Harvard Botany Libraries

3 2044 105 172 415

Price (to Non-FeUows) 17s. 6d.

MAY 1921.

Vol. XLVI

THE

JOURNAL

OF TUK

1

lefi

EDITED BY

F. J. CHITTENDEN, F.L.S., V.M.H. & Kev. W. WILKS, M.A., V.M.H.

The whole of the contents of this ' volume are copyright. For permission to reproduce any of the articles application sJiould be made to the Council.

CONTENTS.

M

Lamartine

The Genua Sedum. By R. Lloyd Praeger, B.A. . Index

Magnolias. By P. C. M. Veitch

Garden Roses. By H. R. Darlington ...

An Account of the Sokoto Garden, Nigeria. By Mrs. R. Yates

Plants introduced from S. America. By W. B. Turrill,

First Early Potatos : old and new varieties compared

Contribution from the Wisley Laboratory :

Pollination of Plums. By A. N. Rawes

Antirrhinums tried at Wisley, 1920

Perennial Asters tried at Wisley, 1920

Peas tried at Wisley, 1920

Second Early Potatos tried at Wisley, 1920

Parsley tried at Wisley, 1919-20

Book Reviews

Notes and Abstracts

Extracts from Proceedings: General Meetings, i; Annual General Meeting,!; Report of Council, 1919, iii; Balance Sheet, 1919, xii; Daffodil, XXV ; Chelsea, xxvi ; York Deputation, xxix ; Cardiff, xxx; British-grown Flower Bulb Meeting, xxxil; Lichfield Deputation, xxxiii; Vegetable Meeting, xxxiv ; British-grown Fruit Meeting, xxxiv; Scientific Committee Meetings, xxxvi ; Fruit and Vegetable Committee Meetings, xlvi; Floral Committee Meetings, liv; Orchid Committee Meetings, Ixxx ; Narcissus and Tulip Com- mittee Meetings, xcvi; Books added to Library, 1920, xcix ; Plants, etc. given to Wisley, cv; National Diploma in Horti culture 1920

General Index

PAOB 1

309 315 323

336 346 351

353 357 368 382 390 395 398 405

cix cx

R.H.S. Office for Advertisements and all Communications, VINCENT SQUARE, S.W.I.

SPOTTISV-^- DNDON.

LIBRARY OF THE GRAY HERBARIUM w.c.

Sold by 1

Entered at St

HARVARD UNIVERSITY THE GIFT OF

^cru^ 3±>o.t^ (L<yVL^

GARTERS TESTED SEEDS

PRIMULA STELLATA FAIRY QUEEN. Award of Merit, R.H.S,, 1921.

A CHARMING variety which is somewhat intermediate between the small-flowered and original Stellata type and the larger flowered florists' form. The plants are par- ticularly robust, bearing large graceful sprays of bloom in the greatest profusion. The individual blooms are creamy white and have a pale rose centre, which gives a very light and pleasing effect when seen in a mass.

Price 1/6, 2/6, 3/6 and 5/- per packet.

List of Seeds for Summer Sowing free on application.

Seedsmen to H.M. the King,

Raynes Park, London, S.W.

JOUENAL

OF THE

ROYAL HORTICULTURAL SOCIETY

ESTABLISHED A.D. 1804

ROYAL CHARTERS A.D. 1809, 1860, 1899

EDITED BY

F. J. CHITTENDEN, P.L.S., V.M.H.,

AND

REV. W. WILKS, M.A., V.M.H.

VOL. XLVI.

1920-21

The whole of the Contents of this Volume are Copyright. For permission to reproduce any of the Article application should be mad£- to the Council of the Society.

LONDON

SOLD BY

THE ROYAL HORTICULTURAL SOCIETY, R.H. HALL, VINCENT SQ., S.W. L

0

Published June 6, 1921

Ipr(ntet) for the IRo^al Ibortlcultural Society

BY

SPOTTISWOODE, BALLANTYNE & CO. LTD.i LONDON, COLCHESTER AND ETON

CONTENTS OF VOL. XLVI.

PAGE

An Accou|it of the Genus Sedum as found in Cultivation. By

R. Lloyd Praeger, B.A i

Magnolias. By P. C. M. Veitch, V.M.H 315

Garden Roses. By H. R. Darlington, F.R.H.S 323

Pioneer Work in Nigeria : The Sokoto Gardens. By Rose Lamartine

Yates, L.C.C., F.R.H.S 336

Plants introduced to Horticulture from Chile and Argentina

(including Patagonia and Fuegia). By W. B. Turrill, M.Sc 346

First Early Potatos 351

Contributions from the Wisley Laboratory :

XXVI. Pollination in Orchards. Self-Fertility and Self- Sterility in Plums. By A. N. Rawes 353

Antirrhinums at Wisley, 1920 357

Perennial Asters (Michaelmas Daisies) at Wisley, 1920 370

Early Peas at Wisley, 1920 382

Second-early Potatos at Wisley, 1920 390

Parsley at Wisley, 1919 and 1920 395

Book Reviews 398

Notes and Abstracts 405

Extracts from Proceedings :

General Meetings i, xxv, xxix, xxxj xxxii-xxxv

Annual General Meeting i

Report of the Council for the Year 1919 iii

Annual Revenue and Expenditure Account xii, xiii

Balance Sheet Vincent Square xiv, xv

Annual Revenue and Expenditure Account Wisley

Gardens xvi, xvii

Balance Sheet Wisley Gardens xviii, xix

Trust Funds xx

Schedule of Investments xxiv

Spring Meeting at Chelsea xxvi-xxviii

Deputation to York Gala xxix

Provincial Summer Meeting, Cardiff xxx

Scientific Committee xxxvi

Fruit and Vegetable Committee xlvi

Floral Committee liv

Orchid Committee Ixxx

Narcissus and Tulip Committee xcvi

Books Presented, Purchased, or Reviewed c

Donors of Seeds, Plants, Books, etc cv

National Diploma Examination, 1920 cix

Index cx

NOTICE TO BINDER.

Volume XLVI. has been issued in one part, containing the 'Journal' proper, paged with Arabic figures, and ' Extracts from the Proceedings paged with Romxn figures.

JOURNAL

OF THE

ROYAL HORTICULTURAL SOCIETY.

Vol. XLVL 1921.

AN ACCOUNT OF THE GENUS SEDUM AS FOUND IN CULTIVATION.

By R. Lloyd Praeger, B.A.

CONTENTS. Part I. Introductory.

PAGE

I. Preliminary . . . i II. Historical ... 3

III. Distributional ... 5

IV. Statistical . . .11 V. Variation . . . 13

PAGE

IX. Characters of the Genus . 21

PAGE

VI. Cultivation and Propa- gation . . .16 VII. Sources of material . . 19 VIII. Notes on the text . . 20

PAGE

X. Description of Species . 26 Index ..... 309

Part II. Descriptive.

PART I.— INTRODUCTORY. I. Preliminary.

It is doubtful if any genus of plants which is widely cultivated is in such a confused state in our gardens and horticultural books as is the genus Sedum. Even in collections where the owners pride themselves on correctness of nomenclature, misnomers and nomina nuda abound ; and common species masquerade under many different names.

This does not arise from any special difficulty in the identification of the species of Sedum. Some of the species, it is true, are variable and in some of their forms not at once recognizable by the uninitiated ;

VOL. XLVI. B

JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

but the majority are stable and distinct plants recognizable at a glance, and more easily diagnosed than, for instance, the Saxifrages, which nevertheless, in gardens, are usually more correctly named.

The confusion among the Sedums appears to be due mainly to the fact that some of them are rampant growers which invade the territory of neighbouring plants and overwhelm them. In nurseries this undoubtedly leads to the intruders being sent out sometimes under the names of the species which they have ousted. The smallest scrap of many of these plants in many cases single leaves will take root and grow, and thus pieces accidentally dropped or carried by wind or other agencies may establish the species at a distance from the parent. Again, some of the species of the rupestre group, notably 5. altissimum and S. Douglasii, have a habit of dropping in autumn numerous short barren shoots, which are rolled about by wind and so on, and take root wherever they find a refuge. There is little doubt that these facts go far to account for the numerous names under which common free-growing Sedums, such as album, acre, sexangulare, reflexum, rupestre, anopetalum, altissimum, and spurium are found in gardens. But a large number of misnomers are due to mere care- lessness.

Another regrettable feature as regards the Sedums is the number of nomina nuda names which belong to no described species which are found in connexion with them. Many nurserymen's catalogues are full of such names. Some are clearly perversions, due to carelessness, of well-known names such, for instance, are Crimea- lense for himalense, and glaciate for gracile ; but the majority seem to be deliberate unlicensed christenings. I have given elsewhere * a list of such of these as I have encountered and suffered from and it is to be hoped that they will disappear from our catalogues. Many of them have not even the merit of being applied consistently to any one species.

Another cause of misnaming among the Sedums is the fact that, like most succulents, these plants dry very badly, often losing all their leaves in the process, and unless killed with boiling water con- tinuing to grow for weeks while being pressed ; herbarium material is thus generally poor and unsatisfactory, often almost useless for comparison with the living plants, and identification is rendered correspondingly difficult. Figures of the species thus assume a special value, and many of the Sedums found in cultivation have never been drawn, while figures of many others are found only in publications inaccessible to the majority of gardeners. For this reason I have been at pains to have a drawing made by Miss Eileen Barnes of every species of which I could obtain fresh material. The descriptions likewise have in every instance where fresh material could be obtained been taken from the plants themselves, and checked with the descriptions given by the original describer and by leading authorities.

* Gardeners' Chronicle, 3rd Ser., 56, 334, 191 4.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 3

The cases in which the descriptions or figures are in whole or part not drawn from living material may be summarized as follows : Fresh material not available

S. ruhricaule, S. Hemsleanum, S. japonicum, S. Zentaro- Tashiroi.

Plants which have not flowered with me, or which died before flowering :

5. chapalense, S. cyaneum, S. dendroideum, S. fnitescens, S. Hallii, S. lenophylloides, S. oaxacanum, S. polyrhizum, S. trul- lipetalum.

Description helped out by dried material :

5. Cockerellii, S. glabmm, S. purpureoviride, S. Stevenianum,

With the design of helping those to whom the technical terms of descriptive botany are unfamiliar, I have prefaced the description of each species with a brief note of the characteristics by which it may be distinguished from its nearest allies. I would like to warn readers that reliance on the figures alone may sometimes lead them astray in a genus so large and complicated ; even if the full description of the plant is not used, a careful study of the short note mentioned is quite necessary if pitfalls are to be avoided.

11. Historical.

As might be expected in a genus of which a number of species, of sufficiently noteworthy appearance, grow in regions associated with early civilizations, species of Sedum were known to the ancient naturalists (e.g. 5. Cepaea, S. maximum, 5. roseum), being referred to by Greek and Latin writers ; these and others were likewise known to the medieval herbalists. Coming to the dawn of modem botany, we find 15 species enumerated in the first edition (1753) of Linnaeus' "Species Plantarum," all of these being European except S. Aizoon and 5. hybridum (both Siberian) and S. verticillatum (Japanese, &c.). In the 4th edition (1799) of the same work the number has risen to 29, mainly by the addition of other European species. In 1828 De Candolle (" Prodromus," 3, p. 401) enumer- ates 88 species of Sedum, some of them tentatively as non saiis nota, but almost all now recognized as good species. De Candolle 's list includes a good many additions from the Caucasus, a few from Siberia, the Himalayas, Japan, North Africa, the United States, and Mexico, and one each from Madeira, Ecuador, and Venezuela. In 1862 Bentham and Hooker (" Genera Plantarum," 1, p. 660) put down the number of known species at 120. This total is increased to 130 in standard works published during the next ten or twenty years, and this figure is raised only to 140 in such standard recent works as ENGLERand Prantl, " Naturliche Pflanzenfamilien," iii. a (1891)

4 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

and Dalla Torre and Harms, " Genera Siphonogamarum " (1901). But as a matter of fact, the number of known species has increased far faster than that. In " Index Kewensis " (1885) some 238 species are Hsted (some of which are now regarded as synonyms or varieties) ; the four supplements which have since appeared raise this number (up to the end of 1910) to 391 ; and a fair estimate would put the total number of species at present known at about 500. The large increase during the last half-century is due mainly to the botanical exploration of Mexico and of Western China (see pp. 8, 10).

The great majority of these 500 species are not, and have never been, in cultivation, and are known (especially the many recently described Chinese species) only from dried specimens. A good many of them are annuals, or of no horticultural value. But there remain many handsome or interesting plants, which one would like to see introduced into our collections. As regards the number of species in cultivation. Dr. Maxwell Masters, in his account of the cultivated Sedums * (1878), lists 65 species ; but a critical examination of his list shows that of these only 44 were certainly examined by himself, or, if not seen, were certainly correct. These are all in cultivation still. Four more were apparently seen by him. Eleven not seen by him I regard as doubtfully correct, and probably re- ferable to species already in his list, while finally six of his species are now to be set down as synonyms, additional species erroneously named, or varieties. These last are :

S. arboreum = S. moranense var. arbor eum. S. Beyrichianum = 5. Nevii var. Beyrichianum. S. Maximowiczii = S. Aizoon. S. pruinatum = S. rupestre. S. sarmentosum = S. mexicanum. S. stoloniferum = 5. spurium.

As the first, fifth, and sixth of these are not in his list under their correct names, the number certainly in cultivation, according to his paper, is raised to 47. At least half a dozen tender species (with which his paper was not concerned) were also certainly in cultivation at that time.

As regards the present list, I went further afield than Dr. Masters, and to the best of my ability ransacked the gardens of the world, till the war put an end to such activities. European gardens yielded a good many species unknown to Masters ; others came from the Himalayas, China, and Japan, while important contributions of Mexican species were received from Washington and New York. As a result I have received and grown a total of 151 species, 13 of which proved to be new to science, and have been described.! I know of only four species which I believe to be at present in cultivation,

Masters, M. T., " Hardy Stonecrops : Sedums." Gard. Chron. N.S. 10, 1878, ii.

I Journal of Botany, vols. 55, 56, 57»

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 5

which I have not seen 5. rubricaule at Washington, S. Hemsleanum at St. Louis, and S. Zentaro-Tashiroi and 5. japonicum (var. senanense) at Tokio. Some other species were in cultivation not many years ago for instance, 5. clavifolium, S. delicatum, S. fili- ferum, S. Painteri, S. semiteres, S. suhmontanum at Washington, and S. Englerianum at Dahlem, but they appear to be now lost.

Some details as to the sources from which the species described in the following pages were derived will be found on p. 19.

In addition to species already in cultivation, some species hitherto unknown in gardens were introduced, thanks to the kind offices of correspondents in Asia and America ; and a few others which had been lost to cultivation, such as S. pruinatum from Portugal and S. lancerottense from Teneriffe, were reintroduced and distributed.

III. Distributional.

The genus Sedum is spread in varying abundance throughout the Northern Hemisphere. The majority of the species inhabit temperate countries, or, if found in lower latitudes, have their homes on the mountains, so that most of them are hardy in our gardens. A few species run very far north, and the genus is represented in Iceland, Nova Zembla, Arctic Siberia, Alaska, and Greenland ; these northern forms belong mostly to the section Rhodiola. Southward, a few endemic species are found in the Philippines ; others reach the Equator on the great mountains of Central Africa ; while in America the genus has spread down the backbone of the continent and has crossed the Equator, the most southern outpost being in Bolivia. Over part of its wide range, the genus exhibits well-marked geogra- phical groups, allied species being concentrated in particular areas for instance, the large Rhodiola group in Asia from the Himalayas to China, the Involucrata group in the Caucasus, the rupestre group in Europe ; on the other hand, the rich Sedum flora of Mexico shows extraordinary variety of forms mostly without close relationship. The succeeding paragraphs briefly describe the Sedum flora of the main areas occupied by the genus ; on pp. 22, 23 will be found notes as to the distribution of the phylogenetic groups into which the genus divides itself.

Etirope.

About sixty species of Sedum altogether occur in Europe, the number increasing generally from the north-west to the south-east. The great.bulk of these are representatives of the section Seda Genuina mostly small creeping plants with very thick leaves, and yellow or white flowers. Among them, the well-marked Rupestre group is characteristically European. About one-third of the total are annual plants ; these are mostly southern, and increase eastward to find their maximum in the region extending from Greece to Persia. Of other sections of the genus, three Telephiums occur S. Telcphium,

6 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

5. maximum, and S, Anacampseros, and one Rhodiola, the ubiquitous S. roseum if we except S. quadrifidum, which spreads from Arctic Asia just into Russia. Almost all the perennial species are in culti- vation, though in some cases very rarely ; a few Balkan and Greek plants are yet unknown in gardens. Some of the annual plants are found in gardens, but the pretty blue S. coeruleum is the only one of value.

In our own islands eight species are undoubtedly native namely, roseum, Telephium, album, anglicum, acre, reflexum, rupestre,. villosum. Several others, such as dasyphyllum and sexangulare, are naturalized. Most of our native species have been spread by human agency much beyond their original native limits.

Literature. Nyman, "Conspectus Florae Europeae," and Sup- plements.

Africa.

The Mediterranean littoral yields a number of the familiar species of southern Europe, and also some endemic plants, such as S. multiceps (well known in cultivation) and the curious 5. tuberosum. A few species occur on the mountains of Abyssinia, and one or two others have recently been discovered as far south as the Equator, on Mt. Ruwenzori. R. Hamet reduces * the nine species which have been described from the interior of Africa to five namely, abyssinicum Hamet, Meyeri- Johannes Engler, ruwenzoriense Baker fil., Epiden- drum Hochstetter, sediforme Hamet. None of these is known in cultivation.

The Atlantic Islands.

Madeira yields three species of Sedum S. farinosum Lowe (possibly an extreme form of the European album) and two yellow-flowered species, fusiforme and nudum, apparently related to those of Central Africa and Central America. The Canaries possess 5. lancerottense (closely allied to the Madeiran nudum) and the Mediterranean annual rubens ; possibly also a third species undescribed (a poor specimen in Herb. Kew.). Of the above, nudum and lancerottense are in culti- vation.

The Caucasus.

The Caucasian region is particularly interesting as being the headquarters of two very distinct sections of the genus the group Involucrata of Marschall von Bieberstein, of which the familiar spurium is a characteristic example ; and the still more distinct little group of the Sempervivoides. Most of these are confin&d to the Caucasus, but a few are found in the adjoining regions of Asia Minor or Persia. The Involucrata number half a dozen species, with roundish, fiat, mostly opposite leaves and red or white flowers. Of these, spurium is very widespread in cultivation, with crimson, pink, or

In litt., Herb. Brit. Museum.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 7

white flowers ; stoloniferum is less frequently seen in gardens, and Stevenianum and proponticum almost unknown. The remaining members of the group, ohtusifoUum, Millii, and involucrahim, from the Caucasus, and Baileyi from China, are not, I believe, in cultivation. The Sempervivoides group includes two remarkable biennial plants, 5. sempervivoides and 5. pilosum, which form dense plump leaf-rosettes like those of the genus Sempervivum, and in their second year produce masses of showy red flowers. Both species are now well known in good collections. For the rest, the Sedums of the Caucasus region, which number some twenty in all, include a few familiar European species maximum, album, acre, sexangulare a few small perennials not found elsewhere gracile, tenellum, and suhulatum, the first of which is in cultivation and some little annual species.

Literature. Lipsky, "Flora Caucasica," 1899 (in Russian). Hamet, " Revision des Sedums du Caucase." Trd. Bot. Sada, Tiflis, 8, Part III., 1908.

Note. Syria, Mesopotamia, and Persia yield a number of Sedums, mostly small annuals.

Siberia and Central Asia.

Just as the Caucasus region is the headquarters of the small and distinct group of Sedums of which the familiar spurium is typical, so we find focussed in Eastern Siberia and the northern parts of China and Japan a compact little group of thick-rooted, flat-leaved, yellow- flowered species the Aizoon group. These include five Aizoon, Selskyanum, hybridum, kamtschaticum, and Middendorffianum of which the first and third were known to Linnaeus, and all have been long in cultivation ; and the two more, Ellacombianum and floriferum, lately described by myself from living material. Only two of the group are not in cultivation 5. Sikokianum and S. Yabeanum, both of Japan. For the rest, the Siberian and Central Asiatic Sedum flora is made up mainly of plants of the Rhodiola and Telephium sections, many of which occur, some of them extending far to the northward. But the main centre of the Rhodiola section lies farther south, in the Himalayan region, and that of the Telephiums south-eastward, in China and Japan.

Literature. Maximowicz, ''Diagnoses Plantarum Novarum Asiaticarum." Bull. Acad. ImpSr. des Sciences de St. Petersbourg, 29, 1883, (Reprinted in Melanges Biologiques, 11.)

The Himalayan Region.

The Himalayas are par excellence the headquarters of the Rhodiola section of Sedum ; not that many species of that group are not found in neighbouring regions e.g. Yunnan but in the Himalayas the Rhodiolas are so abundant as to form a feature of the vegetation of the higher grounds, and only few other Sedums occur, while in Yunnan many other species are found. A good many of the earlier discovered

8 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

Himalayan Rhodiolas are in cultivation, and they are interesting plants. Farther north, in Tibet and Afghanistan, some very pecuUar Sedums occur, such as 5. Balfouri, S. Hohsonii, S. Karpelesae, S. pachyclados, which I group with the Rhodiolas. For the rest, the Himalayan and Tibetan flora includes a few of the Japonica series (which find here their western hmit), a few small annuals, and some miscellanea, such as the Telephium S. Ewersii. Altogether close on fifty species are found within this region, almost all of them being perennials ; about a dozen of them are in cultivation.

China.

In Forbes and Hemsley's " Enumeration of the Plants of China " (which included the area extending from Formosa on the south to Korea on the north), published in 1887, 28 species constituted the list of Sedums. The floral wealth of the interior of China was at that time unknown. Since then the extraordinary results of the botanical explorations of Henry, Wilson, Forrest, and the French missionaries have been pubUshed ; hundreds of new plants have been described, and among them are at least 90 new species of Sedum. Most of these are from the inaccessible western provinces, and have been described almost entirely from dried specimens. Very few of them are as yet in cultivation. Many are small plants of the Japonica section, of no great horticultural importance ; but they include a number of Rhodiolas, and some very interesting plants alHed to the section Telephium, for which two new sections of the genus, Pseudorhodiola * and Giraldiina.t have been created ; one species belonging to the first of these groups (S. yunnanense var. valerianoides) is in cultivation.

The earliest Sedums to come to us from China were spectabile and sarmentosum, and up to the present few have followed them. Not more than 30 of the 120 or so species known to occur in China are at present in cultivation. While the Japonica section cannot be expected to yield much of garden value, we may look for some interesting species among the Chinese Rhodiolas. Some of the Chinese Sedums, such as S. Chaneti and S. limuloides^ are very curious plants indeed.

Literature. Forbes and Hemsley, " Enumeration of all the Plants known from China. ..." Journ. Linn. Soc, Bot., 23. R. Hamet, " Enumeration and Description of Species of Sedum (Plantae Chinenses Forrestianae)." Notes R. Bot. Garden Edinb., 5, 115, 1912. L. Diels, " Catalogue of all the Plants collected by George Forrest . . ., 1904, 1905, 1906." Ibid.,1. 1912-3. R. Hamet, " Enumeration of Crassulaceae collected in China" [by many collectors]. Ibid., 8, 139, 1913.

Japan.

In Japan, the latest census (by Matsumura, 1912) puts the Sedum flora at 25 species, which subsequent additions raise to over

* Pseudorhodiola Diels in Engler's Bot. Jahrbiicher, 29 (1901), P- 360. t Giraldiina Diels in Engler's Bot. Jahrbiicher, 36 (1905), Beibl. 82. p. 48.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 9

30. A few of these, such as Aizoon and kamtschaiicum, are plants which have their headquarters in Siberia ; two others, Telephium and roseum (which occurs in the var. Tachiroi), have a much wider range ; but the majority are endemic. A few of them, such as alboroseum, Sieboldii, and spectabile, have long been known in cultiva- tion, the last two being among the handsomest of the garden Sedums. To the Telephium section belongs nearly one-half of the species represented in the Japanese flora, while an equal number belongs to the Japonica section, which consists mostly of smallish plants with yellow flowers. Of the latter, a ternate-leaved species, S. lineare, is in cultivation in our gardens, and two others, japonicum var. senanense and Zentaro-Tashiroi, are reported as in cultivation in Japan. Of the whole Japanese Sedum flora, one-half is known in gardens.

Literature. Matsumura, " Index Plantarum Japonicarum," 2, Part II., 1912. Maximowicz, " Diagnoses Plantarum Novarum Asiati- carum," v., in Bull. Acad. Imp. des Sciences de St. Petersbourg, 29 (reprinted in Melanges Biologiques, 11), 1884. Making, various papers in Bot. Mag., Tokio, &c.

Note. Formosa yields half a dozen Sedums, and the Philippines several. One of the former, S. formosamtm, is included in the present paper.

The United States and Canada.

Sedums are widely scattered throughout North America, but a larger number is found in the mountainous regions of the west than in the east. Two widespread species, ternatum and pulchellum, long grown in European gardens, were described by Michaux in his "Flora Boreali- Americana " as early as 1803. Another plant found in the Eastern States, Nevii, is also long known in British gardens. From the Western States have come two pretty species, spathuli folium and oreganum ; also two reflexum-like plants of less merit, Douglasii and stenopetalum, and the tall and handsome rhodanthum. Many species found in the Western States do not appear to be anywhere in cultivation, and my efforts to procure them have had only a limited success. The polymorphic Rose-root, S. roseum, which has a circum- polar range, is by American botanists restored to its place as a separate genus (Rhodiola) ; it spreads in varying form along the western moun- tains, and has been split up into half a dozen species. Except for 5. roseum s.s. sent me from Washington, I cannot find that any of the American Rhodiola forms are in cultivation. Altogether about 50 species of Sedum (including some of the " spHt " genera) occur in the States, mostly in the south-western portion. Two European species, annuum and villosum, range in the native state west to Green- land, and are thus included also in the American flora. Several familiar Old World kinds acre, reflexum, spurinm, and Telephium subsp. Fabaria are naturahzed and run wild in the Eastern States.

10 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

The most marked feature of the Sedum flora is the occurrence in the west of a well-marked group of small perennial species with spathulate leaves and mostly yellow flowers, of which S. spathuli- folium and S. oreganum, already referred to, are examples. Some of these have the petals joined together in the lower portion (thus approaching the genus Cotyledon), and have been separated on this account from Sedum by some American botanists ; but I have preferred to retain them in that genus.

Literature. Britton and Rose, " Crassulaceae," in North American Flora," 22, Part I., 1905.

Mexico.

Mexico, which is now known to be extraordinarily rich in Sedums and other Crassulaceae, was until recent years a terra incognita. Two species of Sedum, moranense and oxypetaUim, were described ih 1823 among the plants collected on Humboldt's voyage (vol. 6, pp. 44, 45), and five years later De Candolle included two more, dendroideum and ebracteatum, in his " Memoire sur la famille des Crassulacees " (1828). As a result of herbarium work carried out in connexion with the great " Biologia Centrali- Americana," Hemsley was able, in 1879-88, to enumerate 22 species from Mexico in the first volume of the botanical section of that publication. During the last thirty years the explorations of a number of United States botanists have resulted in the discovery of a surprising number of new and interesting species of Sedum and of closely aUied plants for which new genera have been created, though in a broad sense many of them may be ranked as Sedums ; so that the species known from Mexico is now verging towards a hundred. Living plants of many of these have been sent to the States by their collectors, and are in cultivation at Washington and other places. They are still almost unknown in British and other European gardens, though many of them are handsome and interesting plants, strikingly different in appearance from any of the Old World Sedums. By the kindness of American correspondents, notably Dr. J. N. Rose (the describer of most of the new species) and Dr. N. L. Britton, I have received living specimens of a large number of these species. Plants of all or nearly all of them have now been placed at Kew, Edinburgh, and Dublin (wherever they were not already represented in the collections), and we may hope that these interesting species will now become better known on this side of the Atlantic. They display a remark- able range of form, from stout shrubs several feet in height, such aS^ oxypetaliim and praealtiim, to tiny creeping species like compactum and humifiisum ; the leaves show every variety of shape and size, and the flowers range through almost every hue. Many of the species are striking and decorative plants, such as alamosanum, cupressoides, Stahlii, Palmeri, helium, nutans, pachyphyllum, and versadense.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION:. II

The correlation of the Mexican Sedums with those of the Old World presents difficulties. The shrubby species appear best placed in the Seda Genuina, with which they possess many connecting links. To accommodate another characteristic Mexican group a new section, Mexicana, has been instituted.

Literature. Britton and Rose, " Crassulaceae," in "North American Flora," 22, Part I., 1905, and subsequent papers, mostly by J. N. Rose, in " Contributions from the U.S. National Herbarium," and elsewhere.

Central and South America.

A few species, of no importance horticulturally, occur in Guatemala, to the south of the great Sedum-centre of Mexico. Farther south, we find that the genus has in old days spread along the great back- bone of America, and makes on the Andes its only appearance in the Southern Hemisphere,* a few species being found as far south as Peru, and one as Bolivia.

IV. Statistical.

From the point of view of the gardener anxious to identify a Sedum which is unknown to him, the bringing together, as in the present paper, of all the species in cultivation, instead of helping him, may tend to have the opposite effect, since the comparatively small number of more or less common species (to one of which his plant probably belongs) is buried among a complex of other rarer plants which he is unUkely to encounter. With a view of mitigating this difficulty, I attempt below to indicate the species of most frequent occurrence, and also those at the other end of the scale,. thus :

Species very common in Cultivation.

acre rupestre spurium

album sexangulare Telephium

reflexum

It is probably no exaggeration to say that out of every ten plants (of Sedum) found in British gardens, nine belong to one or other of these species ; and, furthermore, that of every ten names appHed in British gardens to Sedums, five refer to one or other of the seven species above.

Species common in Cultivation.

Aizoon hyhridum oreganum

altissimum kamtschaticum roseum

Anacampseros maximum spectahile anopetalum

* It just reaches the Equator in Africa (see p. 6).

12 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

These are followed in frequency by :

dasyphyllum Lydium populifoUum

Ellacombianum Middendorffianum pulchellum

Ewersii muUiceps Sieholdii

hispanicum Nevii

The above-mentioned species, twenty-seven in all, represent those most usually found in English and Continental gardens, many of them under a multiplicity of names. In identifying a . garden Sedum, if it will not fit the figure and description which are given under the name in the present paper, or if the name under which it was received does' not appear in the Index, it will be well, first, to compare it with the figures and diagnostic notes relating to the seven species first mentioned. If it clearly cannot be matched there, there is a great probability that it belongs to the second or third hst given above.

In many cases the quickest way of " running down " a plant will be found to be to match it roughly by eye by a rapid survey of the illustrations, and then to turn to the description of the suspected species for confirmation. In doing this, the following species may for practical purposes be ruled out, as being extremely rare, and known in Great Britain in only two or three (mostly public) collections :

Species very rare in Cultivation.

adenotrichum

himalense

Selskianum

alpestre

hirsutum

Semenovii

alsinejolitim

longicaule

stellatum

hhutanense

monregalense

Stevenianum

bupleuroides

multicaule

Stribrnyi

cauticolum

Praegerianum

Taquetii

Cepaea

pruinatum

Tatarinowii

dumulosum

pseudospedabile

tiheticum

elongatum

purpureoviride

verticillatum

florijerum

quadrifidum

villosum

gracile

rhodanthum

yosemitense

gypsicolum

ruhroglaucum

yunnanense

heterodontiim

Also all the Mexican species, with the exception of praealtum, moranense, and Stahlii ; and some Indian, Chinese, and Japanese species, including the Japonica series of Maximowicz and a few others :

Celiae

Chaneti

Chauveaudi

Jormosanum

indicum

japonicum

Leblancae

lineare

multicaule

sarmentosum

Someni

trullipetaUim

variicolor

viscosum

Zentaro-Tashiroi

And, lastly, the plants listed on p. 5, which though they are or were in cultivation. I have not succeeded in seeing.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 13

V. Variation.

The species of the genus Sedum present a wide range of size, form, and colour. Minute creeping species are found in both the Old and New Worlds, and many of the annual species are very small ; on the other hand, some of the herbaceous perennials of the Telephium section produce annually stems a yard or more in height, and a few of the sub-shrubby Mexican species are equally tall. As regards duration, about four-fifths of the known species are perennials many herbaceous (that is, dying back to the root in autumn), many evergreen, a few deciduous (that is, having perennial stems but losing their leaves in winter) ; the remaining species are mainly annuals, a few being biennials.

Hairiness is rare in the genus ; and the most constant characteristic is a tendency to succulence, which in many species attains a very marked development, and enables them to live in very dry places. As an example of the amount of water which these plants may contain, a leaf of S. nutans, a Mexican species bearing the largest leaves found in the genus, weighed 75 ounce fresh, and when thoroughly air-dried •02 ounce in other words, ff , or over 97 per cent., of its weight was due to water stored up in the leaf.

The species of Sedum differ much as regards the variability which they display. Some are very stable and constant in character ; many others vary within limits, mostly as regards habit and leaf ; while some are highly variable, and, as regards at least general appearance, differ more from their type than some allied but quite distinct species do from each other. Thus, S. roseum, at once the most variable and the most widely distributed of Sedums, has flowers which range from the normal yellow through red to deep purple, and which may be dioecious or hermaphrodite ; the stem may be stout or slender, a couple of inches or a foot in height ; the leaves green to very glaucous, broadly ovate to linear, entire to deeply toothed. Other conspicuously variable species are S. album, altissimum, anopetalum,reflexum, Aizoon, spurium, Telephium.

Appended are notes of the more conspicuous cases of variation (including " sports ") found among the cultivated Sedums :

Roots varied and often characteristic thick and tuberous (section Telephium especially), woody and hard (section Aizoon), or fibrous.

Root-stock thick and elongate with conspicuous scale-leaves (many Rhodiolas), or spreading laterally into a fleshy mass (other Rhodiolas, Sedastrum), or absent.

Stem very variable as regards form and duration ; perennial and semi- woody (e.g. S. populifolium and many Mexican species), creep- ing and branching indefinitely (Seda Genuina), annual and erect (Telephium, Aizoon, &c.).

Leaves mostly entire, sometimes serrate, never more divided than pinnatifid (S. trifidum) ; spherical or cylindrical to flat, but never really thin ; green or glaucous, rarely hairy or glandular ; sessile or stalked, often spurred at base.

14 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.

Inflorescence mostly cymose, flattish on surface and roundish in outline : sometimes racemose or paniculate. Fig. i (5. lineare) shows a very characteristic and common type, formed of three dicho-

FiG. I. Inflorescence of S. lineare, from above.

tomous branches with a flower in the primary and secondary forks, and a bract subtending each flower.

Sepals regular in European and most Asiatic species, often markedly irregular in Chinese and Mexican plants.

Petals very small and inconspicuous (some Rhodiolas), or relatively large and mostly brightly coloured, patent or seldom erect, entire or seldom fringed.

Stamens normally lo ; 5 in a few species, most of which have no near relationship to each other.

Carpels erect or stellate ; seeds borne in a row along the inner face of the carpel, very seldom (e.g. S. Celiae) in a bunch near the base of the carpel.

Hybrids.

Hybrids are rare in the genus. A notable exception occurs in the case of S. Telephium and its near ally 5. maximum, which cross freely in the wild state and in the garden. Otherwise only a very few hybrids are known.

5. altissimum X reflexum = S. luteolum Chaboisseau (France).

S. acre X sexangulare = S. FUreri K. Wein (Harz Mountains).

S. annuum X sexangulare = S. erraticum Briigg. (Switzerland).

S. annuum X alpestre = S. engadinense Briigg. (Switzerland).

S. atratum X annuum = 5. Derbezii Petitmengin (Maritime Alps) .

S. Aizoon X kamtschaticum. (Wisley, where it was received from a garden as S. kamtschaticum. Also seen at Cambridge.)

5. Telephium X maximum. (Frequent in gardens where the two species are grown.)

Owing no doubt to the fact that the genus is not a popular one among plant-fanciers, we have escaped so far from the production of endless uninteresting artificial hybrids, such as now confuse the allied genus Saxifraga.

ACCOUNT OF GENUS SEDUM AS FOUND IN CULTIVATION. 15

Sports.

Variation in the way of teratology is rare in the genus.

Variegation. ^The best known and most pleasing of the few variegated forms of Sedum are 5. Sieboldii with a gold patch in the middle of each leaf, and S. kamtschaticum with the leaves splashed with gold. A fine variegated form of 5. maximum is figured in " Flore des Serres," 16, t. 1669. Of 5. alboroseum there is a form with a large silver patch in the centre of each leaf, and another with a marginal band of white, or rather of pale green. Two variegated forms of S. acre are in cultivation, one of which has the tips of the shoots golden in spring, the other silver. Of the tender 5. lineare a form has been long in cultivation with the leaves silver-margined.

Of S. Telephium, S. maximum, 5. album, S. spathuli folium, forms are grown in which purple pigment is conspicuously present in the stem and leaves.

In The Garden for 1901, Mr. S. Arnott refers to a small silver- variegated Sedum under the name S. caespitosum. I have not been able to trace this plant.

Fasciation. This monstrosity is rare in Sedum, but one re- markable example is frequent in gardens the " Cock's-comb Sedum," which is a sport of 5. reflexum. From Messrs. Backhouse of York came a smaller plant resembling the last, which is possibly a fasciate S. anopetalum, but no normal branch which might flower has been produced yet. I have received from New York a similar sport of the Mexican S. praealtum ; the last was included a few years ago in Haage and Schmidt's Hst, under the name S. dendroideum cris- tatum. The var. arboreum of S. moranense has a persistent