dif- ferences existing between the individual members of this class of carotinoids may be determim-d. The relative adsorption properties of these pigments offers the most promising method for accomplishing this result but the experimental work would have to be conducted on a very generous scale. The xanthophylls are unquestionably either isomorphic or isomeric forms of the same empirical composition, C40HS602, as Willstatter and Stoll (1913) have pointed out. The author believes that Willstatter and Escher (1912) have already iso- lated pure xanthophyll a in the form of their so-called lutein from egg yolk, as will be discussed more fully in a later chapter. It is not likely that more than four xanthophylls characterize the chloroplastid for the author (1914g) has found only four on applying the chromatographic method to extracts from an entirely different plant than Tswett used, namely, the leaves of alfalfa (Medicago sativa). The possibility of other xanthophylls being present in non- chlorophyllous organs is indicated, however, by a chromatographic analysis which the author (1914g) carried out on the xanthophyll fraction (obtained by the Kraus separation) of the pigments of the carrot root, in which no less than eight distinct yellow or orange zones characterized the chromatogram. The possibility remains to be investigated, however, whether this result was influenced in any way by the method of preparation of the material or other experi- mental steps in the procedure employed. The author regards the adsorption phenomenon of the carotinoids as colloidal so that it may not be possible to secure these pigments in every case in the same ('.[ROTIXOIDS IN Till' PHANEROGAMS 47 decree Dl' colloidal atinre<:ai ion. \Yillstiitter and Mieg found lliat. their crystalline xanthophyll readily entered into combination with solvents I'orminu molecules of solvent of crystallization, which is unquestionably a colloidal combination and might easily influence greatly the adsorption properties of the pigments. The whole adsorp- tion phenomenon deserves a further study using pure preparations of the individual pigments. The xanthophylls arc usually regarded as pigments in which yellow i- the predominating color. Red colored xanthophylls also exist, however. Monteverde (1893) first called attention to a red pigment in the reddish-brown leaves of the young floating pond weed (Potamo- n natans), an aquatic herb widely distributed in Russia, which showed the xanthophyll properties in the Kraus separation. This pig- ment has since1 been called rhodoxanthin by Monteverde and Lubi- menko (1913bl, who obtained it in crystalline form. The pigment appears to be isomeric with the xanthophyll of the chloroplastids, as lycopin is isomeric with carotin. It differs from the usual yellow xanthophyll by dissolving in formic acid with a yellow color, yellow xanthophyll dissolving in this solvent with a green color, according to Monteverde and Lubimenko. Rhodoxanthin also shows spectro- scopic absorption bands with characteristic position, especially in car- bon bisulfide. A comparison of the xanthophyll and rhodoxanthin band- in this solvent, as given by Willstiitter and Stoll (1913) and by Monteverde and Lubimenko, respectively, is shown in the following: -\niitli<>i>lii/ll (W. and S.) Rhodoxanthin (M. and L.) Band I 516-501 nn 575-553 HM- B.i nd II 483-467 " 535-515 " Band III 447-441 " 500-480 " The general solubility properties of rhodoxanthin appear to follow thoM' of xanthophyll very closely. The relation between the empirical constitution of carotin and the xanthophylls is such that the latter may be expressed very simply as carotin dioxides. The character of the oxygen combination, however, i- not dear, for according to the statement of Willstiitter and Micg their cn>t;dline xanthophyll did not show the presence of cither hydroxy, carboxyl or carbonyl groups. The xanthophylls, therefore, cannot be nmple oxidation products of carotin. But these statements iv»-:i?-dinii the character of the oxygen in the xanthophyll molecule p«-ibly should be confirmed, for notwithstanding the fact that it has not yet been found possible to t ran.- form carotin into xanthophyll in 48 CAROTINOIDS AND RELATED PIGMENTS the laboratory, the constant presence of these pigments in the chloro- plastid is very difficult to explain unless the constitution of one type of pigment bears a simple relation to that of the other. The various theories which have been offered regarding the possible functions of the carotinoids in the chloroplastid also fall down unless the caro- tinoids are closely relationcd chemically. Ewart (1915), to be sure, has recently claimed to have succeeded in reducing xanthophyll to carotin in the laboratory. The evidence for this is very unconvincing, especially in view of the fact that Ewart on subsequent study (1918) failed to substantiate any of the other products which he first claimed to have produced from xanthophyll on photo-oxidation. The reduc- tion experiment of xanthophyll to carotin unfortunately was not repeated in the second study. Carotinoids in Etiolated Leaves The yellow chroniolipoids which develop without chlorophyll in the leucoplastids, when plaiiN are grown in the dark, would seem to be closely related to. if not completely identical with those found in the chloroplastids, at lea>t qualitatively, inasmuch as etiolated plants form chlorophyll very rapidly in the li. r>7!t. .Trim. r.H.'ii, Julius Sachs MX.V.I ;i. Ill and Jos. Hoehlll (1S.V.M called (he et iolated pigment leuko- pliyll and chlorogon, respectively. Tin- statement is incorrect. The h'ukophyll of Sachs was a colorless chromogcn in the seeds ;iiid also in Hie etiolated plants which gave rise to the green clilornphyll in Ihe sunlight or on treatment with acids i coin- pare Phipson [isr>s|t, while the chlorophor (not chlorogon as C/apek lias it i of Moehin was the same colorless chromouen. r.oehni dilTered from Sachs in regarding the green acid derivative of the colorless chroinogen as an artificial pi.mnent and the green sun light derivative as the true chloi-ophyll. Koth investigators recognised the existence of the yellow etiolated leaf pigment us, well as, the colorless chroinogen. 50 CAROTINOIDS AND KELATED PIGMENTS either chlorophyll or a closely related fore-runner of one of the chloro- phyllins. Timiriazeff (1875) believed the absorption spectra of alco- holic etiolated leaf extracts to be due to a small amount of chlorophyl- lin admixed with Kraus' xanthophyll. Hansen (1884bi re.irarded the bands in the red as due to chlorophyll. Montevenle 1 1894 1 regarded the substance giving the bands in the red as a forerunner of one of the chlorophyllins and called it protochlorophyll, a view which seems to have been substantiated by the work of (Ireilach (1904). The latter proposes to reserve the name etiolin for this green pigment with properties like chlorophyll which exist- in etiolated leaves in very small amounts, and to use the term in the same sense as Monteverde used the word protochlorophyll. According to Greilach etiolin (proto- chlorophyll I is not ;i constant constituent of the etiolated leaf but appears and then di-appcars during the germination of the seed in the dark. Arnaud (18891, following his earlier (1885) demonstration regard- ing the identity of the yellow leaf pimnent isolated by him with the carotin from carrots, minnled the yellow color of etiolated leaves MS due to the s;iine piinni'nt. No ch< mical proof was offered of this but he determined the quantity of carotin in the etiolated leaves of the kidney bean I /%/*< «/>/x r///[/am I , using a colorimet ric method which will be reviewed in a later chapter. Inasmuch as Arnaud's method of analysis would preclude all but trace- of xanthophylls his result may be regarded as the first proof of the presence of carotin in etio- lated leave-. This \va> confirmed completely by Tmmendorff (1889) the same year. He saponified the alcoholic extracts from etiolated leaves, extracted the carotinoids from the soap with ether and obtained crystals of carotin from the golden yellow extract. He did not suc- ceed in obtaininu crystals from etiolated leaves which had developed only a pale yellow color, hut only from those having a more orange color, but this cannot be interpreted as indicating another pigment in the less pigmented leaves, as Immendorff believed, but must be re- garded as due solely to differences in concentration of pigment, as Kohl (1902f ) has pointed out. Following Immendorff, Molisch (1896), Tammes (1900) and Kohl (1902f) have independently substantiated the presence of carotinoids in etiolated leaves from various plants using microscopic crystalliza- tion methods on the fresh tissues. Inasmuch as our information re- garding the yellow chromolipoids in the etiolated leaf depends at the present time on the observations of these authors and the microchemi- CAROTINOIDS IN THE PHANEROGAMS 51 eal methods which they used, it will In- necessary to state brielly the character and significance of I he methods, reserving a fuller descrip- tion t'nr a later chapter. Frank (1884) first observed that red crystalline needles form in the plastids and between the chlorophyll granules when green leaves are immersed in dilute acids for a time, and then, after washing off the acid, are allowed to remain in distilled water for a still more pro- tracted period. Tschirch (1884 1, who first examined the phenomenon, did not decide the nature of the crystals, but Molisch 1 1890) found the crystals to be identical in properties, although having a more reddish color, with the majority of crystals which he found could be produced by an entirely different method. Molisch's method is to immerse the leaves in dilute (40 per cent by volume) alcohol containing 20 per cent KOH, until the chlorophyll is completely extracted. The process sometimes requires several days. On washing off the green extract with water, and immersing the washed leaves in distilled water for several hours to insure the complete removal of the chlorophyll, it is found that crystals of various forms and colors from yellowish-orange to red have appeared abundantly in the leaf. Molisch proved fairly conclusively the identity of many of the crystals thus obtained with the red-orange crystals which form in concentrated alcoholic leaf extracts, and accordingly decided to call the crystals carotin. Molisch was care"ful to point out, however, that he used the term carotin in the sense of a group of closely related pigments, for he recognized that the crystals formed by his alkali method were not due in all cases to the same pigment. Tammes (1900) and Kohl (1902), however, who greatly extended our knowledge of the presence of carotinoids in the plant kingdom, using the microchemical methods of Frank and Molisch, believed that only one pigment was concerned, namely, caro- tin, and regarded the methods as specific for this pigment. Tswett 1 191 la), however, proved definitely that the crystals obtained by Molisch's method are a mixture of carotinoids, and this has been com- pletely confirmed by van Wisselingh (1915i. Other microchemical methods for carotinoids have been worked out by the two investi- gators just mentioned, and these will be reviewed in. a later chapter. Tswett has stated that Frank's acid method may possibly be specific for carotin. This may well be the case in view of the much greater sensitiveness of the xanthophylls to acids, as van Wisselingh has pointed out, but this investigator who has studied the method closely finds it to be often laborious, requiring sometimes several months for 52 CAROTINOIDS AND RELATED PIGMENTS the crystals to form, and that it frequently fails to show the presence of carotin in plant tissues in which the pigment is known to be pres- ent. By the use of suitable solvents van Wisselingh has demonstrated very ingeniously, however, that it is possible to distinguish the xan- thophyll crystals as a group from the carotin crystals in the mixture formed by the Molisch method in various plant tissues. Returning now to the investigations regarding the chromolipoids in various etiolated plant-, it may be stated that Molisch (1896) demon- strated carotinoids by his alkali method in the etiolated leaves of garden cress (Lcpidiuni *(i) and various varieties of Asphodel. It is apparent that we have as yet only indirect evidence that xanthophylls are pre>ent in the etiolated leaf. ('. A. Schunck (1903) has furnished direct evidence of xanthophylls in an isolated case, namely, the etiolated leaves of the daffodil (Narcissus pseudo-nar- cissus}, using the carbon disulfide separation method which has already been described. A mixture of xanthophylls was found to be present, but Schunck was unable to obtain crystals of chrysophyll (carotin) although he had no difficulty in obtaining them abundantly from alcoholic extracts of the etiolated leaves which had been allowed to turn green in the sunlight. Greilach's (1904) spectroscopic obser- vations of the pigments in etiolated leaves led him to conclude that yellow pigments other than carotin are also concerned in the colora- tion of the leaves. Ewart (1918) states that he has found 8 to 10 parts of carotin to one of xanthophyll in etiolated wheat seedlings. Several investigators have studied the question raised by the last observation of Schunck, namely, what effect greening has on the con- tent of carotinoids in the etiolated leaf. Wiesner (1877a) first studied this point and concluded that the xanthophyll (carotinoids) diminished CAROTINOIDS IN THE PHANEROGAMS 53 during the greening of etiolated sprouted oats. Inasmuch, however, as he used a eolorimetric comparison of the total alcoholic extract of the etiolated plant with the alcoholic xanthophyll layer of the extract from the green plant following the Kraus separation, it is not dif- ficult to account for his results. Arnaud's (1889) quantitative eolori- metric comparison of the carotin content of etiolated and green leaves of the kidney bean, referred to above, led to completely opposite results. Arnaud's data (calculated from his eolorimetric reading) show 34.0 mg. carotin in 100 grams of the dry etiolated leaves, and 178.8 mg. in the same amount of dry green leaves, a result which appears to have been substantiated by the observation of Schunck on etiolated and green daffodil leaves. Kohl (1902f) studied the same question and drew the same conclusion as did Arnaud, namely, that carotin increases during greening. His method of analysis, however, does not permit so exact an interpretation, for he merely compared colorimetrically the total unsaponifiablc pigment extracted from the leaves by alcohol. Kohl's carotin solutions were thus a mixture of carotin and xanthophylls. It is not possible to decide from these observations whether xanthophylls as well as carotin increase in the etiolated leaf during greening. This appears to be the case, however, in view of Ewart's (1918) statement, quoted above, and the fact that xanthophylls are the predominating carotinoids in green leaves as found by Willstatter and Stoll (1913) and Miss Goerrig (1917). An interesting phase of the etiolated leaf pigmentation is that of the most favorable conditions for the development of the carotinoids. Light and temperature are obviously the controlling factors. Wiesner (1877b) observed that potato sprouts, which formed in the light, showed little if any yellow pigment, while those which formed in the dark developed from 30 to 150 per cent more pigment. More interest- ing is the result of Elfving (1882), which was confirmed by Immen- dorff (1889), that carotinoids increased greatly in leaves under con- ditions which depressed chlorophyll formation, i.e., low temperatures (2° to 8° C.) and very diffused light. Carotinoids in Naturally Yellow Leaves Plastids which fail to develop chlorophyll but in which other pig- ments form instead arc called chromoplastids. The pigments of chromoplastids are usually granular, sometimes crystalline and almost invariably yellow to red in color. In the case of some plants the leaf 54 CAROTINOIDS AND RELATED PIGMENTS plastids arc always characterized by an ab-ence of chlorophyll, the leaves being yellow or golden yellow in color. Several investigators have studied the pigmentation of such plants in relation to the yellow chrom'olipoids which characterize the chloroplastid. Thudichinn (1869). many years ago, observed the relation between the pigment of the carrot root and that in the yellow leave- of Coleus, and included them both in his group of luteins. Dippel 11878) found that the spectroscopic absorption bunds of the pi»ment extracted by alcohol from yellow leaves corresponded with those of the yellow pigment which hi1 found to be present in Kraus' cyanophyll layer from green leaf extracts. Dippel called tin- pigment xanthin. Tammes |1900) and Kohl i!902gi lir.-t -ought to show the relation of the pigment of such leave- to carotinoids. Tamme- found that the plastids of yellow leave- gave positive carorinoid reaction- with con. H,SO4, con. HXO... and with HCl containing phenol and with bromine water, when the leaves were tirst dried. Yellow leaves in which the chromoplastids had disintegrated failed to give these reactions. When examined after -ubmittin-j; the leave- to the Moli-ch alkali cry.-tailiza- tion method brownish yellow crystals of variola -hape- were ob-erved in each of the following cases: 1. Aiicuba japonica Thiinb. (Japanese Aokiba). 2. Elaeagnus latifolia li. 3. Euonymous japonicus L. variety vulphwrea '.Japanese .-pindlr tree). 4. Sainbitntfi nii/ni I.., variety awrea. < Kuropean eldi Kohl (1902g) substantiated the-e re-ults, also using the Molisch method, and in addition obtained carotinoids in the following plants with naturally yellow leav 1. Abutilon n