BERKELEY LIBRARY JNIVERSITY OF CALJPORMIA LANDSCAPE ARCHITECT'',;* The Gift of Beatrix Farrand to the General Library University of California, Berkeley REEF POINT GARDENS LIBRARY PUBLICATIONS OF THE ARNOLD ARBORETUM, No. 9 A MONOGRAPH OF AZALEAS RHODODENDRON SUBGENUS ANTHODENDRON BY ERNEST HENRY WILSON AND ALFRED REHDER ISSUED APRIL 15, 1921 PRINTED FROM THE INCOME OF THE WILLIAM L. BRADLEY FUND THE UNIVERSITY PRESS CAMBRIDGE 1921 A MONOGRAPH OF AZALEAS . . PUBLICATIONS OF THE ARNOLD ARBORETUM, No. 9 A MONOGRAPH OF AZALEAS RHODODENDRON SUBGENUS ANTHODENDRON BY ERNEST HENRY WILSON AND ALFRED REHDER ISSUED APEIL15, 1921 PRINTED FROM THE INCOME OF THE WILLIAM L. BRADLEY FUND BY THE UNIVERSITY PRESS CAMBRIDGE 1921 LANDSCAPE ARCHITECTURE Add'l Farrand Gift LANDSCAPE ARCH. LIBRARY PREFACE THE value of Azaleas as garden plants in the eastern United States, where few of the Rhododendrons with persistent leaves can be successfully cultivated, has led to the critical studies of these plants which appear on the following pages. Mr. Wilson's long journeys in eastern Asia has made it pos- sible for him to study all the species of the Japanese Empire and China as wild plants, and to become familiar with the Azaleas growing in Japanese gardens. For many years the Arboretum has been engaged in a field study of the American species, and has been able to place at Mr. Render's disposal the large amount of material which is preserved in this her- barium, and which has been supplemented by that contained in other American collections. The Azaleas of northern Japan and Korea, and of the northern United States, are already growing in the Arboretum, but the gardens of the southern states and of California have still much to gain by the introduction of the species of the southern states and those of southern Japan, China, and Formosa, and this publication will not have accomplished its purpose if it fails to induce the more general cultivation of these plants in the United States and Europe. The art of the hybridizer has produced many interesting and beautiful Azaleas. Imperfect records have made the study of many of these hybrids difficult and uncertain, and this difficulty is increased by the fact that many of them are no longer culti- vated unless, having escaped the change of fashion in plants, they are still growing in English gardens, planted about the middle of the last century, and I venture to suggest to my associates in the English Rhododendron Society that the study of Azaleas in European gardens might add much to the knowl- edge of the origin and value of many of these plants. C. S. SARGENT, Director. ARNOLD ARBORETUM MARCH, 1921. 609 TABLE OF CONTENTS PAGE PREFACE . > * v THE AZALEAS OP THE OLD WORLD. By Ernest Henry Wilson . 1 THE AZALEAS OF NORTH AMERICA. By Alfred Rehder .... 107 DOUBTFUL NAMES 197 OAQ ILLUSTRATIONS ADDITIONS 206 INDEX. 207 THE AZALEAS OF THE OLD WORLD BY ERNEST HENRY WILSON THE AZALEAS OF THE OLD WORLD INTRODUCTION BOTANISTS may hold different views on the classification of the subdivisions of the large genus Rhododendron, but garden- ers and lovers of plants in general have no difficulty in recog- nizing Azaleas as distinct from other groups of the genus. It is hi the popular rather than in the strictly technical sense that the title of " Azaleas of the Old World " is applied here. These plants with few exceptions are sun-loving; and their wide popu- larity among the peoples of the Orient, of America, of Europe and Australia is due largely to the brilliancy of their flowers, their floriferousness and the ease with which many of them can be cultivated. In the Orient they have been favorite garden flowers from very early times and in the Occident for more than a century some have been familiar and valued greenhouse plants, and in recent years it has been found that several are more hardy than they were generally supposed to be. In the Arnold Arboretum such hardy kinds as R. obtusum var. Kaempferi and R. japonicum are among the most satisfactory and most beau- tiful of Asiatic plants. Farther south and in California the old Azalea amoena and A. ledifolia and the newer "Hinodegiri" are much grown in gardens. Others are less known, but I believe that in the near future both in increased variety and in quantity Azaleas will have a much more important place in gardens. In certain groups the hybridists in Europe have wrought wonder- ful results, and the field is still full of promise. In Japan, in the city of Kurume, selection and raising from seed has been in progress among a single group for a century, and these plants are now beginning to find their way into the Occident. With these facts in view the need for a critical survey of the species and forms seems necessary. Through confusion with other species and for lack of accurate names many good plants have become lost from our gardens, and others for the same reason have not been introduced. During my travels in China I greatly admired l 2 THE AZALEAS OF THE OLD WORLD the red-flowered Azalea of the country so abundant on the hills from the coast to the extreme west, from near sea-level to 1300 m. altitude. So common is it that in May much of the country- side is a blaze of red. For years I unhesitatingly accepted this plant as the Azalea indica of Linnaeus, considering the differ- ence in the number of stamens an unimportant triviality. In 1912-13, when working up the Rhododendron material for Sar- gent's Plantae Wilsonianae in conjunction with my colleague, Alfred Rehder, I became conscious that all was not so simple as I had assumed. The question as to what Azalea indica really was began to arouse my interest. On the occasion of the Arnold Arboretum Expedition to Japan in 1914 I devoted much time to the problem, and in Yaku-shima in February collected an Azalea which later I discovered to be the true Azalea indica of Linnaeus. I questioned many botanists in Japan, but none at that time had any definite views on the subject. Nevertheless, by deduction and comparative study I arrived at the conclusion that the group of Azaleas cultivated in Japan under the name " Satsuki" all belonged to Linnaeus' species and that the ma- terial collected in Yaku-shima was its wild parent. On my re- turn to the Arnold Arboretum early in 1915 I critically com- pared the material I had collected with Linnaeus' description and with the old figures on which his species was based, and the proof of this theory was absolute. Japan is so rich in Azaleas, both wild on the mountains and cultivated in gardens, parks and temple grounds, that at the end of 1914 I felt that beyond establishing the identity of Azalea indica I had acquired no more than a moderate working knowledge of the group. My interest, however, was keener than ever. Being favored with another opportunity to visit the Orient I determined to devote as much time as possible to the study of these Azaleas. During 1917 and 1918 I visited many places famous in Japan for Azaleas and gathered specimens and information from all sides. On my visits to Korea, Liukiu and Formosa I collected much new material. The result is that on my various visits to the Chinese and Japanese Empires I have seen growing wild, with few exceptions, every species known from those regions and almost every known variety and garden form. In Japan in 1914 and again in 1917-18 I received in- estimable assistance from Mr. H. Suzuki, President of the INTRODUCTION 3 Yokohama Nursery Company and one of the foremost horticul- turists in Japan. Together we visited Mt. Kirishima and the famous Azalea district of Kurume in Kyushu, also those near Osaka and Tokyo. He introduced me to all the leading special- ists, gave of his own store of knowledge freely, and without his help it would have been quite impossible to have delved so deeply into the cultivated Azaleas of Japan. It is with keen pleasure that I acknowledge my indebtedness to this genial and scholarly gentleman. I know not how many hundreds of speci- mens of my own collecting have been available for this work, and in addition, through the courtesy of the Director of Kew and the Keeper of the Gray Herbarium of Harvard University, I have had on loan all their material, so rich in old types of these Azaleas. In Tokyo, through the courtesy of Professor J. Matsu- mura, I have examined the material preserved in the Herbarium of the Imperial Botanic Garden. Dr. Merrill has kindly loaned me the material preserved in the Herbarium of the Bureau of Science, Manila. To Dr. T. Nakai I am indebted for much useful information relative to certain critical Azaleas to which S. Komatsu has recently given names with brief descriptive notes in Japanese. The study of long-cultivated garden plants is exceedingly complex, and to fathom some of the problems is almost beyond human skill. This statement is as trite as it is true, and I freely confess that I have found the problems presented by many of these Azaleas more intricate than those of the Japanese Cherries. Notwithstanding my unusual opportunities in the field and in the gardens of the Orient and the abundant herbarium material at my command I should have been ill equipped for the task but for the collection of living plants cultivated for nearly three- quarters of a century on the Sargent estate in Brookline, Mass. This collection has been zealously maintained and is rich in old garden types introduced through Fortune and others, some of which have apparently been lost in Europe. Professor Sargent's knowledge of these Azaleas is profound, and it is under his sym- pathetic guidance and help that this little treatise has developed from chaos to order, and to him most largely is due what merit it contains. HISTORY THE cultivation of plants and the development of gardens is ever a sign of a nation's advance toward culture and refinement. How early this art began in the Orient we do not know, though history tells us that plants like the Yulan and Moutan Paeony were highly esteemed garden flowers among the Chinese during the Tang Dynasty (618- 907 A. D.)- It is much to be regretted that we are so ignorant of the early history of China. It is known that during the Han Dynasty (B. C. 206-A. D. 25) the Chinese had intercourse with India and Zan- zibar. If proof were needed of their visiting these countries at an early period it is supplied by the presence in south China and Formosa of such typical African plants as Cactus-like Tree Euphorbias. Such species as E. triangularis Desf., E. neriifolia L. and E. tirucalli L. are common hedge-plants and are naturalised in certain places like Hong- kong, Kowloon and elsewhere in China, and near Takao in Formosa, as they also are in parts of India. In point of fact E. tirucalli was named by Linnaeus from material which came from India. Did we know more about the early travels of the Chinese it is possible that we should find that they were acquainted with America long before its discovery by Columbus. In south China, Formosa and the Liukiu Islands A gave fourcroydes Lem. is naturalised, but this may have been introduced by the Portuguese, and so too may have been Maize and Tobacco. Much more difficult is it to account for Opuntia Dillenii Haw. naturalised and abundant not only on the Yunnan plateaux but in the valley of the Tung River in a remote part of extreme western China. It is significant also that Boym in his Flora Sinensis published in 1656 figures such characteristic American plants as the Pineapple (Ananas saliva Schult. f.), the Sweetsop or Custard Apple (Anona squamosa L.) and the Guava (Psidium guajava L.), which were evi- dently common plants in China when he was there (1642-53). In what era the love for flowers began to manifest itself in Japan is un- certain, but Buddhism in its Chinese form was introduced from Korea about 552 A. D. and from then until the 8th century Korean and Chinese monks and nuns visited Japan for purposes of proselytism. From the 8th century onward it became more usual for the Japanese 4 HISTORY 5 monks to go to China in order to study the doctrines of the best ac- credited teachers at the fountain-head. These monks introduced into Japan a number of favorite trees and shrubs like the Ginkgo, Yulan, Moutan Paeony, Chinese Quince, the Tea-plant, Buddha's Tree (Tilia Miqueliana) , Sophora japonica, Prunus tomentosa, P. japonica, Paulownia tomentosa, such fruit trees as the Peach, Apricot, Persimmon, Pear, Apple, Plum, Cherry, Loquat, the Orange and its relatives and, in all probability, the Chrysanthemum also. Though they probably did not originate the love for gardens these monks must have greatly stimu- lated it. The mutual love of flowers among the peoples of the Orient induced and fostered the introduction of plants from one country to another in the Far East even as it did in the Levant and, in later days, as it has in the Occident. And so we find the history of garden plants intimately associated with the social life of the peoples. We know nothing of early intercourse between the Japanese and Chinese prior to the introduction of Buddhism to Japan. We do know that Chinese monks introduced their favorite flowers to Korea, that Chinese and Korean monks carried them to Japan, and it is safe to assume that these same monks and their Japanese disciples carried to Korea and to China plants of Japan which appealed to their aesthetic sense. In fact the presence in temple-grounds and other sacred places in China of the Cryptomeria, which is endemic in Japan, and of many varieties of Japanese Camellias and Azaleas in Chinese gardens, is positive proof that Japanese plants were long ago introduced into China. The illustrious Venetian, Marco Polo, who lived many years in China during the latter part of the 12th century and to whom we owe the first authentic account of that mighty empire, notices from his own observation many of the vegetable productions of China used for economic purposes. After his memorable travels China was for a long time closed to European access and had been altogether forgotten in Europe when in 1516 the Portuguese first arrived in China. The Portuguese claim the honor of having first introduced the Sweet- Orange from China to Portugal some time between 1545 and 1548. About the end of 1542 the Portuguese adventurer, Mendez Pinto, discovered Japan, first touching at the island now known as Tanaga- shima south of Kagoshima in Kyushu. The Portuguese made their way north as far as Oita in northeast Kyushu where Mendez Pinto in 1543 received a friendly reception from the local Daimyo. The arquebus Pinto carried astounded the Japanese, who had never before seen any explosive weapon. In 1549 St. Francis Xavier reached Japan 6 THE AZALEAS OF THE OLD WORLD and introduced Christianity; he visited Oita, and going north to Hondo established a mission at Yamaguchi, a few miles north of Shimonoseki. This newly introduced religion soon claimed thousands of adherents, much to the alarm of the Japanese Daimyos and military men backed by Buddhist priests, and a period of persecution commenced. The native Christians migrated to Nagasaki, which soon became one of the chief marts of Portuguese trade. In 1596 the Dutch first visited Java and other islands of the East Indies and in 1602 the Dutch East India Company was established. The war which then ensued between the Dutch, Spaniard and Portu- guese for possession of the spice islands lasted until 1610 when the Dutch remained master of these seas. The seat of the Dutch govern- ment was first established on the island of Amboyna, but in 1619 it was transferred to the newly founded city of Batavia in Java. In 1600 a Dutch ship visited Japan, and nine years later the Dutch East India Company sent several vessels to Firando (Hirado), northwest of Naga- saki, where they were well received by the Japanese. In 1611 a formal edict in favor of their trade was obtained. A Dutch factory and also an English factory were established at Hirado the same year. In 1624 Christianity was proscribed in Japan and the land was closed to foreigners. The Portuguese, English and Spaniards, who also had a factory, were expelled and only the Dutch and the Chinese were permitted to carry on trade, and that under galling restrictions. The Dutch factory was removed in 1639 to Deshima, then an islet at the head of Nagasaki harbor but now absorbed into the foreshore. And so Japan secluded herself for over two centuries from contact with the outer world until the advent of the American Expedition in 1853-54 under command of Commodore Perry. All this may seem to have little to do with plants and with Azaleas in particular and yet they are inextricably bound up with the subject. It is to the merchant adventurer that we owe all our first plant introductions from Japan and our early knowledge of the flora. From 1690-92 Engelbert Kaempfer, in the service of the Dutch East India Company, lived in Japan, and in his Amoenitates Exoticae, published in 1712, he gives an admirable account of Japanese plants. He gives good figures of many of these under the vernacular names and among them an Azalea now known as R. obtusum var. Kaempferi Wils. In all Kaempfer mentions twenty-one Azaleas, and it is interesting to note that many of the vernacular names he gives are in common use to-day. C. P. Thunberg in 1775 visited Japan in the capacity of physician to the Dutch Com- HISTORY 7 pany and in 1784 published his Flora Japonica, the first post-Linnaean work on the flora of the Orient. Thunberg mentions the Azaleas under their vernacular names and relegates them all to Azalea indica. Ac- cording to Juel (PL Thunb. 391 (1918)) they represent R. obtusum Planch., R. japonicum Suring. and R. mucronatum G. Don. Lastly in this connection mention must be made of Philipp von Sie- bold, who joined the service of the Dutch East India Company and lived at Deshima from 1823 to 1829. To him we owe the magnificent Flora Japonica; also he introduced about 1830 such ornamental plants as Lilium elegans and L. speciosum from Japanese gardens, and, twenty years later, several others including Mains floribunda, M. Sieboldii and certain flowering Cherries. About 1677 the English East India Company established a factory at Amoy and in 1684, after a conflict with the Portuguese, one at Canton. Through this company many Chinese plants were introduced to India during the 18th century and early in the 19th century to England; they include several Azaleas. However, in China the Jesuit priests were the first to inform us about the flora of the land. In 1790 Joannis de Loureiro, a Portuguese, published his Flora Cochinchinensis and mentions one species of Azalea (A. punctata), but this remains an obscure plant and may not belong to the genus. During the last quarter of the 17th century Jakob Breyne, a merchant of Danzig and a distinguished botanist, visited Holland and saw growing there several famous garden plants of the Orient which he duly records in his Prodromus Plantarum. On page 24, pt. I, pub- lished in 1680, we read " Chamaerhododendron exoticum amplissimus floribus liliaceis. Frutex spectabilis elegans. In horto Beveringiano." This is the plant on which Linnaeus (Spec. 1753) bases his Azalea indica. Breyne does not say from what country this shrub had been brought, but P. Hermann, in his Academici Horti lugduno-batavi catalogue, p. 152 (1687), describes the same plant under the name of "Cistus Indicus Ledi alpini foliis, floribus amplis," figures it on page 153, and reports that it was introduced from Jaccatra, which is Ba- tavia, in Java. No species of Azalea grows in India, and it has been generally supposed that this Azalea had been brought to Batavia by the Chinese, but I am of the opinion that it was taken there from Japan by Dutch traders. In the gardens of Nagasaki and elsewhere, then as now, it was a common garden plant and, moreover, it is indigenous in the warm southern parts of Japan. That at the early date men- tioned plants from Japan were growing in gardens in Holland is certain, 8 THE AZALEAS OF THE OLD WORLD and the Tea-plant was introduced from Japan about 1689. Breyne (Prodromus, I. 4) states that he saw in the garden of Hieronymus van Beverningk in 1678 a fine specimen of the Camphor-tree which had been introduced from Japan. Again, in his Prodromus, II. 66 (1689) Breyne tells of six varieties of Chrysanthemum from Japan which he saw growing in Holland gardens. All these early introductions seem to have been lost. I have carefully looked through many old books of the 18th century and I find no references to any species of Azaleas being in European gardens, and they appear to have been unknown in Europe at the close of the 18th century. The one oriental species was known to Linnaeus through the books. In Batavia, however, several were grown, as we know from the writings of Burmann and Blume. In Aiton's Hortus Kewensis (1810) Azalea indica is stated to be in cultivation in Kew, having been introduced from China in 1808 by the Court of Directors of the East India Company in the ship "Cuffnels" Captain Wellbank. This plant was Rhododendron Simsii Planch., and here began a confusion between Linnaeus' species and the Chinese Azalea which still survives. Through the activities of officers of the East India Company at the close of the 18th and the beginning of the 19th centuries English gardens were enriched by many Chinese plants. No officer was more active in this work than John Reeves, who as Chief Inspector of Tea in the Company's establishment at Canton resided in Macao and Canton from 1812 to 1831. He was either the immediate or indirect source from which our gardens derived Camel- lias, Moutan Paeonies, Chrysanthemums, Roses, Wistaria sinensis and many Chinese Azaleas including R. Farrerae Tate, a pretty Hong- kong species, and also R. phoeniceum G. Don, destined to be of so much value as a stock on which to graft " Indian Azaleas," so called. In 1818 or 1819 Samuel Brooks, a nurseryman at Ball's Pond, New- ington Green, sent out to China Joseph Poole, a gardener, and through him secured among other plants Azalea indica alba (R. mucronatum G. Don) and Azalea indica purpurea plena (R. mucronatum f. plenum Wils.). In 1823 Azalea sinensis (R. molle G. Don) was received from China by Loddiges of Hackney and by William Wells, nurseryman, of Redleaf. In 1832 Joseph Knight, nurseryman, of Chelsea, London, secured Azalea indica variegata (R. indicum f. variegatum DC.) and A. indica lateritia (R. indicum Sweet) brought home from China by Mr. M'Killigan. In 1821 Rhododendron Farrerae was reintroduced and in 1832 R. indicum var. ignescens Sweet, both through Captain Farrer of the East Indiaman "Orwell," who gave HISTORY 9 them to Tate, a nurseryman, of Sloane Square, London. In the Botan- ical Register, t. 56 (1842), is figured a double red-flowered Azalea said to be of Chinese origin and in the possession of William Wells, Esq., of Redleaf . These various Azaleas by hybridization and selection soon became the parents of a number of good plants which are figured in the periodicals of the time. Especially valuable was the Azalea indica variegata which in the hands of Knight, and of Ivery, another English nurseryman, yielded several very ornamental varieties. In 1843 Robert Fortune was sent to China by the Horticultural Society of London, and his travels and explorations there, which ended in 1861, inaugurated a new era in the history of plant introduction from that country. He sent from Chinese gardens at Shanghai and elsewhere to England many Azaleas, including such new ones as Aza- lea obtusa (R. obtusum Planch.), Azalea ramentacea (R. obtuswn f. album Rehder), Azalea crispiflora (R. indicum var. crispiflorum Schneid.), Azalea amoena (R. obtusum f. amoenum Wils.), Azalea narcissiflora (R. mucronatum f. narcissiflorum Wils.), Azalea vit- tata (R. Simsii var. vittata Wils.) and Azalea Bealii (R. Simsii var. vittata f. Bealii Wils.). There is no need to emphasize the value of these new Azaleas. His vittata with lilac flowers and Bealii with red-striped flowers gave an impetus to the raising of new forms and may be said to have initiated an industry which has resulted in the so- called race of " Indian Azaleas." The true R. indicum, whose progeny are not amenable to forcing, dropped out and the forms of R. Simsii, R. mucronatum and R. obtusum seminal, branch sports and hybrids took its place, and to these we owe the familiar greenhouse Azaleas of to-day. There is no necessity for pursuing the subject further except to point out that all the Azaleas from the Orient up to 1860 were cultivated plants introduced from gardens and that the wild parent and habitat of none except R. Farrerae and R. molle were known. The opening of Japan, which followed the signing of treaties first with America in 1854 and then with other powers, soon led to the introduction of a great number of valuable garden plants to America and Europe. Strange to say the Azaleas for which Japan is famous seem to have been passed over for a number of years, probably on the assumption that they were the same as those already introduced to our gardens from China. From 1859-64 C. Maximowicz, the famous Russian botanist, travelled in Japan, and to him we owe our first com- prehensive account of the wild flora of that country. Maximowicz and his Japanese collector, Tschonoski, were among the first to collect 10 THE AZALEAS OF THE OLD WORLD seeds of Azaleas and by this means introduced into Petrograd R. reticulatum G. Don and R. linearifolium var. macrosepalum Mak. Maximowicz also introduced plants of R. linearifolium S. & 7., its variety macrosepalum and R. scabrum G. Don. In 1861 a Hollander, J. B. Groenwegen, claims to have received direct from Japan seeds of R. japonicum Suring., one of the most valuable of all hardy Azaleas. Crossed with the Chinese R. molle G. Don this has given rise to the lovely but short-lived Kosterianum hybrids; all these, again mixed with the so-called Ghent hybrids (R. luteum X American species), have produced the race of Azaleas named X R. mixtum Wils. Under each species and variety the story of its introduction is re- corded, so in this historical chapter I need not repeat it. Curiously enough the most common and one of the most conspicuously beau- tiful Azaleas of Japan, with flame-red flowers and figured by Kaempfer, was ignored until 1892, when Professor Sargent sent seeds to the Arnold Arboretum where it has proved not only perfectly hardy but one of the most valuable exotic plants. This and R. japonicum, raised by Professor Sargent from seeds sent from the Jardin des Plantes in 1876, were the first oriental Azaleas to be raised from seeds in America. Turning to Korea there is little to say, for prior to the beginning of this century the Occident had received no living plants or seeds direct from that country. A common Azalea of Korea, the lovely R. Schlip- penbachii Maxim., found its way to Japan, and from there was sent to England by James H. Veitch in 1893. The first seeds of an Azalea sent from Korea was in 1905 when Mr. J. G. Jack sent to the Arnold Arboretum seeds of R. yedoense var. poukhanense Nakai, a hardy species of compact habit with rose-purple, fragrant flowers. From the Liukiu Islands no Azaleas, either plants or seeds, have been sent direct to the Occident, but the large-flowered R. scabrum G. Don, better known as R. sublanceolatum Miq., endemic there, was long ago brought to Japan, and in Kyushu in particular it is a common garden plant. This Azalea was apparently in cultivation in England when G. Don wrote his General System in 1834, but seems to have been lost soon afterward. Maximowicz in 1864 sent it to Petrograd, but it ap- pears to have been unknown in England when reintroduced by Mr. Notcutt in 1911. The only Formosan Azalea introduced into Europe prior to my visit in 1918 was R. Oldhamii Maxim, which was raised by Messers Veitch from seeds sent in 1878 by Charles Maries. The Philippine Island species has not yet been introduced. DISTRIBUTION OF the thirty-four species of Azaleas enumerated here thirty-three are widely scattered over the countries of the Orient having their northern limit in central Hokkaido, their southern on Luzon in the Philippines; from east to west they are found from the shores of the Pacific Ocean in Japan to the extreme west of China proper. They belong to four well-marked sections of the genus Rhododendron, two of which (Tsutsutsi and Sciadorhodion) are peculiar to the Orient; two (Rhodora and Pentanthera) are represented also in America. The thirty-fourth species is R. luteum Sweet which belongs to the last- named section. It grows in the Caucasus, the Pontus region of the Black Sea and in Lithuania, Poland, and southwestern Russia, and is the only species of the groups here discussed known to grow west of China proper. Geographically the east Asiatic species are found over 45 degrees of longitude and 30 degrees of latitude and altitudinally from sea-level to about 3300 m. They are social plants and where climate and soil conditions are favorable, as in most parts of Japan, of Korea and of China from about the latitude of Shanghai south, these Azaleas are perhaps the most common shrubs. The Chinese R. Simsii and R. Ma- riesii grow also in Formosa, and a variety of the first-named is endemic on the Kawanabe Islands off the northwest coast of the Liukiu Archi- pelago. The Japanese R. serpyllifolium is said to grow on Okinawa, the main island of the Liukius. At least three (R. Schlipperibachii, R. Tschonoskii and R. Weyrichii) of the four Korean species grow also in Japan. One (R. phoeniceum) is unknown in a wild state and is of doubtful standing. The others are endemic in the different countries of the Orient. It is remarkable that in western China only the section Tsutsutsi with three species is represented, and on the Himalayas no member of the four sections is known to grow, yet in these two regions the sub- genus Eurhododendron has its headquarters and is represented by hundreds of species. In Malaysia, too, where the subgenus Vireya is so richly represented, no species of our subgenus is known. The great concentration of species is in Japan. Formosa, considering its small 11 12 THE AZALEAS OF THE OLD WORLD size, is also rich in Azaleas. In China these Azaleas are poorly repre- sented by eight species of which only three are really common and widespread. This is singular in view of the remarkable richness of the Chinese flora in general. In Korea four species grow, but at most only one (R. yedoense var. poukhanense) is endemic and specimens of an Azalea much like this have been collected in Bitchu province, western Japan. In Japan Azaleas reach their northern limits in central Hokkaido, where R. Tschonoskii, R. Albrechtii, R. obtusum var. Kaempferi and R. reticulatum are found. The marked absence of limestone in Japan is favorable to the family Ericaceae which is richly represented there, and on the volcanic mountains Azaleas are astonishingly abundant. The small-flowered R. obtusum, the wild type of which is the form japonicum, is local, being confined to Kiri- shima and to a few other mountains in Kyushu, but the large-flowered variety Kaempferi grows in plenty from the extreme south, northward to central Hokkaido. The same is true of the magenta-flowered R. reti- culatum. Billions of plants of these two species are scattered on the grass and shrub-clad slopes, in thickets, in thin woods and on the margins of forests. In May from Kyushu to central Hondo travelling by train or along the roads one is never out of sight of the blossoms of these Azaleas; in June the same is true for the north until their limit of distribution is reached. Many districts are famous for their Azaleas, and of these I may mention the slopes of Mt. Fuji, the Hakone moun- tains and the Nikko region. In Yamato province, and also round Fuji and Nikko R. japonicum with its clusters of large flame-colored flowers grows in millions. Indeed it is to her Azaleas that Japan owes much for the color which decks her countryside. And what joy it is to ramble through this flower-clad land in spring and early summer. In autumn the leaves of such species as R. japonicum, R. Tschonoskii, R. reticulatum, R. quinquefolium, R. pentaphyllum and R. linearifolium change to rich shades of yellow, orange, crimson and vinous purple. These autumn tints, like the wealth of brilliantly colored flowers in spring and early summer, arrest attention on every side, and small wonder that the Japanese have taken Azaleas into their gardens, parks and temple grounds and cultivated them from time immemorial. The only truly woodland species are R. quinquefolium and R. pentaphyllum, but so adaptable to different conditions are R. reticulatum and R. 06- tusum var. Kaempferi that they too may be considered woodland plants. In dark ravines, by the sides of torrents and in the forests of the Nikko region R. quinquefolium is plentiful, lighting up the dark and DISTRIBUTION 13 solemn depths with its snow-white, almost bell-shaped flowers. Crown- ing a knoll among mixed trees round the Hashimoto Tea House at Chuzenji and high up on the steep, wooded slope behind the Lakeside Hotel near the same place R. pentaphyllum is common, and I shall never forget the day in June, 1914, when I first beheld these bushes covered with masses of bright rose-pink blossoms. These two may be classed as small trees and among the Azaleas of eastern Asia are only approached in size by the pink-flowered R. Schlippenbachii. R. Tschonoskii has the smallest flowers of any species of Azalea and, indeed, in this respect is only approached by the Chinese R. Seniavinii. But in size of plant the real dwarf species is R. serpyllifolium, which is common in the Idzu province, and is also known from Mt. Unzen near Nagasaki, and from Okinawa in the Liukiu Archipelago. Yet individual plants of R. obtusum f . japonicum at alpine limits of vege- tation are equally dwarf. By the sides of rock-strewn torrents high up on mountains on Yaku-shima Island R. indicum, the first-known of all Asiatic Azaleas, forms dense thickets more than a metre high, but in the gardens of Japan it is a dwarf and hugs the ground closely. The lovely R. mu- cronatum with its pure white, fragrant flowers has not been found wild though its pale purple-flowered variety ripense is common by the sides of certain rivers in the island of Shikoku. The pink to rose-colored R. linearifolium var. macrosepalum with sticky leaves and flower-buds and an extraordinarily variable and glandular calyx is abundant in thin Pine-woods and among scrub round Futagawa, Yoshino and near Osaka. Very remarkable is the distribution of the red-flowered R. Weyrichii, which is known only from parts of Shikoku in the east, from the Goto Islands in the Japan Sea, and from the island of Quel- paert off the south of Korea. The small-leafed R. tosaense is confined to parts of the island of Shikoku, and R. Schlippenbachii, which is really a Korean species, grows sparingly on Chokai-san and one or two other mountains in northern Hondo. The slender-branched R. Al- brechtii grows in thickets and on the margins of forests from the Nikko region northward to central Hokkaido, but is nowhere really common. On the foot-hills of Adzuma-yama the remarkable R. nipponicum, with its campanulate white corolla, shreddy, cinnamon-brown bark and fragile fruit, is common in open grass and scrub-clad country, but is not known elsewhere. On Mt. Kirishima in Kyushu R. obtusum f . japonicum grows under alpine conditions, but the most truly alpine species of east Asiatic Azaleas is R. Tschonoskii, widespread in Japan 14 THE AZALEAS OF THE OLD WORLD on high mountains from Shikoku and central Hondo north to central Hokkaido, and in Korea found on the summit of the Chiri-san. All in all the most widely spread, most variable, most spectacularly beautiful and most useful of the red-flowered Japanese Azaleas is R. obtusum and its variety Kaempferi. In Korea, from Chiri-san in the extreme south of the peninsula to fifty miles north of Gensan, grows R. Schlippenbachii. It is partial to rocky places and thickets and often in thin woods it is the dominant undergrowth. On the famous Diamond Mountains in northeastern Korea it is particularly plentiful in mixed, rather open woods and on cliffs; in early summer its myriads of blossoms literally form sheets of pure pink for miles and miles. On exposed cliffs the habit is stiff and sturdy, but in the shade of trees the branches are slender. On Chiri- san it is usually a tree-like bush, four metres tall, and it is on this mountain range that this Azalea reaches its maximum size. On bare, grass-clad and shrub-clad slopes and inline-woods from the latitude of Seoul southward to the island of Quelpaert R. yedoense var. pouk- hanense abounds. In full exposure it ofte forms low mats several yards wide, but among shrubs it is occasionally 1.5 m. tall. This and R. Weyrichii are the only species that grow on Quelpaert, and the last-named is unknown on the mainland of Korea. In China the red-flowered R. Simsii is wild from the coast to the extreme west, the valley of the Yangtsze River being the northern limit of its range. In many places it is extraordinarily abundant, and in May whole hillsides are red with its flowers. This species is wild in the extreme south of Formosa, and a well-marked variety (eriocarpum) is endemic on the Kawanabe Islands off northern Liukiu. This is the only Chinese species known to grow wild outside of China proper, but phylogenetically it and R. obtusum may be considered of common origin, and to these China and Japan owe the wealth of brilliant red color, so dominant a feature in their spring landscapes. The lovely yellow-flowered R. moUe, better known under Sweet's name of R. si- nense, is found on open hillsides from the coast near Ningpo to Ichang in central China. It is a social plant though local in distribution. R. Mariesii grows in thickets and thin woods from Fokien to western Hupeh and also in Formosa, but its near relative R. Farrerae is confined to Hongkong and a few places in the neighboring province of Kwan- tung. On open mountains in western Yunnan the small-flowered R. microphyton is abundant; this is the most alpine of the Chinese species and, with the little known R. atrovirens, is the most western of DISTRIBUTION 15 the section Tsutsutsi. The other two Chinese species are R. Seniavinii and R. Mariae, and these are apparently rare and confined to the east- central and south-central provinces. In Liukiu the gaudy-flowered R. scabrum is endemic, but was long ago taken to the gardens of southern Japan where it is a common plant to-day. This species has the largest flowers of all Azaleas, and in a wild state and in gardens where it flowers freely is strikingly conspic- uous. Unfortunately in the Occident its tendency is to flower sparingly. I strongly suspect that R. phoeniceum, unknown as a wild plant though introduced from Canton about 1824, and the popular stock for grafting the " Indian Azaleas" of western gardens, is specifically not really distinct from R. scabrum. Although eight species grow wild in Formosa, Azaleas cannot be called a prominent feature of the vegetation. The red-flowered R. Old- hamii is widespread, and in districts in the north like Mt. Daiton and round Lake Candidius in the centre of the island it is fairly abundant. On grass-clad mountain slopes east of Ari-san the pretty, pink-flowered R. rubropilosum is plentiful. The others, so far as my personal knowl- edge goes, are local if not actually rare plants. Lastly, on Luzon in the Philippine Islands grows R. subsessik, the most southern representative of the group. The distribution and the number of the species in the four groups are shown in the following table : Teutsutsi Sciadorhodion Rhodora Pentanthera Japan 8 4 3 1 Korea 2 2 Liukiu 3 - - - Formosa 7 1 - - China 5 2 1 Philippines 1 - - - Pontus region, Caucasus, *) Lithuania, Poland and > ' - - 1 southwestern Russia ) CLASSIFICATION MANY of our favorite garden plants of oriental origin came to us from the gardens of China and Japan and in some instances their wild prototypes have not yet been discovered. Often when the wild plant has been discovered it has been found to differ greatly from that already in cultivation, and in consequence has to be designated as a variety or form of the described garden type. Phylogeny and taxon- omy are often opposed, which is regrettable but unalterable. Stu- dents of the floras of the Orient, particularly those of China and Japan, cannot too strongly bear in mind the fact that a great many of the earliest described species are based on cultivated plants. Let me illustrate this by reference to Japan. The Japanese, like the Chinese and Koreans, are lovers of the odd and curious in flowers and trees for which they have an innate love. Nearly every house, from the poorest hovel' to the castle of the feudal lord or his descendants, boasts a few flowers or a garden of some sort. The Japanese love their mountains, many of which are sacred and on most of them shrines have been raised. Pilgrimages to these mountains and shrines are a passion with the people and as mementoes of their visits it is a common practice, and has been for centuries, to bring back some strange or curious plant for their gardens. The result is that Japanese gardens are full of odd and curious plants dwarf plants, weeping plants, fastigiate plants, plants with variegated and curiously incised foliage, plants with abnormal flowers. This love of flowers has been fostered by Buddhist priests, and this has led to mutual exchange of plants among the peoples of China, Korea, Japan and Liukiu. Now if it be properly realised that it was from such gardens really museums of oddities that our early plants were taken and names applied to them the diffi- culty of identifying them with their wild parents will be better under- stood and appreciated. Until after 1860 all the Azaleas introduced from the Orient were from gardens, chiefly those of Batavia, Canton and Shanghai. As the plants flowered they were given specific and varietal names, but no attempt to discover their origin was made. Some were placed under the genus Azalea and others under that of Rhododendron. Salisbury 16 CLASSIFICATION 17 (Prodr. 286 (1796)) was the first to point out that Azalea and Rhodo- dendron could not be maintained as distinct genera. In 1834 G. Don (Gen. Syst. III. 843) placed all the known species under Rhododendron. Indiscriminate naming, however continued, and in 1854 Planchon, in Flore des Serres, IX. (1854), and in Revue Horticole (1854), reviewed the varieties from the Orient, placed all under Rhododendron and multi- plied considerably the specific names. In 1870 Maximowicz (Mem. Acad. Sci. St. Petersburg, se*r. 7, XVI. No. 9 (Rhodod. As. Or.)) monographed the Rhododendrons of eastern Asia, and his work has remained the basis of all subsequent studies. Maximowicz dealt with the subject in a thorough and masterly manner. I differ from him widely in the conception and limits of certain sections and species, and in the systematic position of some varieties and forms, but I am none the less sensible of the great value of his pioneer work. Many will regret the change of names involved but none more so than myself. These changes are as unfortunate as they are unavoid- able. I adhere strictly to the rule of priority as laid down in the Vienna rules, and while this may sometimes be deemed a nuisance its fairness cannot be questioned. Moreover, if we are ever to enjoy fixed order in plant nomenclature it can only be brought about through strict adherence to accepted and established rules. It is a pity that Planchon's name Azalea cannot be used for the sub- genus. It is based on the genus Azalea of Linnaeus, the type of which is Azalea procumbens L., now known as Loiseleuria procumbens Desv. The six species enumerated under Azalea by Linnaeus (Spec. 150 (1753)) belong to two subgenera and one distinct genus, namely, Loiseleuria. With Planchon's name eliminated the oldest valid name for the subgenus is Anthodendron Endlicher. 1 Of the four sections of the subgenus with which we are concerned three were established by G. Don (Gen. Hist. III. (1834)) and one (Sciadorhodion) is proposed here. The section Tsutsutsi G. Don was based on the plant figured by Kaempfer (Amoen. Exot. (1712)), which is R. obtusum var. Kaempferi and is an excellent type of this homo- geneous group. G. Don included R. reticulatum and R. Farrerae, but these have deciduous leaves, no appressed, flattened, bristle-like hairs and lack other features characteristic of Tsutsutsi proper. For this reason I refer them to a new section, Sciadorhodion. G. Don also includes R. molle which belongs to his section Pentanthera. Planchon 1 For full details see page 115. 18 THE AZALEAS OF THE OLD WOKLD in Flore des Serres, IX. (1854), emends Don's name to Tsutsia, includes under it R. molle and R. Farrerae, and adds R. Championae Hook. In the last-named species the flowers develop from clustered, axillary buds, and it belongs to the section Choniastrum Franchet. Maximo- wicz emends the section; calls it Tsutsia, but keeps under it R. Cham- pionae. He also emends Planchon's subgenus Azalea and employs it as a section to include R. Farrerae, R. molle and others; he also places in it R. macrosepalum, which obviously belongs to the section Tsutsutsi. Of the new section Sciadorhodion, R. Schlippenbachii may be taken as the type, and under it we place R. reticulatum and R. Farrerae, referred by G. Don to his section Tsutsutsi, also R. Weyrichii which Maximo- wicz includes in his section Azalea. Together these form a very homo- morphic group. The type of G. Don's section Pentanthera is the Eurasian R. luteum Sweet, but as most of the species are American I leave the discussion of it and also of the section Rhodora G. Don, the type of which is American though a majority of the species are Japanese, to my colleague Alfred Rehder. KEY TO THE ASIATIC SECTIONS OF THE SUBGENUS ANTHODENDRON ENDL. Flowers and leaves from the same bud. Shoots with flattened, appressed, bristle-like hairs; leaves persistent or some- times deciduous, usually dimorphic, scattered on the branches, elliptic to lanceolate or oblanceolate . . . I. Tsutsutsi. Shoots glabrous or villose, without bristle-like hairs; leaves deciduous, not di- morphic, in whorls at the end of the branchlets, scattered only on vigorous shoots, usually rhombic to obovate, rarely ovate . . .II. Sdadorhodion. Flowers from the terminal bud, leaves from lateral buds below. Corolla rotate-campanulate; stamens 8 to 10 III. Rhodora. Corolla funnel-form; stamens 5 IV. Pentanthera. KEY TO THE ASIATIC SPECIES SECT. I. TSUTSUTSI G. DON Bud-scales not viscid; shoots densely clothed with flattened, appressed hairs. Corolla funnel-form-campanulate. Stamens 5 (5 to 8 in R. tosaense). Style villose at base R. breviperulatum, p. 28. Style glabrous at base. Leaves linear-lanceolate to lanceolate or oblanceolate. Leaves crenate-dentate; flowers red to rose-red, more than 2.5 cm. across R- indicwn, p. 22. Leaves entire; flowers lilac-purple, less than 2 cm. across R. tosaense, p. 52. Leaves oval or obovate to lanceolate. Leaves more than 1 cm. long, the midrib beneath covered with hairs and usually with scattered hairs on the whole under surface R. obtusum, p. 29. Leaves 1 cm. to 1 cm. or less long, glabrous beneath, except a few brown hairs on the midrib R. serpyllifolium, p. 53. Stamens 7 to 10. Style villose at base. Leaves narrow-lanceolate to oblanceolate, crenate-dentate; stamens 10 . R. Kanehirai, p. 28. Leaves oval to obovate or lanceolate, entire; stamens 7 to 10. Flowers lilac to magenta; leaves generally elliptic-oblong, acute R. subsessile, p. 51. Flowers pink. Leaves conspicuously dimorphic, obovate to obovate-oblong, abruptly mucronate, 2 to 4.5 cm. long . R. lasiostylum, p. 50. Leaves not conspicuously dimorphic, often widest below the middle, elliptic-ovate to ovate-oblong or elliptic to elliptic-oblong, acute R. rubropilosum, p. 52. 19 20 THE AZALEAS OF THE OLD WORLD Style glabrous. Flowers more than 2.5 cm. across; leaves elliptic to elliptic-oblong or oblong-obovate, from 2 to 5 cm. long R. Simsii, p. 45. Flowers less than 2.5 cm. across. Leaves lanceolate, acuminate R. atrovirens, p. 50. Leaves oval to oblanceolate, 1 cm. or less long . R. Nakaharai, p. 54. Corolla with short, rarely long, cylindric tube and spreading lobes; stamens 5, rarely 4. Style villose at base; flowers white to rose-color . . . R. Seniavinii, p. 57. Style glabrous. Leaves persistent; flowers 5-partite. Corolla rose-purple, leaves not conspicuously dimorphic, oval to lanceolate R. microphytum, p. 57. Corolla lilac-colored, with a long tube; leaves conspicuously dimorphic, lanceolate or elliptic to oblong-obovate . . . . R. Mariae, p. 58. Leaves deciduous; flowers white, 4 to 5 partite; leaves narrow-lanceolate to ovate-lanceolate or oblong-obovate, often prominently 3-nerved R. Tschonoskii, p. 55. Bud-scales viscid on inner surface; corolla wide funnel-form; calyx-lobes green, ample; style glabrous. Shoots densely clothed with flattened, appressed hairs; leaves conspicuously dimorphic; stamens 10. Leaves persistent, calyx-lobes glandular-ciliate. Corolla scarlet to rose-red; calyx-lobes oval to lanceolate; leaves sub- crenulate, coriaceous R. scabrum, p. 59. Corolla rose-purple to magenta; calyx-lobes lanceolate; leaves entire, subcoriaceous R. phoeniceum, p. 61. Leaves deciduous or semipersistent, obscurely crenate-serrate, veins im- pressed above; corolla rose to rose-purple, fragrant, often double; calyx- lobes lanceolate, not glandular R. yedoense, p. 64. Shoots with few appressed, flattened hairs and many spreading, pilose often glandular hairs. Stamens 8 to 10; leaves persistent or semipersistent, not conspicuously dimorphic. Corolla red, not viscid; stamens 10; ovary glandular; leaves not viscid beneath R. Oldhamii, p. 66. Corolla white or pale-purple, viscid outside; stamens 8 to 10; ovary gland- less; leaves viscid beneath R. mucronatum, p. 68. Stamens 5; leaves deciduous or semipersistent, lanceolate to oblanceolate- ovate or linear-lanceolate; corolla pink to pale rosy purple, viscid without, funnel-form or divided into linear segments; ovary glandular R. linearifolium, p. 74. SECT. II. SCIADORHODION REHD. & WILS. Leaves broadest below the middle; stamens nearly equal. Leaves ovate; corolla pale lilac-purple; stamens 8 to 10; fruit conic-ovoid on a curved pedicel R. Farrerce, p. 79. Leaves ovate to ovate-lanceolate; corolla rose-color; stamens 10; fruit short- cylindric, erect R. Mariesii, p. 80. Leaves broadest about or above the middle. Leaves rhombic; fruit cylindric. Stamens subequal; corolla red, rotate-funnel-form; style villose R. Weyrichii, p. 82. KEY TO THE ASIATIC SECTIONS OF THE SUBGENUS 21 Stamens unequal, sometimes declinate; corolla rotate-campanulate. Corolla rose-purple to magenta; style glabrous, villose or lepidote or both; leaves with long scattered hairs when young, glabrescent at maturity; fruit cylindric, furrowed R. reticulatum, p. 83. Corolla white; style glabrous; leaves densely ciliate, short-villose, at least on midrib at maturity; fruit cylindric, terete R. quinquefolium, p. 87. Leaves broadly obovate; corolla pink; style glandular; fruit oblong-ovoid, verruculose R. Schlippenbachii, p. 88. SECT. III. RHODORA G. DON Corolla rotate-campanulate. Leaves obovate to oblong-oblanceolate, pubescent below; flowers 3 to 6, ma- genta; fruit conic-ovoid, bristly glandular R. Albrechtii, p. 91. Leaves elliptic to narrow-elliptic, glabrescent; petiole bearded; flowers 1 or 2, bright rose-pink; fruit spindle-shape, glabrous, verrucose R. pentaphyllum, p. 92. Corolla campanulate, white; leaves obovate to panduriform, subsessile with scat- tered, bristly hairs; fruit oblong-ovoid, verruculose, with thin valves, wavy at the margin when open R. nipponicum, p. 93. SECT. IV. PENTANTHERA G. DON Corolla wide-funnel-form, non-glandular. Stamens as long as the yellow corolla; leaves softly pubescent, often canescent; winter-buds velutinous R. molle, p. 95. Stamens shorter than the orange- or flame-red, rarely yellow corolla; leaves glabrous except on the veins beneath; winter-buds glabrous R. japonicum, p. 99. Corolla tubular with spreading lobes, glandular R. luteum, p. 103. ENUMERATION OF THE SPECIES SECT. I. TSUTSUTSI G. DON Rhododendron section Tsutsutsi G. Don, Gen. Syst. III. 845 (1834), excluding Nos. 30, 31, 36, 39. Azalea Linnaeus, Spec. 150 (1753), as to species No. 1. Tsutsutsi Adanson, Fam. PL II. 164 (1763). Loiseleria Reichenbach, Fl. Germ. Exc. I. 416 (1830). Beverinckia Salisbury ined. ex DeCandolle, Prodr. VII. pt. 2, 726 (1839). Rhododendron section Toutousi Duchartre in Orbigny, Diet. Univ. Hist. Nat. XI. 120 (1849). Rhododendron subgenus Tsutsia Planchon in Fl. des Serr. IX. 75 (1854); in Rev. Hort. 1854, 45. Rhododendron section Tsutsugi Miquel, Fl. Ind. Bat. II. 1059 (1857). Rhododendron section Tsusia Maximowicz in Mem. Acad. Sci. St. Peters- burg, se>. 7, XVI. No. 9, 32 (Rhodod. As. Or.) (1870), excluding No. 17. Azalea subgenus Tsutsutsi K. Koch, Dendr. II. pt. 1, 171 (1872), excluding species No. 6. Shrubs with dense twiggy branches; shoots densely or sparsely clothed with appressed, flattened, red-brown or rufous hairs, occasionally also with spreading soft hairs. Flowers from terminal bud, with leafy shoots from axils of lower scales of same bud. Leaves evergreen or half evergreen, rarely deciduous, more or less villose, rarely glandular. Corolla funnel-form, campanulate-funnel-form, rarely cylindric, with short spreading lobes; stamens 5 to 10, equal or unequal in length, never declinate, puberulous, or pubescent; ovary strigose, sometimes glan- dular. Fruit conic or conic-ovoid, glabrescent or strigose- villose, 5-celled; seeds wingless. This section is confined to eastern Asia, where it is represented by 22 species. Rhododendron indicum Sweet, Brit. Flaw. Gard. ser. 2, II. sub. t. 128 (1833). G. Don, Gen. Syst. III. 845 (1834), excluding varieties. De Candolle, Prodr. VII. pt. 2, 726 (1839), in part. Siebold & Zucca- rini in Abh. Akad. Munch. IV. pt. III. 131 (Fl. Jap. Fam. Nat. 7) (1846), in part. Miquel, FL Ind. Bat. II. 1059 (1857). Maximowicz in Mem. Acad. Sci. St. Petersburg, se*r. 7, XVI. No. 9, 37 (Rhodod. As. Or.) (1870), in part. Franchet & Savatier, Enum. PL Jap. I. 291 (1875), in part. Schneider, III. Handb. Laubholzk. II. 506, figs. 331 1-m, 332 a-d (1911). Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Komatsu in Tokyo Bot. Mag. XXXII. [18] (1918). Azalea indica Linnaeus, Spec. 150 (1753). Burmann, Fl. Ind. 42 (1768). Willdenow, Spec. I. pt. 2, 831 (1798). Persoon, Syn. 212 (1805). Aiton, 22 ENUMERATION OF THE SPECIES 23 Hort. Kew. ed. 2, 1. 318 (1810), as to description and synonyms. Roemer & Schultes, Syst. Veg. IV. 374 (1819), excluding 0. Blume, Cat. Gewass. Buitenz. 44 (1823); Bijdr. Fl. Ned. Ind. 852 (1825). Sprengel, Syst. Veg. I. 628 (1825). Azalea indica var. b. spatulata Blume, Bijdr. Fl. Ned. Ind. 852 (1825). Azalea indica var. d. angustifolia Blume Bijdr. FL Ned. Ind. 852 (1825.) Azalea indica var. e. floribunda Blume, Bijdr. Fl. Ned. Ind. 852 (1825). Azalea macrantha Bunge in Mem. Soc. tr. Sci. St. Petersbourg, II. 115 (Enum. PI. Chin. EOT. 41) (1833). Azalea indica lateritia Lindley in Bot. Reg. XX. t. 1700 (1834). Rhododendron macranthum G. Don, Gen. Syst. III. 846 (1834). Sweet, Brit. Flow. Gard. ser. 2, III. t. 261 (1834). Rhododendron decumbens D. Don apud G. Don, Gen. Syst. III. 846 (1834). Azalea Danielsiana Paxton, Mag. Bot. I. 129, t. (1834). Rhododendron indicum S. ? lateritium DeCandolle, Prodr. VII. pt. 2, 726 (1839). Azalea ? macrantha De Candolle, Prodr. VII. pt. 2, 718 (1839). Azalea decumbens De Candolle, Prodr. VII. pt. 2, 718 (1839). Rhododendron Breynii Planchon in Fl. des Serr. IX. 77 (1854). Rhododendron Danielsianum Planchon in FL des Serr. IX. 79 (1854). Rhododendron lateritium Planchon in FL des Serr. IX. 80 (1854). Rhododendron Sieboldii 0. serrulatum Miquel in Ann. Mus. Ludg.-Bat. I. 33 (1863). Azalea Breynii Andr< in Belg. Hort. XV. 183 (1865). Azalea lateritia Hort. ex Andre" in Belg. Hort. XV. 185 (1865). Rhododendron indicum j. macranthum Maximowicz in Mem. Acad. Sci. St. Petersbourg, se*r. 7, XVI. No. 9, 39 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. II, 292 (1875). Makino in Tokyo Bot. Mag. XXIV. 77 (1910). Matsumura, Ind. PL Jap. II. pt. 2, 461 (1912). Rhododendron indicum y. macranthum a. macranthum genuinum Maximowicz in Mem. Acad. Sci. St. Petersbourg, se"r. 7, XVI. No. 9, 39 (Rhodod. As. Or.) (1870). Makino in Tokyo Bot. Mag. XXIV. 77 (1910). Rhododendron indicum y. macranthum b. lateritium Maximowicz in Mem. Acad. Sci. St. Petersbourg, se>. 7, XVI. No. 9, 39 (Rhodod. As. Or.) (1870). Makino in Tokyo Bot. Mag. XXIV. 77 (1910). Rhododendron lateritium var. brachytrichum Nakai in Tokyo Bot. Mag. XXIX [261] (1915). Rhododendron hangnoense Nakai in Tokyo Bot. Mag. XXIX. [261] (1915). Rhododendron indicum f . lateritium Rehder in Mitt. Deutsch. Dendr. Ges. XXIV. 225 (1916); in Bailey, Stand. CycL Hort. V. 2944 (1916). Rhododendron indicum var. brachytrichum Komatsu in Tokyo Bot. Mag. XXXI. [18] (1918). Japan: Kyushu, prov. Osumi, Yaku-shima, side of torrents, alt. 300-800 m. February 19, 1914, E. H. Wilson (No. 6003); same place, without collector's name (Herb. Bur. Sci. Manila) ; prov. Hizen, Naga- saki, 1873, Capt. Blomfield (Herb. Kew) ; same locality, ex Herb. Lugd.- Bat. 1863, as R. Sieboldii var. minor, in part (Herb. Kew and Herb. Gray); same locality, cultivated, 1863, C. Maximowicz (Herb. Kew, co-type of R. indicum y. macranthum b. lateritum Maximowicz) . Hondo, prov. Shinano, 1864, Tschonoski (Herb. Kew); prov. Sagaim, Hakone, cultivated, June 21, 1914, E. H. Wilson (No. 6888); near Uraga, 1862, 24 THE AZALEAS OF THE OLD WORLD C. Maximowicz (Herb. Kew and Herb. Gray, co-type of R. indicum y. macranthum subvar. a. genuinum Maximowicz) ; prov. Suruga, foot of Mt. Fuji, cultivated, June 13, 1914, E. H. Wilson (No. 6932); prov. Musashi, Yokohama, cultivated, 1862, C. Maximowicz (Herb. Kew and Herb. Gray); Tokvo. cultivated. July 2, 1882, K. Miyabe', same locality (Herb. Bot. Gard. Tokyo, type of R. lateritium var. brachytrichum Nakai); prov. Shimotsuke, Nikko., alt. 600-1000 m. cultivated and sub-spontaneous, June 20 flowers, October 20 fruit, 1914, E. H. Wilson (Nos. 6816, 6818, 7709). Cultivated: Java ex Herb. Ludg.-Bat. 1863, as R. indicum Sweet (Herb. Kew); Hongkong Bot. Gard. May 12, 14, 31, 1919, W. J. Tutcher. The only place where I have seen this species truly wild is Yaku-shima, an island some ninety miles south of Kagoshima and interesting as being the southern limit of the range of a great many Japanese plants. On that island R. indicum Sweet is a common fluviatile shrub growing from a half to two meters high and forming dense masses in open country on the banks of rock-strewn streams. Elsewhere in Japan I found it cultivated only, although in old burial and temple grounds it is often more or less naturalized. Makino says it grows spontaneously in the provinces of Kii and Yamato in Hondo and not so very far from the old capital of Kyoto. In an old burial ground at Nikko the plants were prostrate or nearly so and with their scattered, large red flowers strongly suggested the Rock Cistus, and one saw how very applicable was the name the ancients gave this plant. The plants though often decumbent are naturally upright and very densely branched; the branches are slender but rigid and are clothed with flattened, appressed, strigose, chestnut- brown hairs which in the second and third years lose their color and disappear. The leaves are sub-coriaceous and crowded, short-petioled, narrow-lanceolate, lanceolate to oblanceolate, mucronulate, dark green, rather shining above, pallid or sub-glaucous beneath, ciliolate, usually remotely crenate-dentate, with scat- tered, closely appressed, red-brown setose hairs on both surfaces; in the autumn the leaves change to crimson and vinous purple. The flowers are terminal, solitary or in pairs; the corolla is broad-funnel shape, opening at the same time or im- mediately before annual growth commences, bright red to scarlet, sometimes rose- red; there are 5 stamens shorter than the style but equalling or exceeding the corolla in length. In Japanese gardens this Azalea is a prime favorite and I was told that some- thing like two hundred named forms are recognized. It is planted on or among rocks, singly or in groups, and is used as a garden edging. It bears trimming well, and its naturally low, dense habit makes it particularly useful in the gardens of Japan. Its native name is "Satsuki-tsutsuji," that is Fifth-month Azalea, from the fact that it blossoms in June, which is the fifth month of the year reckoning by the old Chinese calendar. The first mention of this plant in a European book is in Breyne's Prodromus, I. 23 (1680), which I have not seen, but in the edition of 1739 it is mentioned in pt. 1. 16 as " Chamaerhododendron exoticum, amplissimis floribus liliaceis." Breyne was a merchant of Danzig and a distinguished botanist who visited Holland dur- ing the last quarter of the 17th century and in his Prodomus described the re- markable plants he saw there. His "Chamaerhododendron " was growing in the garden of a Mr. Beveringk, but no mention is made of its country of origin. ENUMERATION OF THE SPECIES 25 Hermann in his Academici Horti lugduno-batavi catalogus 152. fig. (1687), de- scribes and figures it under the name of "Cistus Indicus, Ledi Alpini foliis & floribus amplis." He cites Breyne and says the plant came from Jaccatra (that is Batavia, Java). Hermann's figure though crude is easily recognized as repre- senting the plant we are discussing and can in fact be identified with no other species. Doubtless it was taken by Dutch trading ships from Nagasaki to Batavia and from there to Holland. Ray in his Historic, Plantarum II. 1895 (1688) men- tions it, citing Hermann and Breyne. On this plant of Breyne, Hermann and Ray, Linnaeus (Spec. 150 (1753)) bases his Azalea indica so there can be no mistake as to its identity even though Linnaeus gives the habitat as the East Indies, and cites " Tsutsusi Kaempfer Amoen. 845, t. 846," which is Rhododendron obtusum var. Kaempferi Wils. The plant of Breyne and Linnaeus is Kaempfer's " Token, vulgo Satsuki, Cytisus Liliifer autumnalis Tsutsusi congener," of which he men- tions (Amoen. Exot. fasc. V. 849 (1712)) five kinds, including single and double- flowered red and white forms. Linnaeus' species was introduced into England in 1833 by Mr. M'Killigan to Knight's Nurse^in Chelsea under the name of Azalea indica lateritia and flowered there for the first time in May 1834. About the same time forms differing slightly in color were brought from China to England by officers commanding the East India Company's trading ships and named Azalea Danielsiana, R. macranthum and R. decumbens. According to Hovey (Mag. Hort. V. 25 (1839)) Azalea lateritia and A. Danielsiana were introduced from England into Boston, Mass., in 1838. Up to about 1845 color forms, many of seedling origin, of R. indicum were more numerous in our gardens than those of R. Simsii and these with forms of R. phoeniceum and R. mucronatum (Azalea ledifolia) collectively constitute the Azalea indica of the period. It is recorded that on March 13, 1837, ten varieties of Azaka in- dica were exhibited at Ghent by Mr. Charles Deloose and awarded a gold medal. After 1850 the true Rhododendron indicum and its forms rapidly dropped from cultivation and the name was appropriated for the varieties of the Chinese R. Simsii Planch. Aiton in 1810 and Sims in 1812 were first to unite the Japanese and Chinese plants; Sweet in 1832 made the Chinese plant a variety of R. indicum. The rest of the story is told by the synonomy, but that the confusion is complete is proved by the fact that to-day Linnaeus' plant is absolutely unknown to gardens under the name he gave it. In fact it has become more or less lost to cultivation, although its double-flowered form, bakaminaeflorum, is well known and appre- ciated. R. indicum is really a very pretty species and well adapted for rockeries in mild, moist countries, and is about as hardy as R. obtusum f. amoenum Wils. I sent seeds and plants of it from Japan to the Arnold Arboretum in 1914. The Azalea "3. T. Lovett " of American gardens, claimed to be of hybrid origin is nothing but the true R. indicum. Up to about 1850 several varieties of this Azalea were cultivated under the name of "Indian Azaleas" and in all proba- bility hybrids between it and R. Simsii Planch, were obtained. Such old varieties as "The Bride "with white flowers, and Van Houtte's Azaka indica punctulata, A. punctulata variegata, A. punctulata omnicolor (Fl. des Serr. XVI. 17, tt. 1618- 1623 [1865]) are, I think, hybrids between the two species. In the evolution of the large-flowered " Indian Azalea" of commerce of to-day I doubt if R. indicum has had any part. The fact that it cannot be forced by strong heat to blossom out of season is against it, since this is a marked feature of the " Indian Azaleas" of greenhouses to-day. When grafted on the vigorous R. phoeniceum the size of the flowers is increased. A form with white flowers is : Rhododendron indicum f. hakatashiro Millais, Rhodod. 193 (1917). 26 THE AZALEAS OF THE OLD WORLD Rhododendron indicum f. Hakatajiro Komatsu in Tokyo Bot. Mag. XXXII. [45] (1918). This is often seen in old Japanese gardens and has been introduced into England. A form with salmon-red flowers whitish at the base is: Rhododendron indicum f. tanimanoyuki Millais, Rhodod. 193 (1917). Komatsu in Tokyo Bot. Mag. XXII. [45] (1918). The best known to-day of all the forms of R. indicum is: Rhododendron indicum var. balsaminaeflorum Nicholson in Kew Hand-list, pt. 2, 53 (1896); ed. 2, 499 (1902). Bean, Trees and Shrubs Brit. Isl II. 362 (1914). Millais, Rhodod. 193 (1917). Azalea Rollissoni T. Moore in Florist and Pomol. 1878, 35 fig. Garden XVIII. 254, t. 249(1880). Azalea rosaeflora R. Dean in Fl. Mag. XIX. t. 418 (1880). Azalea balsaminaeflora Carriere in Rev. Hort. 1882, 432 t. Nicholson, Diet. Gard. I. 149, fig. 196 (1887). Azalea indica var. rosifiora Render in Bailey, Cycl. Am. Hort. I. 122, fig. 176 (1900). Rhododendron indicum var. rosiflorum Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916); in Mitt. Deutsch. Dendr. Ges. XXIV. 226 (1916). Rhododendron indicum Rollisoni Millais, Rhodod. 193 (1917). This interesting little plant is common in the gardens of Japan, and it is there, as in the West, of very slow growth. It is uncertain whether it was first introduced into England or France. In the Garden, XVIII. p. 254 (1880), it is said to have been " introduced about three years ago by Messrs. Rollisson." Carriere says it was introduced into France from Japan by M. Vie*sener in 1877. On Long Island, New York, it is harcly in sheltered, shady places. Bean says " it has lived out- of-doors for many years at Kew but prefers a milder climate," Millais states that it is "a great favorite for rock gardens, where it does not always flourish." The Japanese name for the Azalea is " Romany o-satsuki." Similar but rather less double is: Rhododendron indicum f. kinnozai Millais, Rhodod. 1913 (1917). An old form is: Rhododendron indicum f. variegatum De Candolle, Prodr. VII. pt. 2, 726 (1839). Azalea, indica var. a. variegata Blume, Bijdr. FL Ned. Ind. 852 (1825). Lindley in Bot. Reg. XX. t. 1716 (1834). Hovey, Mag. Hort. V. 25 (1839). Rhododendron indicum f. Matsushima Millais, Rhodod. 193 (1917). This has white and red striped and spotted flowers and blooms irregularly the year round. In the nursery district round Osaka it is known as " Shiki-takane- satsuki." In Blume's time it was cultivated in Java, and according to Lindley it was introduced into England by Mr. M'Killigan in 1833, to Knight's nursery in Chelsea, London. From this Azalea Knight raised several seedling forms which received names. Such plants as Azalea indica exquisita Lemaire (in Fl. des Serr. III. t. 239 (1847), syn. R. indicum equisitum Lemaire) and Azalea indica Iveryana albo- ENUMERATION OF THE SPECIES 27 cincta Van Houtte (in Fl des Serr. XI. 211, 1. 1180 (1856)), are probably seedlings from Azalea indica variegata. So too are Azaleas " Gledstanesii," " Iveryana," "Charles Encke " and " Reine Louise." In fact this var. variegatum was the parent of most of the best "Indian Azaleas" cultivated up to about 1850. Ac- cording to Hovey the var. variegatum was introduced into Boston, Mass., in 1838, and for many years was a familiar and much admired plant at exhibitions. It is still cultivated in the Holm Lea collection, so also are the above mentioned named forms. These Azaleas were famous in their day and have been grown in this coun- try for more than seventy years. "Gledstanesii" was raised by Metesers Rollis- son of Tooting and exhibited for the first time in London on May 15, 1841. It was imported into this country by Hovey & Co., Boston, Mass., who exhibited it before the Massachusetts Horticultural Society on May 2, 1846. In England the var. variegatum appears to have been lost to cultivation until quite recently, when it was reintroduced under the name used by Millais. A curious and abnormal form is: Rhododendron indicum f . polypetalum Wilson, n. comb. Rhododendron indicum 7. macranthum c. polypetalum Maximowicz in Mem. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9, 40 (Rhodod. As. Or.) (1870). Cultivated: Hongkong Bot. Gard. May 20, 1919, W. J. Tutcher. I have seen the Herb. Kew co-type of this curious plant with its petaloid, deeply laciniate calyx and no corolla. In Japan its vernacular name is " Kin-no-zai." I have not seen this plant nor the following, which has a proper calyx and a deeply laciniate corolla. Rhododendron indicum f . laciniatum Wilson, n. comb. Rhododendron indicum y. macranthum subvar. b. lateritium f. laciniatum Makino in Tokyo Bot. Mag. XXVII. 110 (1910). Makino gives " Shide-satsuki " as the Japanese name of this curious form. Komatsu enumerates several other forms, but the only other it is necessary to mention here is the very distinct: Rhododendron indicum var. crispiflorum Schneider, III. Handb. Laubholzk. II. 506 (1911). Render in Bailey, Stand. Cycl Hart. V. 2944 (1916). Azalea crispiflora Hooker in Bot. Mag. LXXIX. t. 4726 (1853). Lemaire, Jard. Fl. IV. t. 430 (1854). Hovey, Mag. Hort. XXIII. 70 (1857); XXIV. 141 (1858). Azalea indica crispiflora Van Houtte in Fl. des Serr. IX. t. ad p. 79 (1854). Rhododendron crispiflorum Planchon in FL des Serr. IX. 83 (1854); in Rev. Hort. 1854, 67. This interesting variety has thicker leaves than the type and the wavy margins of its corolla give it a distinct appearance. The Azalea indica var. grandiflora Blume, of which there is a specimen in Herb. Kew collected by Thomas Lobb, is similar, but the margin of the corolla is not wavy. R. indicum var. crispiflorum, was introduced into England from China by Robert Fortune to the nursery of Messers Standish & Noble in 1850 or 1851. It was imported from England .to Boston, Mass., in 1855 by C. M. Hovey; soon afterwards a plant was acquired by Ignatius Sargent, Esq., father of Professor C. S. Sargent, and descendants of the 38 THE AZALEAS OF THE OLD WORLD original plant are in the Holm Lea collection to this day. In England it appears to have been lost to cultivation until plants were sent to Kew and elsewhere from Holm Lea by Professor Sargent. Rhododendron Kanehirai Wilson, n. sp. Frutex ramosissimus, 1-2.5 m. altus; rami graciles, ramulosi, juveniles pilis compressis castaneis adpressis vestiti. Folia petiolata, dimorpha; vernalia sparsa, lineari-lanceolata vel oblongo-lanceolata, 2-5 cm. longa, 0.5-1.5 cm. lata, acuta, glanduloso-mucronulata, basi attenuata, obscure crenato-serrata, supra atrovi- ridia, subtus pallida, utrinque pilis lucide brunneis strigosis adpressis vestita, venis supra impressis; folia aestivalia in apice ramulorum aggregata, lineari- oblanceolata vel anguste obovata, 1.5-3 cm. longa, 0.2-0.6 cm. lata, apice acuta vel rotundata, ceterum ut vernalia; petioli 1-5 mm. longi, dense pilis compressis castaneis adpressis vestiti. Flores terminales, bini vel raro solitarii, carminei vel ecarlatini; pedicelli robusti, erecti, 5-8 mm. longi, dense pilis castaneis strigosis vestiti; calyx plerumque bene evolutus; lobi virides, erecti, rotundati vel ovales, 1-5 mm. longi, 1-2 mm. lati, pilis adpressis lucide castaneis ciliati; corolla anguste infundibuliformis, 2.5-4 cm. longa et 2-2.5 cm. lata, glabra; stamina 10, ineaqualia, corolla breviora, filamentis gracilibus infra medium papillose pubescentibus, an- theris 1-1.5 mm. longis; ovarium dense pilis cinereis vel castaneis strigosis ob- tectum; stylus gracilis, 3-4 cm. longus, ad basin pilis paucis compressis strigosis instructus, stigmate parvo capitato. Fructus non visus. Formosa: prov. Taihoku, Urai, in a garden at the police station, April 1, 1918, E. H. Wilson (No. 10,276). This new species is closely related to R. indicum Sweet, which is distinguished chiefly by its smaller leaves, rather larger, usually solitary flowers, its 5 stamens with anthers from 2 to 3 mm. long and by its stouter and glabrous style. The chief morphological difference is the number of stamens, and it may eventually prove that this new species is a decandrous variety of the old Linnean plant. I saw it in the garden mentioned above and was told that the plants had been brought from the higher slopes of the surrounding mountains. It is distinct from anything else I have seen and apparently has nothing to do with any of the species described by Hayata. The flowers vary in color from carmine-red to scarlet. Rhododendron breviperulatum Hayata, Icon. PI. Formos. III. 129 (1913). Kanehira, Formos. Trees, 318, fig. 16 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [14] (1918). Formosa : prov. Nanto, Horisha, planted in Hotel garden, December 6, 1918, E. H. Wilson (No. 11,206). This red-flowered Azalea is well characterized by its 5 stamens and its pilose style; also by its leaves, which are oblong to oblanceolate or more rarely oblong- ovate, with no conspicuous difference between those of the spring and summer. The calyx-lobes are membranous, reddish, from 3 to 4 mm. long, ciliate and pilose without, glabrous within. The corolla is from 2.5 to 3 cm. long and broad; the Stamens are nearly as long as the corolla and have rather large purple anthers. The young shoots and leaves are clothed with red-brown to gray appressed hairs. The plants my specimens came from were from 1 to 1.5 cm. tall and had rather slender branches. I was told that they had been brought from the neighboring mountains. ENUMERATION OF THE SPECIES 29 I cannot be absolutely sure of the correct identification of this plant. It agrees very well with Hayata's description, though the calyx and corolla are larger than he describes them, and the apiculate base of the anthers is by no means very apparent. Rhododendron obtusum Planchon in Fl. des Serr. IX. 80 (1854); in Rev. Hort. 1854, 64. Schneider, III Handb. Laubholzk. II. 506, fig. 332e (1911). Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Jour. Jap. Bot. I. [173] fig. (1917). Komatsu in Tokyo Bot. Mag. XXXII. [39] 1918. Miyazawa in Tokyo Bot. Mag. XXXII. [318] (1918). Azalea indica Thunberg, FL Jap. 84 (1784), in part, not Linnaeus. Juel, PI. Thunb. 391 (1918). Azalea obtusa Lindley in Jour. Hort. Soc. Lond. I. 152 (1846); in Bot. Reg. XXXII. t. 37 (1846). Rhododendron Thunbergii Planchon in FL des Serr. IX. 78 (1854). Rhododendron Sieboldii minor Miquel in Ann. Mus. Ludg.-Bat. I. 33 (1863); II. 164 (1865-66); Prol. FL Jap. 96 (1866-67). Azalea Thunbergii Andre" in Belg. Hort. XV. 184 (1865). Rhododendron indicum 5. obtusum Maximowicz in Mem. Acad. Sci. St. Peters- bourg, ser. 7, XVI. No. 9, 40 (Rhodod. As. Or.) (1870). Franchet & Sava- tier, Enum. PL Jap. I. 292 (1875). Matsumura, Ind. PL Jap. II. pt. 2, 461 (1912). Bean, Trees and Shrubs Brit. Isles, II. 362 (1914). Millais, Rhodod. 193 (1917). Azalea indica var. obtusa Render in Bailey, Cycl. Am. Hort. I. 122 (1900). Rhododendron indicum Kirishima Millais, Rhodod. 193 (1917). Rhododendron obtusum f . a. Honkirishima Komatsu in Tokyo Bot. Mag. XXXII. [39] (1918). Japan: Kyushu, prov. Osumi, Kirishima-jinza in a garden, said to have come from Nishi-Kirishima, May 5, 1918, E. H. Wilson (No. 10,328) ; prov. Hizen, Nagasaki, G. H. Langsdorff (Herb. Kew); same locality, cultivated, 1863, C. Maximowicz (Herb. Kew, Herb. Gray); probably same locality, von Siebold (Herb. Kew, Herb. Gray) ; Hondo, prov. KU, May 3, 1886, H. Mayr-, prov. Musashi, Tokyo, cultivated, May 11, 1883 (Herb. Kew); same locality, May 12, 1904, K. Sakurai; cultivated, Imperial Botanic Garden, Tokyo, March 29, 1914, E. H. Wilson (No. 6338) ; Yokohama, cultivated, Yokohama Nursery Company's grounds, April 30, 1918, E. H. Wilson] Tokyo, Hilgendorf, ex Mus. Bot. Berol. (Herb. Bur. Sci. Manila); prov. Rikuzen, Sendai, cultivated, May 26, 1888, U. Faurie (No. 2296, Herb. Kew); prov. Ugo, Akita, cultivated, May 20, 1888, U. Faurie (No. 2132, Herb. Kew); Japan, without locality, L. Savatier (Herb. Kew) ; without locality or collector's name, ex Herb. Lugd.-Bat. (Herb. Kew, Herb. Gray as R. Sieboldii var. minor Miquel). 30 THE AZALEAS OF THE OLD WOULD It is always unfortunate when a selected or garden form has to do duty as the type of a species. Such is the case here and the result is that the phylogenetic order is inverted. Under the name of " Kirishima-tsutsuji " this plant has for cen- turies been grown in Japanese gardens. By this name it is mentioned by Kaempfer (Amoen. Exot. fasc. V. 849 [1712]), and his description is perfectly accurate. It is a densely branched, twiggy plant, seldom more than a metre high and usually less, and has small, scarlet, slightly scented flowers. Like all its relatives it is exceed- ingly floriferous, and in April and early May, as Kaempfer says, the whole plant appears blood red. In the autumn the leaves are lustrous and assume fine crimson tints. The typical form has rather pointed corolla-lobes, as shown in Lindley's figure, and this is the one so abundantly grown around Tokyo and Yokohama. At Kurume in Kyushu this typical form is called " Hiryu." In the nursery district round Osaka a form with rounded corolla-lobes is grown and sold as the Kirishima Azalea, and it is this form that is grown in many western gardens as the real R. ob- tusum Planch. At Kurume what appears to be this form is called " Hino-tsukasa." A hose-in-hose form of the type is grown around Tokyo as "Yayegiri" and at Kurume as " Yayehiryu." In addition a number of other slight forms are grown, some of which are mentioned by Komatsu. Like a number of other Japanese Azaleas this plant was first sent to western gardens from China. Robert Fortune found it in the Pou-shan Azalea gardens near Shanghai, and sent it to the Horticultural Society of London in 1844. For- tune saw in the same Chinese garden the hose-in-hose (semi-double) form. Fortune's introduction was named and figured by Lindley. The figure shows the obovate, rounded, obtuse, mucronate summer leaves of the previous year and two newly developed elliptic-lanceolate, acute spring leaves. Too little attention has been paid to the fact that the red-flowered species of this section of Rhododendron all have dimorphic leaves. Those which are formed immediately after the flowers open are from lanceolate to ovate or elliptic in shape, acute, light green in color and membraneous in texture. The "spring leaves" are followed by "summer leaves," which are smaller, more or less obovate in shape, obtuse or rounded at the apex, dark green and fairly coriaceous. These spring leaves are normally de- ciduous and the summer leaves persistent, but climate exercises a strong in- fluence, the colder it is the more deciduous the leaves, and vice versa. In the early summer when the larger spring leaves predominate, the plant presents a very different appearance to what it does in late autumn or early spring when the small obovate summer leaves only are present. The overlooking or inappreciation of this simple fact has resulted in much confusion in the classification of the species and forms. Now as to the origin of R. obtusum. Its name " Kirishima Azalea " would appear to suggest its original habitat. Working on this clue I visited Mt. Kirishima and am convinced that the so-called " Kirishima Azalea " is simply a color form of the Azalea so extraordinarily abundant on this mountain and hereunder referred to as f . japonicum. Long ago some pilgrim to this sacred mountain took back a plant or plants as a souvenir, planted it in his garden and from it (or them) has sprung all the plants of this particular Azalea in gardens to-day. The Japanese have always been fond of making pilgrimages to their sacred shrines and mountains, and their houses and gardens are often museums filled with souvenirs of such visits. The Azalea known as amoena is unquestionably of like origin, and so too is " Hinodegiri," a more recent comer but an established favorite in our gardens. Further, the race of Azaleas known as " Kurume Azaleas," with which we are only just now becoming acquainted, is a garden product derived by long cultiva- tion from the same common stock the vari-colored Azalea of Mt. Kirishima, f . japonicum Wils. ENUMERATION OF THE SPECIES 31 A form with white flowers is: Rhododendron obtusum f . album Schneider, III. Handb. Laubholzk. II. 506 (1911). Render in Bailey, Stand. Cyd. Hort. V. 2944 (1916). Azalea ramentacea Lindley in Jour. Hort. Soc. Lond. IV. 291, fig. (1849). Rhododendron ramentaceum Planchon in Fl. des Serr. IX. 81 (1854); in Rev. Hort. 1854, 65. Azalea obtusa alba Mottet, Diet. Hort. Jard. I. 302 (1892-93). Rhododendron (Azalea} Indicum obtusum album, Garden and Forest, IX. 394, fig. 52 (1896). Bean, Trees and Shrubs Brit. Isl. II. 362 (1914). Millais, Rhodod. 193 (1917). Azalea indica var. obtusa f . alba Render in Bailey, Cyd. Am. Hort. I. 122 (1900). Azalea amoena alba Moller's Deutsch. Gartn. Zeit. XVIII. 476, fig. (1903). Cultivated : Hort. Holm Lea, Brookline, Mass.; Hort. Berckmans, Augusta, Georgia, March 18, 1909, April 6, 1914, C. S. Sargent. This is another of Fortune's introductions from China to the garden of the Horticultural Society of London. It was received in May, 1846, from Hongkong, and certainly from a garden (although this is not expressly stated) and doubt- less of Japanese origin. It has the dense, twiggy habit characteristic of its class, pure white flowers with a normal calyx and rather large leaves. In every essen- tial respect it agrees with f. japonicum Wils., of which it is nothing but a selected, cultivated form, as are A. obtusa, A. amoena, "Hinodegiri" and the "Kurume Azaleas." Among the latter are several white-flowered forms with relatively large leaves. Under f . japonicum the origin of this "obtusum" group of Azaleas is dis- cussed, but I may add here that variants of f . japonicum having white flowers have been found wild on Nishi-Kirishima. Such a plant is growing in a garden at the foot of the mountain and from it I gathered specimens. I have been unable to find out just when this plant was introduced into America but in all probability it was very soon after its appearance in England. For more than sixty years it has been cultivated in the Holm Lea collection and from there was sent by Professor Sargent to England, where it appears that it had become lost, and to the Berck- mans and to other gardens. A form with a small corolla and long exserted stamens is: Rhododendron obtusum f. macrostemon Wilson, n. comb. Rhododendron macrostemon Maximowicz in Mem. Acad. Sci. St. Petersbourg, ser. 7, XVI. No. 9, 41, t. 3, figs. 15-20 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 292 (1875). Okubo in Tokyo Bot. Mag. IX. 40 (1895). Schneider, III. Handb. Laubholzk. II. 507, fig. 332 h-k (1911). Millais, Rhodod. 205 (1917). Rhododendron indicum var. macrostemon Okubo & Makino in Tokyo Bot. Mag. XVI. 178 (1902). Matsumura, Ind. PL Jap. II. pt. 2, 461 (1912). Rhododendron Kaempferi var. macrostemon Makino in Jour. Jap. Bot. I. 18 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [43] (1918). Japan : Hondo, prov. Musashi, Tokyo, cultivated, 1862, C. Maxi- mowicz (Herb. Kew, co-type). This garden plant is characterised by its small flowers, minute calyx, and long exserted stamens, which are three times as long as the corolla. It is, however, 32 THE AZALEAS OF THE OLD WORLD more a condition of than a distinct form. In all the forms of R. obtusum, including the large flowered var. Kaempferi, the stamens are variable in length, and are in- constant in this from year to year. In the wild f . japonicum and the cultivated " Kurume Azaleas " the stamens are conspicuously variable in length. The same is true of the/, multicolor. Besides the co-type cited above there is another Japanese specimen in the Kew Herbarium from Maximowicz, collected by Tschonoski in Shinano province, central Hondo, in 1864, and labelled, apparently in Maximo- wicz's handwriting, " Rhod. indicum a. Kaempferi forma subalpina." This differs only in having larger flowers than those of his R. macrostemon. A well-known form with magenta-colored flowers is : Rhododendron obtusum f. amoenum Wilson, n. comb. Azalea amoena Lindley in Paxton, Flow. Gard. III. 81, t. 89 (1852). Hooker in Bot. Mag. LXXIX. t. 4728 (1853). Rev. Hort. 1854, 241, t. Van Houtte in Fl. des Serr. IX. 75, t. 885 (1854). Hovey, Mag. Hort. XXII. 9 (1856); XXIII. 70 (1857); XXIV. 141 (1858). Rhododendron amoenum Planchon in Fl. des Serr. IX. 80 (1854); in Rev. Hort. 1854, 64. Bean, Trees and Shrubs Brit. Isl. II. 341 (1914). Rhododendron Buergeri Miquel in Ann. Mus. Lugd.-Bat. I. 34 (1863); in II. 164 (1865-66); Prol. Fl Jap. 96 (1866-67). Rhododendron indicum c. amoenum d. genuinum Maximowicz in Mem. Acad. Sd. St. Petersburg, s6r. 7, XVI. No. 9, 41 (Rhodod. As. Or.) (1870). Franchet and Savatier, Enum. PL Jap. I. 292 (1875). Matsumura, Ind. PI. Jap. II. pt. 2, 461 (1912). Millais, Rhodod. 193 (1917). Rhododendron indicum e. amoenum b. normale Maximowicz in Mem. Acad. Sci. St. Petersburg, se'r. 7, XVI. No. 9, 41 (Rhodod. As. Or.) (1870). Rhododendron indicum c. amoenum c. Buergeri Maximowicz in Mem. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9, 41 (Rhodod. As. Or.) (1870). Azalea amoena Caldwelli Caldwell apud Van Geert, Icon. Azal. Ind. 45, 1. 17 (1882). Azalea indica var. amoena Render in Bailey, Cycl. Am. Hort. I. 122 (1900). Rhododendron Kaempferi var. amoenum Render in Sargent, Trees and Shrubs, II. 30 (1907). Schneider, III. Handb. Laubholzk. II. 505, fig. 332, f-g (1911). Rhododendron obtusum var. amoenum Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Komatsu in Tokyo Bot. Mag. XXXII. 41 (1918). Azalea amoena Hatsu-giri Millais, Rhodod. 113 (1917). Japan: Kyushu, prov. Hizen, cultivated, Nagasaki, 1863, C. Maximowicz (Herb. Kew, Herb. Gray). Hokkaido, prov. Oshima, Hakodate, cultivated, 1861, C. Maximowicz (Herb. Kew as f. normale Maxim.). Without locality, probably Nagasaki, ex Herb. Lugd.- Bat. (Herb. Kew as R. Buergeri Miquel). Cultivated: Hort. Standish (Herb. Kew, type of Bot. Mag. t. 4728). Arnold Arboretum, May 29, 1884, May 24, 1902 (Nos. 862-1, 862-2). This Azalea, with its small, rich magenta-colored flowers, its small leaves and dense, twiggy habit, is too well known to need further description. It was first sent to England in 1850 or 1851 from Shanghai by Robert Fortune and was soon recognized as the most hardy of its class. It is generally assumed to be of Chinese ENUMERATION OF THE SPECIES 33 origin, but there is no evidence to support this view. On the contrary, I do not think there can be any doubt whatever that it is a Japanese plant and merely a color form of the Azalea so abundant on Nishi-Kirishima, Kyushu, hereunder de- scribed as f . japonicum. Moreover it is simply one of a race of charming Azaleas which for more than one hundred years have been grown and developed in the city of Kurume in the southern island of Kyushu, Japan, and is now becoming known as " Kurume Azaleas." The typical form of amoenum has hose-in-hose flowers and is known at Kurume under the name of " Kochonomai"; it is considered the hardiest of all. The single-flowered form, the f . normale of Maximowicz, is known and more especially around Tokyo, as " Hatsu-giri." The peculiarity of the calyx becoming petaloid and forming a hose-in-hose flower is common to the whole race of " Kurume Azaleas " and is found in every one of the white and colored forms, and by selection and vegetative propagation it becomes fixed. The typical amoenum with hose-in-hose flowers was introduced into America by C. M. Hovey, Boston, Mass., in 1855 and soon established itself as a favorite exhibition plant. There are numerous named forms of this Azalea, and in 1879 Pynaert (Rev. Hort. Belg. V. 277), figures Azalea amoena "Mrs. Carmichael " and mentions five others that were in commerce. In recent years, at least, amoenum has been used in hybridizing with forms of R. Simsii and the result is a beautiful race of small-flowered, hose-in-hose Azaleas of which the well-known " Hexe " ("Firefly" of American gardens) is a familiar example. Near Boston amoenum is not entirely hardy but on Long Island and southward it is perfectly so. The wild form of the species is: Rhododendron obtusum f. japonicum Wilson, n. comb. Rhododendron indicum . amoenum a. japonicum Maximowicz in Mem. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9, 41 (Rhodod. As. Or.) (1870). Rhododendron Kaempferi var. japonicum Render in Sargent, Trees and Shrubs, II. 30 (1907). Schneider, III. Handb. Laubholzk. II. 505 (1911). Rhododendron indicum var. japonicum Makino in Tokyo Bot. Mag. XXII. 56 (1908). Rhododendron kiusianum Makino in Tokyo Bot. Mag. XXVII. 174 (1914); in Jour. Jap. Bot. I [173J fig. (1917). Komatsu in Icon. PL Koisikav. II. 95, t. 132 (1915); in Tokyo Bot. Mag. XXXII [18] (1918). Rhododendron amoenum var. japonicum Bean, Trees and Shrubs Brit. Isl. II. 341 (1914). Millais, Rhodod. 113 (1917). -^ Rhododendron obtusum Miyazawa in Tokyo Bot. Mag. XXII. [318] (1918), not Planchon. . , ^ ^^ ^ Japan: Kyushu, prov. Hizen, Wunzen (Mt. Onsen) near Naga-^-^- Vv ^. saki, 1863, C. Maximowicz (Herb. Kew, Herb. Gray); prov. Osumi, Kirishima-jinza, in a garden, October 1, 1918, E. H. Wilson (No. 10,957); Nishi-Kirishima, alt. 600-1600 m., May 5, 1918, E. H. + Wilson (Nos. 10,329, 10,333); same locality, alt. 1300 m., March 10, ****^^ f^T 1914, E. H. Wilson (No. 6255); same locality, November 1, 1918,^****" + J*W E. H. Wilson (Nos. 11,248, 11,250, 11,255 seeds only); same locality, June 1, 1916, S. Kawagoe; same locality, May 18, 1917, Z. Tashiro; same mountain, Makizono, May 4, 1918, B. Miyazawa; Kirishima-jinza, cultivated, plant said to have come from Mt. Kirishima, May 5, 1918, 34 THE AZALEAS OF THE OLD WORLD E. H. Wilson (Nos. 10,327, flowers white). Hondo, prov. Settsu, cultivated round Osaka, May, 1918, E. H. Wilson (form " Hinode- giri") ; prov. Musashi, cultivated, Hatagaya, Hort. Oishi, May, 1914, E. H. Wilson (80 named forms of " Kurume Azaleas ") Cultivated: Hort. J. S. Ames, 1920 (10 forms of "Kurume Azaleas"); Hort. Kew, May 17, 1909; Hort. Holm Lea, March, 1920 (50 forms "Kurume Azaleas"). This is the phylogenetic type of the race of small flowered Azaleas so plentiful in Japan and of which amoena, obtusa, and "Hinodegiri" are perhaps the best known in western gardens. It is especially abundant on the wind-swept slopes of Nishi-Kirishima, an active volcano in south Kyushu. It is found above the tree-level, growing among lava and in volcanic ashes with coarse grasses and mis- cellaneous low shrubs. At its lowest altitudinal range the leaves are persistent, but higher up they are deciduous. The plants are seldom a metre tall, more usually less than half of this, and quite commonly they are prostrate or hug boulders closely. The habit is normally dense and twiggy, but when sheltered a few strong shoots develop and the plant becomes relatively tall and sparsely branched. The leaves vary greatly in size according to altitude and exposure and are smallest on the plants which annually shed their foliage. On an average plant the spring leaves are oval to elliptic-ovate, from 1 to 2 cm. long and from 0.5 to 1 cm. wide, but they may be smaller or larger. The upper surface is bright green and the under usually pallid. The summer leaves are darker, more coriaceous and from oblanceolate to obovate in shape, with the mucronate apex rounded or acute. Normally the leaves are tufted at the ends of the twigs, but on free-growing shoots they are scattered; both shoots and leaves are clad with appressed, pale gray to yellowish, straight hairs. The flowers vary a little in size, in degree of fragrance, and greatly in color; the most common shades are rosy mauve to magenta, but salmon and salmon-red are plentiful; pink, scarlet and crimson are rather rare. Plants bearing white flowers are occasionally found. It is very floriferous; each twig terminates in a flower-bud, often in a cluster of from three to five, and from each one to three flowers develop in the spring. The bud-scales are dry, nowhere glandular, and are ciliated with pale gray hairs. The calyx is usually well developed with green, oval to nearly orbicular, ciliate lobes; sometimes it is quite small. The corolla lobes may be rounded or pointed and the stamens may be shorter or longer than the corolla; the anthers vary in color from pale yellow-brown to deep magenta, this being correlated with the color of the flowers. The capsule is variable in size and the calyx-lobes are persistent. From the above facts is it apparent that this plant is responsive to varying ecological conditions and that its extreme forms present a very different appear- ance. This is indeed true, nevertheless this form is well marked and, except that there is no definite demarkation between it and the variety Kaempferi, presents no difficulty unless an attempt be made to subdivide it. This I find impossible to do, and fortunately it is not necessary, since all the color forms have received names in gardens and may be classed under the general title of " Kurume Azaleas," from a town in Kyushu where much attention has been paid to their cultivation and development. The type of this form came from Mt. Onsen, a volcano not far from Nagasaki, but I do not know whether it is plentiful there. It also grows on other Kyushu mountains. On Kirishima it is abundant and when in flower forms sheets of color on the bare or grass-clad, bleak slopes and is strikingly attractive. Mt. Kirishima, of ENUMERATION OF THE SPECIES 35 which there are two principal peaks, Nishi and Higashi, is a sacred mountain cele- brated in Japanese mythology. It was on Nishi-Kirishima that the god Ninigi, grandson of the sun-goddess Amaterasu, alighted when he came down from heaven to pave the way for the conquest of Japan. A bronze sword, fixed in the ground, hilt upward, crowns the peak and is considered to be a relic of this divinity. To a place so sacred pilgrimages have been made by the Japanese from time immemo- rial. With their profound love for flowers some of the pilgrims would certainly take back as a souvenir living plants of this charming Azalea. Naturally it was named for the mountain, and in the course of time was distributed widely in the gardens of Japan. It is easily understood that a plant bearing flowers of an un- usual color would be that selected as a souvenir by the average pilgrim. It is such forms that reached gardens first, and so we find the red obtusa the magenta amoena, the white alba to be the earliest known. In the autumn of 1918 I collected a quan- tity of seeds on Nishi-Kirishima for the Arnold Arboretum. These have been widely distributed, and it will be interesting to watch the range of variation in the habit foliage and flowers of resultant plants. I fear, however, that they will not prove so hardy as the var. Kaempferi, but that they will find a welcome in gardens I am quite sure. On visiting Nishi-Kirishima, where between 650 and 1500 m. altitude this form japonicum grows in great abundance, and after seeing the wide range of color among the flowers, I am absolutely convinced that this Azalea is the wild parent of the race known as " Kurume Azaleas." The plant ( R. obtusum Planch.) which, following International Rules, has to do duty as type of this species, also belongs to this race, and of it too there is a hose-in-hose form known as " Yayehiryu " in Kurume. Another of these Azaleas which in recent years has become widely known is the brilliant red-flowered " Hinodegiri." Curiously this form so abundant in Tokyo and Osaka gardens is to-day not grown at Kurume. But good as the forms named above are, they are surpassed by many hardly known in western gardens, and in fact scarcely known in Japan outside of Kurume and a narrow circle of enthusiasts. Millais in his Rhododendrons makes the first reference to the race in a European publication. I first became acquainted with "Kurume Azaleas" in 1914, when at the invitation of Mr. H. Suzuki of the Yokohama Nursery Company and a foremost Japanese horticulturist, I accompanied him on a visit to the nursery district at Hatagaya, a few miles north of Tokyo. There, in the garden of Mr. Oishi, I saw thousands of tiny plants bearing white and colored flowers of nearly every hue. I secured a set of fragments and dried them for our herbarium, and later, in 1916, at my suggestion Mr. John S. Ames, North Easton, Mass., pur- chased a selection of living plants from Mr. Oishi's collection. These arrived in the spring of 1917 and were the first " Kurume Azaleas " introduced into eastern North America. In May, 1917, I saw in the Yokohama Nursery a few large plants of " Kurume Azaleas," trained into umbrella-shape, in full flower and this decided me to visit the place of their origin. This became possible in the spring of 1918 and I was fortunate to have as a companion my friend Mr. H. Suzuki. Kurume is some 800 miles south by west from Tokyo and is quite an important city, but the fame of its Azaleas is likely to make it better known in the near future. The Azaleas are grown in a number of gardens and nurseries, but the oldest and best collection is that of Mr. Kijiro Akashi. This fine old gardener has for more than forty years devoted himself to the development of this race of Azaleas and has raised from seeds and perpetuated by cuttings nearly all the forms in cultivation. Mr. Akashi gave me the history of the race, and it seems that it originated about one hundred years ago with a gentleman named Motozo Sakamoto, who lived at Kurume. The parents came from sacred Mt. Kirishima, but whether brought from there by Sakamoto himself or given to him by some pilgrim is uncertain. At any rate, 36 THE AZALEAS OF THE OLD WORLD Sakamoto cultivated several kinds and raised and selected seedlings, including one he named "Azuma-kagami," from which it is claimed all the pink-flowered forms have descended. After the death of Mr. Sakamoto his collection passed into the possession of Akashi with the result above stated. The original plant of " Azuma- kagami " is still healthy, and I tried but without success to induce Mr. Akashi to part with it. Japanese experts recognize by name more than 250 kinds of Kurume Azaleas, but the differences are often infinitesimal. I made a selection of fifty kinds in duplicate which the following year (1919) safely reached the Arnold Arboretum. The two leading experts, Messers Akashi and Kuwano, at my suggestion selected the following six as the pick of them all: " Takasago " (pale pink, hose-in-hose), " AzumaKagami" (deep pink, hose-in-hose), "Kirin" (deep rose shading to silvery rose, hose-in-hose), "Kumo No Uye" (pure salmon), "Kurai No Himo" (car- mine, hose-in-hose), "Kureno Yuki" (white, hose-in-hose). My companion and I concurred, but had the number been twenty the task would have been much easier. To my mind the " Kurume Azaleas " are the loveliest of all Azaleas. The colors are so pure and exquisite and of every hue and shade from white, pink and salmon to scarlet, crimson and the richest magenta. Many of them are delicately fragrant. The hose-in-hose forms have none of the ugly features of double flowers, but on the contrary they are singularly attractive and the flowers last longer. The stamens, always five, are straight, exserted or included and the anthers vary in color from pale yellow-brown to blackish crimson; the color of the corolla and that of the anthers being correlated. The plants are often trained as low standards with a compact umbrella-shaped crown; less commonly they are dense and globose, or open and irregular in form. They are extremely floriferous, and in season the blos- soms often completely hide the leaves. The leaves are of course dimorphic, but apart from this they vary considerably in size, in their shades of green, in their autumn-coloring and in their degree of persistence. In a great measure these variations are correlated with the color of the flowers, and experts in Japan can with ease distinguish each variety by its foliage and general appearance. It is strange that a race so rich in forms and of such decorative value should have remained un- known to us until now. Yet the explanation is simple. Interior Kyushu is little known to the western world, and even to those Japanese whose home is on the other islands. The feudal system of government which until comparatively recently obtained in Japan created and preserved this aloofness. Further, Kurume is remote from the horticultural centres of Osaka and Yokohama, from whence we have drawn the bulk of our garden plants and where a business is made of growing for export. Nagasaki is much nearer, but in the days of Thunberg and Siebold, Veitch, Maximowicz and Fortune intercommunication was difficult, and for for- eigners impossible. And so it has resulted that the product of Sakamoto's hobby richly developed by Akashi has remained hidden from the outside world until now. During the last decade the fame of the Kurume Azaleas has reached Osaka, Tokyo and other places, and growers have obtained stocks and are propagating them apace. Unfortunately every grower and enthusiast names the plants accord- ing to his fancy and the result in a few years will be chaos. And this is helped by the fact that every slight sport or variant is kept and named and no attempt at selection made. I do not see how it is possible to improve upon the strain grown in Kurume unless yellow could be injected. What is needed is rigorous selec- tion and the reduction of the named forms to fifty or less. In the past seedling selection and preservation of sports by vegetative propagation have been the sole means employed in the evolution of the race of "Kurume Azaleas," but now attempts at hybridising them with the large flowered " Indian " and " ledifolia " types are in progress. This may result in a new race, but whether it will be as lovely and ENUMERATION OF THE SPECIES 37 fascinating as the present one is doubtful. In the matter of their origin Mr. Akashi was careful to inform me that for purposes of policy he had not before set forth the facts so clearly and that more or less distorted versions had appeared in print in Japan. To test the story for ourselves Mr. Suzuki and I on leaving Kurume visited and ascended Nishi-Kirishima and were convinced. In the garden of Mr. Akashi I selected and obtained a double set of the under- mentioned fifty varieties of these Azaleas which I consider the best, and these safely reached the Arnold Arboretum on April 24, 1919. The names are those of the originator, Akashi, and therefore authentic. The plants will be propa- gated and distributed under these names and, owing to the reprehensible habit of changing names by Japanese nurserymen, will probably remain the only set with standard names in existence. No. 1. SEIKAI flowers white, hose-in-hose 2. KURENO YUKI flowers white, hose-in-hose 3. SHIN SEIKAI flowers creamy white, hose-in-hose 4. YOROZUYO flowers white 5. NANI WAGATA flowers white, suffused with salmon-pink 6. TANCHO flowers flesh-color, hose-in-hose 7. HACHIKA TSUGI flowers white, suffused with lavender 8. IROHAYAMA . flowers white, margined with pale lavender 9. Hoo flowers white, tinged with pink 10. STJIYOHI flowers flesh-color 11. TAKASAGO flowers pale pink, the color of apple-blossoms 12. KASUMI GASEKI flowers pale pink 13. BIJINSUI flowers pale pink 14. ASAGASUMI flowers rose-pink, hose-in-hose 15. KIMIGAYO flowers pink 16. AZUMA KAGAMI flowers deep pink, hose-in-hose 17. OSARAKU flowers white, suffused and margined with lavender 18. OTOME flowers blush pink 19. AYA KAMMURI flowers rose-color 20. SHINTOKI No HAGASANE . flowers rose, shading to pink, hose-in-hose 21. SAOTOME flowers rose-color 22. KIRIN flowers deep rose, shading to silvery rose 23. TAMAFUYO flowers white, striped peach-color 24. KIRITSUBO flowers rosy mauve 25. OMOINE flowers pale lavender 26. OINO MEZAME flowers deep rose-color 27. KATSURA No HANA . . . flowers rose-color 28. SHIN UTENA flowers pale salmon 29. KUMO No UYE flowers pure salmon 30. BENIFUDE flowers salmon 31. SUGA No ITO flowers pure pink 32. KASANE KAGARIBI .... flowers dull salmon-red 33. TSTJTA MOMIJI flowers bright red 34. SUETSUMU flowers crimson 35. FUDESUTE YAM A flowers light red 36. IMA SHOJO flowers bright red, hose-in-hose 37. RASHO MON flowers scarlet 38. WAKA KAYEDE flowers red 39. YAYEHIRYU flowers bright scarlet, hose-in-hose 38 THE AZALEAS OF THE OLD WORLD 40. KURAI No HIMO flowers carmine, hose-in-hose 41. AGEMAKI flowers carmine ^42. HINODEGIRI flowers bright crimson 43. AIOI flowers color of almond-blossoms, hose-in- hose 44. SAKURA TSUKASA. . . . , flowers rosy mauve 45. TAMA No UTENA flowers pale salmon 46. GOSHO ZAKURA flowers white, striped peach-color 47. UKAMUSE flowers vermilion, hose-in-hose 48. HINODE No TAKA .... flowers crimson 49. OSARAKU SEEDLING . . . flowers white, suffused and margined with lavender 50. HANA ASOBI flowers red Rhododendron obtusum var. Kaempferi Wilson, n. comb. Azalea indica Thunberg, Fl. Jap. 84 (1784), in part, not Linnaeus. Rhododendron Kaempferi Planchon in Fl. des Serr. IX. 77 (1854); in Rev. Hort. 1854. Rehder in Sargent, Trees and Shrubs, II. 29, t. 113 (1907), flowers here represented with more than 5 stamens; in Bailey, Stand. Cycl. Hort. V. 2944, fig. 3393 (1916). Schneider, III. Handb. Laubholzk. II. 505, fig. 331 g-k (1911). Bean, Trees and Shrubs Brit. Isl. II. 363 (1914). Millais, Rhodod. 197, fig. (1917). Komatsu in Tokyo Bot. Mag. XXXII. [16] (1918). Rhododendron Sieboldii Miquel in Ann. Mus. Lugd.-Bat. I. 33 (1863), exclud- ing varieties; in II. 164 (1865-66); Prol. Fl. Jap. 96 (1866-67). Azalea Kaempferi Andre" in Belg. Hort. XV. 184 (1865). Rhododendron indicum a. Kaempferi Maximowicz in Mem. Acad. Sci. St. Petersburg, s6r. 7, XVI. No. 9, 38 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PI. Jap. I. 291 (1875). Matsumura, PL Nikko, 71 (1894); Ind. PL Jap. II. pt. 2, 460 (1912). Shirasawa, Icon. Ess. For. Jap. II. t. 61, figs. 1-10 (1908). Azalea Indica var. Kaempferi Rehder in Bailey, Cycl. Am. Hort. 1. 122 (1900) ; in Holler's Deutsch. Gartn.-Zeit. XVII. 417, figs. (1902). Rhododendron scabrum var. Kaempferi Nakai in Tokyo Bot. Mag. XXXIII. 208 (1919). Rhododendron scabrum var. Kaempferi f. 2. latifolium Nakai in Tokyo Bot. Mag. XXXIII. 209 (1919). Rhododendron scabrum var. Kaempferi f. 3. tubiflorum Nakai in Tokyo Bot. Mag. XXXIII. 209 (1919). Komatsu in Icon. Koisikav. IV. 7, t. 216 (1918). Rhododendron scabrum var. Kaempferi f . 4. angustifolium Nakai in Tokyo Bot. Mag. XXXIII. 209 (1919). Komatsu in Icon. Koisikav. IV. 31, t. 228 (1918). Japan: Kyushu, prov. Osumi, Yaku-shima February 25, 1914, E. H. Wilson (No. 6082); Mt. Kirishima, western slopes, alt. 1000- 1500 m., March 8, 1914, E. H. Wilson (No. 6228) ; lower slopes of Nishi- Kirishima, alt. 600 m., May 5, 1918, E. H. Wilson (No. 10,334); foothills round base of Nishi-Kirishima, sea-level to alt. 500 m., May 4, 1917, E. H. Wilson (Nos. 10,326, 10,330, 10,334); same general locality, April 29, 1917, Z. Tashiro; prov. Satsuma, Kagoshima, March 15, 1917, E. H. Wilson; same locality, May 6, 1918, E. H. ENUMERATION OF THE SPECIES 39 Wilson (No. 10,343); Togo, March 14, 1914, E. H. Wilson (No. 6278); prov. Hizen, Nagasaki, 1862, R. Oldham (Nos. 505, 507, 508, 607, Herb. Kew, Herb. Gray); same locality, A. C. Maingay (Nos. 872, 882, Herb. Kew); prov. Buzen, Kokura, April 20, 1903, U. Faurie. Shikoku, prov. Tosa, Osaki, June 5, 1889, K. Watanabe (Herb. Gray). Hondo, prov. Yamaguchi, near Onoda, June 17, 1918, E. H. Wilson (Nos. 10,379, 10,380, 10,381); between Habu and Asa, May 7, 1918, E. H. Wilson (No. 10,368A); round Ajisa, May 17, 1917, E. H. Wil- son (No. 8424); prov. Yamato, Yoshino, April 22, 1914, E. H. Wilson (No. 6575); prov. Shinano, above Narai, alt. 1100 m., September 3, 1905, /. G. Jack; without precise locality, 1864, Tschonoski (Herb. Kew); Togakushi-yama, July 9, 1884 (Herb. Kew); prov. Idzu, Oshima, March 30, 1917, July, 1918, E. H. Wilson (No. 8200); prov. Sagami, Hakone, May 12, 1918, E. H. Wilson (No. 10,365); Miya- noshita, August 24, 1892, C. S. Sargent; Kamakura, May 2, 1914, E. H. Wilson (No. 6610); prov. Musashi, Yokosuka, May, 1875, H. N. Moseley (Herb. Kew); Yokohama, 1862, C. Maximowcz (Herb. Kew); Mt. Takao, H. Takeda (No. 89, Herb. Kew); Mt. Mitake, May 19, 1907, K. Sakurai; prov. Shimotsuke, between Nikko and Lake Chuzenji, September 3, 1892, C. S. Sargent; in a garden at Umagaeshi, near Nikko, alt. 1000 m., June 4, 1914, E. H. Wilson (No. 6807); shores of Lake Chuzenji, November 3, 1892, C. S. Sar- gent; Nikko region, alt. 100-1500 m., May 16, 1914, E. H. Wilson (No. 6679); prov. Ugo, Akita, May 17, 1888, U. Faurie (No. 2128 Herb. Kew); Lake Towada, June 25, 1894, U. Faurie (No. 13,272); prov. Mutsu, Shichi-nobe, June 15, 1886, U. Faurie (No. 661); prov. unknown, Mt. Nagi, Mimasaka, May 8, 1903, S. Arimoto (Herb. Gray); without locality, 1866-74, L. Savatier (Herb. Kew, Herb. Bur. Sci. Manila). Hokkaido, prov. Oshima, Hakodate, C. Wright (Herb. Kew, Herb. Gray); same locality, July, 1859, C. Wilford (No. 1017, Herb. Kew); same locality, 1861, M. Albrecht (Herb. Kew, Herb. Gray); same locality, July 29, 1888, T. Tokubuchi; prov. Iburi, Nobori-betsu, June 15, 1893, U. Faurie (No. 10,020, Herb. Kew); Chitose, June 12, 1889, T. Tokubuchi; prov. Hidaka, T. Tokubuchi; Samani-sando, June 19, 1884, K. Miyabe (Herb. Gray); prov. Ishi- kari, Sapporo, July 29, 1888, May 8, 1895, T. Tokubuchi; same locality, October, 1891, K. Miyabe; same locality, May 24, 1903, S. Arimoto (Herb. Gray) ; ex Agricultural College, Sapporo, June 7, 1885. Cultivated: Arnold Arboretum (Nos. 3690, 3690-1, 3690-3, 3690-5, 3697). 40 THE AZALEAS OF THE OLD WORLD This is the Mountain or Hill Azalea (Yama-tsutsuji) of Japan, where it is abund- ant from Yaku-shima in the extreme south to central Hokkaido in the north. Its altitudinal range is from sea-level to about 1600 m., though north of the Nikko region it is much less. In May and June by the wayside and on the hills and mountain-slopes it is most conspicuous with its wealth of brilliant red flowers. It grows among grasses and in thickets, also on the edge of forests and among the undergrowth in thin woods, but it is a sun-loving plant and is seen to best advantage in open thickets on mountain-slopes. The height is from 1 to 3 m., but averages from 1 to 1.5 m., and according to situation it may be densely or laxly branched and form either a broad or a narrow, twiggy bush. In the warmer parts of Japan all the summer leaves and often some of the large spring leaves are persistent, but in the colder parts they are both deciduous. The spring leaves vary in shape from lanceolate or lanceolate-ovate to elliptic, and in size from 1.5 to 6 cm. in length and from 0.8 to 3 cm. in width, and they may be either acute or obtuse; the summer leaves are elliptic to obovate in shape, as is usual in the group to which this plant belongs. The pubescence on the shoot, leaf, calyx, ovary and fruit varies in color from pale gray to rufous. Typically the color of the flowers is red, usually bright red, but it varies considerably and, excluding the forms to be mentioned later, may be any shade from flesh-color to bright red. In size the flowers vary some- what but average from 2.5 to 4 cm. across. The corolla-lobes are rounded or pointed. The calyx is normally well-developed and the lobes are green, ciliate and usually rounded, and from suborbicular to elliptic or ovate in shape, and from 2 to 8 mm. long; usually they are from 2 to 4 mm. long and occasionally they are scarcely noticeable. The stamens may be included or exserted and this considerably affects the appearance of the flowers; the anthers vary in color from pale yellow-brown to blackish purple. The fruit is woody, erect, ovoid and from 0.6 to 1.5 cm. high, and when open the valves are recurved for about one-third their length. In its typical form this variety, with its relatively large clustered flowers each with five stamens, cannot be confused with anything else, but there are interme- diate forms which inextricably link it with R. obtusum. The extremes look widely different, yet on analysis size alone remains, and in this there is every gradation both among the leaves and flowers. The degree of persistence of the leaves, which is dependent upon winter temperature and exposure, strongly affects the appearance of the plants. The leaves, flowers, genital organs and fruit vary in size, and the flowers, anthers and pubescence vary in color. Round the base of Nishi-Kirishima the typical red-flowered var. Kaempferi is plentiful; on the middle and upper slopes of the mountain the typical f . japonicum abounds. In my ascent of this moun- tain and after exhausting every means I could think of I failed completely to dis- cover where one variety ended and the other began. The transition was both gradual and complete. At my request my colleague, Alfred Rehder, has critically examined the mass of material in our possession and is in complete agreement as to the specific identity here set forth. I confess that I should much prefer to keep Kaempfer's Azalea as a species distinct from R. obtusum, but the facts as I inter- pret them will not permit of this. Whether var. Kaempferi or the f . japonicum is best entitled to be considered the phylogenetic type of the species is difficult to determine, but the first named is much more widely distributed, and north of Hamamatsu, in prov. Totomi, central Hondo, is the only representative of the species. Kaempfer's Azalea was first introduced to the Occident by Professor Sargent, who sent seeds of it from the Nikko region and elsewhere in Japan to the Arnold Arboretum in the autumn of 1892, where plants flowered in May, 1897. It has proved both hardy and amenable to cultivation here, and in late May and early June of each year flowers abundantly, producing most vivid and spectacular floral ENUMERATION OF THE SPECIES 41 displays. The flowers last longer and are seen to best advantage when growing in the partial shade of conifers and other evergreen plants. In full sun the flowers bleach and fade more quickly. In the Arnold Arboretum the leaves are quite deciduous. This Azalea is the only kind with red flowers hardy in this Arboretum and is one of the most valuable exotic flowering plants ever brought into cultivation in eastern North America. A form with white flowers is : Rhododendron obtusum var. Eaempf eri f . albiflorum Wilson, n. name. Rhododendron Kaempferi f. album Nakai in Tokyo Bot. Mag. XXIX. [261] (1915). Unknown to me and described in Japanese only. Presumably normal except for its white flowers but possibly the same as the f . monstrosum Wilson. A form with vari-colored flowers is : Rhododendron obtusum var. Kaempferi f. multicolor Wilson, n. f. Japan : Hondo, prov. Yamaguchi, round Ajisu, May 17, 1917, E. H. Wilson (No. 8420 type, 8421, 8422, 8423, 8419, 8418, 9417, 8416); same locality, September 28, 1918, E. H. Wilson (No. 10,959 seeds only); between Habu and Asa, May 7, 1918, E. H. Wilson (No. 10,368); prov. Mikawa, round Futagawa, May 9, 1918, E. H. Wilson (No. 10,347). This form differs from Kaempfer's Azalea in its vari-colored flowers which range from pale pink, salmon or mauve to purple and magenta. Some of the shades are exquisite and scarcely two companion bushes have flowers of identical color. In size the flowers vary slightly, and those of No. 10,368 are from 1.5 to 2.5 cm. across and no larger than those of some plants of the f . japonicum. In fact this new form is the link connecting the var. Kaempferi with the wild f . japonicum. In certain localities this multi-colored form is abundant, as in those cited, and also round Tajimi, near the city of Nagoya, in province Owari. Curiously, where this form is plentiful the typical red-flowered Kaempferi is entirely absent or nearly so. The soil around Ajisu is rotten granite, round Tajimi and Futagawa gravel and brick clay; on the same formation and in the same general neighborhood Kaempfer's Azalea grows in great plenty. On Nishi-Kirishima, where the many- colored f . japonicum and the typical var. Kaempferi both grow, the soil is vol- canic ashes and detritus. It is plain, therefore, that soil has nothing to do with the variation in the color of the flowers, neither has altitude nor exposure. The cause would appear to be innate and not due to external conditions. I collected much seed of this new form from the neighborhood of Ajisu, and this has been distributed in England and in this country by the Arnold Arboretum, and many plants have been raised. If it should prove hardy this form may prove the forerunner of a new race of welcome colors among hardy Azaleas. A form with double flowers is: Rhododendron obtusum var. Kaempferi f . plenum Wilson, n. comb. Rhododendron Kaempferi var. plenum Nakai in Tokyo Bot. Mag. XXIX. [261] (1915). 42 THE AZALEAS OF THE OLD WORLD Japan: Kyushu, prov. Chikugo, Koradai, near Kurume, April, 1918, K. Akashi; without locality, May, 1918, B. Miyazawa. In this plant the calyx is normal but the stamens and pistil are petaloid. Aka- shi' s specimen has rather small flowers salmon-red in color and is from a sponta- neous plant. Miyazawa's is in size and color typical Kaempferi with double flowers. He told me this was also from a wild plant. A form with petaloid calyx forming hose-in-hose flowers and which also has been found wild is: Rhododendron obtusum var. Kaempferi f. Komatsui Wilson, n. comb. Rhododendron Kaempferi var. Komatsui Nakai in Tokyo Bot. Mag. XXIX. [261] (1915). A form with monstrous flowers may be designated: Rhododendron obtusum var. Kaempferi f. monstrosum Wilson, n. f. Korea: prov. Keiki, cultivated, said to have come from Japan, Agricultural Experimental Station, Suigen, May 24, 1917, E. H. Wilson (No. 8482). In this curious plant the corolla is white and has a petaloid calyx, the pistil is abnormal and in most flowers is changed to a staminode, the stigma being changed to a swollen, abortive anther which is often conspicuously aristate. The flowers though hose-in-hose are as large as those of ordinary Kaempferi. Possibly this is the same as Nakai's f . album. A form with abortive or nearly abortive corolla is: Rhododendron obtusum var. Kaempferi f. cryptopetalum Wilson, n. comb. Rhododendron ledifolium S. cryptopetalum Maximowicz in Mem. Acad. Sci. St. Petersburg, ser. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870). Matsu- mura, Ind. PL Jap. II. pt. 2, 462 (1912). Rhododendron Kaempferi f . f . Kinshibe Komatsu in Tokyo Bot. Mag. XXXII. [43] (1918). Japan: Hondo, prov. Kawachi, Yamomoto, cultivated, May 8, 1918, E. H. Wilson; prov. Musashi, Hatagaya, cultivated, April 29, 1914, E. H. Wilson (No. 6593). This is a monstrous form long cultivated in Japanese gardens under the name of " Kinshibe- tsutsuji." The corolla may be entirely suppressed or represented by from 1 to 5 deeply laciniate, strap-shaped lobes. The stamens vary from 5 to 10 (usually 7 to 10) ; the filaments are variable in length, often curled, bright crimson in color, and may be either normal in shape or slightly flattened and petaloid; the anthers are yellow. It has no garden value, being simply a curiosity. As to its specific identity there can be no question, and I am at a loss to understand why Maximowicz referred it to R. ledifolium (R. mucronatum G. Don). ENUMERATION OF THE SPECIES 43 A form with purple flowers and from 5 to 10 stamens is: Rhododendron obtusum var. Kaempferi f. mikawanum Wilson, n. comb. Rhododendron indicum var. mikawanum Makino in Tokyo Bot. Mag. XXIII. 251 (1909). Rhododendron Kaempferi var. mikawanum Makino in Jour. Jap. Bot. I. 18 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [371 (1918). Rhododendron purpureum Komatsu in Tokyo Bot. Mag. XXXII. [161 (1918). Rhododendron poukhanense Komatsu I. c. [12] (1918), as to the Japanese plant, not LeVeille". Rhododendron poukhanense f. obtusifolium Komatsu I. c. [37] (1918). Rhododendron poukhanense f. acutifolium Komatsu I. c. [38] (1918). Rhododendron scabrum Nakai in Tokyo Bot. Mag. XXXIII. 207 (1919), not G. Don. Rhododendron scabrum var. Kaempferi, f. 1. purpureum Nakai I. c. 208. Japan : Hondo, prov. Suruga, near Ashitake-yama, lower slopes of Mt. Fuji, June 20, 1918, H . Suzuki; prov. Musashi, Ome, near Tokyo, May 31, 1918, E. H. Wilson. This is a form of the common Mountain Azalea of Japan having purple flowers and from 5 to 10 stamens. It is found here and there mixed with the type and is occasionally cultivated, being known as the " Murasaki Yama-tsutsuji " (Purple Hill Azalea). It is in cultivation in the Arnold Arboretum from plants I brought from Japan in the spring of 1919. The increase in the number of stamens is in- teresting, and is another instance of the extraordinary range of variation which obtains in R. obtusum. In Tokyo I saw specimens from Hangno which Komatsu refers to R. poukhanense, but it is not that species and I could find no character by which to separate them from Makino' s variety. The material was collected on the Chichibu Mountains in the western part of Musashi province. It has flowers in pairs, red-purple in color, with dark purple anthers; the leaves are rather long, lanceolate and glabrescent, and the pedicels and calyx are clothed with gray pubes- cence. Komatsu's R. purpureum is based on a specimen collected in Yumato prov- ince. It has the foliage of typical var. Kaempferi, pale rose-purple colored flowers each with from 7 to 10 stamens; the calyx-lobes are lanceolate, acuminate and erose. Nakai in the Tokyo Bot. Mag. (XXXIII. 207 [1919]) identified this plant with R. scabrum G. Don. I do not understand how he arrived at this conclusion unless he was guided solely by the color of the flowers. This f. mikawanum never has flowers 3 inches in diameter, which is the size given by G. Don, neither are the leaves coriaceous. It seems to me that the plant described by G. Don as R. scabrum is that which Miquel later described as R. sublanceolatum; from the description I do not see how it can have anything to do with R. obtusum var. Kaempferi or any of its forms. A hybrid race with richly colored flowers of small to medium size is : X Rhododendron Sanderi Wilson, n. hyb. Rhododendron Simsii " Garnet " x R. obtusum Planchon. This is an interesting and valuable race that has been raised and developed at Holm Lea, Brookline, the home of Professor C. S. Sargent. About thirty-five years 44 THE AZALEAS OF THE OLD WORLD ago the gardener, Charles Sander, collected seeds from the Holm Lea collection of "Indian Azaleas' ' and from them raised many plants. Among these when they flowered were two of dwarf, compact habit which produced relatively small flowers (3 cm. diam.) carmine-red or crimson-lake in color. In the Holm Lea collection was the well-known "Indian Azalea Decora" of European origin and introduced into the garden of Marshall P. Wilder, Dorchester, Mass., about 1847. It is highly probable that this was the parent of the two Azaleas above mentioned. The best of these two Sander named " Garnet." This he crossed with an unknown Azalea having small red flowers obtained from M. Atkinson, gardener for Mr. John L. Gard- ner, Brookline, Mass., which to-day Sander thinks was R. dbtusum Planch. After a close study of the race I think this conclusion is correct. The best of the hybrids from the above parents were crossed with R. dbtusum f . album Schneid. and gave rise to forms with rose-pink flowers. In more recent years the well-known " Hino- degiri " has been used with excellent results. The outcome of some three decades of hybridising, selecting and recrossing is a race of beautiful Azaleas of compact twiggy habit with flowers from 2.5 to 4.5 cm. in diameter in varying shades of rose-pink, salmon, red, scarlet and fiery crimson. Among the scarlet and crimsons are deeper and richer shades than I have seen among any of the Azaleas of China and Japan. The influence of R. Simsii forms is seen in the number of stamens, which varies from 5 to 10 and is inconstant on the same individual; also in the size of the flowers and in the ease with which the plants may be forced into blossom during the winter and early spring months. Some of the forms have flowers no larger than those of R. obtusum, though in the majority the flowers exceed in size those of that plant. The race is singularly attractive and highly ornamental, and it needs no expert knowledge to predict for X R. Sanderi a great future. Among the best forms are " Suzuki," scarlet maroon, " Hebe," white with rose-red stripe, " Hermione," deep salmon-red, " Ruby " crimson maroon, " Uncas," dark scarlet, "Helena," rose-pink, "Hilda Hedlund," deep pink, "Holm Lea," crimson- maroon, " Vulcan," salmon-red, " Natalie," salmon, " Muriel," rose-red, " Alice Sargent," bright salmon-red, " Rose Queen," deep rose, " Havemeyer," rose-red, " Brookline," carmine, " Venus," cerise, " Jupiter," fiery-red, " Mars," intense scarlet, "Vivid," scarlet. To this group must be referred the highly appre- ciated and well-known Azalea "Hexe" 1 or "Firefly" of American gardens, which is a hybrid between R. obtusum f. amoenum and R. Simsii "Due de Nassau." This favorite Azalea with rich red, hose-in-hose flowers was raised by Otto Forster, Lebenhof, near Scheibbs, Lower Austria, about 1885, and was first exten- sively cultivated round Dresden. At first it was not particularly esteemed and was used as a stock on which to graft the ordinary "Indian Azaleas" of commerce; later it became one of the most popular varieties, and before the Great War was imported in hundreds of thousands into this country for decorative purposes. It roots freely from cuttings and flowers profusely when quite small. One of its parents, Azalea " Due de Nassau," is an old variety, having been exhibited in London in June, 1862, by Charles Turner of Slough. There are several other hose-in-hose Azaleas of much the same parentage as " Hexe." One of these, " Vuylestekeana," has been crossed at Holm Lea with " Flambeau," a French vari- ety, with crimson-maroon flowers of medium size. The result is a race with relati- vely large (4 to 6 cm. diam.), hose-in-hose flowers varying in color from intense scarlet to crimson-maroon, and perhaps the most highly colored of all Azaleas. 1 Azalea indica (Hexe) Ad. Vanden Heede in Rev. Hort. Belg. XXXI. 49, t. (1905). Azalea Hexe Hort. ex W. Watson, Rhodod. and Azaleas, 67 (1911). Rhododendron indicum Hexe Millais, Rhodod. 193 (1917). ENUMERATION OF THE SPECIES 45 Rhododendron Simsii Planchon in FL des Serr. IX. 78 (1854) ; in Rev. Hort. 1854, 47. Schneider, III. Handb. Laubholzk. II. 506, figs. 331 1-m, 332 a-d (1911). Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Azalea indica Alton, Hort. Kew, ed. 2, 318 (1810), as to the plant, not Linnaeus. Sims in Bot. Mag. XXXVI. t. 1480 (1812). Bentham, Fl. Hongk. 201 (1861), as to the Hongkong plant. Azalea indica a. punicea Sweet, Hort. Brit. 264 (1827). Rhododendron indicum var. ignescens Sweet, Brit. Flow. Gard. ser. 2, II. t. 128 (1832). Render & Wilson in Sargent, PL Wilson. II. 547 (1913). Rhododendron indicum a. puniceum Sweet in Brit. Flow. Gard. ser. 2, II. sub. t. 128 (1832). Azalea indica ignescens Hovey, Mag. Hort. III. 72 (1837). Rhododendron Calleryi Planchon in FL des Serr. IX. 81 (1854); in Rev. Hort. 1854, 66. Rhododendron indicum /3. Simsii Maxim owicz in Mem. Acad. Sci. St. Peters- bourg, ser. 7, XVI. No. 9, 38 (Rhodod. As. Or.) (1870). Franchet in Nouv. Arch. Mus. Paris, ser. 2, VI. 77 (1883); PL David I. 97 (1884); in Bull. Soc. Bot. France, XXXIII. 235 (1886). Rhododendron indicum Hemsley in Jour. Linn. Soc. XXVI. 25 (1889), as to the Chinese plant, not Sweet. Diels in Bot. Jahrb. XXIX. 514 (1900) ; in Notes Bot. Gard. Edinburgh, VII. 392 (PL Chin. Forrest.) (1913). Pam- panini in Nuov. Giorn. Bot. Ital. n. ser. XVIII. 131 (1911). W. Watson, Rhodod. & Azaleas, 79 (1911), in part. Dunn & Tutcher in Kew Bull. Misc. Inform, add. ser. X. 155 (FL Kwangtung and Hongk.) (1914). Millais, Rhodod. 189 (1917), in part. Azales indica var. Simsii Rehder in Bailey, Cycl. Am. Hort. I. 122 (1900). Rhododendron indicum var. Matsumura in Tokyo Bot. Mag. XIV. 69 (1900). Rhododendron indicum var. Tamurai Makino in Tokyo Bot. Mag. XVIII. 102, fig. (1904). Rhododendron indicum 7. macranthum subvar. genuinum forma Tamurai Ma- kino in Tokyo Bot. Mag. XXIV. 77 (1910). Rhododendron indicum var. formosanum Hayata, Icon. PL Formos. III. 134 (1913). Kanehira, Formos. Trees, 321, fig. 9 (1917). China: Hongkong, Happy Valley, November 5, 1903, C. S. Sar- gent; without locality, /. G. Champion (Nos. 122, 123, Herb. Kew, Herb. Gray) ; without locality, C. Wright (No. 195, Herb. Gray, Herb. Kew); April 28, 1914, Mary Strong Clemens (No. 4261, Herb. Bur. Sci. Manila). Kwantung, Canton, March 10, 1835, S. W. Williams (Herb. Gray); same locality, October 27-30, 1916, March 20, 1917, C. 0. Levine (No. 438, 599, Herb. Bur. Sci. Manila); Shiuchow, April, 1919, To Rang P'eng (No. 2761, Herb. Bur. Sci. Manila). Chekiang, Pootoo Island, E. Faber; Ningpo, Ning-kan-jou, April 15, 1877, W. Hancock (No. 32, Herb. Kew); Mokan-shan, alt. 500 m. August 3, 1915, F. N. Meyer (No. 1607); without locality, 1845, R. Fortune (Nos. 72, 152, 153, Herb. Kew, Herb. Gray); hills near Taihu Lake, April, 1881, W. R. Carles. Kiangsu, near Shanghai, 46 THE AZALEAS OF THE OLD WORLD 1915, D. Macgregor. Kiangsi, Killing, alt. 1600 m., July 28, 1907, E. H. Wilson (No. 1682); same locality, 1873, G. Shearer (Herb. Kew). Hunan, Yoku-shan, alt. 70-300 m., February 21, 1918, Dr. Handel-Mazzetti (No. 480). Hupeh, north and south of Ichang, alt. 30-1300 m., May and November, 1907, E. H. Wilson (No. 3474}; same locality, A. Henry (Nos. 782, 1160, Herb. Kew, Herb. Gray); Changyang Hsien, thickets, alt. 1300 m., May and November, 1907, E. H. Wilson (No. 3472) : Changlo Hsien, cliffs, alt. 600-1300 m., May, 1907, E. H. Wilson (No. 3472 a ); Patung Hsien, A. Henry (No. 1416, Herb. Kew); Nanto and mountains to northward, A. Henry (No. 3194, Herb. Kew); Hsing-shan Hsien, alt. 1000-1800 m., May 14, December, 1907, E.JI. Wilson (No. 56JT): " Kao-hien-scian," alt. 800 m., May, 1907, C. Silvestri. Szech'uan, Chungking, 1885, F. S. A. Bourne (Herb. Kew); Kiating Fu, alt. 300-800 m., May, 1908, E. H. Wilson (No. 3475); Mt. Omei, May, 1904, E. H. Wilson (Veitch Exped. No. 5143); without locality, A. von Rosthorn (No. 2148); Yunnan or Kueichou, 1911, A. Hosie (Herb. Kew). Yunnan, Menztsze, alt. 1500-2000 m., A. Henry (Nos. 99QQ b . 99QJT, 9900 e , Herb. Kew, Herb. Bur. Sci. Manila); same locality, alt. 1600-2300 m., April, 1893, W. Hancock (No. 155, Herb. Kew); Mile, A. Henry (No. 9900 d ); " Momyen," June 3, 1868, D. J. Anderson (Herb. Kew); near Tali Fu, Tsang-chan, alt. 2500 m., J. M. Delavay (No. 1122, Herb. Kew); Tali Fu, alt. 2200-2600 m., cultivated, G. Forrest (No. 4173); " Ho-yeh-shui," north end of Lichiang Valley, alt. 2900 m., cultivated by Mossoo natives, May, 1906, G. Forrest (No. _228fc Herb. Kew); without locality, G. Forrest (Nos. T832. 11.824) : i91(T R. P. Maire (Nos. 3713, 3775, Herb. Bur. Sci. Manila).*" ~U, Formosa: prov. Koshun, South Cape, Schmuser ex A. Henry (No. 588 in part, Herb. Kew). Cultivated: " Jardin du Luxembourg, Orangerie, 14 avril, 1822, 27 avril, 1823 " ex Herb. J. Gay (Herb. Kew). This species grows in all the temperate parts of China and in south Formosa, and a variety (eriocarpum Wils.) on the Kawanabe Islands off Liukiu Oshima, but it has not been found in Japan or Korea. It is particularly abundant in the area of the Yangtsze Valley from near Ningpo to Mt. Omei in the far west. Hancock, on a specimen, states that " the Ningpo hills are absolutely crimson in places with this shrub "; the same is true of the Lushan hills round Kiukiang and of the region north and south of Ichang. It delights in rocky places, preferably cliffs, thin dry woods and thickets, blossoms profusely, and its wealth of red flowers makes it one of the most conspicuous of all shrubs. The habit is twiggy and much-branched with a maximum height of three metres, but averaging about half this height. The leaves are variable in size and with the shoots are clothed with appressed, ENUMERATION OF THE SPECIES 47 strigose, gray to shining brown hairs. At low altitudes the leaves are compara- tively large, from lanceolate to ovate-elliptic, and are all persistent; at its upper altitudinal limit the leaves are much reduced in size, and nearly all fall off in the late autumn. The leaves immediately below the winterbuds are small, relatively thick, obovate to oblanceolate and more persistent than the other form of leaves. The flowers are clustered, from two to six together, at the end of the shoots and all I have seen are fairly uniform in color. The corolla is broad-funnel shape, varying in color from rose-red through bright to dark red. The calyx varies from an inconspicuous rim to a length of 5 mm.; the stamens are normally ten, rarely eight, never fewer, as long or nearly as long as the corolla but shorter than the style. In central China it is generally known as the " Yin-shan-hung." Millais says that Forrest states under Nos. 4173, 4173 a , from plants cultivated by Chinese in the Tali Valley, that the color of the flowers is " rose to white." In China I have never seen wild an albino of this species; Forrest's No. 4173 in Herb. Kew is described as having rose-colored flowers. The Formosan specimen cited from A. Henry in Herb. Kew represents Hayata's var. formosanum, the type of which I have seen in Tokyo. It is indistinguishable from typical R. Simsii so abundant in China, but in Formosa known only from the extreme southern end of the island. For nearly a century and a quarter this plant has usurped the name indicum, and in books is almost hopelessly confused with the true R. indicum Sweet, with R. obtusum Planchon and its varieties, and with R. scabrum G. Don (R.sublanceolatum Miquel). The first mention of the Chinese plant I can find is in Macartney's Embassy to China, II. 524 (1797), where Azalea indica is recorded as collected in the provinces of Kiangsi and Canton (Kwangtung) in 1793. In all probability it is the Chinese species that was introduced into Kew " by the Hon. Court of Directors of the East India Company " in the ship " Cuffnels," Captain Wellbank (Aiton, HorL Kew, ed. 2. 318 [1810]), though Aiton's description is of the true R. indicum Sweet. Sims was the first to figure the Chinese species (Bot. Mag, t. 1480 [1812]) under the erroneous name of Azalea indica, and states that it was a rare plant in England and that the only one that had flowered was in the collection of James Vere, Esq. In the Kew Herbarium there are two specimens from the Jardin du Luxembourg collected in 1822 and 1823. These are the oldest cultivated specimens I have seen. Sweet in 1832 gave it a varietal name, and says that the plant was imported in the East India Company's ship " Orwell " from China by Mr. Tate. This Mr. Tate was a nurseryman in Sloane Square, London, and the master of the ship " Orwell " was Captain Farrer, who brought a number of Chinese plants including other Azaleas to England and handed them over to Tate. In Braam's Icones plantarum, issued in 1818, there is a figure (t. 27) of a red-flowered Azalea. The drawing is very crude and, while it probably is meant for R. Simsii, it might represent any Azalea with red flowers. The Chinese Azalea appears to have remained rare in the Occident for several decades. In 1837, under the name of Azalea indica ignescens, it was cultivated in America by Marshall P. Wilder, Dorchester, Mass. Fortune, from 1843 to 1856, introduced into England from Chinese gardens many Azaleas of different kinds, and these gave great impetus to the cultivation of these ornamental plants. The so-called " Indian Azaleas " of western gardens have been originated almost en- tirely from R. Simsii since > about 1850, chiefly in Belgium, but some in France and others in Germany. There is a very considerable literature on "Indian Azaleas" and at least one illustrated book (Iconographie Azalees de L'Inde by Auguste van Geert, published in 1882). Planchon (in Fl. des Serr. IX. 77 [1854]) gives a brief history of them, and enumerates varieties in the collections of M. Margottin, M. Delessert, and M. Michel. Some of them were forms of the true R. indicum, others of R. phoe- 48 THE AZALEAS OF THE OLD WORLD niceum, and a few of R. Simsii; several are recorded as presumed hybrids between R. phoeniceum and R. ledifolium (R. mucronatum G. Don) and between R. phoeni- ceum and R. Danielsianum (R. indicum Sweet) but such hybrids are unknown to-day and probably never existed. In La Belgique Horticole for 1865 (XV. pp. 174-192), there is an excellent article by M. Edouard Andre with a list of vari- eties and the names of the originators. In the various volumes of Flore des Serres are many colored figures and notes chiefly by Louis van Houtte. In fact in all garden periodicals from about 1850 on, there are many references and figures of these plants. However, scientific exactness is wanting and much of their early beginnings is obscure. As far as I can discover the introduction into England of the striped-flowered Azalea vittata by Fortune in 1850 or 1851 led to the raising of seedlings. Further impetus was given by the advent the same years of A. vittata Bealii with red and white striped flowers. Many of the best varieties are branch sports perpetuated by grafting. From the crossing of red and white flowered varie- ties many variegated forms have originated. One of the most successful breeders of " Indian Azaleas " of commerce was Mr. Joseph Vervaene of Ghent, whose work is fittingly commemorated by the lovely Azaleas " Vervaeniana," and " Vervaeniana alba." Other names immutably connected with the raising and perfecting of the "Indian Azaleas" of commerce are Messers Knight & Perry, Ivery and Rollisson in England; L. Eeckhaute, Haerens, Van Houtte, Van Geert, A. Verschaffelt, J. de Kneep and Van der Cruyssen in Belgium; Lesebe, Truffaut, H. de May and Mabire in France; Messers Schulz, E. Liebig, Seidel and Rose in Germany. The varieties are legion, and each year adds materially to the number. All the modern ones have been raised in Belgium or Germany, where they are cultivated by the millions and exported to almost all parts of the world. Prior to 1914 more than two and a half million plants were annually exported from Belgium alone. In Germany a Eurhododendron (Cunningham's White) is much used as a stock for these " Indian Azaleas," but not in Belgium, where R. phoeniceum and its f . con- cinnum are chiefly so employed. According to J. Breck (hi The Horticulturist, I. 512 [1847]), Marshall P. Wilder imported from Germany by way of England and cultivated in his garden in Dorchester, Mass., in 1847, Azalea "Optima" with dark scarlet flowers, Azalea "Prince Albert" with scarlet flowers, and Azalea "Alba Insignis" with large white flowers. The first two were undoubtedly derivatives of R. Simsii and representatives of the " Indian Azaleas " of to-day, and the first intro- duced into America so far as I can discover. The third was probably a form of R. mucronatum. Hovey (Mag. Hart. XIV. 284 [1848]) records that on May 13, 1848, Marshall P. Wilder exhibited before the Massachusetts Horticultural Society Azalea " Decora," another Indian Azalea. A plant of this was acquired by Ignatius Sargent, and its descendants are still in the Holm Lea collection. Among the typical large-flowered forms of " Indian Azaleas " of to-day I can de- tect no influence of any species other than R. Simsii, though without question R. phoeniceum and its forms have been used. Those with smaller flowers, especially the hose-in-hose kinds like " Hexe " and " Vuylstekeana," are without doubt hybrids between forms of R. Simsii and R. obtusum f . amoenum. It is probable also that typical R. obtusum has been employed. R. mucronatum and numerous named forms of it are loosely designated Azalea indica, but in the evolution of the "Indian Azalea" of present-day gardens neither this species nor the true A. indica of Linnaeus have had any part. Rhododendron Simsii var. vittatum Wilson, n. comb. Azalea vittata Fortune, Tea Districts of China, 330 (1852). Azalea indica vittata Van Houtte in Fl. des Serr. IX. t. 886 (1854). Millais, Rhodod. 259 (1917). ENUMERATION OF THE SPECIES 49 Azalea indica vittata punctata Van Houtte in Fl. des Serr. IX. t. 888 (1854). Rhododendron vittatum Planchon in FL des Serr. IX. 82 (1854); in Rev. Hort. 1854, 66. Maximowicz in Mem. Acad. Sci. St. Petersbourg, ser. 7, XVI. No. 9, 49 (Rhodod. As. Or.) (1870). Millais, Rhodod. 259 (1917). Rhododendron vittatum var. b. punctata Planchon, in Fl. des Serr. IX. 82 (1854) ; in Rev. Hort. 1854, 66. Azalea vittato-punctata Lemaire in III. Hort. I. t. 20 (1854). Cultivated: Hort. Holm Lea, May 20, 1918, A. Rehder; Hort. Berckmans, Augusta, Georgia, March 30, 1908, C. S. Sargent. This variety has larger calyx-lobes than is usual in the species, but I have wild specimens of the type with an equally large calyx. The most prominent distin- guishing feature is the color of the flowers. This is white, striped with lilac-purple, often irregularly so; sometimes the flowers are pure white or pure lilac-purple on the same branch, occasionally they are dotted or blotched with color. Planchon says it has been in possession of M. Paillet since 1844, and then adds that it was one of Fortune's introductions to England. Now Fortune did not land in China until July, 1843, so it would seem that there was something wrong with Planchon's date or that it was introduced before Fortune's first visit. In all probability the date is wrong. According to Fortune's account in his Tea Districts of China it was in April, 1850, that he visited the Pou-shan gardens near Shanghai and saw this Azalea in flower. However, he had been to these gardens on his first visit to China (1843-45), but if he sent this variety of Azalea to England on that occasion there is no available record of the fact. In all probability M. Paillet's plant was R. indicum var. variegatum. Fortune's plant did not reach England before 1850 or 1851 so far as I can discover, and its acquisition according to the books led to the raising of many seedlings. According to Hovey (Mag. Hort. XXI. 301 [1855]) an Azalea vittata rosea was exhibited for the first time in Philadelphia on April 17, 1855, by Mr. Buist. It has long been grown in the Berckmans Nurseries, Augusta, Georgia, and in the collection at Holm Lea. The type appears to have been lost to English gardens until recently, when Professor Sargent sent it to Kew and elsewhere. A form in which the flowers are striped with red is : Rhododendron Simsii var. vittatum f . Bealii Wilson, n. comb. Azalea Bealii Fortune, Tea Districts of China, 330 (1852). Lemaire in III. Hort. I. t. 8 (1854). Rhododendron vittatum var. g. Bealii Hort. apud Planchon in Fl. des Serr. IX. 82 (1854); in Rev. Hort. 1854, 66. Azalea vittata Bealii Morren in Belg. Hort. XVI. 1, t. (1866). This is another of Fortune's introductions to Messers Standish & Noble of Sunningdale Nurseries from Pou-shan gardens, Shanghai, in 1850 or 1851, and is characterised by its white flowers striped with red. This plant also has been a useful parent in the evolution of the present-day race of "Indian Azaleas," as shown by the variety " Fiirstin Baro Trysky," for example. It was imported into America by C. M. Hovey, Boston, Mass., in 1855. A variety with white to rose-colored flowers and geographically widely removed from the type is : Rhododendron Simsii var. eriocarpum Wilson, n. comb. Rhododendron indicum var. eriocarpum Hayata, Icon. PL Formos. III. 134 (1913). 50 THE AZALEAS OF THE OLD WORLD Japan : Kawanabe Islands, Suwanose-shima, May, 1917, H. Ushiwo (No. l c ); Takara-shima, May 4, 1917, H. Ushiwo (No. l d , l e ); Taira- shima, May, 1917, H. Ushiwo (No. l f ). These specimens were collected for me by Forestry-officer Ushiwo by courtesy of Dr. Naito, chief of the Kagoshima forestry bureau, and are excellent, though unfortunately unaccompanied by field notes. The flowers of the Takara-shima specimens are white, those of that from Taira-shima pink or rose-colored. The color of the flowers and of the pubescence more readily distinguish this variety from the type than does the shape of the leaves pointed out by Hayata. I am sorry that I did not see this interesting plant in a wild state and am inclined to the view that ultimately it may prove to be a distinct species. The Kawanabe Islands are a number of small rocky eminences which jut out of the sea south of Yaku-shima toward the Liukiu Islands and have a flora like that of Liukiu (see Jour. Arnold Arb. I. 183-186 [1920]). Rhododendron atrovirens Franchet in Bull Soc. Bot. France, XXXIII. 235 (1886). Hemsley in Jour. Linn. Soc. XXVI. 19 (1889). Millais, Rhodod. 124 (1917). The species from western Yunnan is unknown to me, except from Franchet's description. It would appear to be intermediate between R. Simsii Planch, and R. microphyton Franch. It is said to be a large bush or small tree with lanceolate, long-acuminate leaves narrowed to the base and glaucescent on the underside. The flowers are described as red, clustered 3 or 4 together, and each 2.5 cm. or less in diameter. It was discovered by Pere Delavay at " Tchen-fang-ehan " near Ta- kouan in May, 1882, but apparently has not yet been introduced to cultivation. Rhododendron lasiostylum Hayata, Icon. PL Formos. III. 135 (1913). Kanehira, Formos. Trees, 321 (1917). Komatsu in Tokyo Bot. Mag. XXXII [14] (1918). Formosa: prov. Nanto, between Shushu and Horisha, March 12, 1918, E. H. Wilson (No. 10,020). This is a pretty pink-flowered Azalea of which I have seen only one plant. It was growing on top of a rock by the trolley line at about 400 m. altitude, and was about 1 m. high and less broad. The branches are slender and twiggy and when young clothed with the usual appressed, flattened chestnut-brown hairs. The young leaves, petioles, pedicels, calyx and ovary are all densely clothed with chest- nut-brown pubescence changing to gray. The leaves are dimorphic; the spring leaves are lanceolate to ovate-lanceolate and from 2 to 4.5 cm. long and 1 to 1.5 cm. wide, acute and deciduous; the summer leaves are oblong-obovate to oblanceo- late from 2 to 4.5. cm. long and 1 to 2 cm. wide, rounded and mucronate at apex and cuneate at base; both surfaces are clothed with appressed gray to rufous- brown hairs. The flowers are terminal, three or four together at the end of the branchlet. The corolla is funnel-shape, pink, about 2 cm. long with spreading lobes and about 2.5. cm. across, the tube being rather less than 1 cm. long. The calyx is variable in size being either nearly obsolete or with distinct, rounded lobes each a millimetre or more long. The stamens are of unequal length, the longest equalling the corolla-lobes and overtopped by the style. It is with considerable hesitation that I identify my specimens with this species. There are many dis- crepancies but of trivial character the more important being the difference in ENUMERATION OF THE SPECIES 51 size of the calyx-lobes and of the leaves. I have seen the type specimens in Tokyo but the material is so poor that I could arrive at no definite conclusion in the absence of my own material for comparison. Rhododendron subsessile Rendle in Jour. Bot. XXXIV. 357 (1896). Merrill in Govt. Lab. Publ. (Philipp.) XXIX. 40 (1905). Millais Rhodod. 249 (1917). Philippine Islands: Luzon, Prov. Benquet, Baguio, March 1907, 1913, A. D. E. Elmer (Nos. 8595, 14,298); same locality, May- June, 1916, J. K. Santos (No. 38); same province, Mt. St. Tomas, February 18, 1916, E. C. Leano (Forestry Bureau No. 25,125); same locality, March, 1904, A. D. E. Elmer (No. 5799); same locality, July 1, November 29, 1904 R. S. Williams (Nos. 1223, 2001); same locality, May 10, 1904, P. T. Barnes (No. 922); same locality, April, 1908, H. N. Whitford (Forestry Bureau No. 11,090); Mt. Tonglon alt. 2200 m., November 12, 1905, May, 1911, E. D. Merrill (Nos. 4815, 736); same place, August, 1906, H. M. Curran (No. 5032); same place, December, 1906, E. A. Mearns; same place, December, 1908, M. L. Merritt (No. 14,148); Bucao, alt. 2300 m. January 4, 1909, R. J. Alvarez (No. 18,362); Mt. Pulog, January 3 & 8, 1909, T. C. Zschokke and M. L. Merritt (Nos. 18,035, 18,172); Pauai, July, 1907, E. A. Mearns (No. 4275); same place, alt. 2100 m., June, 1909, R. C. McGregor (No. 8420); same place, April 17, 1918, J. K. Santos (No. 31,994); District of Lepanto, January 26, 1909, F. W. Darling (No. 16,573); same district, Mt. Data, November 4, 1905, E. D. Merrill (No. 4606); prov. Abra, February 8, 1909, M. -Ramos (Nos. 7249, 7105); subprov. Bontoc, alt. 1350 m. April 7, 1910, Pere M. Vanoverber (Nos. 351, 364) all these numbers in Herb. Bureau Science Manila. This Philippine species in its pubescence and foliage somewhat resembles, as Dr. Rendle points out, the Japanese R. mucronatum G. Don, but it is non-glan- dular, the flowers are not more than one-third the si v ze and have a minute calyx and a more tubular corolla. Its nearest relative is probably the Formosan R. ru- bropilosum Hayata. It is evidently a twiggy plant and the young branches are clothed with the usual flattened, appressed bristles. The oval to lanceolate per- sistent leaves are from 1.5 to 3.5 cm. long and from 0.7 to 1.2 cm. wide, acute and mucronate, with slightly recurved margins, the midrib impressed above, promi- nent below. On adult leaves the upper surface is clothed with appressed white hairs but when young both surfaces are densely covered with rufous-gray, silky hairs. The flowers are in terminal two to several flowered clusters. The corolla is campanulate-funnel-form, with the tube less than 1 cm. long and short spreading lobes, and is said to be from lilac to violet-purple in color; the stamens vary from 6 to 10, are slightly exserted and are overtopped by the graceful, curved style which is villose at the base. The pedicels, calyx and ovary are densely clad with ferrugineous villose and more or less flattened, straight hairs. The 52 THE AZALEAS OF THE OLD WOULD fruit is conic-ovoid and is densely clothed with rufous, appressed and villose pubescence. It is widely distributed and very common on the highlands of northern Luzon. It is evidently a very pretty species and ought to be introduced into gar- dens. The flowers vary considerably in color and No. 364 from Bontoc is said to have white flowers. No. 351, also from Bontoc, has very slender branches and small, elliptic, acute leaves and is very different in appearance from the other speci- mens. I can find no morphological differences, and it appears to be simply a con- dition; probably the plant from which the specimen was taken was growing in the crevice of a boulder or of a cliff. Rhododendron rubropilosum Hayata in Jour. Coll. Sci. Tokyo, XXX. art. 1, 173 (Mat. Fl Formos.) (1911). Kanehira, Formos. Trees, 326, fig. 11 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [16] (1918). Rhododendron caryophyllum Hayata, Icon. PL Formos. III. 130, t. 23 (1913). Kanehira, Formos. Trees, 318, fig. 17 (1917). Formosa: prov. Kagi, east of Ari-san, alt. 2500-3300 m. February 2 and 4, October 26, 1919, (Nos. 9728, 9779, 10,939); same locality, June, 1914, U. Faurie (Herb. Bur. Sci. Manila). On the grass-clad mountain-slopes east of Ari-san this species is common between altitudes of 2300 and 3300 m. but I did not see it elsewhere. It is a narrow shrub from 1 to 3 m. tall, with twiggy, rigid, ascending branches, which when young are densely covered with appressed, flattened, gray to red-brown hairs. The leaves are persistent, scattered on the free-growing shoots and crowded at the tip of the branchlets; in shape they vary from oblong-lanceolate to elliptic- lanceolate and are from 1 to 4 cm. long and from 0.5 to 1.7 cm. wide, acute, tipped with a glandular mucro and narrowed at base; the upper surface is dark green, with scattered, appressed, pale gray hairs and the underside is pale green and clothed with flattened, appressed, gray to brown strigose hairs, which are especially prominent on the midrib. The petiole is densely clad with appressed, flattened hairs similar to those on the branches. The bud-scales of the flowers are pubescent but not glandular. The flowers are of medium size, pink spotted with rose, short- stalked and borne several together at the end of the branchlet; the corolla is fun- nel-shape, with spreading lobes, and from 1.5 to 2.5 cm. across. Hayata gives the number of stamens as 10 but on my specimens I find 7 or 8 only; they are of unequal length but shorter than the corolla. The pistil overtops the stamens and has a relatively stout, slightly curved style, villose at the base, and the ovary is densely clothed with pale gray soft hairs. The fruit is ovoid-conic, from 0.5 to 0.8 cm. long and densely clothed with shaggy, gray-brown hairs. The type specimen of R. caryophyllum which I have seen, has smaller leaves and flowers than is usual in the species which is due to the fact that the specimens came from a wind-swept slope near the altitudinal limits of the species. My No. 9779 is simi- lar. In late October I found many plants in full bloom and apparently it is pre- cocious in its habit of flowering. I collected seeds, and plants of this species are now growing in the Arnold Arboretum and in several gardens in England. They will not be hardy except in favored localities like Cornwall and California. Rhododendron tosaense Makino in Tokyo Bot. Mag. VI. 53 (1892); in XVIII. 101 (1904). Schneider, III. Handb. Laubholzk. II, 506 ENUMERATION OF THE SPECIES 53 (1911). Matsumura, Ind. PL Jap. II. pt. 2, 465 (1912). Ko- matsu in Icon. PL Koisikav. II. 91, t. 130 (1915); in Tokyo Bot. Mag. XXXII. [17] (1918). Japan: Shikoku, prov. Tosa, round Kochi, sea-level to 1000 m. altitude, November 20, 1914, E. H. Wilson (No. 7801). A twiggy, eglandular shrub growing from 0.3 to 2.5 m. but usually from 1 to 1.5 m. tall, with slender branches clothed with appressed, straight gray to gray- brown hairs. The leaves are scattered on the shoots and free-growing branches and crowded at the end of the branchlets. They are more or less deciduous and change in the autumn to purplish crimson; those formed in the spring are oblance- olate or lanceolate from 1.2 to 3.5 cm. long and from 0.5 to 1 cm. broad and are acute and clothed with appressed, straight, gray hairs; the summer leaves are linear to oblanceolate from 0.5 to 1 cm. long and from 1 to 5 mm. wide. The flowers are produced in clusters of from 2 to 6, but occasionally are solitary. The corolla is lilac-purple, about 3 cm. across. The pedicels are short and the calyx is very small and clothed with long, straight white hairs. The stamens vary in number from 5 to 8, are of unequal length, included and overtopped by the style which is exserted. The fruit is ovoid, about 9 mm. long, clothed with appressed, straight hairs and subtended by the persistent calyx-lobes. The species is abundant on the lower mountain-slopes of Tosa province, Shi- koku, from sea-level to altitudes of 1000 m. On exposed slopes it is dwarf and dense in habit, but in thickets it is tall and loosely branched. The summer leaves are minute and with the slender twiggy branchlets easily distinguish this species from others. I have seen a few flowers, but the color is not attractive though doubtless in spring when covered with blossoms, the plant would have a charm of its own. I saw plenty of it in November, 1914, and collected seeds for the Arnold Arboretum. These were distributed but the plants raised here have not proved hardy. Rhododendron serpyllifolium Miquel in Ann. Mus. Lugd.-Bat. II. 165 (1865-66); Prol Fl. Jap. 97 (1866-67). Maximowicz in Mem. Acad. Sci. St. Petersburg, se> 7, XVI. 42, t. 4, figs. 18-25 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 292 (1875). Hooker f. in Bot. Mag. CXXII. t. 7503 (1896). Boissier in Bull. Herb. Boiss. V. 919 (1897). Matsumura in Bot. Mag. Tokyo XIV. 69 (1900); Ind. PL Jap. II. pt. 2, 464 (1912). Schneider, III. Handb. Laubholzk. II. 507, fig. 333 a-d (1911). Bean, Trees & Shrubs Brit. Isl. II. 378 (1914). Millais, Rhodod. 240 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [15] (1918). Azalea? serpyllifolia A. Gray in Perry, Jap. Exped. II. 315 (1857); in Mem. Am. Acad. n. ser. VI. 399 (1858-59). Masters inGard. Chron. n. ser. XVII. 429 (1882). Japan: Kyushu, prov. Osumi, Mt. Takakuma, March, 1912, K. Toyohira; prov. Hizen, Mt. Unzen, April 14, 1914, K. Sakurai; Naga- saki, cultivated, 1863, C. Maximowicz (Herb. Kew). Hondo, prov. Settsu, near Kobe, April, 1875, H. N. Moseley (Herb. Kew); prov. 54 THE AZALEAS OF THE OLD WORLD Idzu, May 19, 1853, C. Wright (Herb. Gray) ; same place, cultivated, April, 1854, Messers Williams & Morrow (Herb. Gray, type); Mt. Amagi, April 27, 1913, K. Sakurai; prov. Sagami, Hakone, April and May, 1875, H. N. Moseley (Herb. Kew). Cultivated: Arnold Arboretum, flowered in greenhouse, March 14, 1907; Royal Gardens, Kew, April 14, 1896 (Herb. Kew, type of Bot. Mag. t. 7503). During my travels in Japan I did not see this pretty little species in a wild state. On the volcanic mountains of the Hakone region and on those of Idzu province, especially Amagi-san, it is said to be common. This seems to be the northern limit of its range. Matsumura gives two or three other localities in Hondo but I have seen no specimens except one from near Kobe which probably came from a cultivated plant. In the south island of Kyushu it grows on the volcanic Mt. Unzen near Nagasaki, and on Mt. Takakuma in the southern prov- ince of Osumi. Matsumura records it from Okinawa Island in Liukiu but I have seen no material from there. In Japan it appears to be confined to purely volcanic soils. It is also frequently cultivated there, especially in the gardens of the peasantry. According to Messers Williams and Morrow its Japanese name signifies " Rice Azalea" and has reference to the leaves which are minute and smaller than those of any other Japanese species. It was introduced into cultivation in England by Messers Veitch, presumably through their collector Charles Maries, and flowered for the first time in England in 1882. Later it was introduced to America but is still very rare. In the Arnold Arboretum it has not proved hardy. The figure in the Botanical Magazine gives a very good idea of this species. Differing from the type in color of the corolla only is : Rhododendron serpyllifolium var. albiflorum Makino in Tokyo Bot. Mag. XXII. 57 (1908). Schneider, III. Handb. Laubholzk. II. 507 (1911). Komatsu in Tokyo Bot. Mag. XXXII. [15] (1918). Of this I have seen only Makino's type which he says is from the mountains of western Japan. Makino names it in the vernacular the " Shirobana-unzen- tsutsuji." Rhododendron Nakaharai Hayata in Jour. Coll. Sd. Tokyo XXV. art. 19, 153 (Fl Mont. Formos.) (1908). Millais, Rhodod.2W (1917). Kanehira, Formos. Trees, 325, fig. 15 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [14] (1918). Rhododendron serpyllifolium Hayata in Tokyo Bot. Mag. XX. 72 (1906), not Miquel. Formosa: prov. Taihoku, Mt. Shichisei ex Hayata. I did not see this plant growing wild but in Tokyo examined the type specimen and consider it to be a good species. As the author points out it is nearest to R. serpyllifolium Miq. from which it is at once distinguished by its larger flowers, each with 10 stamens. ENUMERATION OF THE SPECIES 55 Rhododendron Tschonoskii Maximowicz in Mem. Acad. Sri. St. Petersburg, se>. 7, XVI. 42, t. 3, figs. 8b-14 (Rhodod. As. Or.) (1870); in Bull. Acad. Sri. St. Petersburg, ser. 3, XV. 230 (1871). Franchet & Savatier, Enum. PL Jap. I. 295 (1875). Boissier in Bull Herb. Boiss. V. 919 (1897). Matsumura, PL Nikko, 71 (1894); Ind. PL Jap. II. pt. 2, 465 (1912). Schneider, III. Handb. Laubholzk. II. 507, fig. 332 l-o (1911). Bean, Trees and Shrubs Brit. Isl. II. 383 (1914). Millais, Rhodod. 256 (1917). Nakai in Tokyo Bot. Mag. XXXI. 244 (1917); Fl. Syl. Kor. VIII. 46, t. 17 (1919). Komatsu in Tokyo Bot. Mag. XXXII. [15] (1918). Azalea Tschonoskii O. Kuntze Rev. Gen. II. 387 (1891). Rhododendron trinerve Franchet apud Boissier in Bull. Herb. Boiss. V. 92 (1897). Rhododendron Tschonoskii a. typicum Makino in Tokyo Bot. Mag. XVIII. 66 (1904). Rhododendron Tschonoskii a. typicum a. pentamerum Makino 1. c. Matsu- mura, Ind. PL Jap. II. pt. 2, 465 (1912). Rhododendron Tschonoskii a. typicum b. tetramerum Makino I. c. Matsu- mura I. c. Rhododendron Tschonoskii /3. trinerve Makino in Tokyo Bot. Mag. XVIII. 66 (1904). Schneider, III. Handb. Laubholzk. II. 507 (1911). Rhododendron Tschonoskii /3. trinerve a. pentamerum Makino 1. c. Matsu- mura I. c. Rhododendron Tschonoskii ft. trinerve b. tetramerum Makino 1. c. Matsumura I.e. Rhododendron Tschonoskii var. tetramerum Komatsu in Tokyo Bot. Mag. XXXII. [15] (1918). Japan: Shikoku, prov. Tosa, Nakatsu-san, June 20, 1889, K. Watanabe (Herb. Gray) . Hondo, prov. Ettchu, Mt. Tateyama, July 23, 1884, (Herb. Kew) ; prov. Shinano, Tsubakura-dake, alt. 1300-1600 m., not common, September 15, 1914, E. H. Wilson (No. 7455) ; Hatta- san, July 13, 1894, U. Faurie (No. 13,356); Mt. Asama, July, 1904, U. Faurie (No. 5831); prov. Shimotsuke, Nikko region, Lake Yumoto, June 22, October 18, 1914, E. H. Wilson (No. 7657); same locality, November 6, 1892, C. S. Sargent; Lake Chuzenji, August 12, 1905, J. G. Jack; Mt. Nyoho, August 3, 1909, K. Sakurai; prov. Uzen, Adzuma-san, alt. 800-2000 m. July 20, 1914, E. H. Wilson (Nos. 7233, 7234, 7235); mountains round Yamagata, July 12, 1889, U. Faurie (No. 4444) ; prov. Rikuchu, slopes of Hayachine-san, alt. 1300- 2000 m. September 28, 1914, E. H. Wilson (No. 7583); Mt. Iwate July 16, 1903, S. Arimoto (Herb. Gray); high mountains, 1865, Tschonoskii ex Maximowicz (Herb. Kew, Herb. Gray, co-type) ; prov. Mutsu, Nambu, Shichinoki, June 23, 1889, U. Faurie (Herb. Kew) ; 56 THE AZALEAS OF THE OLD WOELD Lake Towada, June 25, 1894, U. Faurie (No. 13,279, Herb. Kew); Hakkoda-san, alt. 1600 m. July 5, 1914, E. H. Wilson (No. 7122); same locality, U. Faurie (No. 875, Herb. Kew). Hokkaido, prov. Oshima, Esashi, August 4, 1890, K. Miyabe & Y. Tokubuchi (Herb. Gray); prov. Ishikari, near Sapporo, September 17, 1892, C. S. Sar- gent; Masuzan, July 30, 1897, U. Faurie (No. 636). Korea : prov. South Keisho, Chiri-san, peak Ten-no-bo, alt. 1845 m. November 16, 1917, E. H. Wilson (No. 9595); same place, July 30, 1913, T. Nakai. Cultivated : Arnold Arboretum (No. 3691); Royal Gardens, Kew, July 1, 1910; Gardens, Tokyo, May 31, 1882, K. Miyabe. This is an alpine or subalpine species that grows on most of the high mountains in Hondo from Shinano northward; in Hokkaido it grows sparingly as far north as the mountains round Sapporo. From Shikoku I have seen but one specimen and it is not reported from Kyushu. Out of Japan this species has been found only on the Chiri-san range in south Korea. This is an interesting discovery for the plant does not grow on Quelpaert Island where so many Japanese species are common, neither has it been found elsewhere in Korea. In the Nikko region R. Tschonoskii is quite common, especially round Lake Yumoto, where it grows on shady rocks. On Adzuma-san it is abundant in places fully exposed to the sun and this locality is about the centre of its distribution. On the wind-swept upper- slopes of Hayachine-san it is common but north of this mountain I found it to be rare. The plant forms a broad, densely branched twiggy shrub from 0.3 to 2.5 m. high but is usually from 0.5 to 1.5 m. being taller in the forest shade than on exposed rocks. It prefers rocky ground, but seems to be equally happy in the moist woods of Nikko and on the exposed cliffs and slopes of Adzuma and Haya- chine mountains. Sometimes it forms a broad low mat but is more usually an upright shrub. It is always very twiggy and the young shoots are densely cov- ered with closely appressed, flattened rufous hairs. The short stalked leaves are deciduous, crowded at the end of the branches, and vary in shape from narrow- lanceolate to ovate-lanceolate or broad-oblanceolate, they are acute and mucronu- late and may have one or three prominent nerves; the undersurface is pale and both surfaces have long, appressed, soft, red-brown hairs which are especially noticeable on the young leaves. The flowers are small, with a white corolla, and are four or five partite; they have exserted stamens, and are borne in terminal clusters of from 4 to 6 flowers in each. Except that the leaves vary in size and may have either one or three prominent nerves, and that the flowers are tetram- erous or pentamerous the species is very constant. From one and the same bush on Adzuma-san I gathered specimens which exhibited the whole range of varia- tion on which Makino founded his numerous forms. With its small white flowers R. Tschonoskii is very distinct from all other Japanese species but as a garden plant it has little to recommend it though in autumn the leaves change to rich shades of orange-red and crimson. It was discovered in 1865 by Maximowicz's Japanese collector Tschonoski on the mountains of Rikuchu province, northern Hondo, and is figured and described by Maximowicz. It was introduced into England by Charles Maries who sent seeds to Messers Veitch in 1878; to this country it was introduced by Professor Sargent who sent seeds from the Nikko region to the Arnold Arboretum in 1892. ENUMERATION OF THE SPECIES 57 In 1917 I sent seeds to the same institution from Chiri-san in Korea. The Japanese name for this Azalea is " Shirobana-no Kome-tsutsuji." Rhododendron microphyton Franchet in Bull. Soc. Bot. France XXXIII. 235 (1886). Hemsley in Jour. Linn. Soc. XXVI. 28 (1889). Diels in Notes Bot. Gard. Edinburgh, VII. 309, 392 (PL Chin. Forrest.) (1913). Hutchinson in Millais, Rhodod. 210 (1917). China: prov. Yunnan, Talang, alt. 1600m. A. Henry (No. 11,596); Szemao, alt. 1600 m. A. Henry (No. 12,983); " Che-lin-se, Pe-yen- tsin," May 3, 1916, Simeon Ten (Nos. 86, 86a); " Mt. Lai-chan-po, Pe-yen-tsin," May 2, 1916, Simeon Ten (No. 83), " Ni-kou, Pe-yen-tsin," July 5, 1917, Simeon Ten (No. 417); Yunnan Fu, alt. 2000 m. April, 1916, 0. Schoch (No. 350 g); Tali mountain range, Lat. 25 40' N. alt. 2600-3300 m. May to July, 1906, G. Forrest (No. 4172^ . I have no knowledge of this plant in a wild state but according to notes on the specimens cited above it is an erect, much-branched shrub growing from 0.3 to 2 m. tall. The branchlets are short and twiggy and closely beset with flattened, appressed red-brown bristles. The leaves are persistent, oval to lanceolate in shape, from 0.5 to 3.5 cm. long and from 0.5 to 1.5 broad, acute, obtuse or rounded at the apex which is mucronulate; the upper surface is dark green, the lower pallid and both have scattered hairs on the midrib; the underside and the petioles are clad with red-brown bristles. The flowers are borne several together at the end of the branchlets and in some specimens are so numerous that the stem appears to be clothed with flowers; on others the flower-clusters are sparse. The corolla is rose-colored with crimson markings, with a slender, cylindrical tube from 5 to 8 mm. long and spreading lobes, and varies from 1 to 2 cm. in diameter. The five stamens are exserted and shorter than the style. The calyx is minute or with distinct lanceolate lobes from 1 to 4 mm. long, and like the pedicels and ovary is clothed with straight, shining chestnut-brown hairs. The red-brown pubescence is prominent on this plant. Its small leaves and small flowers resemble those of certain species of the section Lepipherum, rather than those of a species of Tsutsutsi. Where it grows plentifully on open mountain-slopes it must be conspicuous when in bloom, and it would be a desirable addition to gardens even as a pot plant where it would not be hardy. From the altitude given by Forrest (2600-3300 m.) it would appear that this species is not only the most western but the most alpine of the Chinese species of the section. As Franchet points out it is the Chinese analogue of the Japanese R. Tschonoskii Maxim, and like that species exhibits considerable variation in size of leaves and flowers. At first I thought Simeon Ten's No. 417 with minute leaves and small flowers was a distinct variety but after careful comparison and dissection I am satisfied that the distinct appearance is the result of exposed ecological conditions. Rhododendron Seniavinii Maximowicz in Mem. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9, 33, t. 3, fig. 21-24 (Rhodod. As. Or.) (1870). Hemsley in Jour. Linn. Soc. XXVI. 30 (1889). Bretschneider, Hist. European Bot. Disc. Chin. 620 (1898). Millais, Rhodod. 240 (1917). 58 THE AZALEAS OF THE OLD WORLD China: prov. Hunan, Yiin-shan, near town of Wukang, alt. 1350- 1400 m. June 9, 1918, Dr. Handel-Mazzetti (No. 738). This rare and very distinct species is unknown to me in a living state. Dr. Handel-Mazzetti says it is a shrub some 2 m. tall, and that it grows on the margins of forests of broad-leafed trees and has white flowers with the tube of the corolla suffused with rose-color and the upper-lobe spotted with purple. The gray and rufous-gray appressed hairs which clothe the shoots, both surfaces of the young leaves, the under-side of adult leaves, pedicels, calyx and ovary well characterise the plant. It is evidently a much-branched twiggy shrub, and the shoots are clad for a season with the usual flattened, appressed bristles of the group. The leaves are persistent or semi-persistent, crowded at the end of the branches, dark green, glabrescent above with impressed reticulate veins. In shape the leaves vary from oval to lanceolate, being from 1.5 to 6.5 cm. long and from 1 to 2.5 cm. wide, acute or obtuse and mucronate at apex. The flowers are small, clustered from 3 to 10 together at the end of the shoots; the corolla is about 1.5 cm. across, with a nearly cylindric tube hairy on the outside, and spreading lobes. The calyx is small and hidden beneath the gray or rufous straight hairs. The five stamens are exserted, and are overtopped by the style, which has a capitate, lobed stigma. In Maximo- wicz's figure the petioles vary from 0.5 to 1 cm. in length whereas on the specimen cited none exceed 0.5 cm. but this is the only difference I can find and were more material available this would probably disappear. The discoverer of this species is unknown but from what Bretschneider says it seems probable that the original specimen came from the borders of western Fokien, a region still very little known to us. Dr. Handel-Mazzetti's discovery in south- western Hunan is an interesting extension of the range of the species, which has for so long remained obscure. It is not in cultivation. Rhododendron Mariae Hance in Jour. Bot. XX. 230 (1882). Hemsley in Jour. Linn. Soc. XXIX. 27 (1889). Dunn & Tutcher in Kew Bull. Misc. Inform, add. ser. X. 155 (Fl. Hongk. and Kwang- tung) (1912). Hutchinson in Millais, Rhodod. 207 (1917). China: prov. Kwangtung, Ting-woo Mt., May 26, 1918, C. 0. Levine, Canton Christian College Herb. No. 2015 (Herb. Bur. Sci. Manila); Loh-fau-shan, alt. 1000 m. August 10, 1917, C. 0. Levine (No. 1563, Herb. Bur. Sci. Manila); same place, August 13, 1917, E. D. Merrill (No. 11,103, Herb. Bur. Sci. Manila); Tiu-kaen-shan, March, 1918, To-Kang-P'eng, Canton Christian College Herb. No. 2370 (Herb. Bur. Sci. Manila). Known to me only from Hance's description with which the above-cited speci- mens agree exactly. It is evidently a twiggy shrub and the branchlets are clothed with appressed, reddish gray bristles which, changing to gray persist more or less during a couple of seasons. The leaves are dimorphic, clustered at the end of the branches, persistent and for a plant of its group very coriaceous; they are shining dark green and nearly glabrous above, with impressed, reticulate veins, pale below, with raised veins and scattered appressed rufous bristles. The spring leaves are elliptic-lanceolate from 3 to 9 cm. long and from 2 to 3 cm. wide, acute and mucronu- late; the summer leaves are elliptic to obovate from 1 to 3 cm. long and from 0.8 to 1.5 cm. wide, obtuse or rounded at the mucronulate apex. The petioles are ENUMERATION OF THE SPECIES 59 etrigosely pubescent. The flowers are small, crowded in terminal clusters, and fragrant. The corolla is lilac-colored, with a slender tube about 1 cm. long, and small spreading lobes; the 5 exserted stamens are overtopped by the slender style with its capitate stigma. The pedicels are straight from 1 to 1.5 cm. long and with the annular calyx and the ovary are densely clothed with shining, rufous-brown strigose hairs. This species is known only from one or two localities near Canton in south China and is evidently closely related to R. Seniavinii Maxim. So far as I know it has never been introduced into cultivation, although discovered by the Reverend B. C. Henry as long ago as April, 1882. Rhododendron scabrum G. Don, Gen. Syst. III. 846 (1834). Maximowicz in Mem. Acad. Sci. St. Petersbourg, ser. 7, XVI. No. 9, 49 (Rhodod. As. Or.) (1870). Rhododendron sublanceolatum Miquel in Ann. Mus. Lugd.-Bat. II. 163 (1865 66); Prol Fl. Jap. 95 (1866-67). Maximowicz in Mem. Acad. Sci. St. Petersbourg, ser. 7, XVI. No. 9, 35 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 290 (1875). Hemsley in Jour. Linn. Soc. XXVI. 31 (1889). Gard. Chron. XLIX. 342, fig. 157 (1911). Jour. Hort. Soc. Lond. XXXVII. p. cxxxi. fig. 115 (1911). Schneider, III Handb. Laubholzk. II. 504 (1911). Craib in Bot. Mag. CXXXIX. t. 8478 (1913). Bean, Trees and Shrubs Brit. Isl. II. 381 (1914). Millais, Rhodod. 249 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [12] (1918). Rhododendron indicuni Hooker and Arnott, Bot. Voy. Beechey 266 (1841), not Sweet. Matsumura in Tokyo Bot. Mag. XIV. 69 (1900). Rhododendron Jodogawa Hort. apud Hasskarl in Cat. Hort. Bog. alt. 162 (1844). Rhododendron indicum /3. sinensis Miquel in Ann. Mus. Lugd.-Bat. I. 33 (1863-64). Rhododendron ledifolium /3. purpureum Maximowicx in Mem. Acad. St. Peters- bourg, se>. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870), in part. Azalea sublanceolata O. Kuntze, Rev. Gen. II. 387 (1891). Rhododendron indicum var. sublanceolatum Makino in Tokyo Bot. Mag. XVIII. 100 (1904). Matsumura, Ind. PL Jap. II. pt. 2, 461 (1912). Rhododendron sublateritium Komatsu in Tokyo Bot. Mag. XXXII. [12] (1918). Rhododendron liukiuense Komatsu in Tokyo Bot. Mag. XXXII. [12] (1918). Liukiu: Okinawa Island, near Katena, sea-level to 100 m. March 8, 1917, E. H. Wilson (No. 8046); without locality, C. Wright (No. 172, Herb. Gray, Herb. Kew): Takuno-shima, hillsides, April 1, 1853, C. Wright (Herb. Gray). Japan: Kyushu, prov. Satsuma, Kagoshima, cultivated, May 6, 1918, E. H. Wilson (No. 10,344, 10,345); Kajiki, cultivated, May, 1917, Z. Tashiro; prov. Hizen, Nagasaki, 1862, R. Oldham (No. 504, Herb. Gray, Herb. Kew); same locality, G. H. Langsdorff (Herb. Kew); same locality, 1863, C. Maximowicz (Herb. Kew); same lo- cality, cultivated, April 25, 1903, U. Faurie (No. 7873); prov. Chi- kugo, Kurume, cultivated, May 3, 1918, E. H. Wilson. Hondo, prov. 60 THE AZALEAS OF THE OLD WORLD Musashi, Yokohama, cultivated, December 22, 1914, E. H. Wilson (No. 7873); same locality, May 15, 1918, E. H. Wilson. Cultivated: Hort. Holm Lea, Brookline, Mass., April 15, 1915; Hort. Kew, 1912, type of Bot. Mag. t. 8478. This is a laxly branched shrub from 1 to 2 m. high and has relatively stout branches which are clothed in their first year with appressed, flattened, strigose, gray- brown hairs which gradually disappear. The leaves are shining, coriaceous, nearly glabrous except on the veins at maturity, persistent; those formed in the spring are lanceolate to elliptic-lanceolate, from 4 to 10 cm. long and from 2 to 3.5 cm. broad, acute or obtuse and mucronate at the apex, the base broadly cuneate, the margin slightly recurved ciliate and sub-crenulate, the principal veins impressed above and prominently raised on the under surface; the summer leaves are ob- lanceolate, smaller and more coriaceous than the spring leaves and subacute or rounded at the apex, which is mucronulate; when young the leaves have appressed, straight gray-brown hairs more especially on the nerves and an occasional sessile gland on the under-side; at maturity they are very sparsely pubescent except on primary and secondary nerves; the petiole is stout, flattened, somewhat appressed to the shoot, and is clothed with pubescence similar to that on the shoots and leaf veins. The winter-buds are ovoid, acute, clothed with rufous, appressed straight hairs; the inner bud-scales cinnamon-brown, ciliate, viscid on inner surface and re- main on the base of the pedicels of the expanded flowers. The flowers are borne in terminal clusters from 2 to 6 in each. The calyx is green, ample, the lobes usually oval and rounded, about 6 mm. long and 5 mm. broad, sometimes one and occasion- ally all the calyx-lobes lance-shape and 1.2 cm. or more long, acuminate, ciliate and glandular, and covered with appressed, straight, gray hairs. The corolla is rose-red to vivid scarlet, broad-funnel-shape about 6 cm. wide, lobed nearly to the middle; stamens 10, included, overtopped by style. The fruit is erect, ovoid, from 1 to 1.2 cm. long, sparsely hairy, and is subtended by persistent calyx-lobes. This species is endemic in the Liukiu Islands, where it is confined to the middle, or Okinawa, group of islands and to Takuno-shima, the most southern of the north- ern, or Oshima, group. It grows among shrubs, coarse grasses and Pinus luchuensis Mayr, and when in bloom is very conspicuous. It is a common plant in gardens round Kagoshima in south Japan, having been long ago introduced from Liukiu. From Kagoshima it has been taken to Nagasaki and north as far as Tokyo, but it is only in the warmer parts of south Japan that it can be described as a common garden plant. The leaves, flowers and fruit are larger than those of other species of the group; the leaves are also more persistent and more coriaceous. The flowers are normally of an intense scarlet but they vary to rose-red. The calyx-lobes are often very unequal in size on the same flower or flower-cluster and they vary too in degree of pubescence. In habit it is more vigorous but less compact than that of other species of the group. Specimens in Herb. Kew and Herb. Gray labelled R. Sieboldii var. ellipticum ex Herb. Lugd.-Bat., 1863, would appear to belong here though the flowers are abnormally small for the species. The leaves, bud-scales, calyx and number of stamens agree and it may be that the specimens are from a plant flowering out of season or under abnormal conditions. Maximowicz, Miquel and Makino refer to it as the " Chinese Azalea " but there does not appear to be any authority for the statement since the plant itself is un- known in China. To-day in Japan generally it is known as the Liukiu (Riukiu) Azalea, but round Kurume in Kyushu it is called "Hiwoge-yodogawa." Maximo* wicz does not appear to have properly known this species and has confused Wright's Tokuno-shima specimen and others with his R. ledifolium /3. purpureum yet he ENUMERATION OF THE SPECIES 61 states that it was introduced into Petrograd in 1864. In England this Liukiu species does not appear to have been properly known before 1911, when Mr. R. C. Notcutt of Woodbridge exhibited it at the Royal Horticultural Society's Temple Show in London. In 1915 I brought plants from Japan to the Arnold Arboretum. Of all the red-flowered species of eastern Asia except the Formosan R. Oldhamii Maxim, it is the least hardy. In Massachusetts it is a greenhouse plant and blossoms sparsely. Miquel did not know the color of the flowers of his R. sublanceolatum citing it as alba? which it certainly is not for no white form is known even to-day. The color of the flowers varies from intense scarlet to rich magenta. Don gives deep rose-color for the type and the scarlet-flowered form, which is the usual form found growing wild, may be distinguished as f . coccineum Wilson, n. name. It is unfortunate that we have to adopt Don's name for this species yet it is undoubtedly the oldest since his description fits this plant and no other. He says that the leaves are " ovate " a term not often applicable to the leaves of any Azalea, they are, how- ever, broader in this species than in any other. Don cites as a synonym R. maxi- mum Thunb., which is R. Mettemichii S. & Z., but no word of his description has reference to this Eurhododendron. This Liukiu species has been cultivated round Nagasaki and other towns in southern Kyushu I know not how long, but Don is wrong in saying that it is native of Japan, though the mistake is natural when the period he wrote in is considered. Komatsu bases his two species on slight differ- ences in the shape of the leaves, in the color of the corolla and on the length of the corolla-tube. Such trivial differences are unimportant in such variable plants as the Azaleas of the Tsutsutsi section. Rhododendron phoeniceum G. Don, Gen. Syst. III. 846 (1834). Azalea indica var. y. Sims in Bot. Mag. LIII. t. 2667 (1826). Azalea punicea Sweet apud Poiteau in Annal. de From. I. 102, 104 (1829), not Rhododendron indicum a. puniceum Sweet. Rhododendron indicum y. phoeniceum Sweet, Brit. Flow. Gard. ser. 2, II. sub. t. 128 (1832). Rhododendron indicum var. speciosum D. Don in Sweet, Brit. Flow. Gard. ser. 2, III. t. 284 (1835), excluding synonomy. Azalea indica phoenicea Hovey, Am. Gard. Mag. I. 224 (1835). v Azalea indica Rawsonii Loudon, Gard. Mag. n. ser. II. 421 (1836). Azalea Rawsonii Paxton, Mag. Bot. III. 123, t. (1837). Rhododendron indicum y. purpureum De Candolle, Prodr. VII. pt. 2, 726 (1839), in part, not Sweet. Rhododendron ledifolium /3. phoeniceum De Candolle, Prodr. VII. pt. 2, 727 (1839), in part. Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Rhododendron puniceum Planchon in Fl. des Serr. IX. 79 (1854), not Rox- burgh; in Rev. Hort. 1854, 46, 61. Maximowicz in Mem. Acad. Sci. St. Petersburg, ser. 7, XVI. No. 9, 35 (Rhodod. As. Or.) (1870). Schneider, III. Handb. Laubholzk. II. 504 (1911). Azalea phoenicea Hort. Paris apud Planchon in Fl. des Serr. IX. 79 (1854), as a synonym; in Rev. Hort. 1854, 61. Japan: ex Siebold comm. Zuccarini ex Herb. Zucc. 1842 (Herb. Gray, doubtful specimen). Cultivated: Hort. Holm Lea, March 30, 1917, A. Rehder. Although introduced into European gardens as long ago as 1824 and used in such vast quantities as a stock on which to graft the varieties of "Indian Azaleas" 62 THE AZALEAS OF THE OLD WORLD BO-called, the origin of this plant is obscure. It appears to have been sent from Canton, China, to the Horticultural Society's garden at Chiswick, probably by John Reeves but the facts are not clearly known. One of these plants was figured by Sims in the Botanical Magazine (t. 2667), and according to Poiteau it was known in the gardens round Paris in 1829 but was rare. ,No wild species to which it can be referred unhesitatingly has been discovered. In Japan I saw nothing just like it but it is evident that the " Omurasaki-tsutsuji " is a glorified form of it. In its vigorous habit, size of flower and shyness of blossoming R. phoeniceum G. Don resembles R. scabrum G. Don, whilst in the calyx and character of pubescence it suggests R. mucronatum G. Don; its viscid inner bud-scales agree with both species. Possibly it is a hybrid between these two species though I incline to the belief that it is nothing but an extreme form of R. scabrum G. Don. I have not been able to discover just when it reached America but it was cultivated by Hovey & Co., Boston, Mass., in 1835. In England attemps to hybridise this plant with all sorts of species of Rhododendron were made and the books record a number of presumed hybrids of remarkable parentage. For example Paxton's Azalea Raw- sonii is said to be a cross between A. phoenicea and R. dauricum atrovirens but an unbiased examination of the plate reflects nothing but a good colored form of R. phoeniceum. Other presumed hybrids are dealt with hereunder and it is evident that enthusiasm out-did the facts. It is probable that this plant has had something to do in the evolution of the present-day race of " Indian Azaleas." In Belgium R. phoeniceum and its variety concinnum are the principal stocks used in grafting the "Indian Azaleas" of commerce. A form of this is: Rhododendron phoeniceum f . semiduplex Wilson, n. f. Azalea ledifolia var. /3. phoenicea Hooker in Bot. Mag. LX. t. 3239 (1833). Rhododendron kdifolium ft. phoeniceum De Candolle, Prodr. VII. pt. 2, 727 (1839), in part. G. Don and others have merged this into the type but its semi-double flowers make it distinct as a garden plant. The color is poor. I have seen the specimen preserved in Herb. Kew from which the Botanical Magazine figure was prepared but no living material, and do not know if the plant is now in cultivation. In the typical R. phoeniceum flowers are often seen in which the stamens show a tendency toward becoming petaloid. Another form is: Rhododendron phoeniceum f. Smithii Wilson, n. comb. Rhododendron indicum 7. Smithii Sweet, Hort. Brit. ed. 2, 343 (1830). Rhododendron pukhrum Sweet, Brit. Flow. Gard. ser. 2, II. t. 117 (1832). De Candolle, Prodr. VII. pt. 2, 726 (1839). Maximowicz in Mem. Acad. Sci. St. Petersbourg, se>. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870). Rhododendron indicum pulchrum G. Don, Gen. Syst. III. 845 (1834). Azalea indica hybrida Loudon, Gard. Mag. n. ser. I. 326 (1835). Hovey, Am. Gard. Mag. II. 157 (1836). This plant is stated variously to be " the produce of R. ledifolium Hook, im- pregnated by the pollen of the Old Red R. indicum" and " a hybrid between R. phoeniceum and R. ledifolium" but to me it appears to be nothing else than a form of R. phoeniceum. Under Loudon's name it was introduced into America about 1835 and was exhibited for the first time before the Massachusetts Horticultural Society on March 5, 1836, by Hovey & Co. For many years it was a favorite ENUMERATION OF THE SPECIES 63 plant at exhibitions. It is still cultivated in the Magnolia Gardens, near Charles- ton, South Carolina. An old and obscure plant is: Rhododendron phoeniceum f . splendens Wilson, n. comb. Rhododendron phoeniceum var. splendens D. Don in Sweet, Brit. Flow. Gard. ser.2,IV.t.385(1837). This is said to be a seedling raised from R. phoeniceum G. Don pollinated by R. catawbiense Mich, but the figure shows no influence of the latter plant. That it is a seedling may be accepted, but it is certainly nothing more than a good colored form of R. phoeniceum with only from 5 to 8 stamens. A common garden plant in Japan is: Rhododendron phoeniceum var. calycinum Wilson, n. comb. Azalea indica calycina Lindley in Paxton, Flow. Gard. II. 169, t. 70 (1852). Rhododendron calycinum Planchon in Fl. des Serr. IX. 81 (1854) ; in Rev. Hort. 1854, 65. Rhododendron ledifolium /3. purpureum Maximowicz in Mem. Acad. Sci. St. Petersburg, sSr. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870), in part. Franchet & Savatier, Enum. PI. Jap. I. 291 (1875). Matsumura, Ind. PI. Jap. II. pt. 2, 462 (1912). Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Azalea rosmarinifolia var. purpurea Rehder in Bailey, Cycl. Am. Hort. 1. 123 (1900). Rhododendron rosmarinifolium var. purpureum Schneider, III. Handb. Laub- holzk. II. 504 (1911). Komatsu in Tokyo Bot. Mag. XXXII. [351 (1918). Rhododendron rosmarinifolium var. speciosum Makino in Tokyo Bot. Mag. XXVII. 110 (1913). Rhododendron Oomurasaki Makino in Jour. Jap. Bot. I. 18 (1917). Rhododendron Osakazuki Komatsu in Tokyo Bot. Mag. XXXII. [46] (1918) ; in Icon. Koisikav. IV. 9, t. 217 (1918). Japan: Kyushu, prov. Hizen, Nagasaki, cultivated, 1863, C. Max- imowcz (Herb. Kew, Herb. Gray, co-types of R. ledifolium 0. pur- pureum Maxim.); same locality, 1862, R. Oldham (No. 509, Herb. Kew, Herb. Gray; No. 607, Herb. Kew); prov. Chikugo, Kurume, culti- vated, May 3, 1918, E. H. Wilson. Hondo, prov. Choshu, Shimono- seki, Sanyo Hotel garden, cultivated, May 17, 1917, E. H. Wilson (No. 8425); prov. Musashi, Yokohama, cultivated, Hort. Suzuki, May 15, 1918, E. H. Wilson; Tokyo, cultivated, May 11, 1902, K. Sa- kurai; same locality, cultivated, May 12, 1910. As pointed out by Lindley this variety is much superior to the type for it has larger, more richly colored flowers. The corolla is rose-purple with crimson spots. The calyx-lobes are larger and more acuminate but they vary. In Paxton's figure the calyx is glabrous but this is a mistake. It is ciliate and sparsely or densely clothed with long, straight, appressed hairs. Apart from the calyx Paxton's figure is an excellent representation of the Omurasaki-tsutsuji (large purple Azalea) so much grown in the gardens of Yokohama, Tokyo and elsewhere in Japan. At 64 THE AZALEAS OF THE OLD WORLD Kurume in Kyushu it is known as the Kumagai-yodogawa. It is a vigorous free- flowering plant of compact habit of growth, and its large, magenta-red flowers are very handsome. Nothing definite appears to be known in Japan of the origin of this Azalea but it has been cultivated for a long period. Maximowicz 's speci- men undoubtedly belongs here, but those of Langsdorff, Wright and some of Oldham's included by Maximowicz under his var. purpureum belong to R. sca- brum G. Don. Recent botanists seem to have overlooked Lindley's name and excellent figure. According to Planchon this variety was introduced from China by Robert Fortune to the gardens of the Horticultural Society at Chiswick where it flowered for the first time in 1851. A very hardy form with carmine-red flowers much cultivated in Europe and, to a less degree, in this country may be distinguished as: Rhododendron phoeniceum var. calycinum f. Maxwellii Wilson, n. comb. Azalea Maxwelli Hort. Wezelenburg, Cat. 6, fig. (1915). Rhododendron Maxwelli Millais, Rhodod. 208, fig. (1917). Japan : Kyushu, prov. Chikugo, Kurume, cultivated, May 3, 1918, E. H. Wilson. Cultivated: Hort. Bobbink & Atkins, Rutherford, N. J., April, 1920. At Kurume this bright-colored form is known as the " Aka-yodogawa." To this variable species must be added : Rhododendron phoeniceum var. tebotan Wilson, n. comb. Rhododendron Tebotan (Rh. Kaempferi X Rh. rosmarinifoliwri) Komatsu in Tokyo Bot. Mag. XXXII. [46] (1918). This variety has double flowers with small green leaves showing the centre, and is remarkable for its color which is the same as that of Bougainvillaea glabra Choisy, and quite unique among Azaleas. It is an old plant in Japanese gardens but is now rare and Japanese nurserymen substitute for it Azalea " Fuji-manyo " (R. mucronatum f. plenum Wils.). In 1919 I secured in Japan a living plant of the true var. tebotan for the Holm Lea collection where it has flowered. Rhododendron yedoense Maximowicz apud Regel in Gartenfl. XXXV. 565, t. 1233 a-b (1886). Millais, Rhodod. 264 (1917). Azalea Yodogava Grignan in Rev. Hort. 1908, 425 t. fig. 2. Rhododendron phoeniceum Matsumura, Ind. PL Jap. II. pt. 2, 463 (1912), not G. Don. Rhododendron Yodogawa Kunert in Gartenwelt XVI. 163, t. (1912). Millais, Rhodod. 264 (1917). Rhododendron poukhanense var. yodogawa Render in Mitt. Deutsch. Dendr. Ges. XXIV. 225 (1916); in Bailey, Stand. Cycl. Hort. V. 2943 (1916). Rhododendron Poukhanense var. plenum Nakai in Tokyo Bot. Mag. XXXI. 245 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [12] (1918). Azalea Tado Guno Hort. apud Millais, Rhodod. 264 (1917), as a synonym. Rhododendron Matsumurai Komatsu in Tokyo Bot. Mag. XXXII. [13] (1918). ENUMERATION OP THE SPECIES 65 Korea: prov. Keiki, cultivated, Forestry Experimental Garden, Keijyo, May 21, 1917, E. H. Wilson (No. 8422). Cultivated : Arnold Arboretum (No. 7345); ex Hort. E. G. Loder, May, 1898 (Herb. Kew). This double-flowered Azalea, in recent years exported from Japan in some quan- tity to the gardens of Europe and America, is certainly nothing but a monstrous form of R. yedoense var. poukhanense Nakai. My Korean specimens came from a plant which had been found wild not far from Mt. Poukhan and transplanted to the Government Forestry Experimental Garden, and are indistinguishable from others taken from plants cultivated in the Arnold Arboretum. The habit is spread- ing and more lax than in the type except when the latter grows among tall bushes. The flowers are quite double, and the corolla is rose-purple and much less beautiful than that of the type; the calyx is green, often larger than in the type, with oval segments. The leaves are characteristic and have the impressed veins on the upper surface. This form seems to have become known in our gardens comparatively recently and to have been first introduced into Petrograd by Japanese and exhibited at the International Exhibition there in 1884. It is perfectly hardy in the Arnold Arbor- etum. Matsumura gives " Botan-tsutsuji " (Paeony Azalea) as the vernacular name of this plant. The wild type of the species is: Rhododendron yedoense var. poukhanense Nakai hi Tokyo Bot. Mag. XXXIV. (274) (1920). Rhododendron indicum var. Simsii Palibin in Act. Hort. Petrop. XVIII. 150 (Consp. Fl. Kor. 11, 4) (1900), not Maximowicz. Nakai in Jour. Coll. Sci. Tokyo, XXXI. 76 (FL Kor. II) (1908). Rhododendron poukhanense LSveille* in Fedde, Rep. Nov. Sp. V. 100 (1908). Nakai in Journ. Coll. Sci. Tokyo, XXXI. 76 (FL Kor. II) (1908); in Tokyo Bot. Mag. XXXI. 244 (1917); FL Syl. Kor. VIII. 47, t. 18 (1919). Komatsu in Icon. PL Koi&ikav. II. 97, t. 133 (1915). Render in Mitt. Deutsch. Dendr. Ges. XXIV. 225 (1916); in Bailey, Stand. Cycl.Hort. V. 2943 (1916). Rhododendron hallaisanense LeveillS in Fedde, Rep. Nov. Sp. XII. 101 (1913). Rhododendron coreanum Rehder in Mitt. Deutsch. Dendr. Ges. XXII. 259 (1913). Millais, Rhodod. 148 (1917). Korea: Quelpaert Island, sea-level to 1600 m. abundant, October 29, November 6, 1917, E. H. Wilson (Nos. 9396, 9458); same locality April 14, October 25, 1908, June 1909, E. Taquet (Nos. 1090. 1091. 4679, 4680, 2972) ; prov. South Keisho, Fusan, May 18,1906, U.Faurie (No. 670) : prov. Keiki, Kazan, near Suigen in Pine-woods, May 24, 1917, E. H. Wilson (No. 8472); Seoul, April 29, 1894, Miss A. Sontag (Herb. Kew); same place, Mt. Poukhan, September 25, 1905, J. G. Jack; province unknown, " Mt. Pomasa," May 21, 1906, U. Faurie (No. 664). Cultivated: Arnold Arboretum (No. 7254). 66 THE AZALEAS OP THE OLD WORLD This Korean Azalea is usually a compact densely branched shrub from a few inches to a metre high, but in shaded places it may be as much as two metres tall and rather loosely branched; on wind-swept slopes and on the tops of rocks it hugs the ground and forms broad mats. The branches are twiggy and clothed with appressed gray hairs which disappear the second year. No part of the plant is glandular; the winter-buds are ovoid, more or less pointed and covered with appressed, gray-brown hairs. The leaves are quite or partially deciduous according to climate and in the autumn are tinted from orange to crimson. The spring leaves are membranous, oblanceolate to lanceolate or ovate-lanceolate, from 3 to 8 cm. long and from 1 to 2 cm. broad, acute and mucronate at apex, with a narrow wedge-shaped base; they are dark green and have impressed veins on the upper surface, and are pallid below where the principal veins are prominent, ascending and arching inward but not extending to the margins of the leaf; both surfaces are clothed with straight appressed gray to shining brown hairs. The summer leaves are thicker, narrowly oblong-lanceolate, from 0.4 to 1 cm. wide, attenuate at the base; the margins are often obscurely crenate-serrate, and the upper surface is glabrous at maturity except for a few, short appressed, white bristles. The flowers are in clusters of two to several, and are remarkably fragrant, with a corolla rose to rosy purple. The calyx is ample, green and clothed with pubes- cence similar to that of the leaves, the lobes are lanceolate to ovate-lanceolate, from 5 to 10 mm. long, acute, obtuse or rounded. The pistil is exserted but the stamens are included, 10 in number, with purple anthers. The fruit is erect, ovoid, from 7 to 10 mm. long, clothed with strigose hairs and subtended by the persistent calyx-lobes. This is the common Azalea of Korea from about the latitude of Seoul (Keijyo), the capital city southward. It is partial to open country and on grassy mountain-slopes and in thin Pine-woods it forms dense matlike masses from a few inches to a yard high. On boulders in water-courses the growth is low and dense but in thickets the plants are more loosely branched and often two metres high. It grows from sea-level up to about 1600 m. altitude according to latitude and in some places, like the foothills of Chiri-san, it is very abundant. On Quelpaert it grows from sea-level to near the summit of Hallai-san and is common on boulders in the water-courses and on the moorlands. On Poukhan- san behind Keijyo from which it derives its specific name it is rare though there are places not far from there where it is common. This Azalea was introduced to the Arnold Arboretum by Mr. J. G. Jack who sent seeds from Poukhan-san in the autumn of 1905. Plants flowered for the first time in May, 1914. It has proved perfectly hardy here, grows freely and flowers profusely. The flowers are distinctly and pleasantly fragrant and the plant is a decided acquisition to the list of hardy species of its section. Japanese botanists have recorded this plant from several places in Hondo, Japan, and Komatsu (in Tokyo Bot. Mag. XXXII. [38] (1918) ) enumerates fourteen garden forms. I have seen the Japanese material from wild plants referred to this species in the herbarium of the Imperial Botanic Garden, Tokyo, but it is too fragmentary to base a definite opinion upon. That from the Tokyo region has yellow anthers and belongs to R. dbtusum var. Kaempferi f. mikawanum Wils. It certainly has nothing to do with R. yedoense var. poukhanense. That from Bitchu province collected by Z. Yoshino has purple anthers and the veins on the upper surface are indistinctly impressed. It may represent Nakai's variety but the material is too incomplete to judge by. Rhododendron Oldhamii Maximowicz in Mem. Acad. Sci. St. Peters- bourg, se"r. 7, XVI. No. 9, 34 (Rhodod. As. Or.) (1870). Masters in Gard. Chron. n. ser. XVII. 524 (1882). Hemsley in Jour. Linn. Soc. ENUMERATION OF THE SPECIES 67 XXVI. 28 (1889). Henry in Trans. Asiat. Soc. Jap. XXIV. suppl. 57 (1896). J. H. Veitch, Hortus Veitchii, 286 (1906). Matsumura & Hayata in Jour. Coll. Sci. Tokyo, XXII. 218 (Enum. PL Formos.) (1906). Hayata in Jour. Coll. Sci. Tokyo, XXV. art. 19, 153 (Fl. Mont. Formos.) (1908). Matsumura, Ind. PL Jap. II. pt. 2, 463 (1912). Komatsu in Icon. PL Koisikav. II. 69, t. 119 (1914); in Tokyo Bot. Mag. XXXII. [11] (1918). Kanehira, Formos. Trees, 325, fig. 10 (1917). Millais, Rhodod. 220 (1917). Azalea Oldhamii Hort. Veitch. apud Masters in Gard. Chron. n. ser. XVII. 524 (1882). Rhododendron Oldhamii var. glandulosum Hayata in Jour. Coll. Sci. Tokyo XXV. art. 19, 153 (Fl. Mont. Formos.) (1908). Rhododendron longiperulatum Hayata, Icon. PL Formos. III. 138 (1913).- Kanehira, Formos. Trees, 323, fig. 13 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [14] (1918). Formosa: prov. Taihoku, Jyukirin, sea-level to 1000m., April 8, 1918, E. H. Wilson (No. 10,316); Tamsui, 1864, R. Oldham (No. 212, Herb. Kew, Herb. Gray); Sekitei, alt. 100-500 m. March 17, 1918, E. H. Wilson (No. 10,165); prov. Nanto, round Lake Candidius, March 11, December 7, 1918, E. H. Wilson (Nos. 9980, 11,175); be- yond'Musha up to 2900 m. March 3, 1918, E. H. Wilson (No. 10,029); prov. Kagi, beyond Ari-san, alt. 2800-2900 m. February 4, 1918, E. H. Wilson (No. 9778); prov. Koshun, South Cape, Schmuser ex A. Henry (Nos. 588, in part, 588A. Herb. Kew); prov. unknown, " Okasaki," March, 1914, U. Faurie (No. 164, Herb. Bur. Sci. Ma- nila); without locality ex Herb. Imp. Bot. Gard. Tokyo (Herb. Bur. Sci. Manila). This is the common red-flowered Azalea of Formosa, where it is endemic and widely spread from sea-level up to 2800 m. altitude but is more abundant in the north than elsewhere. On volcanic Mt. Daiton, not far from the capital city of Taihoku, it is a feature on the wind-swept upper slopes. On sandstone cliffs east of Taihoku it is common, and in thickets round Lake Candidius in central Formosa it is a conspicuous shrub. On the grass-clad highlands beyond Musha and beyond Ari-san it is plentiful. I did not see it in the extreme south, which is of coral forma- tion, and suspect that Schmiiser's specimen cited really came from the savage territory more to the north. The species is very glandular and very hairy, the leaves shoots, pedicels and calyx being covered with a soft pubescence much of which is long, villose and ferrugineous in color but on wind-swept mountains in winter the pubescence may be bleached pale gray and the plant shaggy in appear- ance. It is a much-branched shrub from 1 to 3 m. tall, bushy in habit with twiggy branchlets densely clothed with glandular, spreading, red-brown hairs among which are scattered flattened, paleaceous hairs spreading and less appressed than is the rule in other red-flowered Azaleas. The leaves are persistent, and vary in shape from elliptic or elliptic-oblong to elliptic-ovate or very rarely to lanceolate; they are from 2 to 8 cm. long and from 1 to 4 cm. wide, rounded or subacute at the 68 THE AZALEAS OF THE OLD WORLD apex which is furnished with a long gland-tipped mucro; the base is narrowed. Both surfaces and the petiole are clothed with pubescence similar to that on the shoots; on mature leaves the upper surface is scaberulous from the presence of the tumid persistent base of deciduous hairs; the veins are impressed above, raised and prominently reticulate below. The flowers are borne at the end of the shoots, and at low altitudes appear every month in the year. The corolla is red, funnel-shape, with spreading, rounded lobes, and from 3 to 5 cm. in diameter. The calyx is green and variable in size, and the lobes may be rounded and less than 1 mm. long or triangular-acuminate and as much as 8 mm. long. The ten stamens are of unequal length, the longest almost equalling the corolla but overtopped by the slender, glabrous, exserted style. The fruit is more or less erect, ovoid, from 6 to 10 mm. long, densely glandular-setose, on pedicels from 1 to 1.5 cm. long. This species varies considerably in size of the leaves and calyx-lobes and some- what in degree of hairiness and glandulosity but is well marked and easily recog- nized among all the species of its section. At its altitudinal limits and on bleak mountain-slopes the leaves, which are scattered on the free shoots and clustered at the end of the branchlets, are tinted blackish purple. It was discovered by Rich- ard Oldham round Tamsui in 1864, and introduced to gardens by Charles Maries who sent seeds to Messers Veitch in 1878. Plants raised from these seeds were exhibited in London in the spring of 1882. Whether plants of this origin are still in cultivation I do not know. In 1918 I sent seeds to the Arnold Arboretum which were distributed in America and Europe. As a greenhouse plant Oldham's Azalea is worth cultivating. In Tokyo I examined the type specimen of R. longiperulatum Hayata and could find no characters by which to separate it from R. Oldhamii. On Mt. Daiton, the type locality given by Hayata for his species, the variable R. Oldhamii is abundant but 1 could find no other species of the group growing there. Mr. Shimada who accompanied me told me it was all R. longiperulatum Hayata so-named from his original specimen collected on this mountain. Rhododendron mucronatum G. Don, Gen. Syst. III. 846 (1834). Azalea rosmarinifolia Burmann, Fl. Ind. 43, t. 3, fig. 3 (1768). Rehder in Bailey, Cycl. Am. Hort. I. 123 (1900). Azalea mucronata Blume, Cat. Gewass. Buitenz. 44 (1823); Bijdr. Fl. Nederl. 853 (1825). Azalea indica alba Lindley, Bot. Reg. X. t. 811 (1824). Loddiges, Bot. Cab. XIII. t. 1253 (1827). Azalea ledifolia Hooker in Bot. Mag. LVI. t. 2901 (1829). Hovey, Mag. Hort. IV. 135 (1838). Azalea liliiflora Poiteau in Annal. de From. 104 (1829). Loiseleur, Herb. Amat. se*r. 2, I. t. 40 (1839). Carriere in Rev. Hort. 1869, 299. Azalea alba Sweet apud Poiteau in Annal. de From. 104 (1829), as a synonym. Rhododendron leucanthum Bunge in Mem. Soc. tr. Acad. Sci. St. Petersbourg, II. 115 (Enum. PL Chin. Bor. 41) (1833). Rhododendron ledifolium G. Don, Gen. Syst. 846 (1834). Siebold and Zuc- carini in Abh. Akad. Munch. IV. pt. III. 131 (Fl Jap. Fam. Nat. II. 7) (1846). Planchon in Rev. Hort. 1854, 48. Boissier, Herb. Boiss. V. 918 (1897). Bean, Trees and Shrubs Brit. Isl. II. 365 (1914). Rehder in Bailey, Stand. Cycl. Hort. V. 2943 (1916). Millais, Rhodod. 201, t. (1917). Rhododendron Burmanni G. Don, Gen. Syst. III. 846 (1834). Rhododendron macranthum var. albiflorum Souchetianum Lemaire in Loiseleur, Herb. Amat. se*r. 2, III. t. 15 (1843). ENUMERATION OF THE SPECIES 69 Rhododendron liliiflorum grandiflorum souchetianum Lemaire in Loiseleur, Herb. Amat. ser. 2, III. sub. t. 15 (1843), as a synonym. Rhododendron ledifolium a. leucanthum De Candolle, Prodr. VII. pt. 2, 727 (1839). Miquel in Ann. Mus. Lugd.-Bat. I. 34 (1863); II. 164 (1865-66); Prol. Fl. Jap. 96 (1866-67). Maximowicz in Mem. Acad. Sci. St. Peters- bourg, ser. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PI. Jap. I. 291 (1875). Matsumura, Ind. PL Jap. II. pt. 2, 462 (1912). Rhododendron rosmarinifolium Dippel, Handb. Laubholzk. I. 421 (1889), not Vidal. Voigtlander in Gartenwelt XV. 650, fig. (1911). Schneider III. Handb. Laubholzk. II. 503, figs. 330 f., 331 c-e (1911). Komatsu in Tokyo Bot. Mag. XXXII. [13] (1918). Azalea rosmarinifolia var. alba Render in Bailey, Cycl. Am. Hort. 1. 123 (1900). Rhododendron rosmarinifolium var. album Schneider, III. Handb. Laubholzk. II. 504 (1911). Rhododendron ledifolium var. album Render in Mitt. DeutschDendr. (?es.XXIV. 225 (1916). Japan: Kyushu, prov. Hizen, Nagasaki, cultivated, April 25, 1903, U. Faurie (No. 5434); prov. Chikugo, Kurume, cultivated, May 3, 1918, E. H. Wilson. Hondo, prov. Choshu, Shimonoseki, cultivated, garden Sanyo Hotel, May 17, 1917, E. H. Wilson (No. 8426); prov. Settsu, near Osaka, cultivated, May 7, 1918, E. H. Wilson; prov. Kishu, May 15, 1911, K. Sakurai; prov. Musashi, Yokohama, culti- vated, 1862, C. Maximowicz (Herb. Kew); Tokyo, cultivated, March 29, 1914, E. H. Wilson (No. 6339); same locality, cultivated, May 15, 1882, K. Miyabe; prov. Ugo, Akita, cultivated, May 20, 1888, U. Faurie (No. 2133); prov. Mutsu, Hirosaki, May 27, 1905, U. Faurie (No. 6786). Hokkaido, prov. Oshima, Hakodate, cultivated, 1855, C. Wright (Herb. Gray, Herb. Kew); same locality, 1862, M. Albrecht (Herb. Kew). Japan: without locality, 1862-63, R. Oldham (No. 506); ex Herb. Lugd.-Bat., 1863 (Herb. Gray, Herb. Kew). China: prov. Hupeh, Ichang, cultivated, A. Henry (No. 3503, Herb. Gray). Cultivated: ex Herb. Bot. Bog. (Herb. Bur. Sci. Manila); Forest Hills Cemetery, Boston, Mass., June 9, 1906, J. G. Jack; Butler Planta- tion, West Feliciana Parish, La., March 28, 1910, C. S. Sargent; Hort. Berckmans, Augusta, Georgia, March 18, 1909, C. S. Sargent; Hort. Kew, May 28, 1901, A. Rehder. This well-known plant is a much-branched compact, wide-spreading shrub usu- ally from 1.5 to 2 m. tall but sometimes over 3 m. high and more in diameter. The shoots in their first year, as well as the petioles and pedicels, are densely clothed with gray to gray-brown, spreading, soft hairs and intermixed with them, espe- cially on the shoots, are few or many appressed and spreading, flattened, bristle-like hairs. On both shoots and pedicels occasional glandular hairs are present; the inner 70 THE AZALEAS OF THE OLD WORLD scales of the flower-buds are also glandular. The spring leaves are light green, deciduous and rather thin in texture, they are lanceolate to ovate-lanceolate, from 3 to 6 cm. long and from 1 to 2.5 cm. wide, acute or obtuse, mucronulate, clothed on both surfaces witlj soft, straight, appressed, gray to rufous hairs; the summer leaves are dark green, numerous, persistent and sub-coriaceous, oblong-lanceolate to oblanceolate-oblong, from 1 to 4 cm. long and from 0.6 to 1.2 cm. wide, obtuse or rounded, mucronate, somewhat rugulose above, prominently veined on the lower surface, their margins more or less recurved; both surfaces are clothed with soft, straight, appressed, gray to rufous hairs, and often mixed with these are spreading, glandular, gray hairs. The flowers are in clusters of one or more at the end of the branch, fragrant, pure white; occasionally an odd branch may have rose-colored flowers, or flowers striped with rose. The calyx is green and ample, the calyx-lobes are erect, lance-shaped, about 1 cm. long, often somewhat erose and more or less densely glandular-pubescent. The corolla is wide-funnel shape from 4 to 5 cm. long and broad; the stamens are usually 10 (occasionally 8) as long as the corolla or nearly so, and are exceeded by the pistil which is slightly exserted. The fruit is conic, with appressed setose hairs, and is shorter than the persistent calyx-lobes which surround it. This Azalea is a favorite plant in Japan and has a place in almost every garden more especially from Tokyo southward, but its wild prototype was discovered only quite recently. 1 It is the " Jedogawa-tsutsuji with white flowers " mentioned by Kaempfer (Amoen. Exot. 848 [1712]), and by Thunberg was included under his Azalea indica as " Yedogawa Satsuke." It was cultivated in Java in the days of Burmann and without doubt was brought there from Nagasaki, Japan, by Dutch trading ships. In China it is grown in the gardens of the wealthy in many parts of the Empire, but how it got there we do not know. According to Lindley (Bot. Reg. X. t. 811 [1824]) it was first introduced into England to the nurseries of Mr. Brookes at Ball's Pond through his collector Joseph Poole who sent it from China in 1819. Like all introductions of this period it came, in all probability, either from Canton or Macao. Probably there were other importations from China in early times and during the last thirty years hundreds of thousands of plants have been sent from Japan to America and Europe. Its fragrant, white flowers and good constitution have made it a favorite and common plant in the greenhouses of the west. In fairly recent years it has been found to be more hardy than was once supposed, and it is now quite frequently seen as an outdoor garden shrub. I have not been able to ascertain the exact date of its introduction into America, but Hovey tells us that it was growing in the garden of J. P. Gushing, Belmont Place, Watertown, Massachusetts, in February, 1838. It is scarcely hardy near Boston but on Long Island, New York, and southward it is perfectly hardy. In Japan it is known as the " Shiro-yodogawa," that is " White Yodogawa." I am familiar with this plant in the gardens of Japan and China as well as in those of Europe and America and in addition to the material enumerated I have seen the many old specimens preserved in Herb. Kew and Herb. Gray, which include several sent from Java many years ago. Every specimen in herbaria that I have seen is definitely stated to have come from cultivated plants. That it is an albino form of some species has long been suspected and it is satisfactory to be able to identify it at last. After an exhaustive study I am of the definite opinion that this old favorite garden plant is a white-flowered form of a plant wild near the banks of rivers on the island of Shikoku, which has lilac-purple flowers and to which Makino (in Tokyo Bot. Mag. XXII. 55 (1908) has given the name R. ripense. 1 Millais states that " Carles collected it on the Seoul mountains in Korea.' This is incorrect; the plant is R. yedoense var. poukhanense Nakai. ENUMERATION OF THE SPECIES 71 The specific name here employed to designate this species is the oldest valid name but it seems to have been quite overlooked. Burmann named the plant Azalea rosmarinifolia in 1768. This is the oldest name but unfortunately it was neglected for over a century and became invalidated under Rhododendron by the R. rosmarinifolium Vidal, a Philippine species, some years before it was taken up for Burmann's plant by Dippel. A form with larger flowers than the type is : Rhododendron mucronatum f. Noordtianum Wilson, n. comb. Azalea japonica alba grandiflora van Noordt Pynaert in Rev. Hort. Belg. XXX. 133, t. (1904) . Reuter in Mailer's Deutsch. Gdrtn. Zeit. XX. 73, fig. (1905) . Azalea ledifolia var. Noordtiana Wittmack in Gartenfi. LIX. 310, fig. 32 (1910). Rhododendron ledifolium var. Noordtianum Render in Mitt. Deutsch. Ges. XXIV. 225 (1916). Rhododendron rosmarinifolium f . Oriukiu Komatsu in Tokyo Bot. Mag. XXXII. [35] (1918). Japan: Hondo, prov. Kawachi, Ikeda, cultivated, May 8, 1918, E. H. Wilson. Pynaert states that this form was raised in the nursery of P. van Noordt & Sons, Boskoop, from seeds collected from some plants introduced from Japan. I can find no difference between it and the form long cultivated in Japan under the name of 0-riukiu-tsutsuji (large Liukiu Azalea). It is considered in Europe to be a very hardy form of the type. Very similar to the last-named form is: Rhododendron mucronatum "Mattapan" Wilson, n. f. Azalea "Mattapan" C. M. Hovey, Mag. Hort. XXVII. 190 (1861). This plant was exhibited for the first time before the Massachusetts Horticul- tural Society on March 23, 1861, by C. M. Hovey, and described as " a new seed- ling, flowers white with a yellowish blotch on the upper petals which are fringed in the way of A. crispiflora." This handsome form is still in the Holm Lea collec- tion and in other gardens round Boston, Mass. In some gardens this Azalea is known by the name of "America." A form with double white flowers is : Rhododendron mucronatum f . narcissiflorum Wilson, n. comb. Rhododendron narcissiflorum Planchon in Fl. des Serr. IX. 82 (1854) ; in Rev. Hort. 1854, 67. Hemsley in Jour. Linn. Soc. XXVI. 27 (1889). Millais, Rhodod. 216 (1917). Azalea narcissiflora Fortune apud Planchon in Fl. des Serr. IX. 82 (1854), as a synonym. Hovey, Mag. Hort. XXII. 9 (1856). Rhododendron ledifolium 7. narcissiflorum Maximowicz in Mem. Acad. Sci. St. Petersburg, ser. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870), in part. Franchet & Savatier, Enum. PL Jap. I. 291 (1875). Matsumura, Ind. PI. Jap. II. pt. 2, 462 (1912). Bean, Trees and Shrubs Brit. Isl. II. 366 (1914). Render in Bailey, Stand. Cycl. Hort. V. 2944 (1916). Azalea rosmarinifolia var. narcissiflora Render in Bailey, Cycl. Am. Hort. 123 (1900). 72 THE AZALEAS OF THE OLD WORLD Rhododendron rosmarinifolium var. narcissifiorum Schneider, III. Handb. Laubholzk. II. 504 (1911). Rhododendron rosmarinifolium f. narcissiflorum a. Shiromanyo Komatsu in Tokyo Bot. Mag. XXXII. [36] (1918). Japan: Kyushu, prov. Chikugo, Kurume, cultivated, May 3, 1918, E. H. Wilson. Under the name of " Shiro-manyo-tsutsuji " this double-flowered form has been cultivated in Japan for a long time probably centuries. It was introduced into England by Robert Fortune who sent plants from China to Messers Standish & Noble about 1850 or 1851. It was imported into this country by C. M. Hovey in 1855. Hovey (Mag. Hort. V. 25 [1839]) speaks of an Azalea ledifolia alba pleno having been introduced into Boston in 1838. The name suggests that the form nar- cissiflorum may have been in gardens prior to Fortune's introduction but I have been unable to find any further reference to the plant mentioned by Hovey. Mil- lais suggests this may be a hybrid R. indicum x R. linearifolium ? but it is difficult to understand why it should be regarded as anything but the common R. mucronatum G. Don (Azalea ledifolia Hook.) with double flowers. The habit, fragrance, calyx and pubescence are identical. Maximowicz refers the double rose- colored form (f . plenum Wils.) to this f . narcissiflorum and Millais does the same but for horticultural purposes it is best kept distinct. A fine garden form of the type is: Rhododendron mucronatum f. sekidera Wilson, n. comb. Rhododendron rosmarinifolium f. Sekidera Komatsu in Tokyo Bot. Mag. XXXII. [35] (1918). Rhododendron Sekidera (Rhod. indicum X Rhod. Kaempferi) Komatsu in Tokyo Bot. Mag. XXXII. [46] (1918). Azalea magniflca Hort. Yokohama Nursery Co. Cat. 1918-19, p. 19. Azalea indica roseum Hort. Andorra Nurseries, May, 1919. Azalea japonica grandiflora superba Hort. Andorra Nurseries, May, 1919. Azalea " Damask Rose " Hort. Andorra Nurseries, May, 1919. Japan: prov. Musashi, cultivated, May 21, 1914, E. H. Wilson (No. 6670). This very handsome plant I saw cultivated in the grounds of the Yokohama Nursery Company. The flowers are as large as those of f. Noordtianum, pure white with the dorsal lobes spotted and splashed with rose, occasionally a flower is rose-colored; the stamens vary in number from 6 to 10 but are usually 10. Under the names Azalea indica roseum and Azalea "Damask Rose" slightly inferior forms were sent to us for determination last year. The true variety is cultivated by Mr. John S. Ames, North Easton, Mass., who imported it in 1917. In 1919 I brought back plants from Japan to the Arnold Arboretum. This most desirable addition to our gardens will probably prove to be as hardy as the typical white- flowered R. mucronatum G. Don. A spontaneous variety and undoubtedly the phylogenetic type of the species is : Rhododendron mucronatum var. ripense Wilson, n. comb. Rhododendron ripense Makino in Tokyo Bot. Mag. XXII. 55 (1908). Schnei- der, III. Handb. Laubholzk. II. 503 (1911). Komatsu in Icon. PI. Koisi- kav. II. 71, t. 120 (1914); in Tokyo Bot. Mag. XXXII. [12] (1918). ENUMERATION OF THE SPECIES 73 Japan: Kyushu, prov. Hizen, Nagasaki, common in gardens, March 19, 1914, E. H. Wilson (No. 6316); prov. Chikugo, Kurume, cultivated, May 3, 1918, E. H. Wilson. Cultivated: Garden of Butler Plantation, West Feliciana Parish, Louisiana, March 28, 1910, C. S. Sargent. I have examined the type of Makino's R. ripense and can find no character by which it differs from R. mucronatum G. Don, except in the rose-purple color of its flowers, and as is well known a branch bearing colored flowers is often found in Don's species. Nevertheless, Makino's discovery is most interesting since it enables us to place definitely the habitat and parentage of R. mucronatum G. Don which has remained obscure for over a century and a half. Makino gives two localities and says that it is found wild by the side of rivers in Tosa and lyo prov- inces in Shikoku, and also that it is frequently cultivated. I have seen it only as a cultivated plant in Japan where it is not by any means such a general favorite as the pure white form. I saw more plants round Nagasaki than elsewhere. At Kurume it is known as the " Momo-iro-yodogawa." Makino gives the vernacular name in Tosa province as " Iso-tsutsuji." There is in cultivation in this country an intermediate and very lovely form in which the corolla is white, delicately flushed with pale lilac-purple and has a few very faint rose-pink spots. As a garden plant it is distinct and desirable and it may be called: R. mucronatum f. amethystinum Wilson, n. f. Cultivated: Hort. Farquhar & Co., Boston, Mass., April 1, 1909, (type). Hort. Andorra Nurseries, Chestnut Hill, Philadelphia, Pa., May 20, 1920, as Azalea japonica alba. Rhododendron mucronatum f. plenum Wilson, n. comb. Azalea indica var. /S. plena Suns in Bot. Mag. LI. t. 2509 (1824). Azalea indica 7. purpurea Sweet, Hort. Brit. 264 (1827). Azalea prolifera Poiteau in Annal. de From. I. 104 (1829). Rhododendron phoeniceum var. p.flore pleno G. Don, Gen. Syst. III. 846 (1834). Azalea purpurea pleno Hovey, Mag. Hort. V. 115 (1839). Rhododendron indicum f . incamatum De Candolle, Prodr. 726 (1839). Rhododendron ledifolium 7. nartissiflorum Maximowicz in Mim. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9, 36 (Rhodod. As. Or.) (1870), in part. Rhododendron ledifolium var. plena purpurea Regel in Gartenfl. XXXV. 565, t. 1233, fig. c-d (1886). Azalea indica purpurea plena Bretschneider, Hist. European Bot. Disc. Chin. 222 (1898). Rhododendron rosmarinifolium f. narcissiflorum /3. Fujimanyo Komatsu in Tokyo Bot. Mag. XXXII. [36] (1918). Japan: Hondo, prov. Kawachi, Ikeda, cultivated, May 8, 1918, E. H. Wilson. Cultivated: Hort. Holm Lea, Brookline, Mass. Round Tokyo and Yokohama this plant is known as Fujimanyo; round Osaka it is the Murasaki-botan-tsutsuji (purple Paeony Azalea). It is an old plant even 74 THE AZALEAS OF THE OLD WORLD in western gardens, having been sent by Joseph Poole from China to Samuel Brookes, nurseryman, Ball's Pond, Newington Green, England, in 1819. The flower is rose- purple, fragrant, quite, or only partially double, and with or without green rudi- mentary leaves showing in the centre of the flower. Hovey tells us that this plant was cultivated by Mrs. Hibbert and exhibited before the Pennsylvania Horticul- tural Society on February 19, 1839. Rhododendron linearifolium Siebold & Zuccarini in Abh. Akad. Munch. IV. pt. III. 131 (Fl. Jap. Nat. II. 7) (1846). Miquel in Ann. Mus. Lugd.-Bat. I. 34 (1863); II. 165 (1864-66); Prol. FL Jap. 97 (1866-67). Maximowicz in Mem. Acad. Sci. Petersbourg, se*r. 7, XVI. No. 9, 34 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 290 (1875). Boissier in Bull. Herb. Boiss. V. 918 (1897). Schneider, III. Handb. Laubholzk. II. 504, figs. 330 g-k, 331 f. (1911). Matsumura, Ind. PL Jap. II. pt. 2, 463 (1912). Bean, Trees and Shrubs Brit. Isl. II. 366 (1914). Millais, Rhodod. 203 (1917). Azalea linearifolia Hooker f. in Bot. Mag. XCV. t. 5769 (1869). Rhododendron macrosepalum var. linearifolium Makino in Tokyo Bot. Mag. XXII. 55 (1908). Komatsu in Tokyo Bot. Mag. XXXII. [34] (1918). Rhododendron linearifolium a. linearifolium Makino in Tokyo Bot. Mag. XXVII. 108 (1913). Cultivated: November, 1867, February, 1869, ex Hort. Standish (Herb. Kew); 1876, ex Hort. Hunnewell (Herb. Gray). This is a monstrous garden form cultivated in Japan but rather rare. The leaves are narrow, strap-shaped, and the corolla is divided into narrow, ligulate segments. An analogous plant is R. indicum f . polypetalum Wils. Maximowicz says it was in- troduced into the gardens of Tokyo from the alpine regions of Sendai in north- ern Japan. This is a mistake. It is unknown in a wild state and was first known from the neighborhood of Nagasaki in Kyushu the southern island of Japan. Its Japanese name is " Seigai-tsutsuji." It was introduced into England by Messers Standish with whom it flowered in November, 1867, and again in February, 1869. It is well figured in the Botanical Magazine. From England it appears to have been sent to this country for it flowered in the garden of Mr. H. H. Hunnewell, Wellesley, Mass., in 1876. According to Bean it is quite hardy at Kew. It is not hardy in the Arnold Arboretum and I do not know whether it is now cultivated in America. Rhododendron linearifolium var. macrosepalum Makino in Tokyo Bot. Mag. XXVII. 108 (1913). Rhododendron macrosepalum Maximowicz in Gartenfl. XIX. 258, t. 662 (1870) ; in Mem. Acad. Sci. St. Petersburg, s6r. 7, XVI. No. 9, 31 (Rhodod. As. Or.) (1870) ; in Bull. Acad. St. Petersburg, se*r. III. XV. 227 (Mel Biol VII. 335) (1871). Franchet & Savatier, Enum. PL Jap. I. 290 (1875). Dippel, Handb. Laubholzk. I. 420, fig. 271 (1889). Makino in Tokyo Bot. Mag. IV. 151 (1890); XXII. 55 (1908). Boissier in Bull Herb. Boiss. V. 918 (1897). Schneider, III. Handb. Laubholzk. II. 503, figs. 330 e, 331 a-b ENUMERATION OF THE SPECIES 75 (1911). Matsumura, Ind. PL Jap. II. pt. 2, 463 (1912). Millais, Rho- dod. 205 (1917). Komatsu in Tokyo BoL Mag. XXXII. [13] (1918). Azalea macrosepala K. Koch, Dendr. II. pt. 1, 180 (1872). O. Kuntze, Rev. Gen. II. 387 (1891). Rhododendron linearifolium /3. macrosepalum forma a. genuinum Makino in Tokyo BoL Mag. XXVII. 108 (1913). Japan: Kyushu, prov. Hizen, roadside, Nagasaki to Mogi, No- vember 12, 1903, C. S. Sargent. Shikoku, prov. Tosa, thickets up to 300 m. alt. November 20, 1914, E. H. Wilson (No. 7794). Hondo, prov. Settsu, near Kobe, April, 1875, H. N. Moseley (Herb. Kew); prov. Kawachi, Yomomoto and Ikeda, May 8, 1918, E. H. Wilson (No. 10,355); prov. Yamato, Koyasan, August 26, 1907, U. Faurie (No. 124); Yoshino, April 22, 1914, E. H. Wilson (No. 6574, in part); prov. Yamashiro, Kyoto, Temple Garden, October 16, 1892, C. S. Sargent; Kyoto, J. Rein, ex Mus. Bot. Berol. (Herb. Bur. Sci. Manila) ; prov. Omi, mountains round Lake Biwa, April 30, 1892, U. Faurie (No. 7820, Herb. Kew) ; prov. Mino, Pine-barrens near Gifu, October 20, 1892, C. S. Sargent; prov. Mikawa, Pine-woods, Futagawa, May 9, 1918, E. H. Wilson (Nos. 10,346, 10,348, 10,349); prov. Shimotsuke " alpe Niko," 1866-74, L. Savatier (No. 766, Herb. Kew). This is the Mochi-tsutsuji (Glandular Azalea) of Japan, where it is common in central Hondo at low altitudes from near Osaka north to Hamamatsu in Totomi province; also in Tosa province, Shikoku. It is partial to gravelly soil and dry situations such as Pine- woods and open thickets afford. A social plant it occurs in great numbers either alone or with R. obtusum var. Kaempferi Wils. At Futa- gawa, for instance, in Pine-woods these two species grow in thousands and in early May when in full blossom they are conspicuous from the railway carriage windows. The specimen from Nagasaki is probably from an escaped plant and there is no other record of its growing in Kyushu. North of Totomi province I have not seen it wild, and I think there is some mistake about it being indigenous in the Nikko region, Maximowicz's statements and Savatier's specimens notwithstanding. It is not mentioned by Matsumura in his Plants of Nikko nor by Hayata in his Vege- tation of Mt. Fuji. As usually seen R. linearifolium var. macrosepalum is a low, rather laxly branched shrub spreading in habit and less than one metre high but in Shikoku I saw it full three metres tall. It has the usual dimorphic leaves and these are mostly deciduous. The shoots are terete and with the petioles are densely clothed with gray to gray-brown pilose (some of them gland-tipped) hairs and a few flattened, spreading bristles which disappear in the second season; the winter-buds are ovoid, acute, densely covered with strigose, yellow-brown hairs. The leaves are mem- branous, scattered on the free-growing shoots and crowded on others, short- petioled, lanceolate-ovate to ovate-elliptic, sometimes lanceolate or ovate, short acuminate, acute or merely mucronulate; those immediately beneath the winter- buds oblanceolate to oblanceolate-oblong obtuse, or rounded, usually mucronulate, the under surface clothed with soft, spreading, gray to gray-brown glandular hairs, the upper surface sparsely glandular hairy, slightly rugulose and hispid; dull green during summer changing to rich vinous purple in the autumn. The flowers are 76 THE AZALEAS OF THE OLD WORLD fragrant, from 2 to 10 in terminal umbels, very variable in size and form. The pedicels are from 2 to 4 cm. long, straight or spreading, densely glandular and pilose. The calyx is green, glandular-pubescent and very variable in size, the strap-shaped segments being from 1 to 3 cm. long and occasionally exceeding the rose-pink to red-purple corolla, which is broad funnel-shape, sometimes 5 cm. but often not more than 3 cm. across; it is glabrous without, and normally larger than the calyx. Stamens normally 5, but at Ikeda I gathered specimens with from 6 to 10 stamens; they are shorter than the pistil and included within the corolla except in abnormally small flowers. The fruit is narrow-ovoid, about 1 cm. long, glandu- lar-pubescent, purplish-brown in color and subtended by the persistent calyx the lobes of which are reflexed, brown without and wine-purple on the inside. There can be no question of this being the phylogenetic type of the species and it is B unfortunate that it should have to rank as a variety. Even in a wild state it is a very variable plant, and a number of forms have been detected and are cultivated in Japanese gardens; these are dealt with hereafter. The calyx, pedicels and inner- most scales of the flower-buds are always glandular, usually densely so and com- monly sufficiently viscid to hold gnats and small flies captive, but this viscosity like everything else in this plant is variable. The flowers are pleasantly fragrant and some of the color forms are very handsome. When the flowers have a com- paratively small corolla and very long calyx-lobes the effect is curious. This Azalea was introduced into Petrograd by Maximowicz who sent seeds in 1863. It flowered for the first time in 1870 and is well figured in Gartenflora. Maximowicz claims that the seeds and his two plants came from the Nikko alps but if this is not a mistake it is strange that no other collector has seen the plant in this locality which is fully two hundred miles north of where it has been seen wild by subse- quent observers. Further, no one else has found it in alpine regions anywhere in Japan. However, there is no doubt that the plant under discussion is Maxi- mowicz's species. What happened to the original introduction I do not know but neither Bean nor Millais mention its cultivation in England. In 1914 I sent to the Arnold Arboretum seeds of this plant from Shikoku some of which were sent to England. Plants raised from this seed in the Arnold Arboretum have not proved very hardy. A form with double flowers is: Rhododendron linearifolium var. macrosepalum f. dianthiflorum Wilson, n. comb. Azalea dianthiflora Carriere in Rev. Hort. 1889, 391; 1891, 60, fig. 18, t. Rhododendron dianthiflorum Millais, Rhodod. 155 (1917). Japan: Hondo, prov. Mikawa, Pine-woods, Futagawa, May 9, 1918, E. H. Wilson (No. 10,350). Cultivated: ex Hort. James Veitch & Sons, June 28, 1904 (Herb. Kew). 1 have seen at Futagawa one or two plants growing wild with double flowers. The pedicels are more spreading than in the type and the calyx-lobes are variable in length. Carriere's figures give a good idea of this form except that the color is brighter and more pleasing than the plate shows. According to Carriere it was introduced into France from Japan in 1889 by M. Wiesener, of Fontenay-aux- Roses. The specimen in Herb. Kew shows that it was cultivated by Messers Veitch in 1904. Bean does not mention it and Millais's account is abbreviated from Car- riere's description. 1 have no knowledge of its being in the gardens of this country. ENUMERATION OF THE SPECIES 77 Another spontaneous form is: Rhododendron linearifolium var. macrosepalum f . decandrum Wil- son, n. f. Japan: Hondo, prov. Kawachi, Yomomoto, roadside, Pine-woods, May 8, 1918, E. H. Wilson (No. 10,355a, type); Ikeda, cultivated, May 8, 1918, E. H. Wilson; prov. Yamato, near Yoshino, dry woods, April 22, 1914, (No. 6574, in part). This differs from the type in having 10 stamens; the Ikeda specimens have from 6 to 10 stamens and that from Yoshino (a fragment) has 6 only. The typical var, macrosepalum is common round Yomomoto where it is known as the Yama-tsutsuji (Mountain Azalea) and growing with it I found occasional plants with 10 or from 6 to 10 stamens in each flower. I have seen none with from 5 to 10 stamens and I do not remember whether the fragment from near Yoshino came from a bush of the ordinary 5-stamen type. An anomalous form is : Rhododendron linearifolium var. macrosepalum f. rhodoroides Makino in Tokyo Bot. Mag. XXVII. 109 (1913). Rhododendron macrosepalum /3. rhodoroides Maximowicz in Mem. Acad. Set. St. Petersburg, ser. 7, XVI. No. 9, 31 (Rhodod. As. Or.) (1870). Fran- chet & Savatier, Enum. PI. Jap. I. 290 (1875). Matsumura, Ind. PI Jap. II.pt. 2, 463 (1912). Komatsuin Tokyo Bot. Mag. XXXII. [34] (1918). Rhododendron ledifolium var. Kochozoroi Komatsu apud Makino in Tokyo Bot. Mag. XXVII. 109 (1913), as a synonym. Japan: Hondo, prov. Kawachi, cultivated, Yamomoto and Ikeda, May 8, 19J.8, E. H. Wilson; prov. Musashi, Okbo, cultivated, May 18, 1909, K. Onuma; Hatagaya, cultivated, April 29, 1914, E. H. Wilson (No. 6596). This is a curious form often seen in Japanese gardens. It has a deeply cleft 5-partite corolla only slightly longer than the calyx-lobes, and greenish-white shaded with rose, exserted stamens with pale purple anthers exceeded in length by the style. Maximowicz raised seedlings from this plant and states that in every in- stance they reverted to typical var. macrosepalum. Makino gives " Kocho-zoroi " as the Japanese name of this plant; in the nursery district near Osaka it is known as " Seikan-tsutsuji." Another anomalous form is: Rhododendron linearifolium var. macrosepalum f. hanaguruma Makino in Tokyo Bot. Mag. XXVII. 109 (1913). Rhododendron ledifolium var. leucanthum f. Hanaguruma Komatsu apud Makino in Tokyo Bot. Mag. XXVII. 109 (1913), as a synonym. Rhododendron macrosepalum var. Hanaguruma Makino in Tokyo Bot. Mag. XXVII. 109 (1913), as a synonym. Komatsu in Tokyo Bot. Mag. XXXII. [35] (1918). Rhododendron linearifolium var. macrosepalum f. polypetalum Makino in Tokyo Bot. Mag. XXVII. 109 (1913), as a synonym. 78 THE AZALEAS OF THE OLD WORLD Japan : Hondo, prov. Kawachi, cultivated at Yomomoto and Ikeda, May 8, 1918, E. H. Wilson; prov. Rikuzen, Sendai, cultivated, May 27, 1888, U. Faurie (No. 2300, Herb. Kew). Cultivated : ex Hort. Ball, March, 1879 (Herb. Kew). In this form the corolla is rose-purple in color, deeply cleft and 5-partite, and the stamens are usually shorter than the corolla-lobes which are from 2.5 to 4 cm. long. Makino gives " Hana-guruma " as the Japanese name; in the nursery dis- trict near Osaka it is known as " Oyeyama-tsutsuji." The specimen in Herb. Kew proves that at one time it was in cultivation in England. Komatsu (in Tokyo Bot. Mag. XXXII. [34] [1918]) enumerates three other forms (" Usuyo," "Amaga-shita" and " Suruga-momyo ") of this species, but these are unknown to me. These are all the species of the section Tsutsutsi that are named or of which I have seen sufficient material. There is, however, in this herbarium specimens (ex Herb. Hongkong Nos. 7030, 5612) from Taimo-shan, Kowloon, Kwangtung prov- ince, of a curious Azalea. It has narrow, linear-lanceolate, acuminate leaves and inconspicuous subsessile flowers with the corolla deeply divided into five narrow segments. It is in every way anomalous, and probably a monstrous condition of some species not yet described, or possibly of the widespread R. Simsii Planch. We have, also, a leafy fragment, collected in Kweichou province by Dr. Handel- Mazzetti (No. 221, July 11, [1917]) which probably belongs to a new species re- lated to R. Oldhamii Maxim. SECT. II. SCIADORHODION REHD. & WILS., n. sect. 1 Rhododendron section Tsutsutsi G. Don, Gen. Syst. III. 847 (1834), as to species nos. 30, 31. Render & Wilson in Sargent, PL Wilson. I. 547 (1913), in part. Rhododendron subgenus Tsutsia Planchon in Fl. des Serr. XI. 75 (1854); in Rev. Hort. 1854, 43, as to R. Farrerae. Rhododendron section Azalea Maximowicz in Mem. Acad. Sci. St. P&ersbourg, s&r. 7, XVI. No. 9, 24 (Rhodod. As. Or.) (1870), in part. Azalea subgenus Euazalea K. Koch, Dendr. II. pt. 1, 179 (1872), as to A. reticulata. Rhododendron section Rhodora Dippel, Handb. Laubholzk. I. 418 (1889), in part. Shrubs with rigid, usually verticillate, often ascending-spreading branches and glabrous or villose shoots without bristle-like hairs. Flowers from terminal buds, with leafy shoots from axils of lower scales of the same bud. Leaves deciduous, in whorls of 3 to 5 at the end of the branchlets, on vigorous shoots scattered, not dimorphic, usually rhombic to obovate, rarely ovate, without appressed bristle-like hairs. Corolla rotate-campanulate, rarely rotate-funnel-form; stamens from 6 to 10 (rarely 5 by abortion), unequal, sometimes declinate. Fruit oblong- ovoid to cylindric, rarely conic-ovoid, often furrowed, villose or glandular, never strigose. This section is found only in eastern Asia, where it is represented by 6 species. Rhododendron Farrerae Tate apud Sweet, Brit. Flow. Gard. ser. 2, 1. t. 95 (1831). G. Don, Gen. Syst. III. 846 (1834). De Candolle, Prodr. VII. 725, pt. 2 (1839). Planchon in Fl. des Serr. XIX. 80 (1854); in Rev. Hort. 1854, 64. Maximowicz in Mem. Acad. Sci. St. Petersburg, se>. 7, XVI. No. 9,25 (Rhodod. As. Or.) (1870). Hance in Jour. Linn. Soc. XIII. 110 (1873). Hemsley in Jour. Linn. Soc. 1 Sectio inter sect. Tsutsutsi et Rhodoram medium tenens, a priori indu- mento piloso vel villoso pallido, non strigoso et fusco, foliis semper deciduis satis magnis, corolla rotato vel rotato-campanulata, ovario villoso vel glabro, non stri- goso, a posteriori gemmis terminalibus flores et folia simul proferentibus bene differt, ab utraque sectione foliis in apice ramulorum subverticillatim congestis quasi umbellam vel umbraculum (graece axids, a/ctdSos, unde nomen) formantibus satis magnis plerumque late obovatis vel rhombico-ovatis distat. In Sargent, Plantae Wilsonianae, I. 548 (1913), we suggested that with their whorled leaves the species here enumerated formed a well-marked group easily distinguished from Tsutsutsi, although agreeing in the character of the winter- buds. Further investigation has brought to light other differences and we are of the opinion that it is best to place them in a distinct section. A. R. & E. H. W. 79 80 THE AZALEAS OF THE OLD WORLD XXVI. 23 (1889); in Kew Bull. Misc. Inform. 1907, 245. Dunn & Tutcher in Kew Bull. Misc. Inform, add. ser. X. 155 (Fl. Kwang- tung and Hongk.) (1912). Millais, Rhodod. 163 (1917). Azalea squamata Lindley in Jour. Hart. Soc. Lond. I. 152 (1846); in Bot. Reg> XXXIII. t. 3 (1847). Walpers, Ann. I. 481 (1848-49). Bentham, Fl> Hongk. 201 (1861). Azalea Farrerae K. Koch, Dendr. II. pt. 1, 178 (1872). O. Kuntze, Rev. Gen. pt. 2, 387 (1891). Azalea squamosa O. Kuntze, Rev. Gen. pt. 2, 387 (1891), as a synonym. Rhododendron Farrerae S. typicum Diels in Bot. Jahrb. XXIX. 514 (1900). China: prov. Kwangtung, Loh-fau-san, August 21, 1917, C. 6. Levine (ex Canton Christian College Herb. No. 1533). This species seems to be peculiar to Hongkong and the neighboring islands and to the adjacent parts of Kwangtung province. It is a low, densely branched shrub with short, rigid, shining brown, verticillate branchlets clothed with ap- pressed, straight and villose hairs when young, becoming glabrous and gray in their second year. The leaves are crowded, usually in threes at the end of the branches, subcoriaceous, deciduous or semi-persistent, glabrescent, ovate, from 2 to 3 cm. long and from 1.2 to 2 cm. wide, acute or sub-acute, with a short mucro, and a rounded base; they are dark green above and pallid below, reticulate, with the principal veins impressed above and raised below. The petioles are from 1 to 3 mm. long and densely villose. The winter-buds are small, broadly ovoid, and are densely clothed with gray or rufous pubescence. The flowers appear before the leaves unfold, either solitary or in pairs at the end of the branches. The corolla is pale to deep rose-color with red-purple spots, from 4 to 5 cm. in diameter; it has a short, narrow, funnel-shape tube and spreacling, undulate lobes. The ring-like calyx is minutely 5-toothed, and pubescent; and the pedicel and ovary are thickly clothed with appressed and villose, ferrugineous or gray hairs. The stamens are from 8 to 10 of unequal length and shorter than the corolla, and are equalled or slightly overtopped by the slender, curved style. The fruit is conic-ovoid, from 0.5 to 1.5 cm. long and from 0.8 to 1 cm. in diameter and densely villose, and is borne on a stout, curved, villose pedicel about 1 cm. long. I have seen this Azalea growing wild in Hongkong and as Hemsley has pointed out it is a very distinct species. It is characterized by its small, ovate leaves, its very short villose petioles and by its relatively large fruit. It was first introduced into England in 1829, by Captain Farrer of the East India Company's ship " Or- well." It was reintroduced by Robert Fortune who in 1844 sent it to the garden of the Horticultural Society of London where it flowered and was renamed by Lindley. I have no knowledge of its having been in cultivation in America, and Millais states that it is a rare plant in English gardens, and hardy only in the south. Rhododendron Mariesii Hemsley and Wilson in Kew Bull. Misc. Inform. 1907, 244. Hutchinson in Bot. Mag. CXXXIV. t. 8206 (1908). Schneider, III. Handb. Laubholzk. II. 496, fig. 327 1-m (1909). Render & Wilson in Sargent, PL Wilson. I. 548 (1913). Bean, Trees and Shrubs Brit. I si. II. 377 (1914). Millais, Rhodod. 207 (1917). ENUMERATION OF THE SPECIES 81 Rhododendron Weyrichii Hemsley in Jour. Linn. Soc. XXVI. 32 (1889), not Maximowicz. Rhododendron Farrerae a. leucotrichum Franchet in Jour, de Bot. IX. 394 ( 1895). Diels in Bot. Jahrb. XXIX. 513 (1900). Millais, Rhodod. 163 (1917). Rhododendron Farrerae a. Weyrichii Diels in Bot. Jahrb. XXIX. 513 (1900). Rhododendron Farrerae y. mediocre Diels in Bot. Jahrb. XXIX. 514 (1900). Rhododendron rhombicum Diels in Bot. Jahrb. XXIX. 514 (1900), not Miquel. Rhododendron Farrerae Kawakami, PI. Formos. 64 (1910), not Tate. Rhododendron shojoense Hayata in Jour. Coll. Sci. Tokyo, XXX. art. 1, 174 (1911). Rhododendron gnaphalocarpum Hayata, Icon. PL Formos. III. 132 (1913). Kanehira, Formos. Trees, 327, fig. 14 (1917). China : without locality, 1845, R. Fortune (No. 122 (2), Herb. Kew); prov. Fokien, April to June, 1905, Dunn's Exped. (Hongk. Herb. No. 2882); prov. Chekiang, Taihu Lake, near Huchau, April, 1881, W. R. Carles (No. 117, Herb. Kew); prov. Kiangsi, Kuling, 1878, C. Maries (Herb. Kew); same locality, alt. 1300 m. July 29, 1907, E. H. Wilson (No. 1681) : prov. Hupeh, Changyang Hsien, A. Henry (Nos. 1422, 5274, 594p,Herb. Kew, Herb. Gray) ; Nanto and mountains to northward, A. Henry (No. 3829, Herb. Kew, Herb. Gray); north and south of Ichang, alt. 300-1300 m. May and November, 1907, E. H. Wilson (No. 606, in part); south of Ichang, alt. 800 m. April and October, 1900, E. H. Wilson (Veitch Exped. No. 29. type); Hsing- shan Hsien, alt. 1300, May 12, 1907, E. H. Wilson (No. 606, in part); prov. Szech'uan, Nanch'uan, A. von Rosthorn (No. 2164, Herb. Chris- tiania). Cultivated: ex Hort. Veitch, April, 1907 (Herb. Kew, type of Bot. Mag. t. 8206). This species is distributed from Fokien and Chekiang provinces of eastern China westward to southeastern Szech'uan. On the Lushan mountains round Kuling it is common and north and south of Ichang it is very plentiful on cliffs and in thickets between altitudes of 300 and 1300 m. Its pleasing pink to rose- colored flowers open before the leaves unfold and in late April and May it is a conspicuous wayside shrub. Curiously this and R. molle G. Don are the only deciduous leafed species in the whole of China except R. Farrerae Tate from Hongkong and the adjacent territory. Maries' Azalea is an upright branching shrub from 1 to 3 m. tall, rather narrow, with verticillate, twiggy, ascending branches which are shining, yellow or chestnut-brown during the first year and afterward pale gray. When young the shoots and leaves are covered with yellow- ish, appressed silky hairs which usually fall away early or may in part persist on the shoot through the entire season. The buds are conic, gray-purple, with ciliate bud-scales glabrescent without and villose on the inner surface. The leaves at maturity are chartaceous, dark green above, pallid below, prominently reticulate, with the principal veins slightly impressed on the upper side and raised on the lower side; they vary from 3.5 cm. to 7.5 cm. in length and from 2 to 4 cm. in width, and are always the broadest below the middle. The prevailing shape is 82 THE AZALEAS OF THE OLD WORLD ovate-lanceolate, with an acute mucronate apex and cuneate base, but occasion- ally they are broad-ovate or elliptic. The petioles are from 0.4 to 1.2 cm. in length, glabrescent, purplish, and often slightly glaucescent on the under-side. The flowers are usually in pairs at the end of the naked branchlets, but are sometimes soli- tary, and occasionally are in 4- or 5-flowered clusters; the corolla is from 4 to 5 cm. across, spotted with red-purple on the posterior lobes; it has a short, narrow, funnel-shape tube and deeply cleft, spreading lobes. The pedicels are villose; the minute, 5-lobed calyx and the ovary are clothed with gray or yellow-brown, straight appressed hairs. The 10 stamens are of nearly equal length and as long as the corolla, the filaments are glabrous, curved and the anthers purple. The style is longer than the stamens and is curved upward and has a capitate, slightly lobed stigma. The fruit is cylindric, from 1 to 1.5 cm. long, slightly furrowed, and clothed with gray or yellow-brown, villose pubescence, and is borne on a straight pedicel. From those species with which R. Mariesii has been confused it is readily dis- tinguished by its leaves. The flowers are of a different color from those of the Japanese species, the corolla is differently shaped and there are other distinctions of importance. In Tokyo I saw the types of Hayata's two species cited and could find nothing by which to distinguish them from R. Mariesii. This Chinese plant was discovered by Fortune, probably in Chekiang province, and afterward by Charles Maries on the Lushan mountains behind Kiukiang in the spring of 1878. It was introduced into Kew gardens through seeds sent from the neighborhood of Ichang in 1886 by Augustine Henry; in 1900 I sent seeds to Messers Veitch from south of Ichang. This Azalea flowered for the first time in the Temperate House at Kew in April, 1907. Bean does not consider it to be hardy. I have not heard of its being cultivated in America. Rhododendron Weyrichii Maximowicz in Mem. Acad. Sci. St. Petersbourg, ser. 7, XVI. No. 9, 26, t. 2, fig. 1-6 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PI. Jap. I. 288 (1875). Makino in Tokyo Bot. Mag. XVIII. 48 (1904). Schneider, 111. Handb. Laubholzk. II. 495 (1909), in a note. Matsumura, Ind. PL Jap. II. pt. 2, 464 (1912). Nakai, Rep. Fl. Quelpaert Isl No. 999 (1915); in Tokyo Bot. Mag. XXXI. 243 (1917); Fl. Syl. Kor. VIII. 44, t. 16 (1919). Millais, Rhodod. 261 (1917), excluding the Chinese plant. Komatsu in Tokyo Bot. Mag. XXXII. [10] (1918). Azalea Weyrichii O. Kuntze, Rev. Gen. pt. 2, 387 (1891). Rhododendron shikokianum Makino in Tokyo Bot. Mag. VI. 43 (1892); in IX. Ill (1895). Japan : Shikoku, prov. Tosa, round Kochi, sea-level to 800 m. al- titude, November 17, 1914, E. H. Wilson (No. 7813). Kyushu, prov. Hizen, near Nagasaki, 1862, R. Oldham (No. 511, Herb. Gray). Korea: Quelpaert Island, between Saishu and the Monastery, 400-650 m. October 30, 1917, E. H. Wilson (No. 9411); same island, Hallai-san, October, 1906, May, 1907, U. Faurie (Nos. 661, 1863 in part, 1864); same place, May, 1909, April 14, June 6, 1908, E. Taquet (Nos. 2970, 4678, 108S). ENUMERATION OF THE SPECIES 83 This little known species has a very curious distribution. On the island of Shikoku it is common round Kochi in Tosa province but does not grow on the mainland of Kyushu, but it is found on the adjacent islands of Amakusa and Goto south and west of Nagasaki. I do not know whether it grows on the island of Tsushima in the Japan Sea probably it does but it is common on the Korean island of Quelpaert although unknown on the mainland of Korea. In Shikoku I found it to be fairly plentiful from near sea-level up to an altitude of 800 m. growing in thickets and open woods. On Hallai-san, an extinct volcano on Quel- paert Island, between 200 and 1000 m. altitude, and especially in thickets by the sides of water-courses and on old lava flows, it is common. This Azalea is a vigorous growing shrub from 1 to 5 m. tall, often decidedly tree-like in habit, with moderately stout ascending to spreading often whorled branches. These are yellow-brown, clothed with ferrugineous tomentum when young, becoming glabrous, gray and finally dark purple in color. The winter-buds are characteristic, being conic and sub-acute and clothed with yellow-brown, matted and straight hairs and are unlike those of other species. The leaves are distinctly petiolate, chartaceous, shining green and deciduous, changing to vinous purple in the autumn; on the vigorous shoots the leaves are scattered and in pairs or whorls but on the branchlets they are clustered in whorls of 3, or in pairs; they vary from broad-ovate to semiorbicular or to rhombic, and in size from 3.5 to 8 cm. long and from 2 to 6 cm. wide; they are abruptly subacute or obtuse at the apex which is mucronate, the base is broadly cuneate or rounded. The principal veins are impressed above and slightly raised on the lower surface which is venulose. When young the petioles and both surfaces of the leaf are covered with rufous-brown pilose hairs but these soon disappear. The flowers are from 2 to 4 in terminal umbels, and open before or at the same time as the leaves unfold. The pedicels, calyx and ovary are densely clad with rufous-brown pilose hairs. The corolla is red, almost brick-red, from 3.5 to 6 cm. across, funnel- shape, with a rather narrow expanding tube and spreading lobes. The calyx is merely a ring with five minute teeth. The stamens vary from 6 to 10, and are of unequal length, and like the curved style and its capitate stigma are in- cluded. The fruit is cylindric or oblong-ovoid, from 1.5 to 2 cm. long, oblique, furrowed and obtusely angled and clothed with appressed, villose pubescence. This species was discovered on the Goto Islands in 1853 by Dr. Heinrich Wey- rich, a surgeon on the Russian warship " Vostok " which formed part of Admiral Putiatin's fleet. On Oldham's specimen the locality cited is Nagasaki but this is a mistake; it probably came from the Amakusa Islands south by west of that city. Makino first found it in Shikoku and Pere Faurie on Quelpaert Island. It was unknown in gardens until 1914 when I sent the seeds from Shikoku to the Arnold Arboretum which were distributed in America and Europe. Plants raised in this Arboretum have not proved hardy but in another Massachusetts garden where they have received the protection of a cold frame they have flowered. In 1917 I collected seeds on Quelpaert and hope that they will produce a hardier type. I have never seen this Rhododendron in flower in a wild state and only a late flower or two on cultivated plants. The color of the flowers is good and where it proves hardy this species should be a welcome addition to gardens. Rhododendron reticulatum D. Don apud G. Don, Gen. Syst. III. 846 (1834). -De Candolle, Prodr. VII. pt. 2, 727 (1839). Rhododendron dilatatum Miquel in Ann. Mus. Lugd.-Bat. I. 34 (1863-64); in II. 164 (1865-66); Prol. FL Jap. 96 (1866-67). Rhododendron rhombicum Miquel in Ann. Mus. Lugd.-Bat. II. 164 (1865-66); Prol. FL Jap. 96 (1866-67). Regel in Gartenfl. X. XVII. 225, t. 586 84 THE AZALEAS OF THE OLD WORLD (1868). Maximowicz in Mem. Acad. Sri. St. Petersburg, ser. 7 XVI. No. 9, 26 (Rhod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 288. (1875). Hooker f. in Bot. Mag. CXIII. t. 6972 (1887). Bean in Gard. Chron. ser. 3, XX. 38, fig. 9 (1896); Trees and Shrubs Brit. Isl. II. 375, fig. (1914). Boissier in Bull. Herb. Boiss. V. 918 (1897). Dalli- more in Garden, LXXII. 267, fig. (1908). Mottet in Rev. Hort. 1909, 79, fig. 28. Schneider, III. Handb. Laubholzk. II. 497, figs. 327 f-g, 328 d-e (1911). W. Watson, Rhodod. and Azaleas, 111 (1911). Matsumura, Ind. PI. Jap. II. pt. 2, 463. (1912). Render in Bailey, Stand. Cycl. Hort. V. 2943 (1916). Millais, Rhodod. 235 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [10] (1918). Azalea reticulata K. Koch, Dendr. II. pt. 1, 179 (1872). Azalea dilatata O. Kuntze, Rev. Gen. pt. 2, 387 (1891). Azalea rhombica O. Kuntze, Rev. Gen. pt. 2, 387 (1891). Rehder in Bailey, Cycl Am. Hort. I. 122 (1900). Rhododendron dilatatum /3. decandrum Makino in Tokyo Bot. Mag. VII. 134 (1893). Rhododendron decandrum Makino in Jour. Jap. Bot. I. 21 (1917). Rhododendron Wadanum Makino in Jour. Jap. Bot. I. 21 (1917). Japan : Kyushu, prov. Satsuma, Kagoshima in gardens, March 4, 1914, E. H. Wilson (No. 6158); Togo, in a garden, March 14, 1914, E. H. Wilson (No. 6266); prov.Osumi, Mt. Kirishima, alt. 100-1 000m. 1917, Z. Tashiro; Nishi-Kirishima, alt. 1000 m. March 11, 1914, E. H. Wilson (No. 6241); prov. Hizen, Nagasaki, Mt. Yuwaja, 1863, C. Maximowicz (Herb. Gray). Shikoku, prov. Tosa, Nano-kawa, May 12, 1889, K. Watanabe (Herb. Gray). Hondo, prov. Choshu, between Habu and Asa, May 7, 1918, E. H. Wilson (No. 10,369); prov. Aki, Miyajima, November 28, 1914, E. H. Wilson (No. 7842); prov. Yamato, Yoshino, April 22, 1914, E. H. Wilson (No. 6573); prov. Yamashiro, Kyoto, April 22, 1914, E. H. Wilson (No. 6568); prov. Shinano, Ogawa, September 5, 1905, J. G. Jack; Nojiri to Nakatsu- gawa, September 6, 1905, J. G. Jack; slopes of Mt. Komaga-take, October 24, 1892, C. S. Sargent; prov. Suruga, lower slopes of Mt. Fuji, alt. 800-1000 m. May 8, 1914, E. H. Wilson (No. 6640); prov. Sagami, Hakone Mts. Kintoki, May, 1876, James Bissett (No. 237, Herb. Kew); prov. Shimotsuke, Nikko region, alt. 1000-1800 m. May 18, October 24, 1914, E. H. Wilson (Nos. 6705, 7694), shores of Lake Chuzenji, September 3, November 7, 1892, C. S. Sargent; same locality, October 25, 1905, J. G. Jack. Hokkaido, prov. Hidaka, Shoya village, August 17, E. Tokubuchi. Province unknown, " Mimasaka," May 10, 1904, S. Arimoto (Herb. Gray). Cultivated: Arnold Arboretum (No. 3427-1). This is a common plant throughout the greater part of Japan. In Kyushu, Shikoku and Hondo it is everywhere abundant; in Hokkaido it is rather rare and ENUMERATION OF THE SPECIES 85 finds the northern limits of its range in Hidaka province. On volcanic ash, lava- flows and on cliffs in many places, as in the Nikko region, lower slopes of Mt. Fuji and on the Hakone mountains for example, it is extraordinarily abundant both in open country, in thickets, on the margins of woods and as undergrowth in thin forests. It is in flower from early in March to late in June according to latitude and altitude and produces in great abundance its rose- to red-purple or rich magenta-colored blossoms. The richest colors are on plants in the open, and in dense shade the color is quite pale. Nowadays magenta is not a popular color in the west but massed and alone save for the varying shades of green the flowers of this Azalea are impressive. The plant forms a much-branched bush or bushy tree from 1 to 8 metres tall, with numerous erect or erect-spreading verticillate, slender but rigid branches, yellow-brown and sparsely or densely hairy when young, usually soon becoming glabrous, and in the second year pale gray. In the shade the branches are more sparse and the plants less compact and shapely in habit. The winter-buds are conic and acute, with rufous- or gray-brown, villose and ciliate scales, the inner ones being viscid. The leaves unfold as the corollas fall, and are clustered in threes or in pairs at the end of the branchlets, and vary in shape from broad-ovate to rhombic or occasionally to ovate, and terminate in a short mucro. In size they vary from 3 to 6 cm. in length and from 2.5 to 5.5. cm. in width, with petioles from 0.5 to 1.5 long. In the shade the leaves are membranous but in the open they are firm and the reticulation is more prominent; they are dark green above, pale even sub-glaucous below, and in the autumn change to vinous or blackish purple or occasionally to yellow with splashes of red-purple. The principal veins are impressed above and raised and usually more or less villose on the under surface. When young the leaves are densely clothed with long, or with short and long, appressed, gray to yellow-brown pilose hairs which dis- appear early from the upper and entirely or partially from the lower surface. The petiole is flattened and usually villose, but may be glabrous and glaucescent. The flowers are borne at the end of the shoots in pairs or solitary or in clusters of three or four. The corolla is from 3.5 to 5 cm. across usually unspotted, with a short tube and spreading lobes which are often divergent and give the corolla a two-lipped appearance. The pedicels are clothed with gray or brownish appressed hairs, and the minute, 5-toothed calyx is ciliate, villose or glabrescent. The stamens are normally 10, in two series, the longer equalling the corolla in length, the shorter half as long, the filaments curving upward; the anthers are purple. The ovary is covered with short straight hairs, or is lepidote, or lepidote and hairy, and the style is hairy or glabrous; the stigma capitate and minutely 5-lobed. The fruit is curved, cylindric, from 1 to 1.5 cm. long, furrowed, lepidote, glabrous, or clothed with gray or brown soft hairs. I have gone very minutely into the degree of variation exhibited by this plant for I realize that those who compare the two figures in the Botanical Magazine will find it difficult to appreciate that they represent the same species and really only differ in the number of stamens. In the array of specimens before me and cited above the variations mentioned are all shown and neither among them nor in the field have I been able to distinguish two species. The distinguishing char- acters emphasized by Maximowicz are inconstant except the number of stamens and I am doubtful if careful examination among many plants would not reveal flowers with five and others with ten stamens on the same plant. I have 10- stamened flowers with a lepidote ovary and style, with a hirsute and lepidote ovary and glabrous style; I have 5-stamened flowers which show exactly the same variation. On both forms I find glabrous leaves and leaves more or less pubescent below especially on the principal veins. Miquel had only leaf specimens and founded his two species chiefly on the difference in shape of the leaves. On every plant of this species as here understood 86 THE AZALEAS OF THE OLD WORLD the leaves tend toward a rhombic shape, in fact this predominates and from this they vary to broad-ovate and occasionally to ovate. An attempt to make leaf- distinctions in this plant of specific value breaks down on the first bush examined. Maximowicz who had flowering specimens and young leaves added to Miquel's description of R. dilatatum the 5 stamens and glandular ovary which subsequent authors have used to distinguish Miquel's two species. That Maximowicz was not justified in this is abundantly proved by the material before me. As to Makino's R. Wadanum I have specimens from Shikoku with slightly spotted flowers but can find no other difference and do not consider that they can be separated even as a variety. It is unfortunate that both Miquel's names should have to be dropped but G. Don's description, brief as it is, leaves no doubt that the plant he had was the same as that later described by Miquel. It has long been accepted that Don's name belonged to one of Miquel's species but the question remained to which? With the acceptance of the view first suggested by K. Koch that both represent one species the question disappears and Don's name becomes the valid, as it is the oldest, name. 1 have taken the form with ten stamens as representing the type since it is by far the more common. According to G. Don this Azalea was first introduced into England from Japan by Messers Knight of Chelsea and evidently about 1832-33, but it must have been lost. It was introduced by Maximowicz into the Petrograd Botanic Gardens in 1865, and is figured in the Gartenflora for 1868 but the color is poor. It was in- troduced into the Arnold Arboretum by Professor Sargent who sent seeds from the Nikko region in the autumn of 1892. The plants have grown slowly but have proved perfectly hardy and each spring for a number of years past have flowered profusely. Apparently it is rare in Europe and is one of the plants which will probably thrive better in eastern North America than in England. The specimen cited from Kintoki on the Hakone mountains is simply an ab- normal condition of typical R. reticulatum G. Don in which the flowers have been retarded and are open when the leaves are well grown. There are two flowers and the pistil of a third, and a small deformed flower on the specimen, and one flower with two unfolding leaves. The leaves have the characteristic rhombic shape and the flowers have from 8 to 10 stamens. Branches bearing late flowers and leaves together may be found occasionally on any species of precocious flowering Rhododendron. A variety with white flowers is : Rhododendron reticulatum var. albiflorum Wilson, n. comb. Rhododendron rhombicum var. albiflorum Makino in Tokyo Bot. Mag. XVIII. 66 (1904). Render in Bailey, Stand. Cycl. Hort. V. 2943 (1916). Makino describes this plant from Tosa province in Shikoku and states that it is very rare. It is unknown to me and is not cultivated in western gardens. The form adopted by Maximowicz as the typical R. dilatatum Miquel, and so considered by subsequent authors, and which is char- acterised by having 5 stamens only may be distinguished as: Rhododendron reticulatum f . pentandrum Wilson, n. nom. Rhododendron dilatatum Maximowicz in M6m. Acad. Sci. St. Petersbourg, se~r. 7, XVI. No. 9, 27 (Rhodod. As. Or.) (1870), in part, not Miquel. Franchet & Savatier, Enum. PL Jap. I. 289 (1875). Matsumura, PI. Nikko 71 (1894); Ind. PL Jap. II. pt. 2, 459 (1912). Boissier in Bull. ENUMERATION OF THE SPECIES 87 Herb. Boiss. V. 918 (1897) . Hooker f . in Bot. Mag. CXXV. t. 7681 (1899) . J. H. Veitch, Hortus Veitchii, 373 (1906). Shirisawa, Icon. Ess. For. Jap. II. t. 61, figs. 23-32 (1908). Schneider, III. Handb. Laubholzk. II. 497, figs. 327 h, 328 a-b (1911). Bean, Trees and Shrubs Brit. Isl. II. 353 (1914). Millais, Rhodod. 155 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [9] (1918). Rhododendron dilatatum a. typicum Makino in Tokyo Bot. Mag. VII. 134 (1893). Japan: Hondo, prov. Shinano, 1864, Tschonoski (Herb. Gray); prov. Suruga, Mt. Fuji, May 12, 1907, (flowers), August 15, 1905, (leaves); prov. Sagami, Hakone Mts., alt. 600-1000 m. April 16, 1914, E. H. Wilson (No. 6448); in a garden, Miyanoshita, April 17, 1914, E. H. Wilson (No. 6429). Cultivated: Royal Gardens, Kew, May, 1880, G. Nicholson. Of the characters proposed by Maximowicz only that of the 5 stamens seems to hold and, as I have already mentioned, I am not convinced that this is con- stant. This form is believed to have been introduced to England by Messers Veitch in 1883, but that it was in cultivation in Kew as early as 1880 is shown by Nicholson's specimens preserved in this herbarium. I have no record of its being cultivated in America though it will probably appear among the seedlings raised from Japanese seeds collected by me in 1914 and distributed by the Arnold Arboretum. Rhododendron quinquefolium Bisset & Moore in Jour. Bot. XV 292 (1877). Franchet & Savatier, Enum. PI. Jap. II. 653 (1879). Matsumura, PI. Nikko, 71 (1894); in Tokyo Bot. Mag. XIV. 69 (1900); Ind. PL Jap. II. pt. 2, 463 (1912). Komatsu in Icon. PL Koisikav. I. 59, t. 30 (1912); in Tokyo Bot. Mag. XXXII. [8] (1918). Bean, Trees and Shrubs Brit. Isl. II. 374 (1914). Rehder in Bailey, Stand. Cycl. Hort. V. 2947 (1916). Millais, Rhodod. 233 (1917). Japan : Hondo, prov. Shimotsuke, Nikko region, alt. 600-1600 m. May 14, May 18, October 20, 1914, E. H. Wilson (No. 6683, type, 6683A, 7676); Lake Chuzenji, October 21, 1914, E. H. Wilson (No. 7676A); same locality, August 11, October 25, 1905, J. G. Jack. This very distinct species is well characterised by its corky, brown bark, its broad-elliptic to obovate verticillate leaves and by its snow white, rotate-cam- panulate corolla. In habit of growth, arrangement and shape of leaves and in the shape of the flowers it superficially resembles R. pentaphyllum Maxim, but the two belong to different sections. In Maximowicz's species the flowers and leafy shoots originate from different buds, the flower-bud being terminal and the leaf-buds lateral below it. In this new species both flowers and leafy shoots issue from the same bud. This Azalea is abundant in shady, rocky ravines in the Nikko region but is elsewhere unknown to me. It is a bush or small tree from 1.5 to 8 m. tall with gray-brown, corky bark, fissured into thick irregular plates; the shoots are glabrous, terete, shining brown the first year and arranged partly in verticils and partly alternate. The winter-buds are narrow-ovoid, acute, with 88 THE AZALEAS OF THE OLD WORLD purple-brown ciliate scales, the basal ones having long aristate points. The leaves are deciduous and unfold at the same time as the flowers open; they are clustered 4 or 5 together at the end of the shoots and vary from broad-elliptic to obovate, the apex is mucronate and the base cuneate, the edges are ciliate and often margined with red-purple; they are more or less villose, especially the lower half of the midrib; the petiole is very short and villose. The flowers are either solitary, or two or three together on slender, glandular-hairy pedicels; the corolla is pure white with green spots within, rotate-campanulate, and the calyx is membranous, with 5 lanceolate or deltoid, glandular-ciliolate segments. The stamens are ten, with greenish filaments of unequal length, dilated and villose at the base. The pistil much exceeds the stamens and is glabrous except at the summit of the ovary. The fruit is stout, cylindric, and from 1 to 1.5 cm. long. According to Bean, this Azalea was introduced into English gardens by Lord Redesdale about 1896, but it appears to be rare in gardens. To America it has been sent by the Yokohama Nursery Company under the name of " Azalea quin- quefolia white " and I have seen it growing fairly well in the garden of Mr. T. A. Havemeyer, on Long Island. It is a woodland plant and needs shade and a cool situation if it is to thrive. In 1905 Mr. J. G. Jack sent seeds from Lake Chuzenji to this Arboretum but the resultant plants failed to prove hardy. The plants raised from seeds sent from the same locality in 1914 have also proved tender. Like many other woodland plants this Azalea is difficult to manage when young. Rhododendron Schlippenbachii Maximowicz in Bull. Acad. Sri. St. Petersbourg, se>. 3, XV. 226 (Mel. Biol VII, 333) (1870); in Mem. Acad. Sri. St. Petersbourg, se>. 7, XVI. No. 9, 29, t. 2, figs. 7-13 (Rhodod. As. Or.) (1870). Herder in Act. Hort. Petrop. 1.347 (1871). Franchet & Savatier, Enum. PL Jap. 289 (1875). . Hemsley in Jour. Linn. Soc. XXVI. 30 (1889). Hooker f. in Bot. Mag. CXX. t. 7373 (1894). Gard. Chron. ser. 3, XV. 462, fig. 58 (1894); XIX. 561, fig. 87 (1896); LV. 9, t. (1914). W. Watson in Garden XL VI. 80, t. 972 (1894); Rhodod. and Azaleas, 112 (1911). Palibin in Act. Hort. Petrop. XVIII. 150 (Consp. Fl. Kor. II.) (1900). J. H. Veitch, Hortus Vritchii, 408 (1906). Komarov in Act. Hort. Petrop. XXIV. 206 (Fl Mandsh. III.) (1907). Schneider, III. Handb. Laubholzk. II. 494, figs. 325 i-1 327 a-b (1909). Nakai in Jour. Coll. Sri. Tokyo, XXXI. 75 (Fl. Kor. pt. 2) (1911); Fl. Syl Kor. VIII. 43, t. 15 (1919). Matsumura, Ind. PI Jap. II. pt. 2, 464 (1912). W. Taylor in Garden, LXXVII. 136, fig. (1913). Bean, Trees and Shrubs Brit. Isl. II. 378 (1914). Rehder in Bailey, Stand. Cycl. Hort. V. 2943 (1916). Millais, Rhodod. 239 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [9] (1918). Azalea Schlippenbachii O. Kuntze, Rev. Gen. pt. 2, 387 (1891). Korea: prov. South Keisho, Herschel Island, 1863, R. Oldham (No. 510, Herb. Gray); Chirisan, alt. 600-1845 m. November 16, ENUMERATION OF THE SPECIES 89 1917, E. H. Wilson (No. 9592); Fusan, September 11, 1905, J. G- Jack; prov. Keiki, East Park palace, Seoul, September 24, 1905, J. G. Jack; Government Forestry Garden, Seoul, cultivated, May 21, 1917, E. H. Wilson (No. 8431); Kazan, near Suigen, May 24, 1917, E. H. Wilson (No. 8472); prov. Kogen, Diamond Mts., July 7, 1918, E. H. Wilson (No. 10,493); prov. South Kankyo, hills round Gensan (Wonsan), September 5, 1903, C. S. Sargent; between Shinkori and Eiko, September 20, 23, 1917, E. H. Wilson (Nos. 9201, 9223); prov. North Heian, altitude 300-1000 m. June 16, 1917, E. H. Wilson (No. 8603); province unknown, " Pomasa," altitude 800 m. May 21, 1906, U. Faurie (No. 66^) ; " Hoangheito," August, 1906, U. Faurie (No. 668); northeast coast, 1854, Baron A. Schlippenbach (Herb. Gray). Northeast Manchuria: shores of Possiet Bay, 1860, C. Maxi- mowicz (Herb. Gray). Japan: Hondo, prov. Ugo, Chokai-san, June 15, 1903, K. Sakurai. Cultivated: Arnold Arboretum (No. 7258). This is one of the commonest shrubs in Korea and in thin woods is often the dominant undergrowth. It just crosses the border into northeast Manchuria on the shores of Possiet Bay and in Japan is known from only two localities in north Hondo. In Korea the southern limit of its range appears to be the Chiri-san range, and Herschel Island in about the same latitude. Curiously it does not grow on Quelpaert Island where the representative species is R. Weyrichii Maxim. On the lower-middle slopes of Chiri-san and the lower slopes of the Diamond Mountains R. Schlippenbachii is extraordinarily abundant. In June these regions are a wonderful sight with literally miles and miles of the purest pink from the millions of flowers of this Azalea. The plant grows from 1 to 5 m. tall and in open places is sturdy and densely branched but in shade the branching is lax. The branches are rigid, erect, verticillate or irregular, covered with curled glandular hairs when young; pale brown the first year, becoming gray and glabrous the second. The leaves are deciduous, thin, clustered in whorls of five at the end of the branches; on strong shoots the whorl may be subtended by a pair of leaves and these by odd, scattered leaves of small size; in shape they are all uniform being more or less obovate (often broadly so), from 5 to 9 cm. long and from 3 to 7 cm. wide, the apex is truncate or rounded, emarginate, with a glandular mucro, and the base is narrowed to a short, broad petiole; the margins are slightly undulate, and both surfaces are at first sparsely pubescent but later are glabrous except on the under- side of the principal nerves. The leaves are dark green above, pale below and in autumn change to yellow, orange and crimson. The flowers are fragrant, borne in terminal 3- to 6-flowered umbellate clusters, and usually open as the leaves be- gin to unfold. The corolla is pale- to rose-pink, broad-rotate, funnel-shape, from 6 to 8 cm. in diameter, with spreading, rounded lobes, the posterior marked with red-brown spots. The pedicels and calyx are clothed with glandular hairs, and the calyx-lobes are green, ovoid, about 5 mm. long and persistent. The stamens, always 10 in number, are of unequal length, the longest equalling the corolla but shorter than the style. The fruit is erect, oblong-ovoid, about 1.5 cm. long, slightly oblique and covered with sessile glands; the seeds are angular, shining dark brown. 90 THE AZALEAS OF THE OLD WORLD This species was discovered by Baron A. von Schlippenbach of the Russian navy on the shores of northeast Korea in 1854, and appears to have been culti- vated sparingly in Japan for many years under the name of " Kurofune-tsutsuji." James H. Veitch in 1893 saw it in a garden in Japan and sent it to England and this was its first appearance in the west. Since then Japanese nurserymen have exported it in small quantities to Europe and America but to this time it is by no means so well known as its beauty warrants. The plants in this Arboretum were raised from seeds collected in Korea in 1905 by Mr. J. G. Jack; they have grown slowly but have been unaffected by winter-cold and summer-drought. For two to three years past they have flowered freely and their large, pure pink blossoms are lovely. This species has the second largest leaves and the most beautiful flowers of its group and is worthy of a place in every garden. I collected a quantity of seed in Korea in the autumns of 1917 and 1918, and this has been widely distributed by the Arnold Arboretum and in a few years this Azalea should be a prominent feature in New England gardens. SECT. III. RHODORA G. DON* Flowers from a terminal bud, the leafy shoots arising from separate, lower, lateral buds; corolla rotate-campanulate, rarely campanulate or two-lipped; stamens 10, rarely from 5 to 7. Leaves deciduous, scattered or whorled. Fruit more or less cylindric, rarely conic-ovoid. Shrubs often of large size, usually with rigid ascending or spreading branches; shoots glabrous or hairy. This section contains 5 species; three of these grow in eastern Asia and two (R. canadense Zabel and R. Vaseyi A. Gray) in eastern North America. Rhododendron Albrechtii Maximowicz in Bull. Acad. Sci. St. Peters- burg, se>. 3, XV. 227 (Mel. Biol. VII. 335) (1870); in Mem. Acad. Sci. St. Petersburg, se>. 7. XVI. No. 9, 30, t. 2, figs. 14-20 (Rhodod. As. Or.) (1870). Franchet & Savatier, Enum. PL Jap. I. 290 (1875). Matsumura, PL Nikko, 71 (1894); Ind. PL Jap. II. pt. 2, 458 (1912). Boissier in Bull. Herb. Boiss. V. 917 (1897). Schnei- der, III. Handb. Laubholzk. II. 496, figs. 325 m-n, 327 e (1909). Millais, Rhodod. 112 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [8] (1918). Azalea Albrechtii O. Kuntze, Rev. Gen. pt. 2, 387 (1892). Rehder in Bailey, Cycl Am. Hort. I. 122 (1900). Japan: Hondo, prov. Shinano, 1864, Tschonoski (Herb. Gray); prov. Musashi, cultivated, Bot. Gard. Tokyo, April 2, 1914, E. H. Wilson (No. 6363); prov. Shimotsuke, Nikko region, alt. 1600-2000 m., June 23, October 15, 1914, E. H. Wilson (Nos. 6858, 7638); prov. Uzen, lower slopes of Adzuma-san, alt. 800-1100 m., July 19, October 24, 1914, E. H. Wilson (Nos. 7208, 7721); prov. Mutsu, Hirosaki, May, 1905, U. Faurie (No. 6785); Hakkoda-yama, alt. 1000-1600 m., July 5, 1914, E. H. Wilson (No. 7083); same locality, October 2, 1892, C. S. Sargent. Hokkaido, prov. Oshima, Fuku-yama, June 3, 1889, U. Faurie (No. 3831, Herb. Kew) ; Kakumi Hot Springs, September 29, 1892, C. S. Sargent; prov. Iburi, Nobori-betsu, July 2, 1914, E. H. Wilson; prov. Ishikari, Sapporo, May 17, October 10, 1903, S. Ari- moto (Herb. Gray); same locality, May 29, 1890, E. Tokubuchi (Herb. Gray); Moiwa-dake, June 15, 1885, K. Miyabe (Herb. Gray); same locality, June 23, 1891, E. Tokubuchi; cultivated, Bot. Gard. Sapporo, August 22, 1905, J. G. Jack. 1 For synonyms and further details see page 119. 91 92 THE AZALEAS OF THE OLD WORLD On the margin of forests and in thickets this species is widely distributed in northern Japan from the mountains round Sapporo in central Hokkaido south- ward to those of Shinano in central Hondo. I am familiar with it from the Nikko region northward, but do not consider it a common plant. It is scattered here and there and is partial to dense thickets on steep slopes of loose soil. It is a shrub of loose and sparse habit, from 1 to 1.5 m. tall, with slender, smooth, purple- brown branches which are brown and furnished when young with curled, partly viscid, gray hairs. The leaves are deciduous, membranous, dark green and change to yellow before they fall; they are scattered on the vigorous shoots, and crowded, forming false whorls of fives, at the end of the branchlets; they are obo- vate to oblanceolate, from 4 to 12 cm. long and from 1.5 to 6 cm. wide, acute or subacute, with a glandular mucro; the base is narrowed to a short, winged petiole; the margins are serrulate and ciliate, the upper surface is sparsely clad with ap- pressed, flattened, hispid hairs and the under surface is more or less densely clothed with short, pale gray tomentum. The flowers are about 5 cm. across, and open in June and early July, according to climate, and immediately before or as the leaves unfold; they are borne in terminal umbels, each containing from 3 to 5 flowers. The pedicels are moderately stout, from 1 to 2 cm. long, and are clothed with curled, glandular, yellowish hairs; the calyx is small, with 5 purple and ciliate lobes; the corolla is rich red-purple, rotate-campanulate, with a short, wide tube and spreading, rounded lobes; the stamens are 10 in number in two series of unequal length, the longest equalling the corolla and over- topped by the curved style, which has a bifid stigma. The fruit is erect, conic- ovoid, from 1 to 1.2 cm. long, purple-brown, and clothed with yellow, viscid, pilose hairs. This species was discovered about 1860 by Dr. Michael Albrecht of the Russian Consulate in Hakodate, and was found by Maximowicz also in the vicinity of Hakodate in 1861. However, it does not appear to have been introduced into western gardens until 1892, when Professor Sargent sent seeds to the Arnold Arboretum from two localities in Hokkaido. Plants raised from these seeds grew for several years in this Arboretum, but never really flourished. I sent seeds from northern Hondo in 1914, and these were distributed by the Arboretum. Plants raised from these seeds have grown fairly well, but it is too early to say whether they will make themselves at home here. There is no apparent reason why this Azalea should not thrive here. The habit is sparse and the flowers are not large, but the color is intense and the plant is graceful and attractive. Rhododendron pentaphyllum Maximowicz in Bull. Acad. Sd. St. Petersburg, ser. 3, XXXI. 65 (MeT Biol. XII. 491) (1887). Mat- sumura, Ind. PI. Jap. II. pt. 2, 463 (1912). Millais, Rhodod. 225 (1917). Komatsu in Tokyo Bot. Mag. XXXII. [8] (1918). Rhododendron pentaphyllum var. nikoense Komatsu in Icon. PL Koisikav. III. 45, t. 168 (1916). Rhododendron quinquefoliwn var. roseum Render in Bailey, Stand. Cycl. Hort. V. 2947 (1916). Japan: Kyushu, prov. Osumi, Tarumizu village, April and May, 1911, S. Kawagoe. Hondo, prov. Kozuki, Mt. Akagi, June 13, 1911, as to flowers, May 17, 1921, K. Sakurai; prov. Musashi, Yokohama, cultivated, April 11, 1914, E. H. Wilson (No. 6396); prov. Shimot- ENUMERATION OF THE SPECIES 93 suki, Nikko region, alt. 1000-2000 m., May 18, May 25, May 31, October 20, 1914, E. H. Wilson (Nos. 6698, 7683); same region, shores of Lake Chuzenji, November, 1892, C. S. Sargent. Cultivated: Hort. T. A. Havemeyer, Glenhead, Long Island, New York, April 23, 1919. This species is distributed in Japan from the extreme south of Kyushu through Shikoku and Hondo to the Nikko region and beyond to Adzuma-san in Iwashiro province but is nowhere abundant. I have seen it wild in the Nikko region only and there, near the Hashimoto Tea House and on the mountain behind the Lakeside Hotel at Chuzenji, it is common. It grows chiefly among deciduous trees and shrubs and, flowering early, its pure rose-pink blossoms are conspicuous from a long distance. My first view of it was across a deep ravine and I shall never forget the charm of that splash of lovely color among the leafless plants of a well-wooded steep slope. This Azalea is a large bush or small tree from 2 to 8 m. tall, with a dense crown of twiggy ascending and spreading branches which are arranged partly in verticils and partly in an alternate manner, and are sparsely pilose when young but soon become glabrous; they are red-brown and more or less angular the first year, later terete and gray. The trunk is clothed with pale gray- brown bark. The winter-buds are short or elongated, acute, with shining purplish brown, ciliolate, rather paleaceous scales. The leaves are clustered in a whorl of 5 at the end of the branchlet and are deciduous, chartaceous in texture, lustrous green in the summer, changing to rich tints of orange and crimson in the autumn; they are oval to elliptic-lanceolate in shape, from 3 to 6 cm. long and from 1.5 to 3 cm. wide, acute and mucronulate, with a narrowed cuneate base; the margins are ciliate and finely serrulate; they are reticulately veined with the midrib villose on both surfaces, especially along the basal half. The petioles are from 3 to 8 mm. long, usually sparsely bearded and glandular. The flowers are solitary or in pairs at the end of the branchlets, and as the flower-buds expand the bud-scales fall. The corolla is unspotted, bright rose-pink in color, about 4 to 5 cm. in diameter, rotate-campanulate, with spreading, rounded, often emarginate lobes. The pedicels are sparsely or densely glandularly pilose or quite glabrous and from 1 to 1.5 cm. long. The calyx is glabrous, membranous, with 5 triangular or deltoid teeth which are ciliate and from 1 to 5 mm. long. The stamens are unequal, 10 in number, included; the base of the filaments is villose and the anthers are yellow. The pistil slightly overtops the stamens and is glabrous. The fruit is spindle- shaped, swollen about the middle, about 1.5 cm. long and 0.8 cm. wide, verruculose, and subtended by the persistent calyx. This lovely plant is very rare in gardens and appears to have been introduced to Europe and America by the Yokohama Nursery Company, who list it under the name of " Azalea quinquefolia pink." It is essentially a woodland species, fond of partial shade. Where it grows at Nikko the soil is volcanic ash and lava overlain with black leaf-mould. I collected seeds in 1914 for the Arnold Arbo- retum and part of them were distributed. The plants raised in this Arboretum have proved tender and difficult to manage, but older plants given a shady wood- land site would probably thrive here as well as they do on Long Island, New York. In Japan its autumn tints are singularly vivid and beautiful. Rhododendron nipponicum Matsumura in Tokyo Bot. Mag. XIII. 17 (1899); in Icon. PL Koisikav. I. 9, t. 5 (1911); Ind. PL Jap. II. pt. 2, 463 (1912). Komatsu in Tokyo Bot. Mag. XXXII. [8] (1918). 94 THE AZALEAS OF THE OLD WORLD Japan: Hondo, prov. Uzen, near Toge, alt. 1000 m., July 20, October 23, 1914, E. H. Wilson (Nos. 7219, 7191). This little known Japanese species is confined to the mountains of north-central Hondo. It is very distinct in its flowers and fruit and also in its papery, cinnamon- brown bark which shreds off and leaves polished brown stems and branches. No other species with deciduous leaves has such attractive bark. In habit it forms an upright, bushy shrub from 1 to 2 m. high, with rigid, red-brown branches clothed with pilose and glandular hairs. The winter-buds are pale straw-color tinged with purple, large, ovoid, with ciliate, eglandular bud-scales. The leaves are scattered, sessile, membranous and deciduous; they have appressed, hispid hairs on both surfaces and the margins are ciliate; in shape they are panduriform, from 6 to 18 cm. long and from 3.5 to 8.5 cm. wide, with the apex rounded or truncate, usually emarginate, rarely obtuse, and are narrowed to the base, which is firmly appressed to the stem and often has a curious shield-like protuberance continued below the point of insertion; the veins are prominent and reticulate on the under side and many are furnished with strigose hairs. In the autumn the leaves are tinted from orange to crimson. The flowers are small and not conspicuous, and open in late June and July with, or even after, the leaves. They are borne from 6 to 15 together in terminal, umbellate corymbs. The corolla is white, tubular or campanulate, from 1.5 to 2 cm. long and from 0.8 to 1 cm. broad, with 5 short, very slightly spreading lobes. The calyx is small, membranous, with lobes of unequal size, and like the slender pedicel is covered with viscid hairs. The stamens are of unequal length, 10 in number and are included, as is the style, which is straight and slightly thickened below the capitate stigma. The fruit is pendent, oval, from 8 to 10 mm. long, and has thin walls which open to the base and have wavy margins. This species in foliage suggests R. Schlippenbachii Maxim., but otherwise is very distinct. On the hills round Toge, near the base of Adzuma-san, in 1914, I found this species to be common in rather open country. The season of flowering was almost past. In the autumn I gathered seeds for the Arnold Arboretum, which distributed them in America and Europe. The plants raised from these seeds have not yet flowered in the Arboretum. This Azalea was unknown to Japanese nurserymen until I gave plants to the Yokohama Nursery Company in 1914. Its bark and foliage is striking, but the size of the flowers and the manner in which they are hidden by the leaves and young shoots detracts from the plant. As a flowering shrub it is one of the least desirable of Japanese Azaleas. SECT. IV. PENTANTHERA G. DoN 1 Flowers from a terminal bud, the leafy shoots from separate, lower, lateral buds; corolla funnel-form-campanulate to funnel-form; stamens 5. Leaves deciduous, scattered on the shoots. Fruit more or less cylindric. Shrubs usually with stiff, ascending branches. Shoots villose or glabrous. This section contains some 15 species and is the most widely distributed of the four sections which we are discussing. Two species grow in eastern Asia, one in Japan and the other in China; one species, R. luteum Sweet, grows on the Caucasus, in the Pontus region in Asia Minor and in parts of eastern Europe. The others are confined to North America, where they are widely distributed. Rhododendron molle G. Don, Gen. Syst. III. 846 (1834). Azalea mollis Blume, Cat. Gewass. Buitenz. 44 (1823); Bijd. Fl. Ned. Ind. 853 (1825). Azalea sinensis Loddiges, Bot. Cab. IX. t. 885 (1824). De Candolle, Prodr. VII.pt. 2, 718 (1839). Fortune, Wanderings in China, 154 (1847); Tea Coun- tries of China 154 (1852) ; Residence among Chinese, 28 (1857) ; Suringar in Gartenfl. LVII. 505 (1908). Rhododendron sinense p. flavescens Sweet, Brit. Flow. Gard. ser. 1, III. t. 290 (1829). Rhododendron sinense Sweet, Brit. Flow. Gard. ser. 1, III. sub. t. 290 (1829). Hovey, Am. Gard. Mag. II. 142, 179 (1836). Hance in Journ. Bot. XVI. 109 (1878). Hemsley in Jour. Linn. Soc. XXVI. 30 (1889), as to the Chinese plant. Suringar in Gartenfl. LVII. 516 (1908). W. Watson, Rhodod. and Azaleas, 112 (1911), in part. Schneider, III. Handb. Laub- holzk. II. 497, figs. 328 g-h, 329 a-b (1911). Render & Wilson in Sargent, PL Wilson. I. 549 (1913). Bean, Trees and Shrubs Brit. Isl. II. 379 (1914). Millais, Rhodod. (1917), in part. Rhododendron sinense a. flammeum Sweet, Brit. Flow. Gard. ser. 1, III. sub. t. 290 (1829). Azalea pontica var. sinensis Lindley in Bot. Reg. XV. t. 1253 (1829), poor figure. Azalea sinensis a. genuina Maximowicz in Suppl. Ind. Sem. Hort. Petrop. 1869, 11. China: prov. Hupeh, stony hills and Pine-woods 15 miles below Ichang on right bank of Yangtsze River, alt. 30-300 m., April 24, 1907 (flowers), January 18, 1908 (fruit), E. H. Wilson (No. 800); same locality, A. Henry (No. 268); prov. Hunan, near Changsha, among shrubs, alt. 70-400 m., April 7, 1918, Dr. Handel-Mazzetti (No. 2342); prov. Chekiang, vicinity of Ningpo, 1908, D. MacGregor; near Chang- hua, rocky banks, alt. 300 m., July 12, 1915, F. N. Meyer (No. 1543); 1 For synonyms and further details see page 124. 95 96 THE AZALEAS OF THE OLD WORLD prov. Kiangsii, Nanking, April 23, 1914, Mary Strong Clemens (4226, Herb. Bur. Sci. Manila). This species is abundant in eastern China, especially on the mountains of the Chekiang province, and appears to have the western limits of its range on the conglomerate hills a little to the east of the city of Ichang where Henry and I col- lected it. The plant is evidently common in the neighborhood of Ningpo, where every collector since Fortune has gathered it. It grows among coarse grasses and shrubs and in thin pine-woods. It is a sturdy, sparingly branched shrub from half a metre to a metre and a half high, taller than broad, and has rich golden- yellow flowers produced in large trusses before the leaves unfold. In habit and floral characters in general it closely resembles its Japanese relative (R. japonicum Suring.), but it has a smaller and less bristly calyx, flower-stalks without or with few bristles, stamens as long or longer than the corolla and winter-buds densely clothed with a short, nearly white, velvety pubescence. The leaves of the two species are conspicuously different. Those of the Chinese plant are densely clothed on the lower surface with soft, nearly white, matted pubescence which persists through the life of the leaf; the upper surface is also softly pubescent but much of this disappears before autumn. In late summer and autumn the leaves appear glaucous below on account of the density of the pale gray pubescence. The leaves of the Chinese species are also larger, up to 15 cm. long and 5.5 cm. wide, more de- cidedly oblong-lanceolate, and only occasionally broadest above the middle. Sweet's figure (Brit. Flow. Gard. ser. 1, III. t. 290 [1892]) clearly shows these essential characters, and if it is compared with Hegel's figure (Gartenfl. XVI. 299, t. 556 [1867]) of the Japanese species the two plants cannot be confused. Since the two may easily be distinguished by the leaves and other characters, behave differently under cultivation and grow wild in widely separated geographical areas, I consider that they are best kept under separate names without discussing the vexed question of what constitutes a species. I have not seen this Azalea in Chinese gardens but it must long ago have been grown in those of Canton, Hanchow and Soochow. Loddiges received the plant from China in 1823 and figured it in the Botanical Cabinet, IX. t. 885 (1824). This plant in all probability was brought by some ship of the East India Company from Canton. Others were introduced through the same agency during the succeeding years but the species appears to have been lost, or nearly so, when it was sent to the gardens of the Horticultural Society of London by Robert Fortune in 1845. Fortune found it in gardens at Canton and wild on the hills near Ningpo, and in his books he mentions this Azalea several times. According to Blume (Cat. Gewass. Buitenz. 44 [1823]) it was introduced into gardens in Java before 1823. Of its first introduction into America I can find no record, but Hovey tells us that it flowered with him, and also in the garden of Mr. J. P. Gushing, Watertown, Mass., in April, 1836. Later it became lost and it was not cultivated in America when in January, 1908, 1 sent seeds from China to the Arnold Arboretum. Some of these seeds were sent to England, where the species appears again to have been lost. In- deed, in western gardens the pure species never seems to have established itself, and like its Japanese relative soon got lost under a mass of hybrid and seminal forms. In this Arboretum the Chinese Azalea is doubtfully hardy, but it is an ex- cellent pot plant and forces well. This tenderness, which is to be expected when the climate of the region in which it grows is taken into consideration, explains its frequent disappearance from gardens. The habit and fine yellow flowers make it a very decorative garden plant but it lacks stamina. Its great use has been in hybridising, notably with the Japanese speqies, from which has sprung a race of hybrids popularly known as " Mollis Azaleas," correctly X R. Kosterianum* ENUMERATION OF THE SPECIES 97 It is a pity to have to discontinue the use of Sweet's name R. sinense in favor of one which has been so much misused, but this cannot be avoided. Blume's description is perfectly good, but this publication seems to have been entirely over- looked, all authors quoting him in his Bijdragen Flora Nederlandsch Indie, which was published two years later. Siebold and Zuccarini in 1846 were the first to apply Blume's name to the Japanese plant and initiated the confusion which has existed ever since. X Rhododendron Kosterianum Schneider, III. Handb. Laubhohk. II. 499 (1911). Azalea mollis x sinensis Pynaert & van Geertin Rev. Hort. Belg. XVII. 121, fig. 18 (1891); XX, 277, t. (1894). Rodigas in Tijd. Boomt. 1893, 183, fig. 25. Rhododendron sinense W. Watson, Rhodod. and Azaleas, 77 (1911), in part, not Sweet. Rhododendron sinense X R. molle Schneider, III. Handb. Laubholzk. II. 499 (1911), as a synonym. Rhododendron molle Millais, Rhodod. 211 (1917), in part, not G. Don, nor Miquel. Rhododendron molle X sinense Millais, Rhodod. 212 (1917). It appears to be the concensus of opinion that Anthony Koster & Sons of Bos- koop, Holland, began to cross R. sinense Sweet and R. japonicum Suring. about 1880 and to raise hybrids in commercial quantities. They were closely followed by Anthony Waterer, Knap Hill, Surrey. But apparently neither were the first in this field, for in the Florist and Pomologist for 1875, p. 282, William Tillery writes: " In 1872 1 had six varieties of the showy hardy Azalea mollis from M. van Houtte named as follows: 'Isabellevan Houtte, 'dark nankeen color, 'Madame Camille van Langenhone,' white striped with rose and carmine, 'Nonpareil,' white bordered with rose, 'Ebenezer Pyke,' half yellow, 'Centi Striatella' and 'Baroness Roth- schild.' I find that all the varieties cross very readily with Azalea sinensis the yellow and white varieties, and I have many healthy seedlings this year from these crosses." This would appear to be the beginning of the race. The work of these pioneers has been emulated by others in Holland, Belgium and England and the re- sult is a wealth of popular Azaleas with lovely flowers in exquisite shades of color. For indoor decoration they are, in normal times, forced by the million hi Europe and America. As garden plants out-of-doors they thrive hi many parts of England and Millais is loud in their praise. While hardy in the gardens of New England the plants are usually short lived, lack constitution and are even less satisfactory than the " Gandavensis hybrids." The tender strain derived from the Chinese species (R. molle) is unfortunately too potent, and though the plants struggle along for a few years they are rarely satisfactory and generally at the end of a few years die. It is a pity, for they are beautiful plants. In Horticulture, XXX. 71 (1919), there is described a hybrid (" Miss Louisa Hunnewell ") raised in 1913 by Mr. T. D. Hatfield, superintendent of the estate of Walter Hunnewell, Esq., Wellesley, Mass., which is a cross between the pure species R. japonicum Suring. and R. molle G. Don obtained from the Arnold Arboretum. It first flowered in 1917, and has an orange-yellow corolla similar to that of "Anthony Koster" but even richer in color; it proved "bud hardy " in the exceptionally severe winters of 1917-18 and 1919-20, and is a free grower. If this new Azalea fulfils expectations it will be a most valuable addition to New England gardens. Another hybrid is : X Rhododendron albicans Waterer apud Zabel in Mitt. Deutsch. 98 THE AZALEAS OF THE OLD WORLD Dendr. Ges. XI. 30 (1902). Schneider, III Handb. Laulholzk.il. 499 (1911). Rhododendron occidental* X sinense Zabel in Beissner, Schelle & Zabel, Handb. Laubholzk-Ben. 380 (1903), as a synonym. Rhododendron motte x ocddentale Schneider, III. Handb. Laubholzk. II. 499 (1911), as a synonym. Azalea occidental "Hybrids" Hort. Wezelenburg, Cat. 9 (1915). Rhododendron occidentak X R. molle x " Ghent Azaleas" Millais, Rhodod. 220, fig. (1917). Many varieties of this hybrid have been raised by Anthony Waterer and by Koster & Sons, among the best being "Superba," "Graciosa," "Exquisite" and "Magnifica." The parentage is, apparently, very mixed; a majority have large white flowers with the usual yellow blotch characteristic of the flowers of the American species, but some have cream and others yellow flowers with a darker blotch. They flower between early June and mid-July and are very hardy. In quite recent years X R. Kosterianum has been hybridised with a variety of forms of X R. gandavense and the result is a race with double flowers, for which I propose the name of: X Rhododendron mixtum Wilson, n. nom. Azalea rustica flore-pleno Hort. ex Vandervoorde in Rev. Hort. Belg. XIX. 232, t. (1893), not Van Houtte. Cat. A. Waterer, 10 (1911-12.) Cat. Wezelenburg, 10 (1915). Millais, Rhodod. 213 (1917). It appears to be safer to give a new name to this race than to risk confusing it with the double-flowered " Gandavensis hybrids " (Azaleas rustique de Gand) first raised by Louis van Houtte, who figures four varieties in Flore des Serres XIX. tt. 2021-2024 (1873), and of which " Graf von Meran," " Narcissiflora " and "Louis A. van Houtte," may be considered types. Typical forms of x R. mixtum are " Murillo " with rose-colored flowers, " Virgile " with yellow flowers, " II Tasso " with red flowers, " Milton " with white flowers, and " Racine " with pale pink flowers. According to Millais (Rhodod. 213 [1917]) there are hybrids raised by Vuylsteke between R. molle and Azalea altaclarense. These are unknown to me. Rhododendron molle G. Don has been crossed with one or more of the broad-leafed Rhododendrons and the oldest of these is : X Rhododendron "Smithii aureum" Paxton, Mag. of Bot. IX. 79, t. and fig. (1842). Gard. Chron. II. 352 (1842). Robinson, Fl and Sylva II. 152, t. (1904). Bean, Trees and Shrubs Brit. Isl. II. 346 (1914). Rhododendron "Hardy Seedling" X Azalea sinensis Paxton, Mag. of Bot. IX. 79 (1842.) " Norbiton hybrids " Hort. ex Masters in Gard. Chron. ser. 3, XIII. 665 (1893). Rhododendron sinense X Eurhododendron spec. Zabel in Beissner, Schelle & Zabel, Handb. Laubholz.-Ben. 379 (1903). I have given the parentage as set down by Paxton but Robinson says " between a yellow form of the Chinese Azalea and Rhododendron caucasicwn, the latter of ENUMERATION OF THE SPECIES 99 which it resembles in habit." This interesting Azaleodendron was raised by a nurseryman at Norbiton, Surrey, named W. Smith, and exhibited before the Horti- cultural Society of London in the spring of 1842. It is not now in cultivation in this Arboretum, not being sufficiently hardy to withstand the winter's cold. A specimen in this herbarium under the name of Rhododendron " Victoria regina " and from an exhibition held by the Massachusetts Horticultural Society on April 19, 1889, evidently belongs here, but whether the name it was exhibited under has any standing I do not know. Another hybrid between a broad-leaf evergreen Rhododendron and R. molle G. Don is: X Rhododendron "broughtonii aureum" Hort. apud W. Watson, Rhodod. and Azaleas 93 (1911). Osborne in Gard. Chron. ser. 3, LI. 53. fig. 29 (1912). Bean, Trees and Shrubs Brit. I si. II. 345, fig. (1914). Osborne in his account of this plant says it " obtains its name from the village of Broughton in Peebleshire, where it was raised by crossing a yellow Azalea and a seedling evergreen Rhododendron." Both Osborne and Bean regard this as the best of its class. It has been in cultivation in this Arboretum since 1910 but does not thrive. More recently M. G. Vander Meulen has raised hybrids between R. molle and the evergreen Rhododendrons " Prince Camille de Rohan " and " Leo- pard " which he names " Edouard AndrS," "Director Rodigas" and "Dr. Masters." Mr. White of Sunningdale has raised similar hybrids. Millais (Rhodod. 214 [1917]) says that they retain their leaves in winter, and are quite hardy and very floriferous. Rhododendron japonicum Suringar in Gartenfl. LVII. 516 (1908). Schneider, III Handb. Laubholzk. II. 1046 (1912). Rehder & Wil- son in Sargent, PL Wilson. I. 549 (1913), in a note. Rehder in Bailey, Stand. Cycl. Hort. V. 2942 (1916). Arnold Arb. Bull. Pop- ular Inform, n. ser. III. 26 (1917); IV. 20, 28 (1918). Komatsu in Tokyo Bot. Mag. XXXII. [6] (1918). E. G. Loder in Rhodod. Soc. Notes, I. No. 4, 197 (1918). Rhododendron molle Siebold and Zuccarini in Abh. Akad. Munch. IV. pt. 3, 131 (Fl. Jap. Fam. Nat. II. 7) (1846), not G. Don. Miquel in Ann. Mus. Lugd.-Bat. 1. 33 (1864); II. 164 (1865-66); Prol. Fl. Jap. 96 (1866-67). Schneider, III. Handb. Laubholzk. III. 499, figs. 328 i-k, 329 c-d (1911). Bean, Trees and Shrubs Brit. Isl. II. 368 (1914). Millais, Rhodod. 211 (1917), in part. Azalea japonica A. Gray in Mem. Am. Acad. n. ser. VI. 400 (1858-59). Hemsley in Garden, XI. 420, t. (1877), in part. Rhododendron molle var. glabrior Miquel in Ann. Mus. Lugd.-Bat. I. 33 (1864). Azalea mollis /3. glabrior Regel in Gartenfl. XVI. 289, t. 556 (1867). Van Houtte in Fl. des Serr. XIX. 177 tt. 2034-2035 (1873). Azalea sinensis 0. glabrior Maximowicz in Suppl. Ind. Sem. Hort. Petrop. (1869) 11. Rhododendron Sinense Maximowicz in Mem. Acad. Sci. St. Petersbourg, se'r. 7, XVI. No. 9, 28 (Rhodod. As. Or.) (1870), not Sweet. Hooker f. in 100 THE AZALEAS OF THE OLD WOULD Bot. Mag. XCVII. t. 5905 (1871). Franchet & Savatier, Enum. PL Jap. I. 289 (1875). Matsumura & Yatabe, Nippon Shokub. 163 (1884). Matsumura, PL Nikko, 71 (1894); Shokub. 250 (1895); Ind. PL Jap. II. pt. 2, 464 (1912). Shirasawa, Icon. Ess. For. Jap. II. t. 62, figs. 19-27 (1908). Hayata, Veget. Mt. Fuji, 66 (1911). Ito, Icon. PL Jap. I. No. 5, 1, t. 17 (1913). W. Watson, Rhodod. and Azaleas, 112 (1911), in part. Azalea mollis Andr6 in III. Hort. XVIII. 132, t. 68 (1871), not Blume. Suringar in Gartenfl. LVII. 505 (1908). Japan: Hondo, prov. Suruga, Mt. Fuji, alt. 600-1000 m., open country, May 8, 1914, E. H. Wilson (No. 6657); same locality, July 28, 1891, K. Watanabe (Herb. Gray); prov. Shinano, Karuizawa, alt. 1000 m. May, 1903, B. Hayata; prov. Shimotsuke, between Lake Chuzenji and Yumoto, open country, alt. 1600-1800 m., June 24, 1914, E. H. Wilson (No. 6872); same locality, September 5, 1892, C. S. Sargent; same locality, August 11, 1905, J. G. Jack; Nikko re- gion, alt. 500-18QO m., May 22, flowers, October 19, fruit, 1914, E. H. Wilson (Nos. 6727, 7670) ; prov. Uzen, foot of Adzuma-san, grassy slopes, alt. 1000 m., October 23, 1914, E. H. Wilson (No. 7192); without locality, June 7, 1913, K. Sakurai; prov. Rikuchu, round base of Hayachine-san, June 4, 1905, U. Faurie (No. 6784). Hokkaido, prov. Oshima, Hakodate, cultivated, 1855, C. Wright (Herb. Gray). Cultivated: Arnold Arboretum (Nos. 861-1, 861-4, 861-5, 861-6); Hort. Kew, May 18 (flowers), August 7 (leaves), 1880, G. Nicholson (No. 807, Herb. Arnold Arboretum). This handsome species is common over a great part of Hondo, the main island of Japan. I am familiar with it from the neighborhood of Kamo, on the Kwan-sai railroad beyond Nara, northward to the foot-hills of Hayachine-san. It is a special feature of the moorlands round the base of sacred Fuji-san and of Ontake-san, and of open grass and scrub-clad places in the Nikko region and elsewhere. Matsu- mura reports it from Higo province in Kyushu but I have not seen it wild anywhere in this southern island. It is cultivated in and around Hakodate in Hokkaido but I can find no record of its growing wild in that island. Wherever I have seen this Azalea wild it was scattered in open grass and scrub-clad country, never in woods or dense thickets and obviously it is sun-loving. It is a sturdy-growing plant, with fairly stout, erect shoots, as much as two metres high and a metre through; usually, however, it is less than a metre high and broad. The flowering season is from about the last week in April to the end of June, according to locality. The flowers appear before the leaves, and the corolla varies in color from orange-red to flame- red or almost red; they are broad funnel-shape, from 5 to 6 cm. across and the flowers are produced 6 to 12 together in umbellate corymbs at the end of shoots. The calyx is small, obtusely 5-lobed with many (rarely few) straight, bristle-like gray hairs; the flower-stalk and annual shoot are also clothed with similar bristles, though occasionally these are wanting on the foot-stalks. The outside of the corolla is velutinous. The stamens, five in number and of unequal length, are shorter than the corolla, with pilose filaments and dark brown anthers. The ovary is villose, and the fruit is erect, oblong-ovoid to cylindric, from 2 to 2.5 ENUMERATION OF THE SPECIES 101 cm. long, grooved, shining brown, often bloomy in color, pilose and villose with scattered hairs, and is subtended by the persistent calyx. The winter-buds are ovoid, acute or subacute, gray-brown and very slightly puberulous, but the scales are margined with a row of white, nearly straight hairs. The leaves are more or less spatulate to oblanceolate, from 5 to 10 cm. long and from 1.5 to 3 cm. broad; when mature they are green on both surfaces, strongly veined and have scattered, appressed, bristly hairs pointing toward the apex of the leaf on the upper surface and on the under-surface of the primary and secondary veins; the margins are ciliate, the apex is tipped by a glandular mucro and the base is narrowed to the short petiole. When unfolding, both surfaces of the leaf, but more especially the lower, are often sparsely covered with a short, soft pubes- cence, but this is soon thrown off. In late autumn the leaves are quite glabrous except for the appressed bristles and the cilia on the margins, and change to red and red-purple. The character of the leaves is important, since it affords an easy and accurate means of distinguishing the species from its Chinese relative (R. molle G. Don). The history of the introduction of this Japanese Azalea into western gardens is considerably involved. A Hollander, J. B. Groenewegen, in Sempervirens, XIX. 532 (1890), states that he received seeds of this plant direct from Japan in 1861, and raised many plants which he afterward sold to a Belgian friend, by whom they were distributed among Belgian nurserymen, including Louis van Houtte. Later these plants were hybridised with other cultivated species and varieties and many beautiful plants were raised. In 1863 Maximowicz introduced seeds to Petrograd, where plants were raised which flowered and were figured by Regel in the Garten- flora, XVI. 289, t. 556 (1867). In 1869 seeds were distributed from Petrograd. Hooker in figuring it in the Botanical Magazine, XCVII. t. 5905 (1871), states that the specimens were communicated by William Bull, nurseryman, Chelsea, with whom it flowered for the first time in March, 1870. Seeds were received at the Arnold Arboretum on April 3, 1876, from the Jardin des Plantes, Paris, and plants flowered in Professor Sargent's garden in the spring of 1880. Seedlings raised from this stock flowered in the Arboretum in 1884. In 1890 Dr. W. S. Bigelow sent seeds from Japan to this Arboretum, and in 1892 Professor Sargent also sent seeds of it in quantity. Plants raised from this latter source have for many years been well established here and seeds from them have been widely distributed. In the Arnold Arboretum R. japonicum is perfectly hardy, grows freely and vigorously and flowers profusely every season. With its large trusses of bright to dull flame- colored flowers it is strikingly handsome and is worthy of a place in every garden. That in the past it has been neglected is undoubtedly due to its having been con- fused with its less hardy Chinese relative and with hybrid and seminal forms of both species. The error was started by Siebold and Zuccarini in 1846 and Miquel followed them; Gray was first to give the Japanese plant specific rank under a new name. Miquel, Regel and Maximowicz in the 60's all recognised it as distinct from the Chinese plant and ranked it as a variety. After this for about forty years botanists merged it under the Chinese species and until Suringar in a critical, painstaking account in the Gartenflora, LVII. 516 (1908) established its right to be considered a good species. From even a cursory glance at the literature it is evi- dent that the nurserymen of Belgium and Holland, at any rate, all along recognised the two plants as distinct. The Japanese name for R. japonicum is " Renge-tsutsuji," and a form with slightly different colored flowers is " Beni-renge-tsutsuji." To the latter the name var. rosea has been given by Ito (Icon. PL Jap. I. No. 5, 4 [1913]). The colors indicated by the two Japanese names are merely extremes and merge in- extricably into each other. Though this Azalea is now grown in some quantity in 102 THE AZALEAS OF THE OLD WORLD the nurseries round Tokyo and Osaka for export it is but little used by the Japanese in their own gardens, probably because it is not amenable to clipping. In western gardens the first to raise and name varieties of R. japonicum from seed was Louis van Houtte of Ghent, Belgium, who published (in the FL des Serr. XIX. 177 [1873]) an article, accompanied by four plates. Enthusiastic on the future possibilities of this plant, he named and described twenty kinds. Among these are some of the best seminal kinds of this Azalea yet produced. The first of them to attain popularity was " Comte de Gomer," which is figured in the Revue Horti- culture Beige, I. 97, t. (1875), by Kerchove de Denterghen under the name of Azalea mollis " Comte de Gomer." In 1879, Messers James Veitch & Sons exhibited it be- fore the Royal Horticultural Society under the same name and it received a First Class certificate. It is figured by Richard Dean in the Floral Magazine, XVIII. t. 367 (1879), and is still among the best kinds raised. Others of the same stock are " Alphonse Lavalle"e," " Baron Edmond de Rothschild," " Comte Papadopoli," " Consul Pdcher " and " Charles KSkuleV' In his article Van Houtte makes no mention of using the Chinese plant but his figure 2033 suggests that plant. His t. 2035, depicting white flowers with yellow blotch on upper petal is presumably the variety " Chevalier A. de R6ali," and suggests an albino form of the Chinese plant rather than of the Japanese species. I have seen no specimen of this, but the leaves should tell its origin. In this connection, however, it is well to remember that although only recently definitely known, the yellow-flowered form (f . aureum) of the Japanese species might have been raised from the seedlings distributed by Groenewegen or from the Petrograd stock. I saw the f . aureum in the British Con- sulate garden in Hakodate in 1914 and the plant might well have been growing somewhere in the town when Maximowicz was there in 1861. In reading Van Houtte's account the inference certainly is that all the forms he mentions and figures are of the Japanese plant, and Van Houtte, of course, knew the Chinese plant well. At this late date it is not possible to be sure, but I incline to the belief that the yellow-flowered form of the Japanese species appeared among the seed- lings raised from Groenewegen's or Maximowicz's seeds. Rhododendron japonicum f. aureum Wilson, n. f . Rhododendron sinense Tanaka, Useful PL Japan, 74, No. 652, fig. (1891), not Sweet. Rhododendron sinense var. flavescens Ito, Icon. PI. Jap. I. No. 5, p. 4 (1913), not Sweet. Japan : Hondo, prov. Musashi, brought from Chichibu mountains, cultivated, Botanic Gardens, Tokyo, May 11, 1917, E. H. Wilson (No. 8414, type); prov. Kawachi, between Yomomoto and Ikedo, cultivated, said to have been brought from round Tokyo, May 8, 1918, E. H. Wilson (No. 10,351). Except for its rich, deep yellow flowers there is nothing to distinguish this* form from the typical R. japonicum. I was informed by Dr. Nakai that this form grows wild mixed with the type on Chichibu mountains in the northwestern part of Mu- sashi province and that the plant my specimens are from came from there. I did not see it wild but Dr. Nakai' s statement was confirmed to me by various horti- culturists. I saw it in gardens in Hakodate in 1914, and in nurseries not far from Osaka in 1918. The winter-buds are pale colored and afford an easy means of dis- tinguishing it from the type. The Japanese name for this plant is " Ki-tsutsuji " or " Ki-renge-tsutsuji " and Tanaka's figure cited above is quite good. As far as I ENUMERATION OF THE SPECIES 103 can discover it was first introduced to America in the spring of 1917 among plants of the type imported from the Yokohama Nursery Company by Mr. J. S. Ames, North Easton, Mass. In 1919 I brought back plants from Japan to the Arnold Arbore- tum. Whether this plant has reached Europe I do not know, but there is no men- tion of it in Millais's recent work on Rhododendrons. In this connection it should be mentioned that Maximowicz (in Mem. Acad. Sri. St. Petersbourg, s6r. 7, XVI. No. 9, 29 (Rhodod. As. Or.) [1870]) speaks of the form with yellow flowers as rare, but it must be remembered that he confuses the Japanese and Chinese species and calls both R. sinense. Maximowicz identifies this plant with the " Rjuku Tsutsusi, lillifera Insulae Luconiae, flore languide luteo, petalis, surrectis, punctis satura- tioribus variegatis " and with " Rjuku Tsutsusi flore igneo, punctis petalorum croceis, apicibus rufis " (Kaempfer, Amoen. Exot. 848 [1712]). What the first named of Kaempfer' s plant is I do not know, neither can I find out the " Insulae Luconiae;" the description, however, would fit our new form (aureum) or the Chinese species. The " Rjuku Tsutsusi fl. igneo " is R. scabrum G. Don (R. sub- lanceolatum Miquel). Personally I doubt if Kaempfer saw or heard of R. japoni- cum or its form aureum, since they are not favorite garden plants in Japan, neither do they grow wild by the road he traversed from Nagasaki to Tokyo. A recently raised hybrid between R. japonicum Suring. and R. canadense Zabel is : X Rhododendron Fraseri W. Watson in Gard. Chron. ser. 3, LXVII. 225 (1920). l Rhododendron luteum Sweet, Hort. Brit. ed. 2, 343 (1830), not Azalea lutea L. Schneider, III Handb. Laubholzk. II. 1046 (1912), not II. 500 (1911). Render in Bailey, Stand. Cycl Hort. V. 2942 (1916). Azalea pontica Linnaeus, Spec. 150 (1753), not R. ponticum Linnaeus, Pallas, Fl. Ross. II. 51, 1. 19 (1788). Andrews, Bot. Rep. 1. 1. 16 (1797). Schmidt, Qster. Baumz. III. 44, t. 169 (1800). Marschall a Bieberstein, Fl. Taur.-Caucas. I. 144 (1808); 111. 136 (1819). -Hovey, Mag. Hort. IV. 276, 367 (1838). De Candolle, Prodr. VII. pt. 2, 718 (1839). t Azalea flava Hoffmannsegg, Verz. Pflanzenkult. 61 (1826). Anthodendron ponticum Reichenbach in Mosseler, Handb. Gewachsk. II. aufl. 1. 309 (1827); Fl. Germ. Excur. I. 416 (1830). Rhododendron flavum G. Don, Gen. Syst. III. 847 (1834), not Pallas. Nyman, Consp. Fl. Europ. 491 (1878-82). Schneider, III Handb. Laubholzk. II. 499, figs. 328 n-q, 329 e-h (1911). Bean, Trees and Shrubs Brit. Isl II. 356 (1914). Millais, Rhodod. 165 t. (1917). Boissier, Fl. Or. III. 972 (1875). Kuzentzof, Fl. Caucas. Crit. IV. pt. 1, 31, 488 (1901, 1906). Rhododendron flavum var. coronarium Sweet, Brit. Flow. Gard. ser. 2, IV. t. 331 (1836). Azalea arborea Linnaeus apud De Candolle, Prodr. VII. pt. 2, 718 (1839), as a synonym. Rhododendron ponticum Schreber apud De Candolle, Prodr. VII. pt. 2, 718 (1839), not Linnaeus, as a synonym. Azalea Pontica a. flava De Candolle, Prodr. VII. pt. 2, 718 (1839). Azalea pontica var. autumnalis K. Koch in Linnaea XVII. 281 (1843). Anthodendron flavum Reichenbach apud K. Koch, Dendr. II. pt. 2, 184 (1872), as a synonym. Schneider, III. Handb. Laubholzk. II. 499 (1911). Azalea pontica "Goldlack" Hesse in Gartenfl. LII. 48 (1903). 1 See page 185. 104 THE AZALEAS OF THE OLD WORLD Caucasus, near summit of Dariel Pass, south side 2000 m. alt., July 24, 1903, C. S. Sargent; Vladikavkaz, April 22, 1902, A. Famine; Mt. Beschtau, June, July, R. F. Hohenacker; prov. Ossetia, Alagir, swamps, May 10, Sept. 12, 1901, R. Marcowicz. Pontus region, "Ci- ganadagh," July 20, 1889, P. Sintenis (No. 1357); Sumila, in woods, July 31, 1889, P. Sintenis (No. 2017). Cultivated: Arnold Arboretum (No. 3159-1) from seeds collected on the Caucasus by C. Schneider, June, August 28, 1917, A. Rehder; Hort. Holm Lea, May and September, 1880; Bot. Gard. Cambridge, Mass., May 31, 1916, A. Rehder; Hort. Kew, May 19, August 9, 1880, G. Nicholson (No. 835); Breslau, Schneitniger Park, June 13, July 23, 1902, C. Baenitz; Hort. Bot. Gottingen, May 22, 1893, A. Rehder (No. 1574). This well-known species is geographically isolated from its relatives and has a rather remarkable distribution. It is native of the Caucasus Mountains, the Pontus region of the Black Sea, of Cilicia, and also of eastern Europe (Lithuania, south- eastern Poland, Volhynia and in southwestern Russia). It is a broad, densely branched shrub from 2.5 to 4 m. high and as much as 6 m. in diameter. The branches are twiggy and clothed when young with glandular, villose hairs. The leaves are scattered, oblong-lanceolate often broadest above the middle, from 4 to 12 cm. long and from 1.5 to 3.5 cm. long, acute or obtuse, mucronate, narrowed to the base, the margin is serrulate and ciliate, and on both surfaces are scattered, appressed, short glandular bristles and short curled pale gray pubescence which later disappears. The petioles, calyx and outside of the corolla tube are glandular. The richly fragrant flowers are borne from 10 to 12 together in clusters at the end of the naked shoots. The calyx is green, deeply divided into 5 oblong or ovate, glandular-ciliate lobes from 2 to 5 mm. long. The corolla is yellow, from 4.5 cm. in diameter with a narrow, cylindric tube from 1.5 to 2 cm. long. The stamens are long exserted but overtopped by the pistil which has a glandular ovary, slender style and capitate stigma. The fruit is cylindric, from 2 to 2.5 cm. long and from 0.5 to 0.8 cm. wide, furrowed and glabrescent. This Azalea was discovered by Tournefort on the eastern side of the Black Sea during his voyage to the Levant (1700-02) and is described in his Corollarium In- stitutionum rei herbariae, p. 42 (1703), as " Chamaerhododendros Pontica, maxima, Mespili folio, flore luteo "; he also figures it in Mimoires de VAcademie royale des sciences, 348 (1704). A long account of it is given in his Voyage du Levant, II. 99 (1718). Buxbaum in his Plantarum minus cognitarum ceniuria V. 36, fig. 69 (1740) gives a crude figure of it under Tournef ort's name. It was on this description and figure that Linnaeus based his Azalea pontica. The plant was introduced into England from the Caucasus by Pallas, who sent seeds to Messers Lee & Kennedy of Hammersmith in 1792, with whom they germinated the following year. In 1796, Mr. Anthony Hove of Warsaw found this Azalea on the north side of the Black Sea near Oczakow in marshy places, on the banks of the rivers Dnieper and Dniester, and on the Asia Minor side of the Black Sea round Trebizond. Mr. Hove sent seeds to Watson, a nurseryman of Islington, and plants flowered in a greenhouse in 1798. According to Schmidt it was found in Volhynia by Dr. Besser, and was in- troduced to the Botanic Garden at Vienna before 1800. I have not been able to ascertain exactly when it was introduced into this country, but according to Hovey ENUMERATION OF THE SPECIES 105 it was a familiar plant in gardens round Boston, Mass., in 1838. In this Arboretum the Pontic Azalea is not properly hardy. This Azalea has been much used in hybridising and crosses between it and various American species have originated the popular " Ghent Azaleas " of gardens which should be known as x R. ganda- vense Render. 1 According to Millais (Rhodod. [1917]) A. Waterer has crossed R. luteum and R. molle and such a hybrid is figured by Millais opposite page 198. A variety with large flowers is : Rhododendron luteum var. macranthum Wilson, n. comb. Rhododendron flavum var. macranthum Bean, Trees and Shrubs Brit. Isl. II. 357 (1914). Millais, Rhodod. 165 (1917). The flowers of this variety are described as being 6.5 cm. wide. There is also the obscure: KHODODENDRON LouREiRiANA G. Don, Gen. Syst. III. 846 (1834). Maximowicz in Mem. Acad. Sci. St. Petersbourg, se*r. 7, XVI. No. 9, 49 (Rhodod. As. Or.) (1870). Millais, Rhodod. 204 (1917). Azalea punctata Loureiro Fl. Cochinch. 113 (1790). This is said to be native of Cochinchina and was described with lanceolate, glabrous leaves scabrous at the margin and with a white corolla dotted with red, as are also the calyx, anthers and ovary. No Azalea answering this description has since been found, and very probably Loureiro made a mistake in referring his plant to that genus. These are all the Azaleas at present known from the Old World but Dumont de Courset (Bot. Cult. ed. 2, III. 337 [1811]) mentions Azalea fimbriata, A. lugens and A. vittata. He gives no descriptions but says the first two are from China, the third from the Indies, and that they are growing in English gardens. Probably they are referable to R. Simsii or R. indicum but it is impossible to identify them with certainty. In Iwasaki's Japanese Herbal, Phonzo Zoufou, issued in 1828, a number of colored pictures of Azaleas are given. These figures are so crudely drawn that I find it impossible to determine, with any degree of certainty, what species and forms they represent. However, in the new edition, now being issued under the editorship of Messers Shirai & Onuma, technical names are given, and in Volume XX (Poisonous Plants, pt. 3) K. Onuma makes three new combina- tions. If his determinations are correct his R. macranthum var. polypetalum should be referred as a synonym to R. indicum f . polypetalum Wile., his R. macranthum f . semperflorens to R. indicum f . variegatum DC., and his R. Nakaharai var. line- arifolium to R. linearifolium S. & Z. 1 See page 178. THE AZALEAS OF NORTH AMERICA BY ALFRED REHDER THE AZALEAS OF NORTH AMERICA INTRODUCTION THE Azaleas, as the term is generally understood, are the deciduous species of the genus Rhododendron, which, as far as concerns the American species, seem well separated from the evergreen Rhododendrons; this explains to some extent the tend- ency of many recent American botanists to consider the two groups generically distinct, but if we take into consideration the Asiatic species, the differences become less pronounced, and therefore it seems better to limit the genus as proposed by Salisbury, D. and G. Don, Maximowicz and most recent authors. In dealing with the American Azaleas we restrict the term to the species of the sections Pentanthera and Rhodora which both belong to the same subgenus and exclude Azaleastrum and Therorhodion, though they, too, have deciduous leaves, but they differ considerably in their inflorescence and are best considered distinct subgenera; they are, moreover, not Azaleas in the usual sense of the word. The subgenus Azaleastrum differs chiefly in its usually one-flowered lateral inflorescences, and Therorhodion in the flowers being borne in terminal racemes at the end of leafy shoots, while in the subgenus Anthodendron to which Pentanthera and Rhodora belong the flowers are borne in leaf- less umbel-like racemes, or rarely solitary, at the end of the shoots of the previous season. I prefer to use the name Antho- dendron first proposed by Reichenbach as a generic name and later used by Endlicher as a subgeneric name instead of Azalea, as it is an older subgeneric name and as, moreover, Azalea of Linnaeus upon which that subgeneric name is based does not belong here but to another genus, as explained on page 116 under Anthodendron. A short history of the genus, as far as it concerns the American species with which we are dealing, may be given here; more detailed accounts will be found under the different species. The 109 110 THE AZALEAS OF NORTH AMERICA first American species recorded is Rhododendronviscosum, of which we find a figure published in 1691 by Plukenet (Phytogr. 1. t. 161, fig. 4) under the name Cistus virginiana flore et odore Peri- clymeni after a drawing made from nature by John Banister, an English missionary in Virginia who took great interest in botany and sent in 1680 a catalogue of Virginian plants to Ray. A few years later Plukenet (Mant. 49) mentioned the plant now known as R. nudiflorum under the name Cistus virginiana Periclymeni flore ampliore minus odorato. Both species were introduced into England by Peter Collinson before 1730 and of R. nudiflorum a good colored plate was published by Trew in 1750 (PL Select. Ehret. t. 48), while the first colored plate of R. viscosum was pub- lished by Meerburgh in 1798 (PL Select. Icon. t. 9). Linnaeus when publishing his Species plantarum in 1753 knew only these two species which he referred to the genus Azalea and named A. viscosa and A. lutea, changing the latter name to A. nudiflora in the second edition of his work. In the same edition, in 1762, he described Rhodora canadensis introduced some time before the year 1756 from Canada to Paris and figured first by Duhamel in 1760 (Sem. PL Add. 10, t. 27, fig. 2). In 1763 Adanson (Fam. PL 164) proposed the name Tsutsusi based on Kaempfer's t. 846 which is R. obtusum var. Kaempferi, and cites " Azalea L." as a synonym; though not an American plant, this is mentioned here, because Small in the North American Flora considers A. indica L., under which Linnaeus cites Kaempfer's tab. 846, the type of the genus Azalea. In 1766 Crantz (Inst. II. 468) pro- posed the new name Hochenwartia for Rhodora, because it was not the Rhodora of Pliny. Gmelin (Syst. Nat. II. pt. 1, 694) in 1791 united Rhodora with Rhododendron, and Salisbury in 1796 (Prodr. 286) first recognized the close relationship of most Azaleas to the genus Rhododendron; he referred the American Azaleas and Rhodora to the genus Rhododendron, retaining the name Azalea for A. procumbensj one of the two species upon which Linnaeus founded the genus. In 1803 Michaux (FL EOT. -Am. I. 150) described two new species, A. calendulacea and A. canescens; the former had been mentioned as long before as 1749 by Golden but without sufficient description. It was men- tioned again hi 1791 by W. Bartram, who discovered it in 1776 and describes its beauty in glowing terms. The second species is possibly identical with one of Aiton's varieties of A. nudiflora INTRODUCTION 111 and its introduction would be in that case prior to 1789. In 1811 Willdenow (Berl. Baumz. ed. 2, 49) described from cultivated plants A. speciosa identical with one of Aiton's varieties of A. nudi flora and the following year Loiseleur Deslongchamps (Nouv. Duhamel. V. 224) described A. rosea also from culti- vated plants. Pursh in 1814 describes as a new species A. arborescens and raises three of Aiton's varieties of A. viscosa and one of A. nudiflora to specific rank. In 1822 D. Don (in Edinb. Philos. Jour. VI. 47) transferred, like Salisbury twenty- six years before, the species of Azalea except A. procumbens to Rhododendron. In this he was followed by Torrey, G. Don, Endlicher, Maximo wicz, Bentham & Hooker and most recent authors, except that Maximowicz, Bentham & Hooker and later authors took up Desvaux's name Loiseleuria for Azalea procumbens. In 1827 Reichenbach, who like the two Dons and Endlicher retained the name Azalea for A. procumbens, proposed the genus Anthodendron (in Moessler, Handb. Gewachsk. I. 244) for A. pontica, A. nudiflora and A. viscosa. G. Don (Gen. Syst. III. 84) in 1834 was the first botanist to subdivide the genus Rhododendron into sections and he proposed the sections Pentanthera and Rhodora with which we are here concerned. After 1814 no new species were added until the year 1855, when A. calif ornica Torr. & Gray (Rh. occidentale Gray) from California was described; it was introduced by Lobb about 1850. The very distinct R. Vaseyi was discovered in 1878 by George Vasey and described the following year by A. Gray; it was in- troduced into cultivation about the same time. In 1891 Greene (m.Pittonia, II. 172) described R. sonomense as a new species, but this is only a form of R. occidentale. In 1892 Otto Kuntze, who took the year 1737 as the starting point for generic nomen- clature and consequently considered Azalea the older generic name, transferred all the species of Rhododendron known to him to Azalea. In 1901 Small (in Bull. Torr. Bot. Club, XXVIII. 360) described as a new species A. Candida now referred to R. canescens as a variety, and in 1903 (Fl. S. E. U. S. 883) he described A. oblongifolia and A. serrulata and separated R. Vaseyi as a distinct genus under the name Biltia. In 1913 (I. c. ed. 2, 1356) he adds two new species A. austrina and A. pruni- folia, and in 1914 (in N. Am. Fl. XXIX. 42) A. prinophylla identical with A. rosea Loisel. (R. roseum Rehd.). In 1917 Ashe 112 THE AZALEAS OF NORTH AMERICA (in Bull. Charleston Mus. XIII. 26) published A. atlantica as a new species and three years later A. neglecta (in Bull. Torr. Bot. Club, XLVII. 581 [1920]) here referred to R. atlanticum as a form. One new species and several new varieties are described in this paper. Of these recent species R. austrinum was introduced into cultivation in 1914, R. oblongifolium in 1917 and R. serrulatum and R. prunifolium in 1918, all through the agency of the Arnold Arboretum. Rhododendron atlanticum has been cultivated for several years by W. W. Ashe in Washington and by Professor Coker in the Arboretum of the University of North Carolina at Chapel Hill. Rhododendron alabamense is the only species not yet in cultivation. Soon after the introduction of the first American Azaleas to England about 1730 they become favorite garden plants and at the beginning of the last century there were already several species and many varieties in cultivation which played an important part among the so-called American plants. By hybridization between themselves and with the Rhododendron luteum introduced at the end of the 18th century and also with evergreen Rhododendrons the garden forms began rapidly to increase in numbers and in 1836 Loddiges cultivated in his nursery near London, according to his catalogue, 107 varieties and hybrids of R. viscosum and 43 varieties of R. nudiflorum and several other species and varieties. The delicate coloring of some and the brilliancy of 'others, combined with a delicate fragrance in several species and the hardiness of almost all these Azaleas, made them very desirable and favorite garden plants both in England and on the Continent. Hybridization in the seventies with R. japonicum added a new race of hybrids which are per- haps even more showy in some respects but weaker in constitu- tion than the older hybrids generally known as Ghent Azaleas. It is, however, doubtful if any of these hybrids surpass in the brilliancy of their flowers the true species like R. calendulaceum and R. speciosum which had almost disappeared from European gardens until recently reintroduced. The American Azaleas are chiefly confined to the Appalachian Mountains and the adjoining regions and to the Atlantic and Gulf coast region; only one species, Rhododendron occidentale,is found in the Pacific states from southern Oregon to southern California. No Azalea occurs south of the United States bound- INTRODUCTION 113 ary; the most southern species are R. serrulatum which reaches middle Florida and R. canescens which extends in Texas nearly as far south. In the North R. canadense is found as far north as Labrador. The western limit of the Atlantic Azaleas runs from southern Quebec to northeastern Ohio, then through southeast- ern Indiana, extreme southeastern Missouri, central Arkansas and extreme eastern Oklahoma to the coast of eastern Texas. The greatest development of the group is on the Appalachian Mountains and as the centre of segregation we must consider Georgia, where eight species, half of all American species, occur. The Azaleas are shrubs from one half to two metres or rarely to 7 metres in height and are essentially inhabitants of the temperate deciduous forest; they are mostly moisture and shade loving shrubs and most species are found on the wooded slopes of hills and mountains growing in scattered individuals inter- mingled with other shrubs and trees. Like most Ericaceae they are calciphobous and therefore are absent from limestone regions, although R. roseum is found in western New York and R. oblongifolium in Oklahoma on limestone soil. In spring when in bloom they form a most conspicuous feature of the woods, particularly species like R. calendulaceum on the mountains from Pennsylvania to Georgia with its brilliant orange or scarlet flowers and R. speciosum in Georgia with bright red or scarlet flowers; also R. nudiflorum, R. roseum and R. canescens with white to pink flowers are conspicuous, the latter being widely distributed in the woods of the southern coastal plain. The late flowering species, as R. viscosum and R. serrulatum, are less showy but deliciously fragrant and inhabit chiefly swamps and moist woods of the coastal plains from New England to Louisiana, though the northern species R. viscosum also grows on the mountains. The related R. arbor escens with white and fra- grant flowers is likewise a mountain plant. Only R. atlanticum is a predominant feature over large areas as it grows in open pine woods of the coastal plain chiefly in North and South Carolina. Most species have a wide distribution, particularly R. nudiflorum, R. canescens, R. viscosum and R. serrulatum', others, like R. speciosum, R. oblongifolium and R. atlanticum, occupy an area not more than about two to three hundred miles in length and R. Vaseyi, R. alabamense and R. pruni- folium are more limited in distribution. 114 THE AZALEAS OF NORTH AMERICA The present study of American Azaleas is based chiefly on the collections of the Arnold Arboretum. For many years Professor Sargent has taken special interest in Azaleas and has succeeded in assembling a large amount of material, particularly from the southern states usually poorly represented in most herbaria. Besides having this material at my disposal, I have consulted the collections of the Gray Herbarium at Cambridge and of the National Herbarium at Washington and the herbaria of the New York Botanic Garden, the Academy of Natural Sciences of Philadelphia, and of the New England Botanical Club ; specimens for further study were kindly loaned to me from most of these institutions and from the herbaria of the State University of Ohio, of the Missouri Botanical Garden and of the Academy of Sciences at Rochester. I also am under obligation for specimens to Professor W. C. Coker of the University of North Carolina, to Dr. K. M. Wiegand of Cornell, Mr. W. W. Ashe of Washington and Mr. John Dunbar of Rochester. I have further had the op- portunity to see most of the species in their native habitats dur- ing various journeys, and the living collections of Azaleas of this Arboretum supplied me with material to study many of the species and a large number of garden forms and hybrids in the living state. The great variability of most species together with the absence of strong morphological characters within this group makes it difficult to assemble the numerous forms into well defined species and varieties. The difficulty is considerably in- creased if one tries to elucidate the innumerable garden forms and hybrids, as few of these forms are adequately described or figured and herbarium or living material of only a small per- centage of them is available. The Azaleas form a distinct subgenus of the Rhododendron for which we propose to use Endlicher's subgeneric name Antho- dendron, as we do not believe that the name Azalea used for it by most authors can be correctly applied to this subgenus, as will be seen from the following enumeration of synonyms and the appended remarks. THE SUBGENUS ANTHODENDRON ENDL. Rhododendron subgen. Anthodendron Endlicher, Gen. 759 (1839), emended. Azalea Linnaeus, Spec. 150 (1753), as to species 1-4. Desvaux in Jour. Bot. Appl. I. 35 (1813). Roemer & Schultes, Syst. IV. 374 (1819), ex- cluding species 12 and 13. Gray, Man. 268 (1848). K. Koch, Dendr. II. 1, 171 (1872). Render in Bailey, Cycl. Am. Hort. I. 119 (1900), ex- cluding A. albiflora, semibarbata, kamschatica. Britton & Brown, III. Fl. 11.558 (1897). Britton, M an. 698 (1901). Small in N. Am. Fl. XXIX. 41 (1914). Anthodendron Reichenbach in Moessler, Handb. Gew&chsk. I. 244, 308 (1827); FL Germ. Exc. 416 (1831). Rhododendron sect. Pentanthera G. Don, Gen. Syst. III. 846 (1834). See also page 124. Rhododendron a. Anthodendron, Endlicher, Gen. 759 (1839). Rhododendron subgen. Azalea Planchon in Rev. Hort. 1854, 43; with Tsusia as distinct subgenus. Maximowicz in Bull. Acad. Sci. St. PStersbourg, XV. 229 (1870); in Mem. Acad. Sci. St. Pttersbourg, s. papilionaceum Sweet, Hort. Brit. ed. 2, 344 (1830). G. Don, Gen. Syst. III. 847 (1834). This form is described by Aiton as having pale reddish flowers with the lower lobe white, and a foliaceous calyx. AZALEA NUDIFLORA 77. PARTITA Aiton, Hort. Kew. I. 203 (1789). De Candolle, Prodr. VII. 717 (1838). Azalea periclymenoides f. partita Pursh, FL Am. Sept. 152 (1814). Rhododendron nudiflorum x- partitum Sweet, Hort. Brit. ed. 2, 344 (1830). G. Don, Gen. Syst. III. 847 (1834). This form is described by Aiton as having pink flowers 5-parted to the base. This is apparently an abnormal form corresponding to Kalmia latifolia f. mon- strosa. It seems to have soon disappeared from cultivation. AZALEA NUDIFLORA \l/. PARVIFLORA Sweet, I. c. 265 (1826), name. Rhododendron nudiflorum j. parviflorum Sweet. L c. 344 (1830), "blush." AZALEA NUDIFLORA PERICLYMENOIDES Loddiges, I. c. (1838), name. AZALEA NUDIFLORA X. POLYANDRA De Candolle, Prodr. VII. 717 (1838). Wood, Classb. Bot. 490 (1870). Azalea periclymenoides ij. polyandra Pursh, Fl. Am. Sept. 152 (1814). 200 THE AZALEAS OF NORTH AMERICA Rhododendron nudiflorum 17. polyandrum G. Don, Gen. Syst. III. 847 (1834). ? Azalea octandra Michaux, Journal, ed. C. S. Sargent, 91 (in Proc. Am. Philos. Soc. (1889)), name only. This form is described by Pursh as having short rose-colored flowers with 10 to 20 stamens and is said to have been found near Philadelphia. Michaux collected his Azalea octandra in or in the neighborhood of Perry County, Pennsyl- vania. AZALEA NTTDIFLORA w. PROLIFERA Sweet, I. c. 265 (1826), name. Rhododendron nudiflorum . proliferum Sweet, I. c. 344 (1830), "blush." AZALEA NUDIPLORA aa. PUMILA Sweet, I. c. 265 (1826), name. Rhododendron nudiflorum aa. pumilum Sweet, I. c. 344 (1830), "blush." AZALEA NTJDIFLORA /3/3. PURPURASCENS Sweet, 1. c. 265 (1826), name. Loddiges, I.e. 1142(1838). Rhododendron nudiflorum /3j8. purpurascens Sweet, I. c. 344 (1830), "flesh- colored." AZALEA NUDIPLORA 77. PURPUREA Sweet, I. c. 265 (1826), name. Loddiges, I. c. (1838). Rhododendron nudiflorum 77. purpureum Sweet, I. c. 344 (1830), "purple." AZALEA NUDIPLORA n\ RUBERRIMA Sweet, I. c. 265 (1826), name. Loddiges, I c. 1142 (1838). Rhododendron nudiflorum ff . ruberrimum Sweet, L c. 344 (1830), "red." AZALEA NUDIPLORA 1777. RUBESCENS, Sweet, L c. 265 (1826), name. Rhododendron nudiflorum 1717. rubescens Sweet, L c. 344 (1830), "pink." AZALEA NUDIPLORA XX. RUPA Sweet, L c. 265 (1826), name. Loddiges L c. 1142 (1838). Rhododendron nudiflorum XX. rufum Sweet, L c. 344 (1830), "red." AZALEA NUDIPLORA w. SEMI-DUPLEX Sweet, L c. 265 (1826), name. Rhododendron nudiflorum vv semiduplex Sweet, L c. 344 (1830), "white." AZALEA NUDIPLORA /3. RUTILANS Aiton, Hort. Kew. I. 203 (1789). De Can- dolle, Prodr. VII. 717 (1838). Azalea rubicunda Dumont de Courset, Bot. Cult. ed. 2, III. 332 (1811), as var. of A . nudiflora. Azalea periclymenoides /3. rutilans Pursh, Fl. Am. Sept. 152 (1814). Rhododendron nudiflorum . rutilans Sweet, Hort. Brit. ed. 2, 343 (1830). G. Don, Gen. Syst. III. 847 (1834). Azalea nudiflora rubicunda Loddiges Cat. 1836 apud Loudon, Arb. Brit. II. 1142 (1838). Azalea nudiflora Be. rubicunda Sweet, Hort. Brit. 265 (1826). Rhododendron nudiflorum 66. rubicundum Sweet, Hort. Brit. ed. 2, 343 (1830). This form is described by Aiton as having deep red flowers with small calyx. Dumont de Courset says that the young branchlets and the leaves are pubescent, the calyx greenish brown and the winter-buds pale yellow in autumn. AZALEA NUDIPLORA SEROTINA Loddiges, I. c. 1142 (1838), name. AZALEA NUDIPLORA . STAMINEA Sweet, I. c. 265 (1826), name. Loddiges, I.e. 1142(1838). Rhododendron nudiflorum St. stamineum Sweet, I. c. 344 (1830), "white." AZALEA NUDIPLORA oo. STELLATA Sweet, I.e. 265 (1826), name. Loddiges, I. c. (1838). Rhododendron nudiflorum oo. stellatum Sweet, I. c. 344 (1830), "white." AZALEA NUDIPLORA VARIA Loddiges, I. c. 1142 (1838), name. DOUBTFUL NAMES 201 AZALEA NUDIFLORA pp. VARIABILIS Sweet, 1. c. 266 (1826), name. Lod- diges, I.e. 1142 (1838). Rhododendron nudiflorum pp. variabilis Sweet, I. c. 344 (1830), "red, white." AZALEA NUDIFLORA