Bena. aie ii rately oe He att livia nae i ye he Pi itferatete) aa ietretar a Teta atePy er neriecaeee A Eee OO fren pape pe i reine ee eer ses y) Cornell University Library Ithara, New York BOUGHT WITH THE INCOME OF THE FISKE ENDOWMENT FUND THE BEQUEST OF WILLARD FISKE LIBRARIAN OF THE UNIVERSITY 1868-1863 1905 RETURN TO ALBERT R. MANN LIBRARY ITHACA, N. Y. CORNELL UNIVERSITY LIBR: RY cin: 3 1924 053 948 935 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924053948935 THE TREES OF GREAT BRITAIN AND IRELAND ‘aLoulnnqiygd ApVT apojavyy ky Susarviq v uo.tyy @TLISVO AYATOHOIH LV UVdAD NONVAAT The Trees of Great Britain & T[reland Henry John Elwes, F.RS. AND Augustine Henry, M.A. VOLUME III CONTENTS List oF ILLUSTRATIONS CEDRUS ; , , F Ceprus Lipani, LEBANON CEDAR CEDRUS BREVIFOLIA, CyPRUS CEDAR CEDRUS ATLANTICA, ATLAS OR ALGERIAN CEDAR CEDRUS DEODARA, DEODAR . LIBOCEDRUS . ; LIBOCEDRUS TETRAGONA LIBOCEDRUS CHILENSIS LisocepRus DoNnIANA LIBOCEDRUS BIDWILLI LIBOCEDRUS MACROLEPIS : LIBOCEDRUS DECURRENS, INCENSE CEDAR CUNNINGHAMIA ‘ CUNNINGHAMIA SINENSIS LIQUIDAMBAR LIQUIDAMBAR STYRACIFLUA, SWEET GUM LIQUIDAMBAR ORIENTALIS. 3 LIQUIDAMBAR FORMOSANA NYSsA ’ ‘ . Nyssa SYLVATICA, TUPELO . Nyssa aquatica, CoTTon GuM NyYSSA SINENSIS, CHINESE TUPELO . SASSAFRAS . z é : SASSAFRAS TzUMU, CHINESE SASSAFRAS SASSAFRAS OFFICINALE, SASSAFRAS . CoryLus. j : : CoryLus CoLuRNA, TURKISH HazEL CARPINUS . : , ; ; CaRPINUS ORIENTALIS : : , CARPINUS POLYNEURA ; CARPINUS JAPONICA . CARPINUS YEDOENSIS 3 ; : CARPINUS CORDATA . : ‘ , CARPINUS LAXIFLORA ; . . CARPINUS CAROLINIANA, AMERICAN HORNBEAM Carpinus BeTULUS, COMMON HORNBEAM . OSTRYA ‘ ‘é ‘ , . ill iv The Trees of Great Britain and Ireland PAGE. OsTRYA CARPINIFOLIA, Hop HoRNBEAM ; : : ; ‘ . ‘ 540 OsTRYA VIRGINICA, IRONWOOD , ; ‘ 543 OsTRYA JAPONICA, JAPANESE Hop HOoRNBEAM ; ‘ 3 ‘ ‘ ; 544 NOTHOFAGUS ‘ : ; 545 NOTHOFAGUS CLIFFORTIOIDES 547 NotHoracus MENZIESII . 548 NOTHOFAGUS FUSCA . 549 Noruoracus Moore! ; ‘ . : ‘ 3 ; : 549 NoTHOFAGUS CUNNINGHAMI . j : ‘ : ‘ : : : 550 NOTHOFAGUS BETULOIDES . . : : : ; ‘ : ‘ 551 NOTHOFAGUS OBLIQUA ; ; : , : i ‘ ‘ ; 552 NOTHOFAGUS ANTARCTICA. : ; : : ‘ 553 ARBUTUS i . 558 ARBUTUS UNEDO, STRAWBERRY TREE : ; ; : . : 559 ARBUTUS HYBRIDA . . : ‘ ‘ : ’ j : : 563 ARBUTUS ANDRACHNE : : : ’ : : i ; é 564 Arsutus MeEnziEsII, MADRONA . : ‘ : : . . 565 SCIADOPITYS . ; d ‘ . ‘ ‘ : ‘ ' 567 SCIADOPITYS VERTICILLATA, UMBRELLA PINE 568 PINUS SYLVESTRIS, SCOTS PINE : . : : : 571 CaRYA . ‘ F : ; ; ; ‘ : : 597 CaRYA OLIVEFORMIS, Pecan Nutr . ‘ ‘ ; 599 CARYA AMARA, BITTERNUT 600 CaRVA ALBA, SHAGBARK HICKORY . : : ‘ ‘ ; : 601 CaARYA PORCINA, PIGNUT . : : ; : ‘ : F 604 CaRYA SULCATA, BIG SHELLBARK . : . . . : ; : 605 CaRYA TOMENTOSA, MOCKERNUT ; : : ; : ; : 606 PLaTANUS . . . E ; ; : : ; : ; 611 PLATANUS ORIENTALIS, ORIENTAL PLANE . j , : F ‘ 614 PLATANUS OCCIDENTALIS, WESTERN PLANE . F . F : : ‘ 627 ACER ; F ‘ 3 : : ‘ : ‘ . : 630 ACER PSEUDOPLATANUS, SYCAMORE . i ‘ ‘ ' ‘ F ‘ 641 ACER CAMPESTRE, COMMON MAPLE . ‘ i i : ‘ : i 651 ACER PLATANOIDES, Norway MapLe ‘ ; ; ‘ : : ‘ 656 AcER LOoBELII ; : F . ; ‘ ; ‘ : ‘ 659 ACER PICTUM F , , ; ; ; ‘ , : : 660 AcER OPALus, ITALIAN MAPLE : ‘ é : ; : 3 3 663 ACER MONSPESSULANUM, MONTPELLIER MAPLE . ‘ : ‘ : ‘ 665 ACER INSIGNE ' ‘ ‘ 4 : ‘ : ‘ . : 667 ACER VOLXEMI ; : 3 j : : : : ‘ ‘ 668 AcER TRAUTVETTERI F : ; : j F ‘ ‘ : 669 ACER RUBRUM, RED MAPLE : ; : i ; ‘ i F 671 ACER DASYCARPUM, SILVER MAPLE . : i : ; i : : 674 ACER SACCHARUM, SUGAR MAPLE . : : : , ; ; : 677 ACER MACROPHYLLUM, OREGON MAPLE : 4 : : ‘ ; ? 681 ACER NEGUNDO, ASH-LEAVED MAPLE 3 F 3 ‘ : : : 684 SEQUOIA i ‘ , , 5 . ‘ ° : é ‘ 687 SEQUOIA SEMPERVIRENS, REDWOOD . 4 ‘ ‘ : F F : 688 SEQUOIA GIGANTEA, WELLINGTONIA . ‘ F : : ; : . 699 LIS) OF PLATES VOL. III Lebanon Cedar at Highclere Castle (from a drawing by Charlotte Lady Phillimore) . Lrontispiece Cedar on Mount Lebanon : Plate No. 127, ¢o face page 456 Lebanon Cedar at Painshill . : ‘ : : ; is 128 iy 460 Lebanon Cedar at Goodwood : ; ‘ ‘ : ay 129 a 460 Lebanon Cedar at Strathfieldsaye » 130 ” 460 Lebanon Cedar at Petworth . : : z ; : = 131 3 462 Lebanon Cedar at Blenheim . 5 132 5 462 Lebanon Cedar at Stratton Strawless . i 133 3 462 Lebanon Cedar at Birchanger i 134 $s 462 Cedar Avenue at Dropmore . ‘ ; ‘ ‘ : 5 135 9 464 Algerian Cedars at Téniet-el-Haad_ . : ; : : - 136 a 472 Algerian Cedar at Ashampstead : . : ‘ ; 5 137 - 474 Algerian Cedar at Fota : : : i : : i 138 4 476 Deodars in the Himalaya. ‘ , : : : 5 139 3 478 Deodar at Bicton . ‘ ‘ : ‘ : ‘ ‘ 140 rr 482 Libocedrus chilensis in Chile j Z ‘ : ‘ 5 141 3 486 Libocedrus decurrens at Frogmore . : ; ‘ ; 5 142 is 492 Cunninghamia sinensis at Bagshot Park : ; Z : i 143 5 496 Liquidambar in America (A), Nyssa sylvatica in America (B) . : s 144 3 502 Nyssa at Strathfieldsaye ‘ ; : : ‘ ‘ 3 145 my 510 Sassafras at Claremont : ‘ ; ; ‘ : i 146 55 518 Corylus Colurna at Wollaton Hall. ‘ F : ‘ 45 147 Fe 524 Hornbeam at Cornbury Park : : : ? . rf 148 5 536 Pollard Hornbeams at Bayfordbury . A 2 . 7 a 149 ‘i 536 Hornbeam at Easton Lodge . é : ‘ : ‘ re 150 i 536 Hornbeams at Weald Park . i : : . : 9 151 95 536 Hornbeam at Gordon Castle . ; ; : ; : 8 152 538 Hop-hornbeam at Langley Park, Norfolk . ¢ . F Pe 153 3 542 Carpinus and Ostrya ; leaves, etc. ; : 3 : j 5 201 5 544 vi The Trees of Great Britain Tasmanian Beech at Fota Evergreen Beech at Bicton Beech Forest in Chile Arbutus at Killarney . Arbutus hybrida at Sedbury Park Sciadopitys in Japan (A), Sciadopitys at Hemsted (B) Scots Pine Avenue at Carclew Scots Pine at Bramshill Scots Pine at Inveraray Castle Scots Pine at Dunkeld Scots Pine in Glen Maillie Scots Pine at Gordon Castle . Scots Pine at Loch Morlich . Scots Pine at Abernethy ‘Scots Pine at Abernethy Scots Pine in Ballochbuie Forest Scots Pine at Ballochbuie Hickory at Bute House Hickories in Syston Park Carya alba at Brocklesby Park Hickory at Kew Carya; leaves, etc. Oriental Plane at Ely Oriental Plane at Corsham Court London Plane at Albury Oriental Plane at Weston Park Western Plane in America (A), Oriental Plane in Syria (B) Sycamore at Colesborne Sycamore at Newbattle Sycamore at Drumlanrig Sycamore at Castle Menzies . Sycamore in Switzerland Common Maple at Cassiobury Common Maple at Langley Park, Norfolk Norway Maple at Emsworth . Norway Maple at Park Place. Norway Maple at Colesborne Italian Maple at Hargham Montpellier Maple at Rickmansworth Red Maple at Bagshot Park . and Ireland Plate No. 154, to face page 550 ” 155 ” 552 » —- 156 » 554 » 157 » 562 » 158 » 564 » 159 » 568 3 160 as 584 5 161 45 586 i 162 is 536 4 163 13 586 x 3600 4% 586 i 164 4 588 5 165 5 588 4 166 7 588 oP 167 oS 588 PF 168 a 588 ‘i 169 5 588 55 170 4 600 i 171 sy 600 %5 173 ss 602 as 172 5 604 35 203 5 610 is 174 55 620 i 175 H 622 ” 178 5 624 rf 367 9 628 Pe 176 35 628 » 179 » 646 si 180 FF 648 ee 181 4 648 3 182 ‘3 648 5 183 Pr 650 re 184 33 654 » 185 » 654 i 186 PA 658 ie 187 Fe 658 3 188 i 658 33 189 P 664 ” 190 3 666 » 177 ” 672 List of Plates Red Maple at Whitton Sugar Maple at Park Place Western Maple in America Acer ; leaves, etc. Acer ; leaves, etc. Acer ; leaves, etc. Redwood Forest in California Redwood at Claremont Wellingtonia at Fonthill Wellingtonia at Aston Clinton Wellingtonia Avenue near Wellington College Nyssa, Diospyros, Sassafras, and Liquidambar ; leaves, etc. Castanea, Liquidambar, Diospyros, Nyssa, and Sassafras ; buds and twigs Nothofagus, Fagus, Castanopsis, and Castanea ; leaves, etc. Platanus, Catalpa, and Liriodendron ; leaves, etc. vil Plate No. 192, 40 face page 672 191 193 205 206 207 195 194 196 197 198 199 200 202 204 bby 680 682 686 686 636 688 696 706 708 708 708 708 708 708 ILLUSTRATIONS Lebanon Cedar at Highclere Castle (from a drawing by Charlotte Lady Phillimore) Frontispiece PLaTE No. Cedar on Mount Lebanon Lebanon Cedar at Painshill we Lebanon Cedar at Goodwood 129 Lebanon Cedar at Strathfieldsaye 130 Lebanon Cedar at Petworth . 131 Lebanon Cedar at Blenheim . : 132 Lebanon Cedar at Stratton Strawless . 133 Lebanon Cedar at Birchanger 134 Cedar Avenue at Dropmore . ; ; : : . F : : 135 Algerian Cedars at Téniet-el-Haad_ . ; : F i ’ ‘ F 136 Algerian Cedar at Ashampstead ; : j , 7 ‘ : ; 137 Algerian Cedar at Fota ‘ . ; : i ‘ : ; : 138 Deodars in the Himalaya. ' é : j ; : : : 139 Deodar at Bicton . . : ‘ : : , : : é 140 Libocedrus chilensis in Chile : ; ‘ ; : : ; ; 141 Libocedrus decurrens at Frogmore . ‘ F : : : ‘ ‘ 142 Cunninghamia sinensis at Bagshot Park ; j ‘ ; é 3 ; 143 Liquidambar in America (A), Nyssa sylvatica in America (B) . ‘ ‘ : ; 144 Nyssa at Strathfieldsaye é f ‘ : : i ‘ . ‘ 145 Sassafras at Claremont ‘ : : ‘ : , : : ; 146 Corylus Colurna at Wollaton Hall. ‘ ‘ ‘ ‘ ; ‘ : 147 Hornbeam at Cornbury Park . , ; ? : ; : ‘ 148 Pollard Hornbeams at Bayfordbury . ; : ‘ ; ; ; F 149 Hornbeam at Easton Lodge . : ‘ ‘ ! ; : : ‘ 150 Hornbeams at Weald Park . ‘ ‘ ' . j i : : 151 Hornbeam at Gordon Castle ‘ ‘ : ‘ ‘ : ; F 152 Hop-hornbeam at Langley Park, Norfolk ‘ : ‘ ; ; : : 153 Tasmanian Beech at Fota_. . ; ‘ : : . : ; 154 Evergreen Beech at Bicton . : ; ‘ ; ; . : ; 155 Beech Forest in Chile 156 Arbutus at Killarney. : 157 Arbutus hybrida at Sedbury Park 158 Sciadopitys in Japan (A), Sciadopitys at Hemsted (B) : . . : . 159 Scots Pine Avenue at Carclew : . : : ou Scots Pine at Bramshill vi The Trees of Great Britain and Ireland Pate No. Scots Pine at Inveraray Castle 162 Scots Pine at Dunkeld 163 Scots Pine at Gordon Castle . 164 Scots Pine at Loch Morlich . 165 Scots Pine at Abernethy 166 Scots Pine at Abernethy 167 Scots Pine in Ballochbuie Forest 168 Scots Pine at Ballochbuie 169 Hickory at Bute House 170 Hickories in Syston Park. . : : ; é 5 ; : 171 Hickory at Kew 172 Carya alba at Brocklesby Park : i : é : . : : 173 Oriental Plane at Ely ; : ; : : i ; : i 174 Oriental Plane at Corsham Court : , : ; : 175 Western Plane in America (A), Oriental Plane in Syria (B). : : : ‘ 176 Red Maple at Bagshot Park . é : , : ‘ F . : 177 London Plane at Albury . : : ‘ : ‘ : é F 178 Sycamore at Colesborne : ; : : ‘ 7 é : . 179 Sycamore at Newbattle 2 : . F : ; ‘ ; : 180 Sycamore at Drumlanrig 181 Sycamore at Castle Menzies . : , : ; ; : . ; 182 Sycamore in Switzerland : : ; : : é i ' ; 183 Common Maple at Cassiobury : ; : : : : ; 184 Common Maple at Langley Park, Norfolk . ; : : ; : . 185 Norway Maple at Emsworth . ‘ ; ; ‘ ‘ . : ; 186 Norway Maple at Park Place : : : ; . ; 3 : 187 Norway Maple at Colesborne : : i : : ‘ : : 188 Italian Maple at Hargham . : : F é 2 : ; 189 Montpellier Maple at Rice anenoiih : F ; ; : : 190 Sugar Maple at Park Place . : 3 : ; : 2 ; ' IgI Red Maple at Whitton ‘ ‘ : : i j : : ‘ 192 Western Maple in America . F ‘ : : ‘ ; ; . 193 Redwood at Claremont : ‘ ; ; : : : : ‘ 194 Redwood Forest in California : : ‘ : : ‘ ‘ ‘ 195 Wellingtonia at Fonthill : ‘ : . ‘ 4 , ‘ : 196 Wellingtonia at Aston Clinton : ‘ F . , i : . 197 Wellingtonia Avenue near Wellington College 3 ; ; ; ‘ ; 198 Nyssa, Diospyros, Sassafras, and Liquidambar ; leaves, etc. . : : 3 : 199 Castanea, Liquidambar, Diospyros, Nyssa, and Sassafras; buds and twigs : : ; 200 Carpinus and Ostrya; leaves, etc. ; F : ; ? , . 201 Nothofagus, Fagus, Castanopsis, and Castanea; leaves, etc. . : : : : 202 Carya ; leaves, etc. . : ; : ; : j ; ‘ 203 Platanus, Catalpa, and Hilesenions leaves, etc. : ‘ : . ; ‘ 204 Acer; leaves, etc. . : ; ; ; ‘ : : , : 205 Acer; leaves, etc. . . : : : : : : : : 206 Acer; leaves, etc. . 3 ; : ; : ‘ : : F 207 CEDRUS Cedrus, Lawson, Agric. Man. 379 (1836); Loudon, A7d. et Frut. Brit. iv. 2402 (1838); Bentham et Hooker, Gen. F2. iii. 439 (1880) ; Masters, Journ. Linn. Soc. (Bot.) xxx. 30 (189 a Larix, Miller, Dict. No. 3 (1724) (ex parte). Pinus, Linnzus, Sp. Pl. 1001 (1753) (ex parte). Pinus, section Cedrus, Parlatore in DC. Prod. xvi. 2, Pp. 407 (1864). Abies, Poiret in Lamarck, Dict. vi. 510 (1804) (ex parte). TREEs belonging to the tribe Abietinez of the order Conifere, with evergreen foliage, borne, for the most part, in tufted flat masses on the ramifications of the branches, which arise irregularly and not in whorls from the stem. Bark dark grey and smooth on young stems and branches; ultimately on old trunks thick and fissuring into irregular longitudinal plates, roughened externally by small scales. Branchlets of two kinds: long shoots bearing in spiral order solitary leaves, and short shoots or spurs with leaves in pseudo-verticels. Buds minute ovoid, with a few brown scales, which persist after the opening of the bud, either sheathing the base of the long shoots or surrounding the annual rings of the short shoots. Long shoot with a solitary terminal bud, prolonging the growth of the branchlet in the following year; and with a few lateral buds solitary in the axils of some of the leaves and usually developing into short shoots. Short shoot with a terminal bud only, which, in the following year, either lengthens slightly the spur and adds to it a whorl of leaves with or without flowers, or occasionally develops into a long shoot. Long shoots, slightly furrowed, between the slightly raised decurrent bases of the pulvini, the free ends of which project and bear leaves, and on older branchlets, from which the leaves have fallen, remain persistent as slight prominences. Leaves, deciduous in the third to the sixth year, variable in length, the shortest on the spurs, articulated at the base, acicular, rigid, sharply pointed, more or less triangular in section, stomatic on all sides; fibro-vascular bundle undivided, hypo- derm thick, with two resin canals close to the epidermis on the lower surface. Flowers, moncecious, terminal, solitary on the older leaf-bearing short shoots. Male flowers, erect, catkin-like, cylindrical, about 2 inches long; anthers numerous, spirally crowded, bi-locular, dehiscing longitudinally ; connective prolonged into an ovate denticulate crest; pollen grains globose, without wings, borne to the female flowers by the wind. Female flowers appearing as small purplish cones, about } inch in length; composed of numerous spirally arranged, closely appressed, irregularly dentate, sub-orbicular scales, each subtended by a short, included, obovate, denticulate pract ; ovules, two on each scale, inverted. III 45! : 452 The Trees of Great Britain and Ireland Ripe cones, solitary, erect, on short stout peduncles, dull-brown, resinous, ellipsoid or cylindrical; rounded, flattened or depressed at the apex. Bracts obsolete or minute and ragged. Scales numerous, closely imbricated, woody, fan- shaped ; upper expanded part thin, transversely oblong, with denticulate rounded or sloping wings, brown-tomentose in greater part beneath, almost glabrous above ; claw thickened, obcuneate, with a raised ridge between the depressions for the seeds on the upper surface, the lower surface being slightly hollowed by the pressure of the seeds of the adjoining scale. Seeds, two on each scale, } to $ inch long, with resin- vesicles on both surfaces, brown, irregularly triangular ; surmounted by a membranous brownish wing, broadly triangular or hatchet-shaped, about twice as long as the body of the seed. Cotyledons, nine or ten. The flowers appear in July or August, the pollen being shed profusely in October. During winter the cones remain small, and only begin to grow in the following April, attaining about half or two-thirds their full size in October of the second year. They are fully ripe in October or November of the third year, 7e. about twenty-six months after the first appearance of the flowers. In their native forests the dissemination of the seed is caused by the autumnal rains, the cones not disarticulating in dry weather. After being soaked with rain, the scales and seeds separate from the axis of the cone (which remains persistent on the branch) and fall to the ground, the seeds with their light wings being blown, when there is a breeze, to a little distance from the parent tree. In England, irregularities occur in the period when the cones dis- articulate, dependent, probably, on the absence of heavy rains in the autumn in certain seasons, Seedling.—Plants raised from seed gathered on Mount Lebanon in 1904, and sown at Monreith in April 1905, averaged 9 inches high in the following September,! and showed the following characters:—Tap-root, about 9 inches long, slender, with a few lateral fibres. Caulicle, 2 inches long, slightly furrowed, glabrous. Cotyledons, ten, sessile, 13 inch long, curved, tapering to a sharp point, triangular in section, the upper two sides stomatiferous, the lower side green and narrow. Young stem glabrous, bearing in a whorl, just above the cotyledons, the first seven leaves, $ inch long, linear, flattened, sharp-pointed, stomatiferous on both surfaces, deeply grooved below, slightly convex above, sharply serrate in margin. Above the whorl, leaves, gradually increasing in size to 1z inch long, arise in spiral order, similarly serrate and stomatiferous, but almost rounded in section ; in addition, the stem gives off at irregular intervals five or six small branchlets. With regard to the different forms of the cedar, which inhabit four distinct and isolated areas, opinions are much at variance as to their rank. They differ more or less in the length of their leaves, and in the size and shape of the cones, cone scales, and seeds, and in the young stage they differ in habit; but in their native forests they all assume, when old, the flattened form which is sometimes erroneously considered to be peculiar to the Lebanon cedar. This is caused by the inflection of the leading shoot, which is followed by a diminution in the rate of vertical growth, the lateral branches at the same time thickening and growing out horizontally. An 1 This growth is quite exceptional in my experience.—(H. J. E.) * Cedrus 453 important difference is the height attained in the wild state, the deodar becoming very tall, the Cyprus cedar remaining short, with the Lebanon and Algerian cedars intermediate in size. They differ in their period of vegetation. At Kew the deodar is the first to put forth young leaves in spring ; the Lebanon usually follows a fort- night later; and the Algerian generally comes out last, after an interval of a few days. They may be correctly considered geographical races of the same species ; but for arboricultural purposes it is most convenient to rank them as distinct species. CEDRUS LIBANI, Lezanon Cepar Cedrus Libani, Barrelier, Planta, Icon. 499 (1714); Loudon, Avd, et Frut. Brit. iv. 2402 (1838); Ravenscroft, Pinet. Brit. iii. 247 (1884); Kent, Veitch’s Man. Conifere, 415 (1900). Cedrus patula, Koch, Dendrol. ii. 268 (1873). Pinus Cedrus, Linneus, Sp. PZ. 1001 (1753). Larix Cedrus, Miller, Gard. Dict. ed. viii. No. 3 (1768). Larix patula, Salisbury, Zrans. Linn. Soc. viii. 314 (1807). Abies Cedrus, Poiret in Lamarck, Dic. vi. 510 (1804). Leading shoot of young trees erect or slightly bent, not pendulous. Branchlets not pendulous, glabrous or with slight short pubescence. Leaves up to 1} inch in length, broader than thick. Cones large and broad, ellipsoid, 3 to 44 inches long, 12 to 24 inches wide; scales 2 inches or more in width, with the claw inflected almost at a right angle. VARIETIES 1. Var. axgentea, Antoine et Kotschy, /ter. Czhc. No. 417. Trees with glaucous foliage, growing wild in the Cilician Taurus, intermingled with the ordinary form. This variety appears in cultivation, but is rarer than the glaucous form of C. atlantzca. 2. Var. decidua, Carriére, Contf. 372 (1867). Leaves deciduous. A tree of this kind, slow in growth and bushy in habit, was obtained by Sénéclauze in 1851. Kent mentions one growing at Westgate near Chichester." Webster reports’ another, 65 feet high, growing on Lord Derby’s property in Kent, and said to be in perfect health, though from its bare appearance in winter it has often been supposed to be dying. 3. Var. ¢ortuosa. On the lawn of a private house at Dulwich, belonging to the Dulwich College estate, there is a remarkable cedar, a photograph of which was sent to Kew in 1903. The stem and all the branches are spirally twisted. The Lebanon cedar is variable in habit, and numerous supposed varieties are mentioned by Beissner, as xana,* a dwarf form ; s¢rzcta, narrowly pyramidal in habit ; pendula, with pendulous branches and branchlets; and viridis, with bright green shining foliage. 1 But on writing to Captain Norman, who was the authority for this, he tells me that the tree is now dead, and that in his opinion the deciduous habit, which was regular and unfailing, was due to constitutional weakness, caused by uncongenial surroundings, in proof of which he states that another tree at the same place raised from a seed taken from the same cone, was much more robust and showed no abnormal tendency.—(H. J. E.) 2 Hardy Coniferous Trees, 27 (1896). 3 A specimen of the dwarf cedar, only 4 feet high and of considerable age, is growing in grounds adjoining one of the oldest parks at Hemel Hempstead. The branches are flattened, horizontal, and very close together, giving the plant a dense, stiff appearance. —Gard. Chron. xix. 563 (1896). 454 The Trees of Great Britain and Ireland Supposed hybrids between the Lebanon and Atlantic cedars have been recorded,' but on insufficient evidence. (A. H.) DISTRIBUTION The best account of the Cedar of Lebanon known to me is the classical paper by Sir Joseph Hooker published in the Natural Hestory Review, vol. ii. p. 11 (1862), and as this gives a careful summary of the facts bearing on the specific identity of the forms of cedar, I summarise it as follows :—In the autumn of 1860 Sir J. Hooker went to Syria in company with Captain Washington, Hydrographer of the Navy, and Captain Mansell, R.N., and arrived at Beyrout on 25th September. The party proceeded to the Lebanon, where Captain Mansell made a detailed survey of the basin where the cedars grow, at the head of the Kedisha valley, 15 miles from the sea in a straight line. At that time the other groves were apparently unknown, though Professor Ehrenberg informed Sir Joseph Hooker that he found many trees in forests of oak on the road from Bsharri to Bshinnate. The Kedisha valley at 6000 feet elevation terminates in broad, flat, shallow basins, and is two or three miles across and as much long. It is three or four miles south of the summit of Lebanon, which is about 10,200 feet in height, the chapel in the cedar grove being about 6200 feet. The cedars grow on a portion of the moraine which borders a stream, and nowhere else; they form one grove about 400 yards in diameter, and appear as a black speck in the great area of the corrie and its moraines, which contain no other arboreous vegetation, nor any shrubs but a few small barberry and rose bushes. The number of the trees is about 400, and they are disposed in nine groups, corresponding with as many hummocks of the range of moraines ; they are of various dimensions, from 18 inches to upwards of 40 feet in girth; but the most remarkable and significant fact connected with their size, and consequently with the age of the grove, is that there is no tree of less than 18 inches girth, and that no young trees, seedlings, nor even bushes of a second year’s growth were found. Calculating from the rings in a branch of one of the older trees, now in the Kew Museum, the younger trees would average 100 years old, the oldest 2500, both estimates no doubt being widely far from the mark. Sir Joseph goes on to say, that the word cedar as used in the Bible applies to other trees, and he doubts whether the cedar of Lebanon is the one which supplied the timber used in building Solomon’s temple. He thinks that the cypress or the tall fragrant juniper of the Lebanon (/unzperus excelsa) would have been not only much easier to procure, but far more prized on every account.? Between individuals from the Lebanon and the common Asia Minor form there is said to be no appreciable difference by those who have examined both, but there are two distinct forms or varieties in Asia Minor, one having shorter, stiffer, and more silvery foliage than the other; this is the silver cedar, C. arvgentea, of our gardens. Northern Syria and Asia Minor form one botanical province, so that the Lebanon groves, 1 Beissner, Wadelholekunde, 301, 302 (1891). 2 But at a later period Sir J. Hooker changed his opinion on this subject, and believed that the wood used by Solomon and by Nebuchadnezzar in buildings was the Lebanon cedar. Cedrus 455 though so widely disconnected from the Taurus forest, can be regarded in no other light than as an outlying member of the latter. After speaking of the Algerian cedar and the deodar, Sir Joseph says that it is evident that the distinctions between them are so trivial, and so far within the proved limits of variation in coniferous plants, that it may reasonably be assumed that all originally sprang from one. There are no other distinctions whatever between them of bark, wood, leaves, male cones, anthers, or in their mode of germination, growth, or hardiness (but this has not been confirmed during the severe winters of a later date in England). Though the difference in the shape of the scales and seeds of Deodara and Libani are very marked, they vary much, many forms of each overlap, and further transitions between the most dissimilar may be established by the _ inter- calation of seeds and scales from C. atlantica. Sir Joseph accounts for the difference in the habit of the three forms in a great measure by the climate of the three localities: the most sparse, weeping, long-leaved cedar is from the most humid region, the Himalaya; whilst a¢/antica, the form of most rigid habit, corresponds with the climate of the country under the influence of the great Sahara desert. No course remains, then, but to regard all as species, or all as varieties, or Deodara and atlantica as varieties of one species, and Lzdanz as another. The hitherto adopted and only alternative of regarding Lzdani and atlantica as varieties of one species and Deodara as another species must be given up. Ravenscroft, in Pinetum Britannicum, gives a very full account of the cedars of Lebanon from various sources, with four good illustrations from photographs taken by F. M. Good of Winchfield, and there are many points in his account worth referring to. Mr. Ridgway of Fairlawn, who visited them in 1862, says’ that there is a young tree 50 yards west of the chapel, of exactly the same form and habit as a deodar in his park near Tonbridge. It has the same graceful drooping habit, the same light silvery green, and none of the usual rigid horizontal form of the cedar. He says the remainder of the race of trees vary from 20 to 25 feet in girth; some are as tall and straight as poplars, some not above 20 feet high, and gnarled and stunted. Ravens- croft gives in a table the facts relating to the number of trees found in the accounts of various authors who have written on the Cedars of Lebanon, commencing with Belon in 1550 and ending with Canon Tristram in 1864. Of the older ones there were 28 in Belon’s time, which are now reduced to about half that number. There is a gap of some centuries—Ravenscroft says probably more than 1000 years—between the cedars of the second size and the older ones, and again a very long interval of growth between all the young trees, which are now about 4oo. I do not find any reliable information, taken from an actual count of the number of rings in any of the old Cedars of Lebanon, as to what their possible age may be. Ravenscroft has gone very carefully into the question of the age of the Cedars of Lebanon, which, he says, may be from 4000 to 5000 years old; and he further gives a table based on 200 measurements of cedars of all ages in England, which shows that the average growth in height in England is about 1 foot per annum for trees up to sixty years 1 Gard. Chron. 1862, p. 572. 456° The Trees of Great Britain and Ireland old, and from 6 to 9 inches in trees of 100 to 200 years. He gives the average breadth of the annual rings per annum in trees of from twenty to fifty years as from 34 to of lines, and in trees from 60 to 200 years as 34 lines. In Gardeners’ Chronicle, xii. 204 (1879), S. R. Oliver writes :—‘‘ And now about the cedars themselves. The guardian told us that there are exactly 385 trees, large and small, but the smallest must be at least from 50 to 80 years old, and no younger trees are springing up—a fact to which it would be well to draw the attention of the public. At this time of year (28th September) innumerable seeds, which are scattered everywhere beneath the trees from the fallen and expanded cones, are germinating, scattered by the wind; these germinating seeds extend far beyond the actual area covered by the remaining trees; and if it were not that they are trodden under foot, or, what is still more destructive, eaten up by the goats, a few decades of years would soon see a fair sprinkling of healthy young cedars enlarging the borders of the grove. In 100 years the grove would be increased into a wood, and five centuries hence the wood would have become a forest. At present, for want of proper protection against the goats and thoughtless tourists, the present grove is dwindling away, and another generation will exclaim against our supineness in thus allowing a relic of the past to die out prematurely. For asmall sum of money a stone wall might be built, enclosing the area of the cedar grove sufficiently well against goats. Future travellers ought to be warned by the guardian to confine their steps to certain paths, so as not to injure the young trees; and stringent precautions should be taken against the dis- figurement of the trees now existent, by the cutting of names, tearing down of branches for the cones, etc. It would be easy to build such a wall so as not to be an eyesore or disfigurement, by taking advantage of the sinuosities of the numerous small valleys which permeate the vicinity. I am sure that many travellers would contribute small donations should a subscription list be opened for such a purpose. As the property of the cedars belongs to the Patriarch of the Maronites, by name Butross Massaad, who resides by the Dog river, not far from Beyrout, it would be necessary to obtain his co-operation, and I hope, through the aid of the Consul- General for Syria, to have an interview with him on the subject before I leave the neighbourhood. Most of the single trees of antique growth average 20 to 30 feet in girth at about 6 feet from the ground, but the enormous fathers of the forest are in reality a congeries of two, three, or even more trees which have grown so closely together as to coalesce and actually form a single trunk. Among the younger specimens twin and triplet trees are rather the rule than the exception, and this will explain such a girth as Dr. Wartabet measured round the largest tree on the slope north of the Maronite chapel overlooking the ravine, viz. 48 feet. This tree is by no means one of the oldest, but is at its full growth of maturity, and in vigorous health. The hoar, gaunt, and withered trunks of greatest antiquity are around the usual camping ground at the S.E. corner of the group.” Dr. A. E. Day wrote to me as follows on their actual condition more recently, in a letter dated Beirut, Syria, 9th November 1903 :— 1 Rustem Pasha informed Sir W. Thiselton Dyer that he had built a wall to protect the young cedars from grazing, but at a later period this was broken down. NONVdIAT LNOOW NO NVaaO ‘Lol ALvIg Cedrus 457 » ‘To the best of my knowledge there are five groves of cedars in Lebanon. The best known one, and that containing the oldest trees, is one in northern Lebanon above Bsharri. [Plate 127, from a photograph by Dr. Van Dyck, shews one of these trees.] The condition of that has, I think, not changed much in thirty years. I am sure that no new trees have grown up in it. A few of the oldest ones have lost branches, or have entirely perished. The grove is a favourite resort in summer for Syrians and for foreigners. A few hours south and west of Bsharri is the village of Hadeth-el-Jubbeh, or Hadeth, as it is often called, though there are a number of Hadeths in Lebanon. Within a half-hour to the south of Hadeth is a fine grove of young trees which, I think I have been informed, was started and has been preserved by a Greek or Maronite bishop. The remaining three groves are near each other, on the western slope of the main ridge of Lebanon, the most northern one being a few miles south of the Beirut-Damascus road as it crosses the ridge. The most northern of the three is above the village of Ain-Zahalta, the next is above Bartk, and the third is above Maasir, each being known by the name of the village near it, being also the property of that village. The smallest grove, but that containing the oldest trees, is that of Madasir. The Barik grove is the most extensive of all the five in Lebanon, and contains many young trees in all stages of growth. Most of the trees are upon a very steep slope, but in the upper part of the grove there are various knolls and hollows, affording a few charming spots for camping. I am sorry to say that this fine grove suffers much from being cut. The people of Barfik obtain from it roof-beams and wood for fuel, and I am informed that they are discussing selling a large part of it to be felled for pitch. I have failed to find a single large tree in the Barak grove which has not been cut off, with the result that several branches have taken the place of the principal stem. The ordinary Arabic name of the cedar is ‘Arz,’ but the natives of the villages near the three southern groves call the tree ‘ Ubhul.’” The cedar is also found in the Taurus and Anti-Taurus ranges in Asia Minor, extending from the province of Caria’ in the west to near the frontier of Armenia in the east. It forms a considerable part of the coniferous forest, which, in a few scattered localities, covers the mountains between 4000 and 7000 feet. It is usually associated with Ep as ue 13 0 17 6 1865. : . 15 0 16 0 I5 II 16 5 22 1 1880. ‘ . 16 0 17 3 7 23 18 oO 1895 . : » 16 7 we 18 9 ‘i 26 oO 1900. , . 16 9 18 oO I9 5 I9 5 27 1 1904 . ; . 17° 6 ig 2 19 5 I9 9 a7 3 At Langleybury, Herts, a large cedar® was growing in the grounds of 1 History of Windsor Great Park and Windsor Forest, Plate 14. ® Figured in Gard. Chron. xxvi. 521, f. 102, and 553, f. 109 (1886). 3 Gard. Chron. xiv. 392 (1880). think ay = PLATE 131 LEBANON CEDAR AT PETWORTH WIPHNATI LV UVGHO NONVAAT ‘cfl ALVIg LEBANON CEDAR AT STRATTON STRAWLESS UVAONVIOUIA LY YVGHO NONVIAT Cedrus 463 E. H. Loyd, Esq., in 1880, which at 4 feet from the ground measured 22 feet 4 inches in girth, with a height of 107 feet. At Chart Park, Deepdene, Surrey, a tree 95 feet high is 19 feet 3 inches in girth, and divides at 12 feet up into ten upright stems. At Chorleywood Cedars, near Rickmansworth, there are seven very fine cedars, standing on high ground, which form a landmark in the country, and are said to measure about 23 feet in girth. Another at the same place was recently struck by lightning, and cut down. At Beechwood, near Dunstable, the seat of Sir Edgar Sebright, Bt., there are some very fine old cedars, of which the largest, as I am informed by Miss F. Wool- ward, measures 100 feet by 28 feet 4 inches, with a spread of branches 46 feet across. Another, 90 feet by 23 feet, has branches from 50 to 60 feet long. At Chiswick House there are a number of very fine cedars still surviving, though not so many as when the late Mr. Barron, Superintendent of the Royal Horticultural Society’s Gardens, measured them in 1882. The two largest trees at that time were 16 feet and 18 feet in girth, and when I saw them in 1904 the two largest were 16 feet 5 inches and 18 feet 5 inches. These are supposed to have been planted about 1720, but are nothing like so fine as many trees at a greater distance from London. ’ One of the most remarkable cedars in England, on account of its habit, stands in what was probably a dense grove of tall silver firs near the site of the old house at Stratton Strawless, the home of Robert Marsham, who planted it when 1} feet high, in 1747. When described by Grigor’ in 1841 its stem was 44 feet high, free from branches, and 12 feet 2 inches in girth at 6 feet. His plate shows that it has changed but little now. When Mr. Birkbeck showed it to me in April 1907 it was about 80 feet high and 163 feet in girth, and though some branches in the crown had been broken off, it looks remarkably vigorous (Plate 133). A fine tree of the same type, but not equal to the last mentioned, is in a sheltered part of the grounds at Gosfield Hall, Essex, the property of Mrs. Lowe. It is 80 to go feet high, by 144 feet in girth, with a clean stem up to about 60 feet, and a flat, spreading crown of branches at the top. A cedar which is growing at Birchanger Place, near Bishop-Stortford, for a photograph of which I am indebted to the owner, T. Harrison, Esq. (Plate 134), is strikingly different in habit, and of its type is one of the most beautiful and perfectly shaped in England. It is about 60 feet high and 17 feet in girth, the branches covering an area at least 100 yards in circumference. Another tree of the same type, but not so symmetrical, grows at Billing Hall, the seat of Valentine Cary Elwes, Esq., near Northampton, and measures about 60 feet by 19 feet 5 inches. The branches, which spread over an area about 100 paces round, are supported by a great number of wooden props. In the west of England this tree does not attain the same size and beauty as in the drier counties, the largest I have seen in Devonshire being at Bicton, which is about 214 feet in girth, At Castlehill, in the same county, there is a tree about 80 feet by 14 feet 9 inches; and at Sherborne Castle, in Dorsetshire, there are a 1 Eastern Arboretum, p. 84, plate opposite p. 104 (1841). 464 The Trees of Great Britain and Ireland number of fine trees, the largest of which I found to be about 105 feet by 16% feet, dividing at about 15 feet up into five or six tall, straight stems. In Wales I have seen none remarkable for size except a tree at Maesleugh Castle which is about 100 feet by 16} feet, with a clean stem about 20 feet high. In Lincolnshire and Yorkshire, even where the soil is good, the cedar does not attain the same dimensions as in the south of England, but it ripens seed at least as far north as Syston Park, where there are some trees near the house in an exposed position at an elevation of about 500 feet above sea-level, which show remarkable variation in colour. When J saw them on 16th June one was only just opening its buds, and looked quite black in comparison with others whose new leaves were well out and of a very glaucous colour. This colour is reproduced by their seeds, for two young trees raised from them, which were kindly given me by Sir John Thorold, are so glaucous that every one who has seen them in my nursery has mistaken them for C. atlantica glauca, while two seedlings of C. atlantica from Cooper’s Hill are not distinguishable from C. Lzdanz. In Cumberland there are two splendid cedars at Eden Hall, the seat of Sir R. Musgrave, Bart., which, according to a paper’ by Mr. Clark of Carlisle, were supposed to be 270 years old, and one of them measured 86 feet by 223 feet, the other 86 feet by 21 feet, with a spread of 101 feet in diameter. At Alnwick Castle, Northumberland, there is a tree in the wood near the Duchess bridge, measuring 69 feet by 7 feet 3 inches. The finest avenue of cedars I know in England is that at Dropmore, of which I give an illustration taken from a photograph made in 1903 (Plate 135). This avenue is said’? to be composed of Lebanon cedars planted probably about 1844, and if really so young as this, is a very remarkable instance of the rapid growth of the cedar in this country. There is, however, some doubt as to whether they are Cedars of Lebanon or Algerian cedars, and though I have made inquiries from Mr. Fortescue I cannot ascertain with certainty their origin. The best account I know of the Cedar of Lebanon in Scotland is given in the Transactions of the Horticultural Society, vi. 429, in 1826, by Mr. J. Smith, then gardener to the Earl of Hopetoun, and as this shows the rate of growth of the cedar to be, even in that latitude, greater than that of any other tree, I quote it as follows :— ‘“‘The extensive pleasure grounds at this place were laid out about the year 1740, and in that and the subsequent years a great variety of curious ornamental trees was planted, which are now of considerable size, and in great beauty and perfection: among these are three cedars, which were planted in the year 1748. The two largest are growing in a favourable deep soil, which although not wet inclines to be moist; the third is on a gravelly soil, beside a rill of water. Their situation is well sheltered, and about 100 feet above the level of the sea. In the year 1797 the third tree was the largest, and Dr. Walker,’ who noted its size at that date, ascribes its superiority to the wetness of its situation. He has 1 Trans. Eng. Arb. Soc. 1887, p. 135. ® Gard. Chron. xxv. 138, fig. 52 (1899). 5 Essays on Natural History, 69 (1808). AMYONdTOUG LY ANNAAV UVAAO Cedrus 465 ca stated that it was 5 feet and 1 inch in circumference, but omitted to mention at what height from the ground this measurement was taken. In 1801 the dimensions of these trees, as well as of other kinds planted at the same period, were taken; the observations were repeated in 1820, and I am now enabled to add the present size of those which had been before noticed, as well as some others of different kinds but of the same age, which were not before attended to. The circumference of the trunks is taken in all cases at three feet above the ground, and it will be seen by comparing the different measures how much the cedars have exceeded all the other trees :— 1801, 1820. 1825. Ft. In. Ft. In. Ft. In. First cedar . ; ‘ ' IO O 13. If 14 Oo Second cedar ‘ : : 8 6 Io ot EE: of Third cedar : : : 7 10 9 «69f 10 68 Sweet chestnut . : ; 10 =I i: I2 Oo Beech : : ; ’ 9 4 Oo 11 10 3 Sycamore . : ‘ : 8 II 9 «7 9g Ir” I visited Hopetoun, the seat of the Marquess of Linlithgow, in April 1904, and found that two of these cedars still survive in good condition, the larger being about 80 feet high and 23 feet 8 inches in girth, the other about 88 feet by 13 feet. There is a fine cedar at Biel, East Lothian, the seat of Mrs. N. Hamilton Ogilvy, which is said to have been planted in 1707 by Lord Belhaven, to com- memorate the Union of England and Scotland. According to Mr. S. Ross’ it was, in 1883, 75 feet high by 174 feet in girth; but I am informed by Mr. T. Muir that it is now 85 feet high by 19 feet 9 inches at 1} feet from the ground, with a spread of ror feet. At Moncreiffe House near Perth, the seat of Sir R. Moncreiffe, there is a well-shaped tree, which Hunter*® mentioned as bearing many cones and measuring 66 feet by 11 feet. In 1907, when I saw it, it was about 80 feet by 144 feet at 3 feet from the ground. At Dupplin Castle, the seat of the Earl of Kinnoull, there are two cedars of which the best shaped measures 86 feet by 16 feet 10 inches, and the other is 18 feet 8 inches in girth at 3 feet. At Murthly there are two good trees, which, though probably not much over seventy years old, measure 74 feet by 9 feet 3 inches and 70 feet by 10 feet 6 inches respectively. The best I have heard of in the west of Scotland are one at Mount Stuart in Bute, which Mr. Renwick tells me is 64 feet by 8 feet 3 inches, and another at Erskine House, near Renfrew, which is 62 feet by 1o feet at 14 feet from the ground. In the N.E. of Scotland it also grows well; there are two very fine trees at Beaufort Castle. According to the measurements given me by Mr. G. Brown the largest of these is 73 feet by 22 feet 8 inches at 3 feet from the ground, dividing at five feet into four large stems, which measure from 9 to 11 feet in girth, The other is the same height and 16 feet in girth, At Brahan Castle there are also some fine cedars. 1 Woods and Forests, Dec. 26, 1883, p. 59- 2 Woods, Forests, and Estates of Perthshire, p. 135 (1883). 466 The Trees of Great Britain and Ireland In Ireland the Lebanon cedar has been rarely planted in comparison with its frequency in England; and Henry has not seen any large trees except one at Carton, which in 1903 was 93 feet by 14 feet 9 inches, and is said to have been the first planted in Ireland; and six fine trees’ at Anneville near Dundrum, Co. Dublin, the largest of which was 143 feet in girth in 1904. There is an excellent article on cedars by Dr. Masters in the Gardeners’ Chronicle for Oct. 17, 1903, giving an illustration of the historic tree in the Jardin des Plantes at Paris, about which many incorrect statements have been published. Carriére? gives 1736 as the date at which it was planted, from seed brought from England by Bernard de Jussieu in 1735. From this seed was also derived the cedar at Montigny (Seine et Oise), and the one at Beaulieu, near Geneva. Carriére states that the cedars at Geneva produce seeds so freely that but for the scythe of the mower it would form forests on the shores of the Lake. In a letter from M. Maurice de Vilmorin I learn that the Montigny cedar® is now probably the best in France. About 1855 it was 7 metres in girth at two metres from the ground, and it is now 7.90 metres at the same height. There is another tree at Vrigny, the residence of M. Duhamel de Monceau, near Pithiviers, Loiret. His notes of 1874 state that this tree, planted in 1744, had suffered much from the frost of 1870-71, when two-thirds of its branches were frozen. It measured about 8 metres in girth. I saw a very fine cedar in the grounds of M. Philippe de Vilmorin at Verriéres, near Paris, in May 1905, which measured 87 feet by 13 feet; and also visited the tree in the grounds of Madame Chauvet at Beaulieu, near Geneva, which is now considered to be the finest on the Continent, though not equal to several English trees. It is a well-shaped spreading tree about 100 feet high, though difficult to measure exactly, and 16 feet in girth, with a spread of 102 feet. TIMBER What is called cedar in commerce is usually the wood of Cedvela odorata, a tree found in the West Indies and Central America. The wood of the so-called pencil cedar, Juniperus virginiana, is also often known as cedar,’ and this can be distinguished at once by its colour and smell from the true cedar. A case was recently tried in London with regard to the quality of the cedar used in panelling a room at Packington Hall, in which it was stated in evidence by a so-called expert that there were three kinds of cedar known in the trade, “ English grown, pencil cedar, and Californian cedar,” “ the latter used for inferior work.” This is a not unusual instance of the gross ignorance which prevails in England among users of timbers as to their names and native countries, and this ignorance has led to many costly lawsuits. The Lebanon cedar grows so fast in England under favourable circumstances that the wood is of a much softer character than it is in Syria, but it may be used for 1 These are said by Loudon, 4rd. e¢ Frut. Brit. i. 114 (1838), to have been brought direct from the Lebanon by an ancestor of Lord Tremblestown, and to be the oldest in Treland. 2 Traité Conif. 78 (1867). 3 An account of it in Revue Horticole, 1907, p. 465, gives the dimensions as 105 feet high by 24 feet in girth at one metre from the ground. * In the Eastern States it is known as red cedar, but this term is applied to 7zuya plicata in the Pacific States, Cedrus 467 many purposes of internal decoration ; and the best instance of such use that I know is at Broom House, Fulham, the residence of Miss Sulivan. This lady, having a cedar blown down on her lawn, had it cut into boards, of which there were sufficient to floor and panel the whole of a good-sized drawing-room. When the wood is carefully selected, its pale pink colour and handsome figure make it very ornamental. Its value in commerce is, however, low, because neither the supply nor the demand is regular; and the cost of removing and sawing up large cedar trees is so great, that I was offered a tree containing over 300 cubic feet for nothing if I could get it away; and the Earl of Powis told me that some large trees which were blown down at Walcot were unsaleable, and were eventually used as a cheap material for the kennels of the United Foxhounds. (Hajek CEDRUS BREVIFOLIA, Cyprus Cepar Cedrus brevifolia, Cedrus Libant, Barrelier, var. brevifolia, Hooker, Journ. Linn. Soc. (Bot.), xvii. 517 (1879); Beissner, Nadelholzkunde, 300, fig. 75 1 (1891). Resembling C. Libani in characters of leading shoot and branchlets, but with very short leaves, not exceeding 4 inch in length. Cones smaller than those of C. atlantica, which they resemble in other respects. The cedar was discovered in the mountains of Cyprus in 1879 by Sir Samuel Baker, whose specimens were described in the same year as Cedrus Libant, var. brevifolta, by Sir J. D. Hooker, who considered this form to agree more closely with the Algerian than the other cedars, resembling it in the small size of the cones and in the general characters of the foliage. The best account? of this cedar forest is by Sir Robert Biddulph, who wrote as follows in 1884 to the Director of Kew :— “The cedar forest occupies a ridge on the principal watershed of the southern range, and about 15 miles west of Mount Troodos. The length of the forest is about 3 miles, its breadth very much less. A few outlying cedar trees were visible on neighbouring hills, but on the ridge they were quite thick, and probably many thousands in number. I took the height above the sea by an aneroid barometer, and found it to be 4300 feet. The trees are very handsome and in good condition, but comparatively young. The smallest seemed to be from ten to fifteen years old; the largest, I am told by the principal forest officer, are probably not over sixty or seventy years. The worst feature is that there were no seedlings or young trees under ten years; and indeed this is the same with regard to the pine forests. It would seem as if the great influx of goats has been comparatively recent. I. made a tour through the heart of the forest last August. I started from a point on the west coast, and from thence ascended to the main watershed, and kept along the top till I reached Mount Troodos, taking three days to do it. The country through which we passed on the first day was perfectly uninhabited, and a mass of hills and forest, chiefly Pinus hale- 1 Beissner’s figure represents a ripe cone, collected on Mt. Troodos by Herr v. St. Paul. 2 Published in Mature, xxix. 597 (1884). Cf. also Proc. R. Geog. Soc. xi. 709 (1889). Ill 468 The Trees of Great Britain and Ireland pensis and the Ilex. The trees were in very great number, but there was a scarcity of young trees, and most of the old ones had been tapped for resin. On the second day we passed through the cedar forest, and the same sort of country as before, the Pinus Laricio beginning at an altitude of 4000 feet. We got as far as the monastery of Kykou that day, and the next day I continued along the watershed to the camp at Troodos. Our road as far as Kykou was a mere track on the side of the hill, in some parts rather dangerous, and we had to lead our ponies on foot, in many parts very steep. The difficulty on the road is the want of water at that elevation. We halted the first night at a beautiful spring, but we had to carry with us food for man and beast for the whole party, muleteers, etc. The scenery was wild and romantic. This spot is the centre of the moufflon ground ; three of them were at the spring when we approached it. It gave me a clearer idea of the forests of Cyprus than J ever had . before.” Madon, who wrote for the Government in 1881 a report’ on the forests of Cyprus, states that none of the trees were then apparently over eighty years old ; but that all were in a vigorous state of vegetation, with numerous. young trees of every age covering the soil. In addition to the main forest, three outlying clumps were seen by Madon,—one on the other side of the Ogostina valley, a group of forty-four very young trees at the Kykou monastery, and a third group much lower down. He considered that the cedars formerly covered the whole of the mountain heights from Machera to Livrami, being limited below by the zone of the olive tree. The timber can be recognised in the houses at Campo and in the carvings of the Kykou monas- tery, showing that the tree was formerly felled for building purposes. Madon noticed what has been confirmed by other observers, that the foliage varied in tint, most of the trees being glaucous. Hartmann, who has recently visited Cyprus, reports’ that the trees are remark- able for their broad, umbrella-like crowns, and average about 40 feet in height, 6 feet in girth, and 100 years in age. (A. H.) I am informed by Mr. C. D. Cobham, Acting Chief-Secretary to Government, in a recent letter, that the Cyprian cedars now occupy an area of about 500 acres in the centre of the Papho Forest, of which the summit, Tripylos, is 4640 feet above the sea, The cedars are mixed with pines and Ilex. There are also a few young trees at Kykou Monastery, a few in the vineyards at Chakistra, and one good specimen tree at Pedoullas. This last was purchased by the Government to preserve it from being cut for building material. There are a number of seedlings in the cedar forest, but these do not seem to have been affected by the exclusion of goats, as animals avoid the cedar when they. can find other food. The largest tree in the forest is in Argakis Irkas Teratsa, near the Kykou goatfold. It stands about 60 feet high, and measures 11 feet 6 inches in girth at five feet from the ground. A photograph of this tree is so precisely like a Lebanon cedar standing on my own lawn, which I see as I write, that I need not reproduce it. I may add that some cones sent from Cyprus in February 1905 were smaller than the cones from Syria ' Parly, Paper: Encl. 2 in Cyprus, No, 366, of 1881, p. 28. 2 Mitt. Deut. Dendr. Ges. 1905, p. 181. Cedrus 469 or those grown in England. Though at the time I did not think they were mature, yet the seeds contained in them have germinated and produced young plants, which in July were just putting forth their second whorl of leaves, but by the following May had increased very little in size, being much smaller than those of the same age from Swiss and English seed. Plants were raised at Kew from seed received in 1881; and two, now growing in the cedar collection at Kew, have attained only 6 feet in height, and are remarkable for their singularly short leaves and stunted bushy appearance. A number of them were killed by the winter, having been planted out when too young, which seems to show that this variety is more tender than the Lebanon tree. (FL. GE) CEDRUS ATLANTICA, Arttas or ALGERIAN CEDAR Cedrus atlantica, Manetti,! Cat. Plant. Hort. prope Modiciam, Suppl. Secundum, 9 (1845); Ravenscroft, inet. Brit. iii, 217 (1884); Kent, Veitch’s Man. Conifere, 409 (1900); Masters, Gard. Chron. x. 425, f. 53 (1891). Cedrus africana, Knight, Syn. Conif. 42 (1850). Cedrus Libani, Mathieu, More Forestitre, 564 (1897). Pinus atlantica, Endlicher, Syn. Conif. 137 (1847). Pinus Cedrus, Linneeus, var. atlantica, Parlatore, DC. Prod. xvi. 2, p. 108 (1864). Abies atlantica, Lindley and Gordon, Journ. Hort. Soc. v. 214 (1850). Young trees stiffer in habit than the Lebanon cedar, and with an erect leader. Branchlets not pendulous, covered with short dense pubescence. Leaves up to an inch (occasionally in cultivated specimens 14 inch) long, usually as thick as or thicker than broad. Cones shorter and more cylindrical than in C. Lzbanz; scales 14 inches in width, claw inflected. VARIETIES Var. glauca.—In the cedar forests of Algeria a certain proportion of the trees have glaucous foliage, the leaves being marked above with conspicuous white stomatic bands; but there is no other difference, and no foundation exists for the opinion, first mooted by Jamin,’ that the glaucous variety constitutes a distinct species.* The glaucous tint is an essentially unstable character,‘ trees occurring in the wild state in which glaucous leaves appear only on some of the branches. This variety often arises in cultivation. Beissner® mentions several varieties, which have been obtained in cultivation, as pyvamidalis, columnaris, and fastigiata,® characterised by peculiarities of habit ; and a variegated form in which the foliage of the young shoots is yellowish,’ but so far 1 Manetti gives the name only without any description, in the second supplement to his catalogue (1845), and not in the first supplement (1844) as usually stated. Endlicher first described the Atlas cedar from plants 6 inches high, sent in 1847 by Manetti from the Royal Gardens at Monza (Modicia) near Milan. 2 Decaisne, Rev. Hort. ii. 41 (1853). Cf. Gard. Chron. 1853, p. 132. 3 Cedrus argentea, Renou, Ann. Forest. iti. 2 (1854). 4 Cf. Fliche in Mathieu, Flore Foresti®re, 564, note 2 (1897). 5 Nadelholzkunde, 304 (1891). 6 Var. fastigiata, a pyramidal form, with branches ascending like those of the Lombardy Poplar, originated as a seedling in Lalande’s nursery at Nantes. Cf. Gard. Chron. vil. 197 (1890). 7 Var. aurea, young foliage of a rich golden colour, which changes to the normal green of the species in the second year. This variety is mentioned by Kent, /oc. czt. 470 The Trees of Great Britain and Ireland as we have seen these are not distinguishable as they get older. At Glasnevin there is a remarkable tree about forty years old, of which the stem is erect for about 25 feet, and beyond this bends over almost horizontally, extending laterally outwards for almost 12 feet ; and Elwes saw one of very slender and pendulous habit at Angers in France. DISTRIBUTION This cedar occurs in Algeria and Morocco. In the latter country its distribution is still scarcely known, though it was in Morocco that the Atlas cedar was first discovered. Philip Barker Webb visited! Tangiers and Tetuan in the spring of 1827, and from a native received branches of cedar which had been collected in the im- penetrable mountains of the province of El Rif, where there were said to be vast forests. Webb’s specimens are preserved in the museum at Florence, where I saw them in December 1906. His discovery was published in an article? by De Candolle in 1837. Dr. Trabut? states that the tree occurs in the mountains behind Tetuan ; and it is supposed ‘ to exist to the south-east of Fez, where the traveller Rohlfs states that he saw larch growing. In Algeria the cedar® forms a considerable number of isolated forests, none of them of great extent, at altitudes between 4000 and 6900 feet. The tree appears to be indifferent to soil, as it grows both on limestone and on sandstone formations. No meteorological observations have been regularly taken in the cedar forests ; but in general, where the tree flourishes, snow lies for several months during winter, the temperature descending to 5° Fahr., and frost prevailing until May. In summer the weather is dry with moderate temperatures. In the following detailed account I have supplemented my own observations by consulting both the special pamphlet® concerning the cedar, published by order of Governor-General Cambon, and M. Lefebvre’s excellent book’ on the forests of Algeria. ‘The chief forests are those in the vicinity of Ouarsenis, Téniet-el-Haad, and Blida, and in the Djurdjura range in the province of Algiers ; and those on Mt. Babor, in the Mdadid mountains south of Sétif, and in the Aurés and Belezma mountains near Batna. The forest* of Ouarsenis, the most westerly in Algeria, lies in the mountains south of Orléansville. Here the cedar, mostly in mixture with Quercus Ilex, only covers an area of 250 acres. The forest near Blida, which is often visited by tourists, as it lies near the railway not far from Algiers, is 1700 acres in extent, and consists of cedars either growing pure or in mixture with the evergreen oak ; and it is, gener- ally speaking, in a poor condition. In the Djurdjura range, extending in an inter- rupted band on both slopes for nearly 4o miles, are the remains of an ancient forest, most of the trees either growing singly or in small groups on rocks and precipices, 1 Gay, Bull. Soc. Bot. France, iii, 39 (1856). 2 Bibliotheque Universelle de Gendve, 1837, PP. 439, 440. 3 Les Zones Botaniques de ? Algérie, 7 (1888). 4 Lefebvre, Les Foréts de Cedre, 1 (1894). 5 A fine picture of a forest in Algeria is given in Garden and Forest, viii. 335, f. 47 (1895). 8 Les Foréts de Cédre (Alger-Mustapha, 1894). 7 Les Foréts de l Algérie, pp. 406-421 (Alger-Mustapha, 1900). 8 Hutchison, Zrans. R. Scot. Arb, Soc. xiii. 211, states, but does not give his authority, that cedars were cut here, the diameter of which was so great, that it was necessary to join two saw-blades, each 6} feet long, in order to fell the trees. Cedrus - AgT * between 4900 and 6500 feet; but on the Haizer peak M. Britsch saw a few trees on the north slope as high as 7100 feet. The forest on Mt. Babor is of no great extent, but is an interesting one, con- sisting of a mixture of cedar, Quercus Mirbeckit, and Abies numidica, and will be described in our account of the last-named species. The brigadier in charge of this forest informed me that he had measured there a cedar 62 feet in girth. In the mountains of Maadid there are four distinct forests, generally speaking in bad condi- tion, and yielding scarcely any timber, though in one of them, called Ouled Khellouf, there are said to be some very large trees. The forests which are the most important from every point of view are those in the west near Téniet-el-Hdad, and those in the east in the vicinity of Batna, visited by me last January. The cedar occurs around Batna, both on the Aurés range and its spur Belezma. The forest of Sgag is 23 miles distant from Batna and covers 1200 acres. Between Batna and Biskra, about 20 miles north of the latter place, the forest of Djebel Lazereg is 1350 acres in area, and is noted for producing a peculiar kind of cedar timber, pink in colour and with a juniper-like odour. A very fine forest of considerable extent, 28,000 acres, lies around Mt. Chélia, the highest point in Algeria, 7500 feet altitude, 43 miles to the south-east of Batna; but it was practically inaccessible in January. In one part of it, the forest of Beni-Oudjana, 44,666 trees have been marked for felling, estimated to contain 3,615,000 cubic feet of timber, which will be offered for sale by the Government in the course of the present year. I visited the forest of Belezma, which is only 12 miles to the north-west of Batna. The whole wooded area here under government control is 140,000 acres in extent; but of this the cedar occupies only 22,000 acres, ascending the mountain to its summit, 6900 feet, and descending on northern slopes to 3600 feet, and on southern slopes to 4300 feet. The forest was badly treated in former years, whole tracts of the finest trees having been clean cut away and the timber used in building the town of Batna. The drought which prevailed from 1875 to 1881 caused serious damage to the remaining trees, and many died, most of which, except those that have been lately felled, are still standing, Felling is done regularly every year, only dead trees being removed. The sapwood of these has rotted away, but the heart- wood remains quite sound and unaltered. This timber is mainly used for railway sleepers, though some has been utilised in house-building and for making wood pavement and furniture. None of it appears ever to have been exported; and it is a great pity to see such excellent wood utilised only for rough purposes. The price obtained for it is as it stands very low, 1d. to 2d. per cubic foot ; yet it is fairly accessible, as the haulage to Batna is very cheap, but the rate by railway from there to Philippeville, the nearest seaport, is 15s. per ton. It snowed very heavily during my stay at the forester’s house near the top of the mountain; but so far as I could see, the cedar only grows here in a dense condition in the young stage, there being in the ravines fine stands of cedars 30 feet high, which are slightly mixed, like the rest of the forest, with Quercus Ilex, Juniperus thurifera, and Juniperus phenicea, These young trees are narrowly 472 The Trees of Great Britain and Ireland pyramidal in form, with erect stiff leaders; but in slightly older trees the leader begins to incline over on one side, and the branches to thicken and elongate, and this process being continued, eventually the tree assumes when old the habit of Lebanon cedars, as we see them in English parks. In other parts of the forest the older trees are more or less scattered with the same admixture of junipers and evergreen oak, the undergrowth being Phillyrea and broom. The cedar appeared to be slow in growth, the annual shoots of young vigorous trees not exceeding three or four inches in length. From observations made in one section of this forest the tree shows at different ages the following dimensions :— Height of Market- AGE. Diameter. able “raber: Total Height. Feet. Inches. Feet. Feet, 125 years : : : ; 2 7 46 98 160. 4; 2 114 524 105 200 _,, : : . : 3 3 59 115 255» : : . : 4 3 59 125 305», 4 11 59 125 An official document, which I saw at Batna, gave the total number of sound trees over 40 inches in girth as 265,500, estimated to contain between ten and eleven million cubic feet of timber, the total timber in the forest, young and old trees, cubing 16,000,000 feet. In addition, there is still standing 900,000 cubic feet of dead timber. In a few spots, as in the Chellala-Bordjen section, there are rather dense stands of old trees, which run to 7000 cubic feet per acre; but there are large tracts in parts of the forest which have scarcely 150 cubic feet to the acre. The tree produces seed abundantly every two or three years ; and regeneration is good in favourable situations, as in northerly ravines. The cones? disarticulate in November, after the autumnal rains, but if the weather is exceptionally dry, do not open. Seedlings appear under dense cover, but in such situations grow slowly, and do much better in the partially open places between large trees. The wide- spreading branches which the tree ultimately produces show, I think, that in old age it requires a great deal of light, and tends to grow in a more or less isolated condition ; but until middle age the trees bear crowding without injury. In the bare parts of the mountain, where the trees were cut away many years ago, artificial planting has been tried on a small scale, and has succeeded on northern slopes when two-year-old seedlings have been planted in autumn. Plants put out in the spring on the southern slopes have died of drought, which is the great enemy to both artificial and natural regeneration. The forest of Téniet-el-Haad is about a day’s journey from Algiers—four hours by rail and thence seven hours by the coach to the town, which is distant from the cedars about an hour’s walk. The mountain-range runs in a N.W.-S.E. direction, the cedars ascending to the summit of the crest, 5900 feet, and descending on the north side to 4250 feet, and on the south side to 4900 feet, there being a zone of 1 Only the central part of the cone contains good seed. In January the basal scales of many cones were still remaining around the central axis, the other scales having fallen much earlier. PLATE A CEDARS AT TENIET EL HAAD ALGERIAN Cedrus ae a Quercus Ilex below, with which the cedar slightly mingles. The cedar forest occupies 2300 acres, four-fifths of this being on the north slope and one-fifth on the south slope, and consists of a mixture in varying proportions of cedar and Quercus Mirbeckit, the latter a beautiful tall tree with semi-evergreen foliage, often attaining 12 feet in girth, This mixed forest is nowhere very dense, except where there are young stands, and grows upon sandstone—the undergrowth being chiefly Rosa and Rubus, with Juniperus in the lower zone. The tallest cedar does not, I believe, exceed over 120 feet ; and the largest, which I measured and photographed (Plate 136, 8), are La Soltane, 98 feet high by 24 feet in girth, and Le Massaoud (Plate 136, a), 108 feet by 23 feet. Trees of peculiar shape are common; one, 108 feet by 19 feet, dividing into two stems at eight feet up; and another, Le Cédre Parasol, which stands on a rocky promontory, being a low tree with a peculiar broad- shaped umbrella-like crown. Around the forester’s house, Le Rond Point, at 4600 feet, there is a plateau of some extent, with many fine old trees having the habit of the Lebanon cedar as we see it in England.1 No felling is done at present in this forest, which is rapidly improving in value owing to the entirely successful natural regeneration, cedars being present in all stages of growth. The wood of the cedar, though without resin-canals, contains a quantity of resin, which gives it a peculiar, penetrating, and distinctive odour.2 At Batna, fibanol, a kind of resin, is obtained by distillation of the sawdust of old trees. This product is very valuable in the treatment of inflammation of the mucous membranes, and is said to be curative in influenza. Cedar wood contains a large amount of white sapwood, 25 to 50 annual rings, with a brown or brownish-yellow heartwood. The heartwood is homogeneous and fine in the grain, and takes an admirable polish. It lasts indefinitely, trees which were cut down fifty years ago in the forest at Batna remaining still on the ground quite sound, and when not exposed to the air is imperishable. Pieces of cedar wood have been found in tombs which are supposed to belong to the Punic period, and portions of ancient mosques built of cedar are in perfect condition. Placed in water, the heartwood becomes very hard; and vats made of it, which have been buried in sand for thirty years, are not only well preserved, but cannot be cut by an axe. The wood of dead trees can be used at once, but that of living trees requires to be seasoned carefully for six or twelve months. Though the timber is used in building, it is rather heavy for that purpose, and has no great elasticity or resistance to flexion under a heavy weight. It is, however, well suited for the finest kinds of cabinet-making. (A. H.) CULTIVATION The seed ripens in most seasons in England at least, as well as that of the Lebanon cedar, and will sometimes come up naturally near the parent trees, as at Cooper’s Hill near Windsor, from whence I transplanted two self-sown seedlings to my own garden. 1 An excellent illustration in Garden and Forest, viii. 335 (1895), shows the flat-topped habit of mature trees in their native forest. ; 5) 2 The odour disappears after exposure to the air for a few years, and is not noticeable in the cedar furniture which is so common in the houses at Batna and Téniet-el-Haad. Cf. Lefebvre, Les Foréts de 1’ Algérie, 350 (1900), 474 The Trees of Great Britain and Ireland When staying at Heythrop Park, Oxfordshire, in March 1901, I went out ona morning when the frost was so hard that the hounds could not hunt till noon, and found seeds which had germinated on the ground beneath a glaucous cedar. The radicles were protruding from the seeds, in some cones which had not fallen ; I took them home and planted them, and have now several healthy young trees about a foot high. I also sowed a quantity of imported seed in the open field, where they germinated well, but the plants were all destroyed by mice, frost, and drought in the first season, though seedlings raised in the nursery stood the winter without protection. As the seed can be procured in quantity at a cheap rate from Messrs. Vilmorin of Paris, I should recommend its being sown in a frame and protected for two or three years, after which it will require two to three years more in the nursery before planting out. . The tree seems to like lime in the soil, and will, in my opinion, prove a valuable timber tree if planted in open woods, in warm, dry soils, sufficiently close together to prevent its branches from developing too much, and possibly if mixed with beech it might thrive better than alone. As regards the relative rate of growth of the Atlas and Lebanon cedars we have the evidence of M. André Leroy, the well-known nurseryman of Angers, who, in the Belgique Florticole, 1867, p. 59, gives the following measurements :— LEBANON CEDAR AtTLas CEDAR on Height. Rew Height. Metres. Centimetres. Metres. Centimetres. I year ‘ fe) 6-8 I year. oO 10-15 2 years fe) 12-15 2 years fe) 20-30 A cay oO 18-25 a 4 oO 40-50 45 : re) 36 4) Sy ‘ I ° 5» o 50 5 oo» I 75 GO as ro) 75 oO 2 50 7 oe I fe) ae gs 3 and upwards, After seven years of age, he states that the annual growth was often more than one metre, and mentions a tree only twelve years old, from seed, which was one metre in circumference (I presume at the ground). He also says that it is easier to trans- plant, and endures exposure and bad soil better than the Lebanon cedar, and believes that it will prove a valuable tree for planting on barren wastes where nothing else will thrive. These remarks, no doubt, will apply better to the soil of Central France than to England, but I have the highest possible opinion of the hardiness of the tree, and have found it endure the damp, cold, and early and late frosts of the Cotswold hills in a way that few other conifers will do. So far as my experience goes, however, it is not a tree which can be transplanted without some care in a small state, and when it has had its roots cramped in small pots, as is often done by nurserymen for con- ALGERIAN CEDAR AT ASHAMPSTEAD Cedrus 475 venience of sale, is rather apt to die. I am not aware that it has ever yet been tried in quantity under forest conditions; but, so far as I have seen, it is not subject to insect or fungoid diseases which attack and kill the deodar. ’ Many of the grafted trees of the glaucous variety, which are usually sold by nurserymen, are one-sided and unsightly objects, for a good many years after planting at any rate; and though it is claimed by some that grafting, if properly done, does not permanently disfigure the tree, yet I would always prefer seedlings. Even if not quite so glaucous in colour as the best of the others, a certain number of this tint will generally appear among them. The date at which the Algerian cedar was first introduced to this country is somewhat uncertain; but it must have been subsequent to 1844, and if any older ones exist they cannot be recognised with certainty. Several trees appearing older than this have been supposed to be African, on account of their habit and cones, but there is nothing on record to prove it. According to Ravenscroft, the oldest of which we have an exact record were raised at Eastnor Castle in 1845, from cones gathered by Lord Somers himself at Téniet-el-H4ad. In December 1860 the tallest of these was 184 feet; in December 1866, 31 feet. When I measured it in 1906, it was 77 feet by 8 feet 1 inch. REMARKABLE TREES The tallest tree that I have measured in England is at Linton Park, Kent, and is a glaucous tree, which, from its shape, seems to be grafted, though there is no evidence of this. It was 80 feet high in 1902. The largest recorded at the Conifer Conference in 1891 was at Mulgrave Castle, Yorkshire,! the seat of the Marquess of Normanby. It was then 66 feet by 5 feet 10 inches. Mr. Corbett informs me that it is now 72 feet by 8 feet 4 inches. On Ashampstead Common, Berks, there is a handsome and well-grown tree which has grown up in a semi-wild condition among other trees, and which was 63 feet by 64 feet when I last saw it in 1907 (Plate 137). At Ashridge there are several fine glaucous trees, raised from seeds, which were brought. by Earl Brownlow, in 1862, from Téniet-el-Haad ; the best of them already measures 58 feet by 6 feet. At Merton Hall, Norfolk, there is a very well-shaped tree measuring 60 feet by 6 feet. At Bicton there is a fine tree measuring 68 feet by 7 feet 6 inches. At Coldrinick, in Cornwall, there is a well-shaped tree which, in 1905, was 64 feet by 5% feet. At Heanton Satchville, North Devon, I saw a healthy young tree ina shrubbery, which was clear of branches to 20 feet up, and though 48 feet high, was only 2 feet 7 inches in girth, showing the ability of this cedar to thrive without much space, even in a climate so much damper and cooler than that of Algeria. At Tortworth there is a cedar about 50 feet high with very short leaves, and remarkably fastigiate habit, which seems to belong to the variety named /astzgzata. In Scotland I have not seen any so large as in England; but the tree grows 1 A tree at Grimston, near Tadcaster, Yorkshire, reported in 1900 to be 70 feet high and 13 feet in girth at three feet from was said to be sixty-five years old, is probably a Lebanon cedar. Cf. Gard. Chron. xxviii. 210 (1900). the ground, which A III 476 The Trees of Great Britain and Ireland well at Murthly and other places. At Smeaton-Hepburn, a tree,’ planted in 1847, was, in 1902, 69 feet high and 63 feet in girth. At Fordell, in Fifeshire, the property of Lord Buckinghamshire, 1 am informed by Mr. Sibbald that a number of cedars were planted by Mr. Fowler, then head gardener, 42 years ago on a damp sandy soil and well sheltered by other trees. The average height of the Algerian cedars in 1906 was 48 feet, with an average girth of 4 feet 4 inches, and of the deodars 33 feet by 34 feet. The majority of them are in good health, though the Algerian have made by far the best trees, and as the soil and climate of Fifeshire do not seem to be so favourable to the growth of trees generally as those of Perthshire, Morayshire, or parts of Ross-shire, this seems to prove that the tree may be planted in Scotland with good hopes of success. The finest Atlas cedar in Ireland is at Fota, and is of the glaucous variety. It was planted, according to Lord Barrymore, in 1850, and measured in 1904 83 feet high by 7 feet 7 inches in girth (Plate 1 38). At Carton, the seat of the Duke of Leinster, a tree, which is, from its habit, apparently an Atlas cedar, was, in 1903, 80 feet high by 9 feet in girth, At Powerscourt a glaucous specimen was in the same year 50 feet high by 5 feet in girth. In the south of France and North Italy this tree grows better and faster than in England. Perhaps the best that I have seen are in the public garden at Aix en Savoie, where there is a grove of splendid trees 90 to 95 feet high, though only planted in 1862. They average 6 to 7 feet in girth, and there are many self-sown seedlings near them. On the shores of the Lago Maggiore the tree succeeds per- fectly, several fine trees in the grounds of the Villa Barbot near Intra being 90 feet or over, and one 74 feet in girth. It seemed to me likely to become a most valuable forest tree in this region. (H. J. E-) CEDRUS DEODARA, Deopar Cedrus Deodara, Lawson, Agric. Man. 381 (1836); Loudon, Avd. et Frut. Brit. iv. 2428 (1838); Brandis, Forest Flora, 516 (1874), and Indian Trees, 691 (1906); Ravenscroft, Pinet. Brit. iii. 225 (1884); Masters, Gard. Chron. x. 423, f. 52 (1891); Kent, Veitch’s Man. Conifera, 411 (1900). Cedrus Liban, Barrelier, var. Deodara, Hooker, Himal. Journ. i. 257 (1854), Wat. Hist. Rev. ii. 11, tt. 1-3 (1862), and FZ. Brit. Ind. v. 653 (1888); Collett, Hora Simlensis, 486 (1902) ; Gamble, Ind. Timbers, 710 (1902). Cedrus indica, Chambray, Arb. Res. Contf. 341 (1845). Pinus Deodara, Roxburgh, Hort. Beng. 69 (1814). Abies Deodara, Lindley, Penny Cycl. 9 (1833). Young trees with pendulous leader. Branchlets always pendulous, grey and densely pubescent. Leaves up to 2 inches long, as thick as broad. Cones large and broad, ellipsoid, 4 to 5 inches long by 3 to 4 inches in diameter, rounded at the apex ; scales 2 to 24 inches wide, with claw not inflected, usually less tomentose than in the other cedars. 1 Sir A. Buchan-Hepburn in Proc. Berwick Nat. Club, xviii, 210 (1904). AT FOTA CEDAR EP RIAN 1 ALC Cedrus 477 VARIETIES A considerable number of varieties have arisen in cultivation, ten being mentioned by Beissner.? 1. Var. albo-spica. Growing shoots during spring and early summer of a milky- white colour. Trees of this kind at Dropmore? are pyramidal in habit, and make splendid growth. At Grayswood,* Haslemere, a bushy form with this peculiar foliage has been noted. 2. Var. robusta. Branchlets stout; leaves longer and thicker than in the ordinary form. 3. Var. crasstfoha. Branches short and stout ; branchlets not pendulous ; leaves short and thick. 4. Var. verticillata. Branchlets whorled. 5. Var. fastigzata. Fastigiate in habit. 6. Varieties with variegated foliage and with bright yellow leaves have also been noted. The glaucous tint has appeared in cultivation, and is met with in the wild state. A very glaucous tree at Castlewellan has been named var. xzvea.t Trees with thin, shining, deep green foliage have been distinguished as var. wzrzdis. (A. H.) DISTRIBUTION The deodar is found in the Western Himalaya; and extends eastwards to the Dauli river in Kumaon, occurring at 4000 to 10,000 feet, most common at 6000 to 8000 feet. It extends westwards through Kashmir to the Peiwar forests in the Kuram valley of Afghanistan. According to Gamble, from whom I take the most of the following account, it is a gregarious tree, but rarely forms pure forests, though exceptions are met with, generally in the form of sacred groves; usually it is associated with Prcea Morinda and Pinus excelsa, and three species of oak in their various zones. Sometimes the silver fir (4dces Pendrow) accompanies it, but more rarely ; the cypress (Cupressus torulosa) in its favourite localities joins it; the yew is often found under it; and at low elevations it mixes with Pzxus longzfolza. Among other trees commonly found with it may be mentioned Betula alnoides, Populus ciliata, sculus indica, elm, hazel, hornbeam, maples, bird-cherry, holly (Llex dipyrena), Prerts ovalifolia, and rhododendron; while among the shrubs commonly found in deodar forests may specially be noted species of Berbers, Indigofera, Desniodium, Cotoneaster, Euonymus, Salix, especially Salix elegans, Viburnum, Lonicera, Parrotia, and rose, while Clematis montana, Vitts semicordata, and ivy, are frequently met with climbing over and festooning its branches. In the outer ranges the deodar forests chiefly clothe the northern and western slopes of the ridges, while in the interior hills, to which the rainfall of the south-west monsoon still reaches, they are found on all aspects, but less pure. Beyond the region of the south-west monsoon the deodar is still found, but gets 1 Nadelholskunde, 307, 308. 2 Gard. Chron. xxxvii. 44, 76 (1905). 3 Gard, Chron, xxxvii. 59, 105 (1905). 4 Ibid, xxv. 399, fig. 146 (1899). 478 The Trees of Great Britain and Ireland gradually scarcer, and in such places its companions may be Pzmus Gerardiana and Quercus Ilex. The deodar can attain a very great size! Thomson? mentions one near Nachar, on the Sutlej, that was 354 feet in girth. Dr. Stewart measured one at Kiuarsi in the valley of the Ravi that was 44 feet at 2 feet, and 36 feet at 6 feet from the ground ; another about 900 years old was 34} feet in girth. Minniken records a tree at Punang, in Bashahr, that was 150 feet high and had a girth of over 36 feet, the clean bole being 45 feet long. Dr. Schlich measured a tree in the Sutlej valley 250 feet high with a girth of 20 feet. In the Dumrali block in the Tehri-Garhwal leased forests a fallen tree was unearthed 90 feet long and over 7 feet in diameter, which had been dead for at least 100 years, and was, when it fell, probably 550 years old. When cut up it gave 460 metre-gauge sleepers. I am indebted to Mr. J. H. Lace for the illustration (Plate 139) representing a group of deodars in the Himalayas. A great section in the corridor of the forest school at Dehra Duin is 23 feet in girth, with 665 annual rings. The number of annual rings to the inch varies much according to the elevation and rainfall, but averages about 8 to 12, though in the Kuram valley Bagshawe found an average of about 21. As an ornamental tree there are few in the world that can compare with the deodar. From the Lebanon cedar and the Atlas cedar it differs somewhat in appearance, but even to an expert, in the collections of Europe, it is not always easy to recognise to which of the three species a given specimen belongs. Roughly, however, the deodar is distinguished by means of its drooping branches and its longer needles. Two well-marked varieties are recognisable in the forests, the one with dark green, the other with silvery foliage. The latter variety, well known in European collections, is found wild in ravines at a comparatively low level. Gamble saw it in Jaunsar, in the upper Dharagadh, in ravines at from 4000 to 6000 feet, and believes that the variety comes true from seed. Deodar trees are often lopped for litter, and if the leading shoot is not damaged, the tree grows on well enough ; when the leading shoot is cut or damaged, the tree shows a great tendency to form others; and frequently several erect shoots, with the appearance of young trees, may be seen growing up straight from its branches. The deodar may be almost said to produce coppice shoots, for, as Brandis remarks, if only a small branch be left to a stump, it will send out shoots and grow well, eventually, perhaps, forming a new tree. In close forests deodars flower and seed rather sparsely ; for good seed bearers we have to look to the old trees on dry ridges, where they can get a large amount of sunlight. When the seeds are ripe the cones break up and the scales fall; the winged seeds are then carried by the wind for a short distance. It may be interesting to record the result of the examination of an average cone by Mr. B. B. Osmaston in October 1900. He found in the top part 25 scales, with 50 bad seeds; 1 Webber, in Forests of Upper India, 331 (1902), says: ‘‘I have seen deodars 40 feet in girth and 250 feet high, the age of which must be 1000 years or more”; and Pakenham Edgworth informed Bunbury that he had measured deodars 46 feet in girth. Cf. Lyell, Liye of Sir C. J. F. Bunbury, ii. 238 (1906). * Western Himalaya and Tibet, 64 (1852). AY DEODARS IN THE HIMAL Cedrus 479 in the middle 100 scales, with 90 good and 110 bad seeds; in the lower part 94 scales, with 188 bad seeds—the whole cone, therefore, giving 219 scales, with 438 seeds, of which 90 were good. CULTIVATION The best account we have of the introduction of the deodar is given by Ravenscroft, who states that the Hon. Leslie Melville sent seeds! in 1831 which were sown at Melville in Fifeshire, at Dropmore, and elsewhere. Lord H. Bentinck sent some to Welbeck in 1832, but it was not until 1841 that the Right Honourable T. F. Kennedy, then at the head of the Woods and Forests, took steps to procure seed in large quantities from the Himalayas. His proceedings are described at great length in the 7hirty-first Report of the Commissioners of Woods, pp. 168-172, and pp. 440-454 (1853), and further in the 74:rty-fourth Report (1856), pp. 87, 88, and pp. 120-122. From this it appears that 60,000 seedlings were distributed in the spring of 1856 amongst the New, Dean, and Delamere forests, and a further 40,000 were sent out in the following autumn. I am indebted to Mr. E. Stafford Howard, C.B., for information as to the results of these experiments as given in letters from the Hon. Gerald Lascelles and the late Mr. P. Baylis. The former says :—‘‘I have made search for any records of the planting of the deodars, but can find nothing worthy of quotation. It is a fact that it was very largely planted here, as we can see for ourselves,—more, however, as an avenue or ornamental tree than, strictly speaking, for timber. Large quantities were raised in the nursery at Rhinefield, which at that time was managed by one Nelson, who in a small book speaks of the very large experience he has had in raising and transplanting deodars. The tree is, however, a failure by reason of the way in which it suddenly dies off, unaccountably, when it is about forty or fifty years old. There are some notable successes, such as the grove at Boldrewood? and others, but I must have cut hundreds which had died off suddenly.” Mr. Baylis wrote on 8th May 1905: “I cannot give much definite information on the subject, though Crown Keeper Smith remembers some deodars being planted about 1857 along the sides of the rides in the High Meadow estate; but large numbers of these have perished, and there are no very fine trees among those that are left. A ride along the top of the Churchill enclosure was also planted about the same time with similar trees; but many of these also have died, and I cannot say that any of them have thriven well, though one tree has occasionally borne cones. I think that the climate here is too cold and damp for them to thrive, and that they cannot stand the damp cold of our winters in the Forest, though on the slopes of the Malvern hills they flourish fairly well.” This liability of the deodar to die after attaining considerable size has been often noticed, and, so far as I have observed, is most common on soils which are poor in lime. 1 A tree raised from these seeds was planted near the Director’s Office at Kew, and had attained a height of 32 feet in 1864. It became diseased and was removed in 1888. Cf. Kew Hand List of Conifere, xiv. (1903). 2 The best deodar at Boldrewood is now 64 feet high. 480 The Trees of Great Britain and Ireland The Earl of Ducie informs me that in 1854, and for several years afterwards, he planted many deodars at Tortworth, both on the old red sandstone and on the mountain limestone. Many of these have perished after thirty to forty years’ growth, without any apparent reason, except that in one case where only six out of about ninety remain, it is probable that they were infected with disease by the dead roots of beech trees which previously occupied the ground. Very few deodars at this place seem likely to attain a great age, and contrast unfavourably with the Cedar of Lebanon. But at Miserden Park, in the same county, on a dry oolite limestone, at an elevation of at least 600 feet, a line of deodars about sixty years old have remained healthy, though their growth here is much slower than at Tortworth. At Poltimore, near Exeter, there is a fine avenue of deodars which were planted in 1851-52, and have grown to an average height of 70 to 80 feet in 1906, most of them being extremely vigorous, but there are several blanks in this avenue. The cause of these deaths is explained by Mr. R. L. Anderson in a note published in the Quarterly Journal of Forestry, i. 216, who states that the fungus now known as Armillaria mellea, Vahl., was present on the roots of one of these deodars; and as the best means of checking its spread to other healthy trees, recommends trenching the ground round the affected tree, digging up and burning its roots, and scattering gas lime over the ground where they have been. At Castle Menzies, in Perthshire, of a number of deodars, which were planted by the late Sir R. Menzies about 1852 to commemorate the birth of his son, on soil which was too wet to suit them, though Zsuga albertiana and Picea sitchensis have succeeded very well close by, several are dead, and all are more or less stunted, though one of these trees measuring 74 feet in girth was successfully transplanted in February 1907, and had not lost a leaf when I saw it in the following July. I have not myself gathered any ripe seed of the deodar in England, but there is a tree growing in Kew Gardens between the main gate and the Director’s office which measures 37 feet by 4 feet 8 inches, and was raised from seed produced in 1861 or 1862 by a tree at Killerton, and sent by the late Sir Thomas Acland to Kew in February 1868. Mr. Smith, the then Curator of Kew, was so much impressed by the good quality of the soil from the top of Killerton Hill in which this tree was raised, that two truck loads of it were sent to Kew. The earliest record’ of the deodar producing fruit in England is of a tree at Bury Hill, near Dorking, which produced cones in 1852, when it was 28 feet high. Cones have also been borne on trees at Dropmore,? Sunninghill,’ Bishopsteignton ? near Teignmouth, Enys’ in Cornwall, and Fota’ in Ireland. Seedlings have been raised from home-grown seed at Rozel Bay’ in Jersey and at Bicton.® A deodar in Kew Gardens produced cones in 1887, according to a note in Gardeners’ Chronicle, ii. 248 (1887), where it is stated that the production of cones on this species in this country has hitherto been a rare occurrence. At The Coppice, Henley, the seat of Sir Walter Phillimore, Bart., and at Shiplake House, the 1 Gard, Chron. 1852, p. §82, and x. 279 (1891). 2 Ibid. x. 423, 435, 436, 492, 679 (1891). 3 Joid, 1869, p. 1279. Cedrus 481 residence of Miss Phillimore, there are deodars coning profusely at present, probably on account of the hot summer of 1906. At White Knights Park, Reading, there is a seedling now about 8 feet in height, and supposed to be 16 years old, which germinated on a vine border, the seed having come from a tree which measures 75 feet in height and 10 feet in girth. In India the cones are often much damaged by the larvee of a Pyralid moth which eats out the seeds, and the saplings are attacked by the well-known fungus Trametes radiciperda, which spreads underground through the roots from tree to tree. The leaves are also attacked by Uredinous fungi, especially by Accdium cedri, Barclay, which forms small yellow spots and causes them to fall. ’ As regards the comparative hardiness to severe winter frosts of the three cedars we have valuable evidence’ collected by Mr. Palmer in 1860-61. Reports were received from no less than 211 places in England, Scotland, and Ireland. “The winter of 1860-61 was the most severe that has happened since its intro- duction. It was a winter such as had scarcely any parallel for severity in the memory of man, and unless some general change of climate should take place, it may be looked upon as exceedingly improbable that any cold of greater intensity should again visit us. The effect of that winter upon the deodar may therefore be taken as a safe guide in judging of its suitableness for our climate; what the effect was we are, as already mentioned, enabled, through the kindness of Mr. Palmer, to state with accuracy. Mr. Palmer's record of observations shows that the deodar is by no means so hardy a tree as the larch, and also that it is the least hardy of any of the cedars. There is no instance of any of the larches reported to him having been injured by the cold of 1860; while out of the deodars growing at 211 places in Great Britain and Ireland, plants were killed at 55, and were uninjured only at 80, having been more or less injured at the remaining 76, a percentage of frailty much greater than we should have anticipated. The Cedar of Lebanon and the Cedrus atlantica proved more hardy, and about equal between themselves. The following summary will show the actual results of Mr. Palmer’s report on all three :— Total Places Proportion of cared gs Not injured. Injured. Much injured. _—Kiilled. Killed and i i Much injured. Cedrus Deodara . 211 80 50 26 55 I in 24 Cedrus Libani_. 81 51 19 6 5 tin 74 Cedrus atlantica . 74 48 19 2 5 I in 103 It may be interesting to notice in what proportion the three different parts of the kingdom suffered. It was as follows :— Total Places Proportion of ted Not injured. Injured. Much injured. Killed. Killed and sd asi Much injured. Scotland . ; 64 19 26 14 5 1 in 34 England . . 142 61 24 13 50 I in 24 ” Ireland. : 4 3 I 1 Published in Ravenscroft, Prve/, Brit. iii. 242 (1884). 482 The Trees of Great Britain and Ireland REMARKABLE TREES The two finest deodars, as regards size and symmetry, that I have seen in Great Britain are at Bicton, where cones were produced, according to Pznet. Brit., - as long ago as 1858. One of these on the lawn measured in 1902 was 80 feet by 11 feet 8 inches (Plate 140). The other is near the ornamental water in a more sheltered situation, and was then go feet by 9 feet 1 inch. Another of about the same height at Beauport has an erect top, and looks as if it might become much taller. The tallest reported at the Conifer Conference was at Studley Royal, and was then 70 feet by 74 feet; but when I visited that place I saw no very large tree of the kind. At Dropmore there is a handsome tree which in 1905 was 77 feet by 8 feet 10 inches, and had many of the woody knots embedded in the bark that are sometimes seen in the cedars, It is said to have been planted’ in 1834. | At Westonbirt, a tree, planted by the late Mr. Holford, about 85 feet by 8 feet 9 inches, is one of the largest and best shaped that I have seen. A deodar of peculiar habit at Linton Park, Kent, reported to be 79 feet high, is figured -in Gardeners’ Chronicle, December 12, 1903, fig. 159. At Barton there is a fine tree branched to the ground, which in 1904 was 76 feet by 93 feet. At Highclere there is a handsome tree about 75 feet by 8 feet 4 inches, which was planted by the then King of Spain in 1844. At Williamstrip, on rather heavy soil, there is a healthy tree of 72 feet by 8 feet. At Ombersley Court, near Worcester, there is a very fine tree 84 feet by 8 feet 4 inches, which has the erect habit of a¢/antica; but the drooping branchlets show it to be a deodar. At the Frythe, near Welwyn, Herts, a large deodar was cut down some years ago; and from the side of the stump there is now (1906) a young tree springing up, quite vigorous and healthy, and about 25 feet high. At Chart Park, Surrey, there is a tree 89 feet by 8 feet 11 inches; and adjoining this place, in the Tunnel Park, Deepdene, there is another fine tree 77 feet by 9 feet, both measured by Henry in 1905. At Fulmodestone, Norfolk, a tree planted in 1861 was in 1905 66 feet by 7 feet 4 inches in girth. At Shiplake House, near Henley, a tree, planted in 1852, was 73 feet by 7 feet 9 inches in 1905, and is bearing numerous cones in the present year. A deodar, growing on Haddington Hill, near Wendover, at 800 feet elevation, is 63 feet by 5 feet 10 inches. There are many trees of from 60 to 70 feet in other parts of England, but we have seen none which call for special notice. In Scotland the deodar is only hardy in the warmer parts of the country, and does not seem to have attained anything like the same dimensions as in England or Ireland. At Poltalloch, notwithstanding the wet and windy climate, it grows fairly well and has attained over 50 feet. At Rossdhu, on Loch Lomond, it is even taller. In Perthshire there are good specimens at Abercairney, Castle Menzies, and Dunkeld, which seem to have been planted after the great frost of 1860-61, which ' Gard, Chron, xxv. 138 (1899). 2 ? = A a ae 140. PLATE DEODAR AT BICTON Cedrus destroyed so many of this tree in the north. The tree at Abercairney is remarkably weeping in habit, and measured, in 1904, 51 feet high b feet 8 inches in girth, The best that we know in this county is nent . i Murthly, which is older and bore cones in 1892. It grows well at Gordon Castle, where there is a tree about 50 feet high, and as far north as Dunrobin in Sutherlandshire. At Conan House, Ross-shire, there is a healthy tree 47 feet by 9 feet 9 inches. At Leny, near Callander, there is a very old-looking but rather stunted deodar, which may have been introduced by the distinguished Indian naturalist Buchanan Hamilton, grandfather of the present owner, but when I saw it in 1906 it was only about 45 feet by 7 feet. At Smeaton-Hepburn, a tree! planted in 1841, when it was 24 feet high, measured in 1902, 55 feet in height and 6 feet 7 inches in girth. The finest deodar in Ireland is growing at Fota, Co. Cork, and measured, in 1903, 84 feet high by 7 feet 2 inches in girth. At Coollattin, Wicklow, there are two trees, one of which measured, in 1906, 53 feet by 6 feet 10 inches. At Hamwood, Co. Meath, a tree, supposed to have been planted in 1844, was 74 feet by 734 feet in 1905. At Mount Shannon, Limerick, there is a tree 66 feet by 8 feet 5 in. in 1905. At Emo Park, Portarlington, a tree measured, in 1907, 61 feet by 7 feet 4 inches, and was thriving; but in the dry climate of Queen's County, the deodar as a rule is not a satisfactory tree. 48 3 TIMBER The timber is the most important of any in North-Western India, and supplies a large quantity of railway sleepers, bridge, and building timber. Gamble says that it is rather brittle to work, and does not take paint or varnish well. It has also a very strong odour which, although pleasant in the open air, is not so in a room. It is extremely durable, probably with cypress (Cupressus torulosa) the most durable of Himalayan woods. Stewart mentions the pillars of the Shah Hamadin Mosque at Srinagar in Kashmir, which date from 1426 a.p., and were quite sound when he wrote. Its grain is so straight that the logs can be split into boards, which are afterwards trimmed with an adze; and shingles for roofing, according to Webber,’ stand the changes of climate for centuries without any sign of decay. , The weight of well-seasoned dry wood of average growth is about 35 pounds per cubic foot, branch wood being very much heavier and more full of resin. Oil is extracted from it by distillation, which is a dark brown, strong, and unpleasant smelling fluid, said to be a good antiseptic, and serves to coat the inflated skins known as ‘‘mussucks” used for crossing the Himalayan rivers. (H. J. E.) 1 Sir Archibald Buchan-Hepburn in Proc. Berwick Nat. Club, xviii. 210 (1904). 2 Forests of Upper India, 41 (1902). III LIBOCEDRUS Libocedrus, Endlicher, Syn. Conif. 42 (1847); Bentham et Hooker, Gen. Pi. iii. 426 (1880); Masters, Journ. Linn. Soc. (Bot.), xxx. 19 (1892), and Gard. Chron. xxx. 467 (1900). Heyderia, Koch, Dendrologie, ii. 2, p. 179 (1873). Calocedrus, Kurz, Journ. Bot. xi. 196 (1873). Thuya, Baillon, Hist. Pl. xii. 34 (1892). EVERGREEN trees with aromatic odour, belonging to the tribe Cupressinez of the order Coniferze, closely resembling Thuya in habit and other characters, the branches as in that genus ending in frondose ‘‘branch-systems,” which are flattened in one plane and three- to four-pinnately divided, with their axes bearing scale-like leaves in four ranks. On the main axes the leaves are often remote by the lengthening of the nodes ; on the lateral axes they are closely imbricated, and vary in the different species in size and form, as detailed in the three sections below. In seedling plants the leaves are always linear-lanceolate and spreading. Flowers: moncecious with those of the two sexes on different branchlets, or rarely dicecious, solitary, terminal. Male flowers oblong, subsessile, with six to twenty stamens decussately opposite on a slender axis; filaments short, dilated into broadly ovate or orbicular scale-like peltate connectives, which bear usually four sub-globose anther-cells, two-valved and opening on the back. Female flowers oblong; subtended at the base by several pairs of leaf-like scales, which persist slightly enlarged under the fruit; composed of four or six decussately opposite acuminate bracts; lowest pair small, unfertile; next pair above fertile, bearing at the base two erect ovules on a minute accrescent ovular scale ; uppermost pair when present unfertile. Cones small, pendulous or erect, ripening and letting out the seed in the first year, persistent empty on the branchlets in the second year. Scales decussate, four or six; the lowermost pair short, thin, often reflexed ; the next pair long, thickened, woody, widely spreading at maturity, marked externally close to the apex by the shortly acuminate or long-beaked tip of the bract; third pair, when present, con- nate into an erect median partition. Seeds, two or one by abortion on each of the two fertile scales, with two lateral wings, one broad, oblique, nearly as long as the scale ; the other short, narrow, or rudimentary; cotyledons two. Eight species of Libocedrus have been described, remarkable for their distribu- tion over widely separated areas in the two hemispheres. Three sections may be distinguished :— 484 ; Libocedrus 485 I. Ultimate branchlets on mature trees tetragonal, bearing leaves all alike and uniform in size. 1. Libocedrus tetragona, Endlicher. Chile, Patagonia. Leaves spreading. 2. Libocedrus Bidwith, Hooker. New Zealand. Leaves closely appressed. II. Ultimate branchlets flattened, with leaves of two kinds ; lateral boat-shaped, median flat and appressed. A. Median and lateral leaves equal in length. 3. Libocedrus decurrens, Torrey. Oregon, California, W. Nevada. Leaves green on both surfaces. 4. Libocedrus macrolepis, Bentham et Hooker. China, Formosa. Leaves glaucous on the lower surface, with white stomatic bands. B. Lateral leaves much longer than the median leaves. 5. Lzbocedrus chilensis, Endlicher. Chile. Median leaves minute, rounded at the’apex, with a conspicuous gland. 6. Lzbocedrus Doniana, Endlicher. New Zealand. Median leaves ovate, acute, mucronate, scarcely glandular. The two following species, imperfectly known and not introduced, will only be mentioned here. They belong to the last section :— 7. Lebocedrus papuana, F.v. Mueller.” New Guinea. 8. Libocedrus austro-caledonica, Brongniart et Gris.2 New Caledonia. LIBOCEDRUS TETRAGONA Libocedrus tetragona, Endlichler, Syx. Contf. 44 (1847); Lindley and Paxton, Hlower Garden, i. 47, f. 32 (1850); Kent, Veitch’s Man. Conifere, 256 (1900). Libocedrus cupressoides, Sargent, Silva N. Amer. x. 134 (1896). Thuya tetragona, Hooker, London Journ. Bot. iii. 148, t. 4 (1844). Pinus cupressoides, Molina, Saggio Sulla Storia Naturale del Chile, 168 (1782). A tree® attaining in South America, though rarely, a height of 160 feet. Branchlets tetragonal. Leaves equal in size and uniform in shape in the four ranks ; those on the ultimate branchlets about 7, inch long, adnate only at the base, the remaining part free and spreading ; ovate, acute, or rounded at the apex, keeled on the back, concave and glaucescent above; those on primary axes larger, adnate for the most of their length, the apices only being free and spreading. Cones on long branchlets, less than 4 inch long, brown. Scales four, minutely 1 Trans. Roy. Soc. Victoria, i, 32 (1889). 2 Bull. Soc. Bot. France, xviii. 140 (1871). 3 This tree has been confused by travellers with /2tzroya patagonica, which has very similar foliage when old. In the former, the leaves gradually taper to a rounded or acute apex; in the latter they are broadest in their upper third, close to the rounded apex. The cones are entirely different. 486 The Trees of Great Britain and Ireland pubescent on the margin, each bearing above the middle on the back a lanceolate, subulate, erect, incurved spine; the two smaller scales lanceolate; the two larger scales oblong, each bearing a solitary seed ; the larger wing oblique, obovate, obtuse, twice as long as the seed, the shorter wing narrow. (A. H.) This tree inhabits the western slopes of the Andes of Chile from latitude 35° southwards, and was collected by me in February, 1902, on the west end of Lake Nahuel-Huapi at two to three thousand feet. It was growing both on swampy ground, where it attained a considerable size, and on the steep hill-sides above Puerto Blest. The natives of the district call it Alerce,’ which is the usual name in South Chile for Fitzroya patagonica, and use it for making long straight thin shingles, which seem to be extremely durable. Owing to the inaccessible nature of the country and the scarcity of inhabitants, little or no timber has as yet been cut in the dense forests which clothe the shores of this large and picturesque lake. Judging from the climate, which is severe in winter, this beautiful tree should be hardy in the west and south-west of Great Britain and Ireland. According to Dusen and Macloskie,’ it ‘is common in Western Patagonia, extending through Fuegia to Cape Horn, rising up to the snow-line in the mountains, and met with of all sizes, from 2 to 160 feet high. As a rule it never forms forests, but grows either in small thin groves or sparingly mixed with Wothofagus betuloides and Drimys Wintert. It was introduced by W. Lobb? in 1849, but is excessively rare in cultivation, the only specimen we have seen being a small tree 15 feet high, in 1906, at Kilmacurragh, Co. Wicklow. This tree is narrowly pyramidal in habit, with bark scaling off in long papery ribbons. (H. J. E.) LIBOCEDRUS CHILENSIS Libocedrus chilensis, Endlicher, Syn. Conif. 44 (1847); Lindley, Journ. Hort. Soc. v. 35 (1850); Lindley and Paxton, Mower Garden, i. 48, £. 33 (1850); Kent, Veitch’s Man. Conifera, 252 (1900). Thuya chilensis, Don, in Lambert, Pinus, ii. 19 (1824); Hooker, London Journ. Bot. ii. 199, t. 4 (1843). Thuya andina, Poeppig et Endlicher, Mov. Gen. et Spec. iii. 17, t. 220 (1845). A tree, attaining in Chile 50 feet in height, usually with a short trunk branching into a compact pyramidal head, or becoming at high altitudes a dense shrub. Branch- lets compressed, slender; leaves scale-like in four imbricated ranks, those of the lateral ranks much longer than the others, boat-shaped, free at the apex, and spread- ing for one-third their length, keeled, acute, marked above and below with a white stomatic band; median leaves, minute, appressed, rounded at the apex, the dorsal with a prominent gland. 1 Sir W. T. Thiselton-Dyer suggests that this is no doubt a Spanish corruption of the Arabic £/ Arz, a name which seems to include any coniferous tree, e.g. Cedrus Libani and Pinus halepensis, According to Pearce, the tree producing the valuable alerce timber is Futzroya patagonica. Cf. Hortus Vettchiz, 46 (1906). 2 Scott, Princetown Univ. Exped. Patagonia, viii. 6, 18, 142 (1903). 3 Gard, Chron. 1849, p. 563. ~\ ae eS . ee PLATE I4I. LIBOCEDRUS CHILENSIS IN CHILE Libocedrus 487 Cones’ on short branchlets, $ inch long; scales four, each with a minute pro- jecting point below the apex, bright brown, two larger fertile and two smaller unfertile. Seeds one or two on each of the larger scales, oblique, with a narrow short wing on one side below, and an oblique broad oval wing on the other side above, the two wings being upper and lower, rather than lateral in position, (A. H.) A tree, said by Bridges—who was the first to send home seeds to Low of Clapton in 1847—to attain occasionally 80 feet in height. It grows on the lower slopes of the Andes of Southern Chile, from lat. 34° southward to Valdivia; and was collected by me in the valley of the Rio Limay below Lake Nahuel-Huapi at 3500 to 4500 feet. Here it grows scattered on grassy hillsides or in open groves, and is a graceful tree of 50 to 60 feet in height. A photograph of our camp in this valley, taken by Mr. Calvert, gives a good idea of its appearance (Plate 141). Though from the climate of the region in which it grows, this tree ought to be hardy in the warmer parts of England, and though in Mr. Palmer’s tables a small number of trees seem to have survived the frost of 1860-61, as at Bishopstowe, Nettlecombe, Southampton, and even at Keir in Perthshire, yet by far the greater number of the plants introduced in 1847 were killed; and it is now very rare in cultivation ; but seems, though slow in growth, to thrive at several places. By far the largest specimen I have seen is at Whiteway near Chudleigh, Devon, the property of Lord Morley, which in 1907, according to the measurements of the gardener, Mr. Nanscawen, was 46 feet 8 inches by 53 feet. We have also seen specimens in England at Blackmoor, Hants, the seat of Lord Selborne; and in Ireland at Castlewellan, the largest tree there being 20 feet high in 1903; at Powerscourt, where in 1906 there was a tree 28 feet high by 3 feet 3 inches, with the bark scaling off in long, narrow, papery slips, the habit being much wider than that of Z. decurrens, with ascending branches; and at Kilmacurragh, Wicklow, where there is a tree 25 feet in height. (H. J. E.) LIBOCEDRUS DONIANA Libocedrus Doniana, Endlicher, Syn. Conif. 43 (1847); Kirk, Forest Flora New Zealand, 157, tt. 82, 82a (1889); Kent, Veitch’s Man. Conifere, 254 (1900); Cheeseman, Mew Zealand Flora, 646 (1906). Libocedrus plumosa, Sargent, Silva N. Amer. x. 134 (1896). Dacrydium plumosum, D. Don, in Lambert, Pinus, ed. 2, Appendix 143 (1828). Thuya Doniana, Hooker, London Journ. Bot. i. 571, t. 18 (1842). A tree, attaining in New Zealand too feet in height and 15 feet in girth, with reddish, stringy bark scaling off in ribbons. Branchlets flattened, with leaves similar in shape and arrangement to those of L. chzlenszs ; lateral leaves adnate in the lower half, free and spreading in the upper half, acute, mucronate, green and shining above, glaucescent with a white band below; median leaves appressed, ovate, acute, mucronate, scarcely glandular. Cones about 4 inch long; scales four, each with a lanceolate acuminate, erect, 1 Cones ripened on young trees at Les Barres in France in 1900. Pardé, Ard. Nat. des Barres, 31 (1906). 488 The Trees of Great Britain and Ireland incurved spine above the middle on the back; two lower scales half the size of the others, acute ; two upper scales rounded at the apex, each bearing one seed, which has two lateral wings, one short and narrow, the other broad and entire or sub-dentate. This tree occurs in the North Island of New Zealand, in forests from Mongonui southward to Hawke’s Bay and Taranaki, at elevations from sea-level to 2000 feet, usually rare and local. Kawaka is the native name, and it is also known as the New Zealand Arbor Vite, the dark red wood, beautifully grained and durable, being used in cabinet-making. It is occasionally seen in conservatories ; the only tree growing in the open, that we know of, being one at Powerscourt, which was 20 feet high and 18 inches in girth in 1903. (A. H.) LIBOCEDRUS BIDWILLI Libocedrus Bidwilli, J. D. Hooker, Flora New Zealand, i. 257 (1867); Kirk, Forest Flora New Zealand, 159, tt. 82a, 83 (1889); Cheeseman, Mew Zealand Flora, 647 (1906). A tree similar to Z. Donzana, but smaller, attaining a maximum of 80 feet in height and 12 feet in girth; but often bushy at high altitudes and on peat-bogs. Branchlets on young trees like those of Z. Donzana, but more slender ; on old trees tetragonal, sth to 74th inch in diameter, clothed with densely imbricated, minute, scale-like leaves, uniform in size and shape in the four ranks, closely appressed, boat- shaped, ovate, acute, green in colour. Cones like those of Z. Donzana, but smaller, 4 to £ inch long. This tree occurs both on the North and South Islands of New Zealand, from Te Aroha mountain and Mount Egmont southward to the Foveaux Strait, not un- common in hilly and mountain forests at 800 to 4000 feet elevation. It is known as cedar or Pahautea, and has soft, red, and rather brittle wood. This species has not apparently been introduced, though judging from its occurrence higher in the mountains and more southerly in latitude than Z. Donzana, it ought to be hardy in the milder parts of the British Isles. (A. H.) LIBOCEDRUS MACROLEPIS Libocedrus macrolepis, Bentham et Hooker, Gen. Pl. iii. 426 (1880); Kent, Veitch’s AZan. Contfera, 255 (1900); Masters, Gard. Chron, xxx. 467 (1901); Henry, Garden, Ixii. 183, with figure of tree (1902). Calocedrus macrolepis, Kurz, Journ. Bot. xi. 196, t. 133 (1873). A tree, attaining in China 100 feet in height, broadly pyramidal in habit, with whitish, scaly bark. This species resembles Z. decurrens in foliage—the frondose branch-systems being, however, more flattened, and the leaves thinner in texture and larger at the corresponding stages of growth than in that species—the best mark of distinction being the glaucous tint of the leaves beneath, Staminate flowers oblong, Libocedrus 489 tetragonal ; stamens sixteen to twenty. Cones on very slender branchlets (which are modified in being tetragonal, with minute appressed leaves uniform in the four ranks), about 2 inch long, purplish or dark brown, roughened externally by longi- tudinal ridges; scales six, resembling those of L. decurrens, but smaller and with blunter minute processes. Seed, one on each of the two middle scales ; two-winged, with the larger wing broader in the middle and more obtuse than in the Californian species. This species occurs in the forests of Southern Yunnan in China, at 4000 to 5000 feet, but is rarely met with wild, and only in ravines near water-courses. It was dis- covered by Anderson near Hotha in 1888; and was subsequently seen by me wild, near Talang, and frequently planted in temples. It is known to the Chinese in Yunnan as Poh or Peh; and the wood is much esteemed, especially that of logs often found buried, the result of inundations in past times. Specimens of this species, so far as one can judge by the foliage alone, have been sent to Kew from North Formosa by Bourne. The Chinese Libocedrus was introduced by Mr. E. H. Wilson, who collected seeds when he was paying me a visit at Szemao in the autumn of 1899. Young plants,’ raised at the Coombe Wood Nursery, have beautiful, glaucous, large, flat foliage, the apices of the leaves being tipped with very fine, long, cartilaginous points. They may also be seen in the temperate house at Kew. The tree would probably be hardy in Cornwall and the south-west of Ireland, and being highly ornamental, is worth a trial in warm, sheltered spots. (A. H.) LIBOCEDRUS DECURRENS, INcEnsE CrEpar Libocedrus decurrens, Torrey, Smithsonian Contrib. vi. 7, t. 3 (1854); Sargent, Sz/va WV. Amer. x. 135, t. 534 (1896), and Zvees V. Amer. 73 (1905); Kent, Veitch’s Man. Conifera, 253 (1900) ; Mayr, Premdland. Wald- u. Parkbdume, 315 (1906). Thuya Craigana, Murray, Botan. Exped. Oregon, 2, t. 5 (1853). Thuya gigantea, Carritre, Rev. Hort. 1854, p. 224 (in part) (not Nuttall). Heyderia decurrens, Koch, Dendrologie, ii. 2, p. 179 (1873). A tree, attaining in America 180 feet in height and 21 feet in girth, with a straight stem tapering from a broad base. Bark nearly an inch thick, light cinnamon-red, irregularly fissuring into ridges covered with appressed flat scales. Leaves shining green, each set of four equal in length, adnate for most of their length to the branchlets, but free at the tips, which end in fine cartilaginous points ; about + inch long on the conspicuously flattened secondary and tertiary axes, increasing to 4 inch on the main axes, which are only slightly flattened: those of the lateral ranks boat-shaped, gradually narrowing to an acuminate apex, keeled and glandular on the back, covering in part the median leaves, which are obscurely glandular and flattened, with broadly triangular cuspidate apices. 1 A seedling is figured in 4x. of Bot. xvi. 557, fig. 30 (1902), concerning which Sir W. Thiselton Dyer says :—‘‘ The primitive leaves are not very different from the cotyledons, with which they are serially continuous ; but after a time there is a complete change in the form and disposition of the foliar organs.” 490 The Trees of Great Britain and Ireland Flowers appearing in January at the end of short lateral branchlets of the previous year; staminate, } inch long; pistillate, with ovate, acute, greenish-yellow scales, subtended at the base by two to six pairs of slightly altered leaves, which persist yellowish, sharp-pointed and membranous at the base of the fruit. Cones about an inch long, pendulous, reddish-brown, on short branchlets with ordinary leaves. Scales six; lower pair short with a reflexed process; middle pair long, lanceolate, gradually narrowing to a rounded apex, below which is a minute deltoid spreading or reflexed process, and concave on the inner surface at the base, with depressions for the seeds; upper pair connate into a thick, woody, median partition, slightly longer than the fertile scales, crowned by three minute spines. Seeds four, two collateral on each of the middle scales ; body, 4 to 3 inch, lanceolate, pale brown, containing liquid resin, marked with a white hilum on each surface at the base; wings two lateral, one short and narrow, the other oblique, produced above the seed, nearly as long as the scale, rounded at the narrow apex, and about one-third as broad as long in the middle widest part. Seedling.—Seedlings sown at Colesborne in spring were about 3 inches high in August, and had a slender tap-root, about 5 inches long. Caulicle, 14 inch long, terete, brownish, glabrous. Cotyledons, two, 1% inch long, linear, nearly uniform in width, rounded at the apex, green beneath, marked above with numerous in- conspicuous stomatic lines. Primary leaves variable in number, first pair opposite, succeeded by three or four whorls in sets of four each, or only one or two whorls are produced ; linear, 3 inch long, tapering to an acuminate apex, glaucous on both surfaces with indistinct stomatic lines. Above the primary leaves the stem gives off branches, and produces scale-like small leaves, arranged in four ranks, and intermediate in character between the primary leaves and the adult foliage. (A. H.) DIsTRIBUTION Libocedrus decurrens was discovered by Fremont in 1846 on the upper waters of the Sacramento river. It was introduced in 1853 by Jeffrey, who collected for the Oregon Botanical Association of Edinburgh; and his specimens were named by Murray Zhuya Craigana in honour of Sir W. Gibson Craig, one of the members of the association. Carriére confused the tree with Thuya gigantea; and for some time there was great confusion in the nomenclature of the two species. Lzbocedrus decurrens is the name now universally adopted. According to Sargent, the distribution extends from the north fork of the Santiam river in Oregon, lat. 44° 50’, southward along the western slopes of the Cascade Mountains, and through the Sierra Nevada in California, occasionally crossing the range into Western Nevada; also along the Californian coast ranges from Mendocino county to the San Bernardino, San Jacinto, and Cuyamaca Mountains, reaching its most southerly point on Mount San Pedro Martin, half-way down the peninsula of Lower California. Sargent states that it is rather rare in Oregon, ascending to 5000 feet, and in the Californian coast ranges, where it rises to 5000 to 7000 feet ; being most abundant and of largest size in the sierras of Central California at 3000 to 5000 feet, thriving best on warm, dry hillsides, plateaux, and Libocedrus 491 the floors of open valleys, usually growing singly or in small groves, often mixed with Prnus ponderosa and black oak. Henry saw it in Oregon on the eastern spurs of the coast range near Kerby ; and found it common on the road from there south-west across the Siskiyou range into Northern California, where it grew near Gasquet’s Inn, about twenty miles inland from Crescent City on the coast. In these localities it occurred scattered on dry, sunny hills, in situations similar to that of Pus ponderosa, at 2000 to 3000 feet altitude, and was not seen in shaded, moist ravines. The trees here are broadly pyramidal in habit, not assuming the columnar form of English cultivated trees, and of no great size, the largest measured being 123 feet by 11 feet 1 inch. Plummer, in his report on the Cascade Forest Reserve, where a good illustration is given, on p. 102, of a grove of this tree, says :—“ The incense cedar is almost always hollow-trunked or dry-rotted at the heart, even though the tree may have every outward appearance of perfect health. The wood has been very little used for any purpose but fuel or fencing, and is not cut when better is obtainable. It is said by Rothwell and Rix to ascend the mountains as high as 5750 feet.” Sudworth in his report on the Stanislaus and Lake Tahoe forest reserves’ says that it is here an abundant tree at between 3500 and 5500 feet, but extends from 2000 to 7000 feet. Mature trees are from 80 to 100 feet high, and 4 to 7 feet in diameter, attaining these dimensions in from 100 to 200 years. Large trees, as shown by a photograph (plate cxiii. of Sudworth), are almost always rotten at heart. Reproduction by seed is good and abundant almost everywhere, especially in the drier situations. The largest trees I have seen were on the lower slope of Mount Shasta at about 4000 feet, where I measured a tree 130 feet by 12 feet 7 inches which had been left standing when the surrounding forest was cut. Here it grew in company with Douglas fir, Aédzes concolor, and A. magnifica, on dry soil, and though the fruit on ist September was fully formed the seeds were not ripe. The average size of the trees here was go to 100 by 8 or g feet. Prof. Sheldon says that it attains 100 to 150 feet high by 3 to 7 feet diameter, but such dimensions are not common. CULTIVATION When raised from seed it is somewhat slow in growth at first, but in good nursery soil soon makes a well-rooted plant, which is proof against the worst spring and winter frosts, and seems hardy on heavy soil and in damp situations, where Thuya plicata is sometimes injured when young. It produces very little seed in this country, and these do not always mature, and in consequence is usually propagated by cuttings. Though it seems doubtful at present whether this tree can be looked on as a timber tree in England, yet on account of its rather stiff and formal habit it is well suited for the formation of small avenues, and when planted close together, as at Ashridge Park, forms a dense shelter without any clipping. 1 Washington, 1900. III G 492 The Trees of Great Britain and Ireland REMARKABLE TREES The finest tree that I know in England is the one figured (Plate 142) which grows in the grounds at Frogmore. This was planted, as I am told by Mr. M‘Kellar, by H.S.H. the Princess of Hohenlohe on 16th March 1857, and must be about 54 years old. It has been stated on a photograph taken for the late Hon. Charles Ellis to be 82 feet high, but when I measured it in 1904 I could not make it more than 65 feet, and being forked at about five feet from the ground its girth was about 9 feet. Another very fine tree grows close to the house at Bicton, which in 1900 was 60 feet by 7 feet 7 inches; and at Killerton there is a tree 55 feet by 5 feet 5 inches. At Orton Hall, near Peterborough, the tree succeeds very well on rather heavy soil, which does not suit many conifers, and here a tree 60 feet by 6 feet 9 inches has borne fruit, from which Mr. Harding, gardener to the Marquess of Huntly, has raised seedlings, some of which are now 9 feet high; smaller ones which he sent me are growing at Colesborne. At Hardwicke, near Bury St. Edmunds, one of the healthiest young trees I have seen, which was only planted in 1873, is already 48 feet by 4 feet 5 inches. At Crowsley Park, Oxfordshire, a tree planted about 1850 was, in 1907, 53 feet high by 8 feet 1 inch in girth, dividing into two stems at 10 feet from the ground, but forming a very narrow column. At the Wilderness, White Knights, Reading, an extremely narrow tree is 60 feet high by 4% feet in girth, At Nuneham Park, Oxford, there is a fine tree in the pinetum, which is 58 feet by 7 feet. At Bayfordbury, Herts, the best specimen is 52 feet by 5 feet 9 inches. In Herefordshire the best specimen I know of is at Eastnor Castle, which forks at about 6 feet, and measured in 1906 53 feet by 7 feet 6 inches. There is a nice avenue of it in the grounds at Ashridge Park, and also a circle consisting of 32 trees at only 1 yard apart, which were planted by Earl Brownlow thirty-five years ago, and are now about 35 feet high. Other remarkable trees which we have seen are at. Fulmodestone, Norfolk, 58 feet by 5 feet 11 inches in 1905; at Highnam, Gloucester, 50 feet by 5 feet 3 inches in 1905; at Beauport, Sussex, 53 feet by 6 feet 2 inches at 2 feet up, dividing into two stems, a conical tree, with extremely dense foliage, in 1904; at Dropmore, a large tree not measured. At Coldrinick, Cornwall, there is a tree which Mr. Bartlett informs us, was, in 1905, 51 feet by 63 feet. Mr. R. Woodward, jun., measured in 1906 a tree at Wexham Park, Stoke Pogis, 56 feet by 3 feet. At Salhouse, Norfolk, Sir Hugh Beevor measured, in 1904, a tree 57 feet by 6 feet 8 inches. Ai fine specimen at Tittenhurst, Sunninghill, is figured in Gardeners’ Chronicle, xxxvi. 284, fig. 127 (1904). In Scotland the tree is not so common, though specimens 40 to 50 feet high are growing in various places; the tallest reported at the Conifer Conference or 1891 was at Torloisk in Mull, and then measured 37 feet in height. At Smeaton-Hepburn, East Lothian, a tree planted in 1843 was measured by Henry in 1905 as 53 feet by 5 feet 4 inches, A tree at Keir, Perthshire, seen see LIBOCEDRUS DECURRENS AT FROGMORE Libocedrus 49 3 by Henry, was 42 feet by 4 feet 10 inches in 1905. At Brahan Castle, Ross-shire, Col. Stewart Mackenzie of Seaforth informed us in 1904 that he had a tree 4 feet 10 inches in girth, height not stated. In Ireland, Lzbocedrus decurrens is rare in cultivation. At Stradbally Hall, Queen’s County, a fine tree measures 53 feet high by 53 feet in girth. There is a tree at Fota 45 feet high, dividing into two stems at 2 feet from the ground. At Churchill, Armagh, a fine healthy specimen, growing in sand, was 45 feet by 4 feet 10 inches in 1905. At Adare a tree measured, in 1903, 47 feet high by 7 feet 9 inches in girth. In North Italy this tree grows larger than in England and ripens seed freely, which it rarely does here. At Pallanza, in Rovelli’s nursery, I measured a splendid tree over 90 feet high by 9 feet 3 inches in girth. Another on the Isola Madre was 90 feet by 9 feet 10 inches, from which I gathered seed in October 1906, which have produced a good crop of seedlings. It also ripens seed and grows well in the climate of Paris, and also at Les Barres, and has produced self-sown seedlings at Thiollets (Allier).’ The largest I have seen in France is at Verriéres, near Paris, a handsome and well-shaped tree, which measured, in 1905, 50 feet by 5 feet 5 inches, and is figured on plate vii. of flortus Vilmorinianus (1906). thle J. Ee) 1 Pardé, Ard, Nat. des Barres, 32 (1906). CUNNINGHAMIA Cunninghamia, R. Brown, in Richard, Conif. 80, t. 18 (1826); Bentham et Hooker, Gen. 77. iii. 435 (1880); Masters, Journ. Linn, Soc. (Bot) xxvii. 304, fig. 18 (1889), and xxx. 25 (1892). Belis,) Salisbury, Trans. Linn. Soc. viii. 315 (1807). Jacularia, Rafinesque, in Loudon, Gard. Mag. viii. 247 (18 32). Raxopitys, Nelson (Senilis), Prnacee, 97 (1866). A Genus, belonging to the Conifer, with only one known living species,” and doubtfully represented in the fossil state.° Cunninghamia is considered by Bentham and Hooker, and by Masters, to be a member of the family Araucariez; but it is placed by Eichler* in Taxodinez. Seward and Ford, who have lately published an exhaustive monograph * of Araucaria and its allied genus Agathis, agree with Eichler that it has no close relationship with those genera. It appears, however, to be a connecting link between the Araucariez and the Taxodinez ; and mainly differs from Araucaria, some species of which it closely resembles, in foliage, in having three ovules on the bract, and not one only, as in that genus. The generic characters are given in the following detailed account of the species :— CUNNINGHAMIA SINENSIS Cunninghamia sinensis, R. Brown, Joc. cit. (1826); Lambert, Genus Pinus, ed. 2, t. 53 (1832); Loudon, Av). et Frut, Brit. iv. 2445 (1838); Murray, Pines and Firs of Japan, 116, figs. 216- 224 (1863); Kent, Veitch’s Man. Conifere, 292 (1900); Shirasawa, Jeon. Ess. Forest. Japon, text 23, t. 9, ff. 1-24 (1900). Cunninghamia lanceolata, W. J. Hooker, Bot. Mag. t. 2743 (1827). Pinus lanceolata, Lambert, Genus Pinus, ed. 1, t. 34 (1803). Belts jaculifolia, Salisbury, oc. cét. 316 (1807). Belis lanceolata, Sweet, Hort. Brit. 475 (1830). 1 This name, though the earliest, is not adopted on account of its close resemblance to the genus Bellis, used for the daisies. 2 While the above was passing through the press, there has been received at Kew a specimen of a new species of Cunninghamia, lately discovered in the mountains of Formosa at 7000 feet altitude. This species, which will shortly be published by Mr. Hayata, differs from C. szmenszs in having shorter leaves, acute and not acuminate at the apex. Mr. Hemsley is inclined to think that a specimen, preserved in the Herbarium, which was collected on Mt. Omei, in Western China, by Faber, is possibly a third distinct species. 7 3 Engler u. Prantl, Natur. Phlansenfamil. ii. 85 (1889). Cunninghamites, an allied fossil genus, has been found in the Keuper and Chalk deposits in Saxony, Bohemia, Westphalia, Southern France, and Greenland. Cf. Schimper a. Schenck, Palaontologie, 283 (1890). 4 The Araucariee: Phil. Trans. Roy. Soc., vol. cvili. p. 308 (1906). 494 Cunninghamia 495 An evergreen tree, attaining in China 150 feet in height and 18 feet in girth of stem, with brownish bark scaling off in irregular longitudinal plates, and exposing a reddish cortex beneath. Branches at first in pseudo-whorls, afterwards given off irregularly, Young branchlets sub-opposite or in pseudo-whorls, covered with green epidermis ; older shoots brownish except for the green leaf-bases. Leaves persistent alive five to seven years, afterwards remaining dry and dead for many years on the branches and even upon the stem; densely and spirally arranged on the branchlets, but twisted on their bases so as to be thrown into two lateral spreading ranks ; narrowed at the base and decurrent on the shoot to the insertion of the next leaf; rigid, more or less curved, narrowly lanceolate, acuminate, 1 to 2 inches long; upper surface dark green, concave with slightly raised margins ; lower surface convex, with a green midrib and two white stomatic bands, the stomata in several regular lines ; sharply and finely serrate ; with one resin-canal beneath the single unbranched fibro- vascular bundle. Staminate flowers, five to ten in an umbel at the apex of a branchlet ; the umbel surrounded at its base by numerous triangular imbricated serrulate bracts; each flower a spike-like cylindrical column of spirally crowded stamens ; each stamen con- sisting of a slender stalk with an ovate serrulate connective, from which hang three longitudinally-dehiscing anther-cells. Female flowers, single or three or four together at the apex of a branchlet; erect ovoid cones, composed of numerous spirally im- bricated lanceolate mucronate bracts in a continuous series with the leaves; lower bracts sterile, resembling leaves but with thickened bases; ovular scale on the upper fertile bracts visible only as a slight projection; ovules three on the base of each bract, reversed. Fruit, an ovoid-globose brownish cone, about 14 inch long, composed of thin woody scales, which are the bracts of the flowers increased in size and hardened, but otherwise little altered; loosely imbricated, serrate in margin, broadly ovate or reniform, with a cusped apex often reflected outwards. Seed-scale visible only as a transverse narrow membranous fimbriated projection on the inner surface of the woody bract, below its centre and above the seeds. Seeds three on each bract, about } inch long, brown, oblong compressed, surrounded by a membranous narrow wing. Cotyledons two. The cones persist for a year or more on the branchlets after the escape of the seed; and are occasionally proliferous, the elongated shoot above the cone producing leaves and growing to be several inches in length.’ Seedling.—Seedlings sown at Colesborne in spring were about 3 inches high in August, and had a short flexuose tap-root, provided with a few lateral fibres. Caulicle brownish, terete, glabrous, 1 inch long. Cotyledons two, about } inch long, coriaceous, entire, linear, with a median groove beneath. Young stem glabrous, ridged by the decurrent bases of the leaves. Leaves numerous, spirally arranged on the stem, $ to 1} inch long, soft in texture, linear, curved, broad at the base, whence they taper gradually to a fine bristle-pointed apex, serrulate in margin, green above, marked beneath with two narrow white stomatic bands. In Cunninghamia, as in Araucaria, root-suckers are often produced, which grow 1 Cf. Woods and Forests, 1884, p. 593, and Garden, xxix. 173 (1886). 496 The Trees of Great Britain and Ireland into young trees, close to the parent stem. Coppice shoots are also produced freely from the stools in China, when the trees are felled. Cunninghamia! was discovered in 1701 by J. Cunningham in the island of Chusan ; and his specimens, preserved in the British Museum, were early described by Plukenet.? The first accurately scientific description, however, is due to Lambert, and was based on specimens brought home by Sir G. Staunton, who accompanied Lord Macartney’s embassy to China in 1793. The tree has been known to the Chinese from the most ancient times, being mentioned in their earliest classical writings. It is called sha, a name, however, which is often applied also to Cryptomeria and other conifers yielding valuable timber, It was introduced* by William Kerr from Canton into Kew Gardens in 1804; but no trees of that date now exist there. Probably most of the existing trees in England were raised from seed collected by Fortune about 1844. Cunninghamia is widely spread throughout the central, western, and southern provinces of China, extending southwards from Szechwan, Hupeh, Kiangsi, and Kiangsu to Yunnan and Kwangtung. It is usually a tree of mountain valleys, requiring a hot summer and considerable humidity to thrive; and ranges in altitude from sea-level to 5000 feet, occurring in Central China below the zone, which, in the high mountains, is covered by silver fir and spruce. There appear to be large forests of it in the interior of Hunan and Fokien, judging from the vast quantities of its timber which are exported from there. In Fortune's time it was abundant on the islands of Chusan and Pootoo, but was rare in Hongkong, where the only wild trees of this species grew as isolated specimens in the Happy Valley. Fortune,‘ in 1849, passed through fine forests of Cunninghamia in the mountains of Northern Fokien, many of the trees being 80 feet in height, and perfectly straight; and he noticed variations in the tint of the foliage. He met with dense woods in the Snowy Valley and other parts of Chekiang, but the trees were usually young, and not remarkable for size. Mr. E. H. Wilson informs me that there are magnificent forests of Cunning- hamia in Western Szechwan. One which he specially noted in the Upper Ya Valley extended for fifty miles between 2000 and 5000 feet altitude, the best trees ranging from 100 to 150 feet in height, and from 15 to 18 feet in girth ; and when growing in close stands, with straight stems clean to 4o feet or more, the branches above being short, slender, and horizontal. In the open the trees have much longer pendulous branches. The foliage is occasionally glaucous. Where trees had been cut down, new growth was being everywhere produced by shoots from the stools. Mr. Wilson mentions the common use of the timber in China for house-building purposes generally, and for the masts and planking of native craft. The bark is also used in the mountains for roofing houses. In the Chien Chang Valley in 1 In a note in King’s Survey of the Coasts of Australia, ii, 564 (1826), R. Brown states that he requested Richard to change the name Be/is, given by Salisbury, into Cunninghamda, in honour of both J. Cunningham, the discoverer of the tree, and of the collector Allan Cunningham. 2» Amaltheum Botanicum, i. t. 351, f 2. (1705). 8 Ailon, Hortus Kewensis, v. 320 (1813). hes 4 Wanderings in China, 379 (1847); Zea Countries, ii. 178, 212 (1853); Residence among the Chinese, 189, 277 1857). Cunninghamia 497 Szechwan, owing, according to tradition, to earthquakes some two centuries ago, landslips occurred which have buried whole forests in certain places beneath the soil. The dead timber is now being dug out, and is in an excellent state of preservation, being redder and more fragrant than the ordinary timber. It is known to the Chinese as fragrant Cunninghamia, Zszang-sha, and sells for extra- ordinary prices, selected thick planks for coffins often being worth £12 to £60 a piece. The wood, according to Mayr,’ is extraordinarily light, with a broad sap-wood and a dark yellow heart-wood. It is used extensively in the coast ports of China for making tea-chests. Cunninghamia appears to be confined as a wild tree to China; but it is occa- sionally planted® in Japan, the Loochoo Islands, and Formosa. M. Hickel has lately received seeds from Tongking, but these may have been gathered from cultivated trees. (A. H.) REMARKABLE TREES The growth of this tree in England depends mainly on the amount of heat in summer, which in most places is evidently insufficient; and though it endures severe winter frosts without much injury on well-drained soil, it suffers much from wind and frost in spring. It rarely ripens seed in this country, the only case I know of being a tree at Penrhyn Castle which is now dead, but from whose seed some young trees were raised. The best of these, when I saw them in 1906, was about 10 feet high. The tallest trees of this species that we know are at Killerton, where, in 1904, there were two which measured 62 and 60 feet in height by 4 feet in girth. One of these has since been cut down, its branches having become ragged, and a section sent to the Kew Museum shows the age to be at least 63 years. Another, at Bicton, was, in 1906, 56 feet by 4 feet ro inches, also rather ragged in its branches. There is a tree at Highnam, in Gloucestershire, about 25 feet high. At Heanton Satchville, the seat of Lord Clinton, in North Devon, there is a slender but healthy-looking tree 50 feet by 3 feet, and another one which has thrown up a shoot from the stool. At Escot in South Devon, the seat of Sir John Kennaway, Miss Woolward measured one in 1905, 45 feet high. At Pencarrow in Cornwall Mr. Bartlett showed me a tree, planted by Sir W. Molesworth in 1850, which was in 1905 40 feet by 4 feet 8 inches, and one of the healthiest that I have seen; and there is a smaller tree, 30 feet by 4 feet, at Coldrinick, in the same county. Coming farther east there is a splendid tree at Bagshot Park, the seat of H.R.H. the Duke of Connaught, which, when I saw it in May 1907, was no less than 47 feet high by 7 feet in girth, and 48 yards in the circumference of its branches, Being on very well drained soil, and well sheltered by other trees, it has suffered 1 Fremdlind. Wald. u. Parkbdune, 285 (1906). 2 Cf. Hayata, in Zokyo Bot. Mag. xix. 50 (1905). 498 The Trees of Great Britain and Ireland little from frost and wind, and is the handsomest and best-shaped tree’ I have seen (Plate 143). At Beechlands, near Lewes, the seat of Captain Rose, I am told by Mr. Chisholm that there is a Cunninghamia 50 feet high by 5 feet 1 inch in girth, forked near the top. It bears many cones, which, however, do not produce fertile seeds. At Grayswood, Haslemere, a tree planted in 1882 is 30 feet high by 2 feet 7 inches in girth, but has not a very thriving appearance. Another at Redleaf, near Penshurst, Kent, the seat of Mrs. E. Hills, though forked near the ground, has one good trunk 47 feet by 5 feet 4 inches, and healthy foliage. At Langley Park, near Norwich, the seat of Sir Reginald Beauchamp, there is a tree 35 feet by 3 feet, which though healthy looking has grown but little for many years. At Tittenhurst,? near Sunninghill, there is a fine healthy tree over 20 feet in height. At Bayfordbury, Herts, Cunninghamia, though planted several times, has never succeeded, being much injured by spring frosts, and only one specimen, a few feet high, survives. The most northern point at which I have seen the tree growing in England is in the sheltered Duchess’ garden at Belvoir Castle. This, I was told by Mr. Divers, was planted in 1844, and in 1907 measured 39 feet by 3 feet 2 inches; but Mr. Fenner informs me that there is one 32 feet high at Holker Hall, Lancashire. In Scotland, as might be expected, there are no trees of any great size. At Brodick Castle, in the Isle of Arran, a tree, which was planted about the year 1858, had only attained, according to the Rev. Dr. Landsborough,’ 10 feet high in 1895, and never throve. There was formerly a tree at Smeaton-Hepburn, East Lothian, which died about five or six years ago after a drought. In Ireland, Cunninghamia is a very rare tree. There is one in Mr. Walpole’s garden at Mount Usher in Co. Wicklow, which was in 1903 31 feet high by 2 feet 2 inches in girth. It was supposed to be then about 28 years old. In Mr. Acton’s arboretum at Kilmacurragh, in the same county, there is a thriving specimen, which Henry measured in 1903 as 25 feet high by 14 feet in girth. Around Paris* the tree always looks suffering, the leaves turning yellowish and assuming a burnt aspect ; but it grows well at Les Barres,’ and fructifies annually. In North Italy the climate evidently suits the tree much better, as I saw, in the grounds of the Villa Ceriana near Intra, a tree 76 feet by 7 feet 4 inches, produc- ing cones freely in 1906, from which I have raised healthy seedlings. At Locarno,° on the northern end of Lake Maggiore, a tree planted fifteen years is 23 feet in height. (He Je Es) : John Smith, in Records of Kew Gardens, 290 (1880), states that a Cunninghamia, possibly the same tree as the one mentioned above, bore cones at Bagshot in 1838. 2 Gard. Chron. xxxvi. 284 (1904). 3 Trans. Bot. Soc. Edin. 1896, xx. 527. * Mouillefert, 7raité des Arbres, ii, 1336 (1898). ° Pardé, Ard, Nat. des Barres, §7 (1906). ® Christ, Hore de la Suisse, 77 (1907). LIQUIDAMBAR Liquidambar, Linnzeus, Gen. Pl. 463 (1742); Bentham et Hooker, Gev. Pl. i. 669 (1865); Engler u. Prantl, Léanzenfam. iii. pt. 2, 123 (1891). Decipvous trees belonging to the order Hamamelidee. Leaves alternate on long shoots, crowded and almost fascicled on short shoots, long-stalked, simple, palmately lobed, glandular-serrate. Stipules two, attached to the petiole near its base, lanceo- late or subulate, caducous or persisting throughout the summer. Flowers moncecious, or in rare cases polygamous, in heads subtended at the base by caducous bracts. Staminal heads, globose or elongated, several in a raceme on an erect axis, which is subterminal ; each head composed of numerous stamens, interspersed with minute scales, without corolla or calyx ; filaments slender ; anthers basi-fixed, oblong-obcordate, dehiscing longitudinally. Pistillate heads solitary, on long pendulous stalks, arising in the axils of the uppermost leaves, composed of numerous confluent flowers, the ovaries embedded in the axis of the inflorescence ; calyces minute, united together and with the ovaries, and bearing on their summits each four or more stamens, with usually aborted anthers; corolla absent ; ovary two- celled, each cell with numerous ovules ; styles two, recurved, stigmatic above on their inner surface. Fruit : a woody spherical head, composed of numerous capsules, consolidated together. Capsule with two valves, opening above to let out the seeds, each valve terminating in a beak (the hardened woody persistent style); calyx persistent, either minutely tuberculate or produced above into long spines. Perfect seeds, angled, winged above, one or two in a capsule, the remaining ovules having aborted. Most of the capsules, however, contain only numerous minute unfertile seeds without wings. The leaves of Liquidambar resemble strongly those of certain maples; but in the latter they are always opposite, and not alternate or fascicled as in the former. Moreover, stipules or their scars are present on the petiole near its base in Liquid- ambar, and are absent entirely in Acer. Three species of Liquidambar are well known, and occur in cultivation. Besides these there are apparently two species,’ wild in China, which are imperfectly known and not introduced. 1 These are :— 1. Liguidambar Rosthornii, Diels, Flora von Central China, 380 (1901), a small tree occurring in Szechwan ; flowers and fruit unknown. It resembles in foliage LZ. ortentalis. 2. Liguidambar sp., Hemsley, Journ. Linn, Soc. (Bot.) xxiii. 292 (1887). fruits, were sent to Kew from Hankow by Consul Alabaster. Judging from the imperfect material, this is a distinct species. Mr. E. H. Wilson has recently observed a species of Liquidambar, growing on the plain near Kiukiang, in Kiangsi, which is probably the same. Cf. Gard, Chron, xiii. 344 (1907). III 499 H Specimens, consisting of detached leaves and 500 The Trees of Great Britain and Ireland The species in cultivation are :— 1. Liguidambar styracifilua, Linnzus. North America. Shoots glabrous. Leaves large, usually five-lobed, only occasionally lobulate in margin ; under surface glabrous, except for dense tufts of pubescence in the axils of the main nerves at the base, and occasional minute tufts at the junctions of the lateral and main nerves. 2. Liguidambar ortentals, Miller. Asia Minor. Shoots glabrous. Leaves small, five-lobed, margin with large lobules; under surface quite glabrous. 3. Liguidambar formosana, Hance. China, Formosa, Tonking. Shoots pilose. Leaves large, usually three-lobed ; under surface pilose, without conspicuous axil-tufts. LIQUIDAMBAR STYRACIFLUA, Sweer Gum Liquidambar styracifiua, Linneus, Sp. Pl. 999 (1753); Loudon, A7d, e¢ Frut. Brit. iv. 2049 (1838); Oliver, in Hooker, /con, Plant. xi. 13 (1867); Sargent, Selva N. Amer. v. to. t. 199 (1893), and Zvees NV. Amer. 340 (1905). Ligquidambar macrophylla, Oersted, Am. Cent. xvi. t. 10 (1863). A tree, attaining in America 160 feet in height and 17 feet in girth. Bark deeply and longitudinally fissured, with broad ridges covered by thick corky scales. Young shoots green, glabrous. Leaves (Plate 199, Fig. 7) large, averaging 6 inches broad and 5 inches long, variable in form, cordate or almost truncate at the base, five-nerved, palmately and deeply cut into five oblong-triangular acuminate lobes, the terminal lobe largest, the basal lobes smallest, rarely lobulate ; serrations shallow, non-ciliate ; upper surface dark green, shining, glabrous ; lower surface light green, shining, glabrous except for dense tufts of pubescence in the axils of the nerves at the base and occasional minute tufts at the junctions of the lateral and main nerves. Petiole glabrous, slightly grooved on its upper side, dilated at the base, near which are two scars indicating where the lanceolate stipules have fallen off in early summer. Fruiting heads, about 14 inch in diameter, hanging on the tree during winter after the fall of the seeds in autumn, calyx margins with irregular small tubercules; capsules with two stout style appendages, forming woody spines, one terminating each valve. Perfect seeds few, with short terminal wing; imperfect seeds numerous, minute, angled, without wings. The branchlets* of many trees of this species are remarkable for their corky wings, which begin to develop in the second season and increase in width and thickness for many years. These wings occur on lateral branches, on the upper side only, in three or four parallel ranks; but on vertical branches they are borne irregularly on all sides. Trelease? observed in the case of Liquidambar trees 1 See Miss Gregory in Botanical Gazette, xiii. 282 (1888). ? Garden and Forest, 1890, p. 195. Liquidambar oa growing in Tower Grove Park, St. Louis, that about half the trees either showed no sign of the corky wings or in some cases only a slight trace of them. In Kew Gardens the same difference is noticeable in trees of the same age growing close together, some being without corky-winged branchlets, while others have them much developed. The leaves usually turn a most brilliant colour in autumn, the tint being red purple, or yellow. IDENTIFICATION In summer the maple-like but alternately-placed leaves are unmistakable. In winter (Plate 200, Fig. 2) the following characters are available : Twigs moderately stout, slightly angled, greenish, glabrous ; lenticels scattered, prominent. Leaf-scars alternate, obliquely set on projecting pulvini, arcuate or semicircular, marked by three bundle-dots. Terminal bud about § inch long; lateral buds smaller, varying in size, and directed outwards from the twig at an angle of about 45°; all ovoid, acute at the apex, and composed of six to seven imbricated scales, which are green with brown margins, vaulted on the back, shining, glabrous, ciliate, and often minutely cuspidate at the apex. Short shoots are numerous in this species, and, unlike the long shoots, are pubescent. All the shoots show at the base ring-like marks, indicating where the accrescent scales of the terminal bud of the preceding year have fallen off in spring. VARIETIES Though Oersted considered the Mexican and Guatemalan trees to constitute distinct forms, no varieties have been clearly made out. The species occurs over a wide extent of territory and in diverse climates; and certain differences are observable in the shape, size, and pubescence of leaves in wild specimens; but these scarcely warrant the division of the species into geographical forms. In dry regions in Mexico the under surface of the leaf is covered with dense pubescence. Leaves with only three lobes occur on adult trees in Mexico and Guatemala; but as three-lobed leaves are frequently borne on young shoots of the common form, this peculiarity scarcely merits the rank of a variety. (A. H.) DISTRIBUTION The Liquidambar or Sweet Gum,’ as it is usually called in the United States, has a very wide range of distribution. Its most northerly station is, according to Sargent,” near Newhaven, Connecticut, where it only grows near the coast as a small tree, 40 to 60 feet high. Farther south it extends westwards as far as S.E. Missouri and Arkansas, and in the south to Florida and Texas, reappearing on the mountains of Mexico and Guatemala. In the maritime region of the South Atlantic States and in the Lower Mississippi basin it is one of the most abundant 1 Also known as Red Gum. 2 Garden and Forest, ii. p. 232. 502 The Trees of Great Britain and Ireland forest trees, but only attains its full size and perfection in deep rich swamps and river bottoms. I have seen it of immense size in the Lower Wabash Valley in Southern Illinois, where Ridgway measured a tree no less than 164 feet high by 17 feet in girth with a clear stem 80 feet long, and another 137 feet high by 113 feet in girth, which was 94 feet to the first branch. Plate 1444, taken from a photograph for which I am indebted to the U.S. Bureau of Forestry, represents the tree (Example M) mentioned in Proc. U.S. Nat. Mus. v. 67, by Ridgway, which was 124 feet in girth at the base, 78 feet to the first limb, and contained 7888 feet board measure. It grew two miles from Mount Carmel on land now cleared. Such trees, however, are now hardly to be found except in very inaccessible places. On the coast region of North Carolina, Ashe and Pinchot give its dimensions as 100 feet high and 5 or 6 feet in diameter. The largest that I saw in the Eastern States was a tree in the Clifton Park, near Baltimore, which was 71 feet by 5 feet 9 inches. In New England, near Boston, Sargent says that it suffers from frost in severe winters, and I saw none in cultivation so large as those in England. I found it in a very different and more beautiful form in the mountains near Jalapa, Mexico, at about 4000 feet elevation, where in the month of March in open forests its leaves were conspicuous by their scarlet colour, but the trees were not of extraordinary dimensions. In America it grows mixed with Nyssa, Liriodendron, maples, and oaks. Ashe says that it fruits annually or every other year, but that much of the seed is abortive, and that it springs up commonly on damp hillsides and bottom lands, and also shoots from the stool after the trees have been felled. History AND CULTIVATION According to Loudon, this tree was first mentioned by Francis Hernandez, a Spanish naturalist, who published a work on the natural history of Mexico in 1651 at Rome. In 1681 it was sent home by Banister to Bishop Compton, who planted it in the Palace Gardens at Fulham. It had become common in cultivation in Michaux’s time, but he says that even in France it had never produced seed. In Northern Italy it grows well, and I found a good-sized tree on the Isola Madre in Lake Maggiore, which bore seed, from which I have raised plants, Though this tree will grow to considerable size in the warmest parts of England, and on account of its beautiful autumnal tints is highly ornamental, yet it requires a much greater degree of heat and moisture than our climate affords to bring it to perfection, and has been somewhat neglected by nurserymen on account of its tenderness when young. I have raised it from imported seeds, which do not keep well when extracted from the fruits, but the seedlings grow so slowly that the more common way of raising it is from layers. It does not transplant well, and requires a good deal of moisture in the soil and a warm, sheltered situation. Its branches are easily broken by the wind, and though it does not come early into leaf, is often injured by late frosts. VOINAWVY NI VOILVATAS VSSAN VOINANV NI MVENVAINOIT a Liquidambar 503 REMARKABLE TREES The largest trees mentioned by Loudon were at Strathfieldsaye (64 feet) and at Syon (59 feet), the latter tree being reported in 1849 to measure 84 feet by 4 feet. We cannot identify either of them now; but at Syon there is a tree, leaning considerably to one side, which was about 75 feet by 6 feet in 1904. The tallest which I have seen is at Godinton, the property of G. Ashley Dodd, Esq., near Ashford, Kent, which in 1907 was 82 feet by 6 feet, a piece estimated at 12 feet long having been broken off the top; and the next to it is one at Petworth, which Sir Hugh Beevor measured in 1894, 84 feet by 5 feet 7 inches; another tree at the same place, 7 feet 6 inches in girth, has been damaged at the top by wind. Miss Woolward tells me of a fine tree at Escot, Devonshire, the seat of Sir John Kennaway, which was referred to by Bunbury as the largest known to him, and in 1905 measured 75 feet by 7 feet 8 inches. At Cobham Hall, Kent, there is one which I measured as 80 feet by 5 feet 9 inches ; and at Broom House, Fulham, there are two trees on the lawn of about the same height and over 6 feet in girth. At Barton,’ Suffolk, there are four trees, which were planted in 1825-26, the two largest measuring, in 1904, 71 feet by 5 feet 6 inches and 52 feet by 3 feet 2 inches. At Arno’s Grove, Middlesex, a tree drawn up in a plantation, measured by Henry in 1904, was 83 feet by 3 feet 10 inches. A large tree which we have not seen was reported’ to be growing on the lake side at Chevening Park, near Sevenoaks, Kent. At Arley Castle there is a tree 65 feet by 4 feet 3 inches. In Scotland we have no records worth mentioning, though the species exists in the south-west. In Ireland there is a good tree at Fota, which in 1903 measured 57 feet high by 8 feet in girth. TIMBER Though neglected until recent years this tree is now very largely cut for timber in the Mississippi valley, and has been introduced to Europe under the name of satin walnut. Owing to its low price it has been tried, under the name of red gum, for street paving with very bad results, though, according to Stone,’ it is very resilient, and if creosoted may be a useful wood for this purpose. A careful investigation of the mechanical properties of this wood was made by A. K. Chittenden of the U.S.A. Bureau of Forestry in 1905, from which I take the following :—‘‘ Red Gum is perhaps the commonest timber tree in the hardwood bottoms and drier swamps of the Southern States, growing best on alluvial soil of great fertility, which is liable to heavy floods in winter and spring, and often covered with water from January till May. In the best situations it reaches a height of 150 feet and a diameter of 5 feet. It reproduces well only where there is sufficient light, as the seedlings will not bear shade. It also sprouts readily from the stump up to about fifty years of age, but such shoots rarely form large trees. The demand for 1 Bunbury, Ardoretum Notes, 28. 2 Garden, xxxviii. 208 (1890). 3 Timbers of Commerce, 113 (1904). 4 U.S. Dept. of Agriculture, Bureau of Forestry, Bulletin, No. 58 (1905). 504 The Trees of Great Britain and Ireland the timber has increased rapidly during the last few years, owing to the increasing scarcity of better timber, and about 75 per cent. of the best grades, ‘Nos. 1 and 2 clear heart,’ are exported to Europe for furniture and inside fittings. It is said to make very good flooring, and is now largely used for railway waggon box boards, the price in the U.S.A. being about 27 dollars per 1000 feet for firsts and seconds, as compared with 41 dollars for cypress. From 1900 to 1902 much of the wood was cut into 3-inch by 9-inch planks, to be used for cutting paving blocks in London, but in 1902 the market for this gave way, and the mills are now trying to introduce this wood as a paving-block material in the United States, where several large cities were in 1905 considering the use of this wood. The qualities necessary for a good paving block are durability, close grain, and the power of resisting abrasion. These qualities are found in red gum.” A very unfortunate experiment was made in Whitehall in the autumn of 1901, when the Corporation of Westminster accepted the tender of an American contractor to pave this street with ‘‘red gum.” The surveyor seems to have supposed that red gum in America was the same as red gum in Australia, where the name is applied to several species of eucalyptus, which have a good reputation for street paving. Be this as it may, the paving wore out so soon that a large proportion was taken up again in July 1902, and a long and costly lawsuit followed. The contractor alleged (1) that the defects arose from the bad foundations of the road; (2) from excessive watering ; (3) from stones having been forced into the pavement; and the case was not finally settled till October 1905. Mr. Weale tells me that an inferior quality of this wood containing much sap- wood is also known in the trade as “hazel pine.” ‘Satin walnut” is worth whole- sale from 2s. to 2s. 3d. per cube foot, and “hazel pine” only rs. 3d. to 1s. 6d. In colour the former is a light fawn, often marked with a rich dark stripe; but is so deficient in strength and durability, and even when well seasoned is so liable to warp and twist, that it is only used for the cheapest classes of furniture. Michaux says that though much inferior to black walnut and cherry, it was used a good deal in his time in America for picture-frames, bedsteads, coffins, and furniture. Red gum is now much used for veneer in the United States. It furnishes 17 per cent of all the veneer produced, the quantity in 1905 being over 187 million square feet.’ I brought from St. Louis a slab of this timber cut from a tree of 30 inches diameter, of which the sapwood was about 6 inches thick and much paler in colour. Though cut 4 inches thick this plank cracked badly in drying ; and it will evidently be a very difficult wood to dry without warping. It has a very close, fine grain, and takes a good polish. (H. J. E.) 1 U.S. Dept. Agric. Forest Service Circular, No. §1 (1906). Liquidambar 505 LIQUIDAMBAR ORIENTALIS Liquidambar orientalis, Miller, Gard. Dict. No. 2 (1768); Oliver, in Hooker, fcon. Plant. xi. 13, t. 1019 (1867); Hanbury, Sccence Papers, 139, with figure (1876); Bentley and Trimen, Medicinal Plants, ii. No. 107, t. 107 (1880). Ligquidambar imberbe, Aiton, Hort. Kew. iii. 365 (1789); Loudon, Ard. e¢ Frut, Brit. iv. 2053 (1838). A tree attaining in Asia Minor 40 to 60 feet in height. Bark longitudinally fissured, with corky irregularly quadrangular scales on the ridges, the orange- coloured inner bark visible in the fissures. Young shoots glabrous. Leaves (Plate 199, Fig. 6) small, averaging 3 inches wide by 24 inches long, palmately cut about half-way into five oblong triangular acute lobes, the upper three lobes usually with one to four lobules ; base truncate or widely cordate; margin with shallow glandular serrations ; upper and lower surfaces quite glabrous in cultivated trees, but with axil tufts of pubescence at the base of the under surface in wild specimens. Petiole glabrous, swollen at the base, and bearing near its insertion two minute triangular stipules. Flowers and fruit similar to those of Lzguzdambar styractflua, but smaller. Fruiting head about 1 inch in diameter; capsules with more slender beaks than in the preceding species ; calyx slightly tuberculate and not spiny. In winter the twigs resemble those of the American species, but are more slender, with smaller leaf-scars and buds, which are reddish and have six glabrous ciliate scales ; short shoots glabrous. This species does not apparently develop corky ridges on the branches. DISTRIBUTION Ligquidambar orientalis is known to occur wild only in the south-western part of Asia Minor lying opposite to the island of Rhodes, and in Cilicia, near Alexandretta. It forms woods of considerable extent in the district of Sighala, near Melasso, and in the vicinity of Budrum, Mughla, Djova, Ughla, Marmoriza, and Isgengak. According to Maltass, who obtained specimens for Hanbury, there is a fine forest of this species between the village of Caponisi and the town of Mughla, many trees attaining 40 feet in height, while in other forests, according to native report, they were as high as 60 feet.’ Liquid storax, a balsamic resin, obtained from the inner bark of the tree by boiling it in water, is exported in considerable quantity from Smyrna and other Levantine ports, the bulk of this product going to China and India, where it is known in commerce as rose maloes,” Liquid storax is used to a small extent by druggists in this country, and is one of the ingredients of ‘“ Friar’s Balsam.” 1 Elwes passed through this district in 1874 on the way from Makri to Ephesus, but saw no trees of any size. This is a very hot country in summer, myrtle, oleander, and arbutus being the common shrubs. 2 Rose maloes is a corruption of rassamala, the Javanese and Malay name for A/tingta excelsa, Noronha, a tree allied to Liguidambar, which yields by incisions in the bark a sweet-scented resin. Cf. Bretschneider, Bor. Stnzcum, iii. 464 (1895). 506 The Trees of Great Britain and Ireland CULTIVATION The Oriental Liquidambar was introduced into France about the middle of the eighteenth century by the French Consul at Smyrna, and speedily passed into England, where it was cultivated in 1759 by Miller. It grows very slowly in this country, where it is very rarely seen in cultivation. There is a tree in Kew Gardens, about 15 feet high, the age of which is unknown. According to Nicholson it was 1o feet high in 1884. It has a twisted, crooked trunk, dividing about 6 feet up into two main stems. The branches are numerous and drooping, the habit of this tree being in marked contrast to that of a tall Liguidambar styracifiua close beside it, and probably results from the young branchlets being continually killed by the frost. A larger and very old tree at White Knights, near Reading, in the grounds of Mr. J. Heelas, was in 1904 about 25 feet high by 3 feet 4 inches in girth, and was decayed at the top, with many dead branches and a hole in the butt close to the ground, This tree is commonly cultivated in the Mediterranean region ; and Mr. Hickel, Inspector in the French Forest Service, informs us that there is a very large specimen, rivalling in size the American species, in the square near the railway station at Montpellier. In the park at Baleine’ (Allier) there is a tree 75 feet high by 7 feet in girth. Elwes measured a tree in the Jardin des Plantes, Paris, which was 40 feet high; but was told that it did not ripen seed; and in the Botanic Garden at Padua he saw a tree about 50 feet high by 4 feet in girth, which in May had abundant fruit of the preceding year upon it, but could find no seeds in them. (A. H.) LIQUIDAMBAR FORMOSANA Liquidambar formosana, Hance, Ann. Sc. Nat. 5 série, v. 21 5 (1866), and Journ. Bot. 1870, P. 274; Oliver, in Hooker, Leon. Plant. xi. 14, t. 1020 (1867); Hemsley, Journ. Linn. Soc. (Bot.) Xxill. 297. : Liquidambar acerifolia, Maximowicz, Mé. Biol. vi, 21 (1866) and viii. 419 (1871). Liguidambar Maximowicsii, Miquel, Ann. Mus. Bot. Lugd, Bat. iii, 200 (1867). A tree® attaining, in China, 80 feet in length and 15 feet in girth. Young shoots with scattered long hairs. Leaves (Plate 199, Fig. 8) widely cordate at the base, usually with three broad oblong-triangular acute or acuminate lobes, the outer lobes occasionally giving off two short additional lobes ; margin, occasionally lobulate, sharply serrate, ciliate; palmately three-nerved with two strong lateral nerves; upper surface dull with scattered long hairs; lower surface light green 1 Pardée, Arbor. Nat. des Barres, 208, note 1 (1906). 2 The peculiarities of the buds, leaves, and stipules have been fully described by Lubbock, in Journ. Linn. Soc. (Bot.) XxX. 495 (1894). Liquidambar 507 with dense long pubescence. Petiole pilose, with two subulate, persistent, pubescent, glandular stipules. Fruiting heads spiny, 14 inch in diameter, each capsule surrounded by several long spines arising from the calyx, and resembling the two indurated styles which terminate the valves. Perfect seeds few, or absent in many capsules, with narrow short wings. This species is widely distributed over the central and southern provinces of China, and occurs also in Tonking, Hainan,’ and Formosa. In Hupeh, where it has not been seen over 1000 feet altitude, the tree is valuable, as its timber is used for making the Hankow tea-chests. The Chinese call it Féng tree.’ It is doubtful if it will prove hardy, and is extremely rare in cultivation in Europe, the only plant known to us being one in Kew gardens, which is trained against a wall, and is interesting for its beautiful foliage, which lasts till late in November. It was introduced by seeds sent by Consul Alabaster from Hankow in 1884. (A. H.) 1 Swinhoe, Journ. Bot, i. 257, says it is the commonest tree in the mountain forests of Hainan. Hance, (oc. cz¢., says that at Canton old stumps buried beneath the soil sucker freely. 2 It yields a resin, éng-hsdang ; and a caterpillar, which feeds on its leaves, produces a coarse kind of silk, used for fishing-lines. III I NYSSA Nyssa, Linneus, Gen. Pl. 308 (1737); Bentham et Hooker, Gen. Fl. 1. 952 (1867); Harms in Engler u. Prantl, Péanzenfam. iii. 8, 257 (1898). Tupelo, Adanson, Fam. Fi. ii. 80 (1763). Ceratostachys, Blume, Bijdr. Fl. Ned. Ind. 644 (1825). Agathisanthes, Blume, loc. cit. 645. Daphniphyllopsis, Kurz, Journ, Asiat. Soc, 1875, ii. 201. Decipuous trees or shrubs belonging to the order Cornacee. Leaves alternate simple, stalked, with margin entire or remotely one- to four-toothed, without stipules. Branchlets with discoid pith. Flowers small, dicecious or polygamous, borne at the summit of axillary peduncles, the staminate flowers numerous in heads, umbels, or short racemes, the pistillate and perfect flowers solitary or aggregated in two- to eight-flowered heads, umbels, or short racemes. Staminate flowers: calyx short, flat or cup-shaped, five- to seven-toothed or entire; petals five to seven or ten to fourteen ; stamens five to ten, inserted on the margin of an entire or lobed disc; filaments slender, anthers oblong. Pistillate flowers: calyx campanulate or urceolate, five-toothed or entire; petals four to five, seldom three or six to eight ; stamens absent or equal in number to the petals and alternating with them, bearing fruitful or barren anthers ; ovary coalesced with the receptacle, crowned above by a disc, one- rarely two-celled, one ovule in each cell; style one, recurved, stigmatic along one side near the apex. Fruit a drupe, oblong or ovoid, urceolate at the apex; flesh thin, oily; stone bony, thick-walled, terete or compressed, ridged or winged, one- or rarely two-celled, containing one seed, which has a membranous testa and copious albumen. Cotyledons flat and leafy. The alternate stalked simple leaves, entire and ciliate in margin; and the branchlets with true terminal buds, without stipules or their scars, showing on section the peculiar discoid pith, are characteristic of Nyssa. Seven species of Nyssa have been described :—Wyssa sessiliflora, Hooker, a tree attaining 60 feet in the Himalayas and Java; has not been introduced and would probably not be hardy in England. Myssa sinensis, Oliver, has recently been intro- duced from Central China. The remaining five species are natives of Eastern North America. Nyssa acuminata, Small, a species imperfectly known, is a small shrub growing in pineland swamps in Georgia. Myssa Ogeche, Marshall, a tree of moderate size, occurring in river swamps in South Carolina, Georgia, and Florida, is unknown in cultivation outside of its native home, and would probably not grow in England. Nyssa biflora, Walter, a small tree, growing in ponds, from North Carolina to 508 Nyssa 509 Louisiana, is probably only a variety of yssa sylvatica, Marshall; and no trees referable without doubt to it are known to us in England. Nyssa sylvatica, Marshall, and yssa aquatica, Marshall, occur rarely in cultivation in England. NYSSA SYLVATICA, Tureto, Pepreripce, BLack Gum Nyssa sylvatica, Marshall, Arbust. Am. 97 (1785); Sargent, Silva NM. Amer. v. 75, t. 217 (1893), and Zrees V. Amer, 707 (1905). Nyssa multiflora, Wangenheim, Wordam. Holz. 46, t. 16, f. 39 (1787). Lyssa villosa, Michaux, #. Bor. Am. ii. 258 (1803); Loudon, Avd. e¢ Frut. Brit. iii. 1317 (1838). A tree, occasionally attaining in America 100 feet in height and 15 feet in girth. Bark thick and deeply fissured longitudinally. Young shoots glabrous or with short, erect pubescence. Leaves (Plate 199, Fig. 2, leaf from a tree in Arnold Arboretum, U.S.; and Fig. 9, leaf from a tree at Kew) extremely variable in shape and size, obovate, oval or elliptical; base tapering or rounded, apex acuminate or acute, margin entire or repand and ciliate; upper surface glabrous, dark green, usually shining; lower surface glabrous or with slight pubescence on the midrib and principal veins. Petiole channelled or winged, glabrous or pubescent, 4+ to 1 inch long. Flowers on pubescent peduncles, appearing after the leaves; staminate flowers numerous, stalked and in crowded clusters; pistillate flowers sessile, two to fourteen ina head. Fruit ovoid, bluish-black,.4 to 2 inch long ; stone terete or more or less flattened, with ten to twelve indistinct ribs. Seedling. —The caulicle, glabrous, terete, and about 2 inches long, ends in a long flexuose whitish tap-root, which gives off numerous lateral fibres. The cotyledons are ovate-lanceolate, rounded at both base and apex, about 14 inch long by inch broad, on petioles 4 inch long, slightly coriaceous, entire in margin, pale beneath, glabrous, pinnately veined. The stem, reddish and pubescent, gives off alternately the true leaves, which are oval, with a cuneate base and acuminate apex, entire or one- to two-toothed and ciliate in margin, pale and glabrous on the under surface with the exception of some pubescence at the base of the midrib, and with a pubescent petiole. The preceding description was drawn up in the summer of 1905, from a seedling at Colesborne, raised from seed gathered by Elwes at Boston at the end of the preceding September. IDENTIFICATION Nyssa sylvatica, with leaves quite glabrous or pubescent only on the midrib and principal veins beneath, is readily distinguishable from yssa aguatica, with leaves grey and pubescent all over the under surface, and with one or two teeth often on the margin. VVyssa sznensis, which resembles in foliage Myssa sylvatica, is dis- tinguished by the appressed pubescence of the shoots. In winter WVyssa sylvatica (Plate 200, Fig. 5) shows the following characters :— 510 The Trees of Great Britain and Ireland Twigs slender, glabrous, or with slight pubescence near the tip only ; stipule-scars absent. Leaf-scars small, crescentic, set somewhat obliquely on slightly prominent pulvini, surrounded by a narrow raised rim, marked with three bundle dots. Buds conical, pubescent, and acute ; scales five or six, imbricated, pubescent, ciliate, reddish or greenish; terminal bud larger than the lateral buds which arise at an angle of about 45°. Pith solid, but interrupted by transverse woody partitions, showing on longitudinal section a ladder-like appearance. The inner scales of the bud are accrescent ; and the base of the shoot is marked by ring-like scars, indicating where these scales have fallen off in the preceding spring. VARIETIES This species is extremely variable in leaf, both in wild specimens and cultivated trees. This is well shown in the Strathfieldsaye tree, the leaves of which vary from a long elliptical acuminate to a short broad obovate obtuse outline; some are quite glabrous, whilst others are pubescent on the midrib and principal veins beneath. Usually the leaves are very shining above and coriaceous; but in a tree growing at Kew in a wood, they are dull above and thin in texture. In some specimens there are numerous glands on the under surface of the leaf; whilst in others, as in a specimen growing in the Arnold Arboretum collected by Elwes, no glands are visible. The fruit is also variable, being either terete or flattened. The tree occurs in America in very diverse stations, both on wet soils and on dry mountain slopes ; and this may explain the remarkable extent of its variation. Var. dzflora, Sargent, Szlva N. Amer. v. 76, t. 218 (1893). Nyssa biflora, Walter, Ft. Carol. 253 (1788); Loudon, Ard. e¢ Frut. iii. 1317 (1838); Sargent, Trees N. Amer. 709 (1905). Leaves smaller than in the type, very narrow, glabrous and glandular beneath, quite entire in margin. Fruit with an oval, flattened stone, narrowed at both ends and prominently ribbed. This variety is a small tree, rarely more than 30 feet high, growing in ponds on the pine barrens near the coast from N. Carolina to Louisiana. It usually has a trunk with a swollen base, and appears to be a form of the species which has adapted itself to life in water. The cultivated trees mentioned by Loudon as being Wyssa éifora were all probably Vyssa sylvatica of the typical form. (A. H.) DISTRIBUTION Nyssa sylvatica is found in North America from Southern Ontario, where it grows to a good size near Niagara, and in New England, where I saw it in the neighbourhood of Boston 60 or 70 feet high, westwards to Central Michigan and South-Eastern Missouri, and southwards to Florida and Texas. It attains its largest size, according to Sargent, in the southern Appalachian Mountains, growing as high as 100 feet with a maximum girth of about 15 feet.’ It is found generally in wet soil on 1 But Ridgway measured a black gum in Wabash Valley, 125 feet high by 13 feet in girth, and 64 feet to the first limb. A AT STRATHFIELDSAYE Zz, - YN 2) Nyssa Sil the borders of swamps ; but in the south grows also on high wooded mountain slopes. It is very variable in form, sometimes branching close to the ground; but oftener has a stout straight trunk, covered with light brown deeply furrowed bark, which is often curiously divided into hexagonal scales. Plate 1448 shows the trunk of a tree in America. The upper branches are twiggy and usually crooked. The glossy green leaves are rarely disfigured by fungi or insects, and turn to deep red in autumn. An excellent illustration of a group of trees growing near a pond in Massachusetts is given in Garden and Forest, ii. 491, which resemble in habit the Siberian or Japanese larch; and this is the form which the trees often assume in low swampy ground in New England. Another figure in the same journal, vii. 275, fig. 46, shows the habit of a tree growing in drier ground in Pennsylvania. CULTIVATION Nyssa sylvatica was in cultivation at Whitton, near Hounslow, in 1750. It is, when well grown, a very distinct and beautiful tree, the brilliant scarlet assumed by its leaves in autumn rendering it a very desirable ornament for the park or pleasure ground. Sargent says that one reason why this tree is not more generally planted is that its long roots with few rootlets make it difficult to transplant, and that it must be either planted out when quite young or frequently transplanted in the nursery. Those which I have raised from seed grew slowly the first year, but seemed to ripen their young wood better than many American trees. When pricked out singly into pots in the following spring they all died. We have seen very few specimens in this country, the only one of great size being the tree’ at Strathfieldsaye, which measured in 1897 74 feet high by 5 feet 5 inches in girth. It grows on rather heavy soil. This tree was reported by Loudon to be about 30 feet high in 1838, and is probably over roo years old (Plate 145). It produced seed in 1906 which appeared to be mature. There is a tree at Munden, near Watford, the seat of the Hon. A. Holland Hibbert, which has a short bole of 44 feet long with a girth of 3 feet 3 inches, dividing into two stems, the branches of which are very spreading, forming a crown of foliage 38 feet in diameter ; the total height is only 20 feet. Mr. Daniel Hill of Watford, who kindly sent these measurements, says that the fork has been leaded over ; and it is possible that the tree lost its leader early from some accident, and in consequence subsequently assumed its present peculiar habit. At White Knights, near Reading, there was a large tree of this species which was cut down some years ago; and there are now many suckers arising from the roots.2. There is another tree at Bicton about 35 feet high by 3} feet, which in August 1906 had full-sized fruit upon it which seemed likely to ripen. 1 The girth of this tree given in Gard. Chron. xxvi. 162 (1899), is evidently erroneous, 14 feet 104 inches being a misprint for 4 feet 10$ inches. 2 Schenck, in Bzltmore Lectures on Sylviculture, 56 (1905), says that in the forest old trees are often surrounded by an abundance of seedlings ; but on abandoned fields it seems to come up from sprouts and not from seeds, 512 The Trees of Great Britain and Ireland There are three small trees in Kew Gardens, the largest about 20 feet high, growing in a densely wooded part close to the Arboretum Nursery. A tree growing in the garden at Harpton in Radnorshire, at an elevation of 700 feet above sea-level, was in 1905 274 feet high by 2 feet 8 inches in girth, The owner, Sir Herbert E. F. Lewis, Bart., who kindly sent us particulars, has not noticed during the last forty years any considerable increase in the size of this tree. Its leaves turn bright yellow in autumn. TIMBER The wood seems to be unknown in commerce, and is not mentioned by any of the English writers, but Sargent says it is very durable under water and used for keels of boats, and being extremely difficult to split, is also used for yokes, rollers, wheel-hubs, and pumps. Sections of it in Hough’s American Woods, Pt. I. No. 9, show a pale or reddish-brown wood of very close texture, somewhat resembling sycamore in appearance. (Hi. J. Be) NYSSA AQUATICA, Cotron Gum, TuprELo Gum Nyssa aquatica, Marshall, Arbust. Am. 96 (1785); Linnzeus, Sp. PZ. 1058 (ex parte) (1753); Sargent, Silva N. Amer. v. 83, t. 210 (1893), and Zrees NW. Amer. 711 (1905). Nyssa uniflora, Wangenheim, Mordam. Holz. 83, t. 27, f. 57 (1787). Nyssa denticulata, Aiton, Hort. Kew. iii. 446 (1789). Nyssa tomentosa, Michaux, F2. Bor. Am. ii. 259 (1803). Nyssa angulisans, Michaux, oc. cit. Nyssa grandidentata, Michaux f., Hist. Arb. Am. ii. 252, t. 19 (1812); Loudon, A7d, et Frut. Brit. iii, 1319 (1838). A tree, attaining in America 100 feet in height, with a trunk 12 feet in girth above the greatly enlarged base. Bark thick, longitudinally fissured, and roughened on the surface by small scales. Young shoots pubescent towards the tip, becoming glabrous below in summer. Leaves (Plate 199, Fig. ro) elliptical or ovate-oblong, base rounded or tapering, apex long-acuminate; margin entire or repand, ciliate, often with one to three or more triangular teeth, usually ending in a bristle; upper surface dark green, glabrous ; lower surface greyish in colour and with a scattered, fine pubescence; petioles more or less pubescent, 1 to 14 inch or more in length. Flowers on long, slender, pubescent peduncles: staminate flowers pedicellate in dense clusters, with a cup-shaped, obscurely five-toothed calyx and oblong short petals rounded at the apex ; pistillate flowers solitary, with long, tubular calyx, ovate minute spreading petals, and included stamens with small mostly fertile anthers. Fruit solitary, on long, drooping stalks, oblong, dark purple, about an inch long; stone obovate, rounded at the apex, pointed at the base, flattened, with about ten wing-like ridges. Nyssa 513 IDENTIFICATION. (See under WVyssa sylvatica) In winter, specimens from the tree at White Knights showed the following characters :—Twigs stout, pubescent near the tip, glabrescent elsewhere. Leaf-scars slightly oblique on prominent pulvini, almost orbicular or obcordate, notched in the upper margin, surrounded by a slightly raised rim, and marked by three conspicuous bundle-dots. Lateral buds minute, globose, two-scaled, reddish, shining, glabrous, arising in the notch of the leaf-scar. Terminal buds nearly globose, short and broad, with four to five thick, pubescent, reddish scales, keeled on the back and apiculate at the apex; in December the three outermost scales had dropped the apiculus and showed a truncate apex with a terminal scar. The base of the shoot is marked by ring-like scars as in Vyssa sylvatica. DISTRIBUTION Nyssa aquatica is found growing in swamps throughout the coast region of the United States, from Southern Virginia to Texas, and in the Mississippi valley, in Arkansas, Southern and South-Eastern Missouri, Western Kentucky, and Tennessee, and in the valley of the lower Wabash River in I]linois. An interesting account of the peculiar habit of this tree, as observed in the swamps of Arkansas, is given by Coulter... Occurring in company with Taxodium distichum, wherever the ground is inundated with water, the trunk develops an enlarged, dome-like base, often of immense size. A tree only 45 feet high, of which a figure is given, had a swollen base 55 feet in girth at the point where the roots entered the ground. When the water-supply is scanty the base is only slightly enlarged; and trees growing in dry soil show no swelling of the trunk. Coulter saw numerous seedlings of Nyssa, and concludes that it is gradually ousting from the swamps the Deciduous Cypress, which rarely seeds itself. Wilson® states that around the swollen base of these trees in the swamps there are masses of roots extending 6 to 8 inches above high-water line, each root going vertically up out of the water, and after a sharp bend going down into the water again. He compares these roots, rising above the water for purposes of aeration, with the knees of Taxodium. CULTIVATION Nyssa aquatica was cultivated*® by Collinson near London in 1735. It is now scarcely known in cultivation in England, the only tree which we have found being- one at White Knights Park, Reading, the residence of T. Friedlander, Esq. It is a slender tree, about 36 feet by 2 feet 2 inches, which looks of considerable age and is not vigorous in growth. Loudon‘ states that most of the trees which he saw at White Knights in 1833 were planted between 1790 and 1810; and one was a fine specimen * of Vyssa aquatica, perhaps identical with the tree now living. 1 Report Missouri Bot. Garden, 1904, xv. 56, plates 18, 19. 2 Proc. Philadelphia Acad, Nat, Sc. 1889, p. 69. 3 Aiton, Hort, Kew, iii. 446 (1789). 4 Gardeners’ Magazine, ix. 664 (1833). 5 This tree is not referred to by Loudon in his large work, published in 1838. 514 The Trees of Great Britain and Ireland Michaux states that it endures the climate of Paris, and does not exact in Europe as moist a soil as it constantly requires in the United States. (A. H.) TIMBER According to Holroyd,’ it has only recently been possible to market the timber of this tree, and under a fictitious name, so great has been the prejudice against this and others known as gums. Formerly when lands bearing tupelo and cypress were logged, the cypress alone was taken, and tupelo trees from 2 to 3 feet in diameter were left, because the lumbermen considered them to be worthless. At present, however, tupelo timber is extensively cut in Alabama, near Mobile, as well as in Southern and Central Louisiana. The best grades closely resemble the Yellow Poplar (Liviodendron). The wood has a fine uniform texture, is moderately hard and strong, not elastic, but very tough and hard to split, and easy to work with tools. It is not durable in contact with the ground, and requires much care in seasoning. It is now extensively used for house flooring and indoor finish. Mr. Weale informs me that it has a tendency to warp and split which cannot be prevented by any known process of seasoning; and only a small quantity has as yet been imported to England, in the form of boards, which are worth from Is. 9d. to 2s. per cubic foot, and are used by the makers of cheap furniture. But he thinks that if it was sent in boards as well planed as those of the so-called Hazel Pine, it would be more attractive, and its consumption would increase. (H. J. E.) NYSSA SINENSIS, Curnest Tureto Nyssa sinensis, Oliver, in Hooker, Jéon. Plant. t. 1964 (1891). A tree, attaining in China 4o feet in height. Young shoots covered with a dense appressed white short pubescence, retained in the second year. Leaves (Plate 199, Fig. 1) elliptical, base tapering, apex acuminate, margin entire and ciliate ; upper surface dull, dark green, and glabrous except for some slight pubescence on the midrib towards the base ; lower surface light green, shining, pilose on the midrib and chief veins and occasionally on the veinlets ; petiole, 4 to 8 inch long, pilose. Flowers, on long slender axillary peduncles, pedicellate, crowded in racemose clusters. Staminate flowers with a minute calyx, narrow oblong petals, and five to ten stamens on a fleshy disc. Pistillate flowers imperfectly known, but with bifid style and glabrous ovary. Fruit in clusters of about three, on short pedicels at the ends of long (two to three inches) erect or ascending pubescent peduncles ; oblong, bluish, 3 inch long ; flesh scanty ; stone with ten inconspicuous longitudinal ribs. This is a rare tree, occurring in mountain woods in Central China, in the western part of Hupeh, and on the Lushan Mountains, near Kiukiang, in Kiangsi.” It was discovered by me in 1888, and was subsequently found by Mr. E. H. Wilson, who sent home seed to Messrs. Veitch in 1902, from which a single plant has been raised at Coombe Wood, where it is perfectly hardy so far. (A. H.) 2 : aa pas U.S. Dept. Agric., Forest Service Circular, No. 40 (1906). 2 E. H. Wilson in Gard. Chron, xlii. 344 (1907). SASSAFRAS Sassafras, Nees ab Esenbeck u. Ebermaier, Handb. Med. Pharm. Bot. i. 418 (1830); Bentham et Hooker, Gen. Pi. iii. 160 (1880). Decipvous trees belonging to the order Lauracez, with alternate pinnately-veined simple leaves without stipules. Flowers dicecious or rarely perfect, in few-flowered racemes in the axils of bud-scales at the ends of the previous year’s shoots. Calyx six-lobed, the lobes in two series, imbricated in bud ; petals absent. Staminate flowers ; stamens nine in three series, the three inner ones each with two stalked glands at the base; anthers opening with four valves. Pistillate flowers with flattened ovate pointed or slightly two-lobed staminodes, or occasionally with fertile stamens like those of the male flowers; ovary ovoid, glabrous, superior, one-celled; ovule solitary, suspended ; one style elongated with a capitate stigma. Fruit an oblong- ovoid, one-seeded dark-blue berry, surrounded at the base by the enlarged and thickened calyx-limb, and supported on pedicels much thickened above the middle. The genus comprises only two species, one occurring in North America and the other in China. SASSAFRAS TZUMU, Curnese Sassarras Sassafras Tzumu, Hemsley, in Kew Bull. 1907, p. 55, and in Hooker, Jcon. Plant. t. 2833 (1907). Litsea laxiflora, Hemsley, Journ. Linn. Soc. (Bot.) xxvi. 383, t. 8 (1891). Lindera Teumu, Hemsley, of. cit. 392 (1891). This species grows sparingly in China in mountain woods at 3000 to 5000 feet elevation, south-west of Ichang, in the province of Hupeh; near Kiukiang in Kiangsi; and inland from Ningpo in Chekiang. It attains a height of 50 feet and yields a timber esteemed by the mountaineers, who call it the ¢zu-mu or huang ch‘tu tree. Resembling very closely the American species in the characters of the foliage and inflorescence, it was considered by Prof. Sargent’ and Mr. E. H. Wilson to be indistinguishable. Mr. Hemsley, however, points out certain differences in the floral organs, which entitle it to rank as a distinct species. The flowers are slightly smaller than those of the American tree, and are pubescent within and not glabrous as in that species. The male flowers have three staminodes alternating with the glandular row of stamens and a prominent pistillode, which are wanting in Sassafras 1 Trees N. Amer, 336 (1905). III 515 K 516 The Trees of Great Britain and Ireland officinale. The female flowers have twelve staminodes in three rows of six, three, and three; only six staminodes in two rows of three each occurring in the American species. There is a tree of this species, 10 feet high, growing in the Coombe Wood nursery, which was raised from seed sent by Wilson in 1900. It has made wonderful growth during the past summer, and is very handsome. It differs from the American species in having glabrous non-ciliate leaves, which are very lustrous on the upper surface; and the young branchlets are also devoid of pubescence. SASSAFRAS OFFICINALE, Sassarras Sassafras officinale, Nees ab Esenbeck u. Ebermaier, /oc. cit.; Bentley and Trimen, Afedicinal Plants, iii, 220 (1880). Sassafras Sassafras, Karsten, Pharm. Med. Bot. 505 (1882); Sargent, Silva WM. Amer. vii. 17, tt. 304, 305 (1895), and Zvees N. Amer. 337 (1905). Sassafras variifolium, O. Kuntze, Rev. Gen. Pl. ii. 574 (1891); Sargent, in Bot. Gazette, xliv. 226 (1907). Laurus oes Linneus, Sp. P2. 371 (1753); Loudon, 47d. et Frut. Brit. iii. 1301 (1838). A tree, attaining in America 90 feet in height and 18 feet in girth. Bark,’ according to Sargent, dark red-brown, deeply and irregularly divided into broad scaly ridges. Young shoots green or reddish, pubescent when young, becoming glabrous, remaining green in the second year. Leaves (Plate 199, Fig. 5) deciduous, entire, or two- to three-lobed; the entire leaves oval with an obtuse apex and cuneate base ; the others obovate, with a large triangular or oblong lobe on one or both sides, directed forwards and outwards ; margin entire or repand, ciliate; upper surface dark green with a scattered short pubescence; lower surface pale with a long pubescence, often falling by the end of summer; petiole, 1 to 2 inches long, pubescent. The nerves are pinnate, the two lowest arising near the base of the leaf, running nearly parallel with the margin, and ending in the lobes when these are present. Berry * gives an account with illustrations of the extraordinary variation which occurs in the leaves of wild trees growing in America. He has found leaves with four, five, and even six lobes. SEEDLING Out of some seed gathered by Elwes at the Arnold Arboretum late in Septem- ber and sown at Colesborne in October 1904, only one germinated in the following June, and the seedling showed the following characters in August :—The cotyledons remain in the seed-case, the young stem emerging between them after the splitting of the seed into two halves. The terete glabrous and reddish stem first gives off alternately two minute scales, which are succeeded by true leaves ; the first, } inch long, arising 14 inch above the ground, is half-oval in shape, one side of the leaf 1 In cultivated trees in England the bark is grey and fissured into longitudinal narrow ridges. 2 Bot, Gazette, xxxiv. 426 (1902). Sassafras 517 being scarcely developed, entire in margin, and on a short stalk about } inch long. The second leaf, # inch long, is obovate-spathulate, entire in margin, very unequal- sided, rounded at the apex, and tapering at the base. Succeeding leaves (six in all being produced by August) are oval, 14 to 24 inches long, stalked, unequal-sided, pinnately-veined, slightly undulate in margin; pale green and glabrous, with a raised midrib beneath. IDENTIFICATION In summer Sassafras is readily distinguishable by the aromatic leaves of different shapes, entire and two- to three-lobed, and by the branchlets, without stipules or their scars, remaining green for two or three years. In winter (Plate 200, Fig. 6) the following characters are available :—Twigs glabrous, green, shining, brittle, and strongly aromatic in odour when broken; lenticels few and inconspicuous ; pith wide and mucilaginous. Leaf-scars alternate, oblique on prominent pulvini, very small, semicircular with a raised rim, and showing a transverse band of minute coalesced bundle-dots. Terminal buds ovoid, with a long sharp beak ; external scales, four to five, imbricated, slightly pubescent, ciliate, green, often ridged or veined. Lateral buds minute, arising from the twigs at about an angle of 45°. Base of the shoot marked by ring-like scars, indicating where the scales of the previous season’s terminal bud have’ fallen off. (A. H.) DISTRIBUTION Sassafras occurs usually in rich, sandy, well-drained soil; and is widely spread in the eastern half of the United States, crossing into Canada in Southern Ontario. The northern limit passes through the southern parts of Maine, Vermont, and Ontario to Central Michigan, whence the western limit is continued through Eastern Kansas and the Indian Territory, to the valley of the Brazos river in Texas. On the eastern side it extends from Maine to Central Florida. In the South Atlantic and Gulf States it often takes possession of abandoned fields. In America the tree is very handsome at all seasons of the year, the light green foliage of summer turning delicate shades of yellow, orange, and red in autumn. The fruit, which is abundantly produced in some years, is showy, the berries dark blue in colour contrasting with the scarlet cups in which they sit. The tree produces root- suckers very freely. In New England the Sassafras does not often become a tree of considerable size. Emerson’ states that it rarely reaches 30 feet in height by a foot in diameter, and Michaux says that near Portsmouth, N.H., it is only a tall shrub rarely exceeding 15 to 20 feet high. But near Boston it sometimes grows much larger, and Emerson mentions one which grew at West Cambridge in 1842, and measured nearly 60 by 8 to 9 feet, with a clean straight stem 30 feet long. This tree was felled in order, as he says, “to allow a wall to run in a straight line.” But such vandalism as this, which a generation ago was common in New England, is now disappearing ; and great care is taken of the few surviving old trees of the original forest. Tree 1 Trees and Shrubs of Massachusetts, p. 359. 518 The Trees of Great Britain and Ireland wardens are appointed in most parishes, who are often ladies; and I am indebted to one of the most enthusiastic and active of them, Miss Emma G. Cummings of Brookline, Mass., for showing me some of the large Sassafras trees which still survive in the suburbs of Boston. These form a group on a slope on the south side of Covey Hill, the smallest being 6 feet in girth, and the largest 9 feet 7 inches and over 50 feet high. But these are far inferior to the trees in the forests of the south and west, where Ridgway measured, in the Wabash valley, a Sassafras 95 feet high by 7% in girth, and where, he says, it sometimes attains 12 feet in circumference. CULTIVATION The Sassafras was one of the earliest American trees introduced into England, having been cultivated in 1633 in a garden near London.’ The tree is propagated by seeds, which should be sown as soon as ripe, and by suckers and root-cuttings. When large it is difficult to transplant, as the thick fleshy roots are scantily provided with rootlets. Cobbett,? who gave an interesting account of the Sassafras, and was very enthusiastic in its praise, found that the seeds rarely if ever come up in the first year, and apparently often lie over for two years. Fresh seeds gathered by me in the Arnold Arboretum and sown in autumn, only produced one seedling in the first year, and no more have since germinated. This seedling though kept in a green- house grows very slowly, and at three years old is only 10 inches high. But though the tree is now rare in England there is no reason why it should not be grown on rich sandy soil in those districts where the summers are warm and dry, if young trees can be procured and established. REMARKABLE TREES The only really fine specimen of this species that we have seen in England is in the garden at Claremont, the seat of H.R.H. the Duchess of Albany. This is a handsome, healthy tree which in 1907 measured 48 feet by 6 feet 8 inches at 1 foot from the ground. It forks low down, and the main stem is 4 feet 10 inches at 5 feet. This tree flowers freely in the month of May, but Mr. Burrell has observed no seeds on it (Plate 146). A tree formerly grew at Beeston Hall, near Norwich, which Grigor states to have been 38 feet high in 1840, but this, as I am informed by Mr. Wall, the gardener there, died and was taken down about 1808. There are four small trees in Mr. Friedlander’s garden at White Knights Park, Reading, which appear to be suckers from the roots of an older one now dead; and in the adjoining properties, White Knights and the Wilderness, there are also trees of which the tallest is about 35 feet by 2 feet 10 inches. There isa younger tree in Mrs. Robb’s grounds at Goldenfield, Liphook, and a small one in Kew Gardens planted by Sir W. T. Thiselton-Dyer. There is also a healthy young tree at Tortworth. ! Gerard, Herball (ed. Johnson), 1§24 (1633). 2 Woodlands, Nos. 489 seg. (1825). PLATE. 146. SASSAFRAS AT CLAREMONT Nn Sassafras 519 The trees reported by Loudon to be growing in his time at Syon and at Croome cannot now be found. MEDICINAL PROPERTIES An interesting article on this tree by Prof. Sargent, with a figure of the trunk of an old one on Long Island showing the peculiar bark, is given in Garden and Forest, vil. 215; and from this I take the following :— The Sassafras is one of the most interesting trees of eastern North America. The last survivor of a race which at an earlier period of the earth’s history was common to the two hemispheres, it is the only tree in a large family which has been able to maintain itself in a region of severe winter cold. Towards the middle of the sixteenth century the French in Florida heard from the Indians wonderful accounts of the curative properties of a tree which they called Pavame, and which for no obvious reasons the Europeans called Sassafras. The tree and its virtues were first described by the Spanish physician, Nicholas Monardes, in his Natural History of the New World, published in Seville in 15609. The reputation of the roots and wood as a sovereign cure for most human maladies soon spread through Europe, and extraordinary efforts were made to procure them. To collect Sassafras was one of the objects of the English expedition which landed in Massachusetts in 1602, and eight years later Sassafras is mentioned among the articles to be sent home, in the instructions of the English Government to the officers of the young colony in Virginia. For nearly two centuries the reputation of Sassafras was maintained, and many medical treatises have extolled its virtues, though now it is generally recognised as simply a mild aromatic stimulant. Recently the thick pith of the young branches has been found to yield a mucilage useful to oculists, as it can be combined with alcohol and subacetate of lead without causing their precipitation. The oil of Sassafras, obtained from the wood and roots by distillation, is used to perfume soap and other articles ; and perhaps after all the most useful product of the Sassafras tree is the yellow powder prepared from the leaves by the Choctaw Indians of Louisiana, used to give peculiar flavour and consistency to ‘‘Gumbo filé,” one of the best products of the Creole kitchen. TIMBER - The wood has little or no economic value and is unknown in Europe. Michaux says that it was never seen in the lumber yard, and was only occasionally used for joists, rafters, and bedsteads; and that it is not attacked by beetles on account of the odour, which it preserves as long as it is kept dry. Ashe says it is light, soft, weak, brittle, and coarse-grained, very durable in contact with the soil, and apt to crack in drying. But the unusual orange-brown colour of the heartwood seems to me to give it a value for ornamental carpentry, if it can be procured of sufficient size. (H. J. E.) CORYLUS Corylus, Linneeus, Sp. P/. 998 (1753); Bentham et Hooker, Gen. Fi. iii, 406 (1880); Winkler, in Engler, Pfanzenreich, iv, 61, Betulacee, 44 (1904). Dectbuous trees or shrubs, belonging to the order Betulaceez. Leaves alternate, distichous on the branchlets, stalked, simple, penninerved, doubly serrate ; stipules two, caducous. Buds composed of numerous imbricated scales, corresponding to stipules. Flowers moncecious, arising from buds on the branchlets of the previous year. Male flowers in cylindrical catkins, appearing in autumn; fascicled, or two to five on a common peduncle ; composed of numerous imbricated bracts, each bearing on its inner side two partly adnate bracteoles and four stamens, without a perianth ; filaments bifid, each branch bearing a single anther cell, tufted with hairs at its apex. Female flowers in buds resembling those which contain leaves only, but distinguish- able in spring by the projecting styles. The lower scales of the buds bear leaves in their axils, the flowers, few in number, arising only in the axils of the uppermost scales, each scale bearing two flowers. Each flower, surrounded at the base by two minute bracteoles, more or less deeply cut and forming an involucre, consists of a two-celled ovary, surmounted by a short, denticulate perianth and two long styles; each cell containing one ovule. Fruit, in clusters at the end of the short leafy branch into which the bud has developed ; a one-celled, one-seeded nut, the remains of the other cell and ovule, which have aborted, being visible in its upper part. The nut is contained in a leafy involucre, open at the summit, and variously lobed or dentate. Seed without albumen; cotyledons thick, fleshy, containing oil, remaining on germination underground. Eight or nine species of Corylus are known, all natives of northern temperate regions, and mostly shrubs or small trees. Only one species, Coryus Colurna, attains the dimensions of a timber tree, and comes within the scope of our work. CORYLUS COLURNA, ConstanTINOPLE or TurKIsH HazeL Corylus Colurna, Linneus, Sp. Pl. 999 (1753); Loudon, Ard. e¢ Frut, Brit. iii. 2029 (1838) ; Willkomm, Forstliche Flora, 377 (1887); Winkler, of. cit. 50. A tree of moderate size, attaining 60 to 80 feet in height and 7 to 8 feet in girth of stem. Bark of trunk grey, thick, and scaling off in small irregular plates. 520 : Corylus 5 ia | Twigs brittle, the young shoots glandular pubescent, those of a year old glabrous and brown in colour, the bark of older shoots becoming corky. Leaves 3 to 5 inches long by 2 to 4 inches wide, broadly oval, ovate, or obovate, deeply cordate at the base, acuminate at the apex, doubly serrate or with large serrate teeth, dark green above, lower surface lighter green and sparingly pubescent, with glandular hairs on the principal nerves and midrib; nerves usually eight pairs; petiole 4 to 1 inch long, glandular pubescent or glabrescent. Catkins' 1} to 3 inches long. Fruits crowded, three to ten in number, long, compressed, pubescent towards the apex. Involucres tomentose with intermixed glandular hairs, deeply and irregularly divided into linear, acute, stiff, long-pointed segments, which are either entire or toothed, exceeding in length two to three times the nut. SEEDLING The germination resembles that of the oak,-the cotyledons, which are short- stalked, plano-convex and obovate, remaining in the seed and not being carried above ground. Caulicle stout, terete, tapering, ending in a long tap root with numerous branching fibres. Stem stout, terete, covered with numerous scattered glandular hairs, giving off an inch above the cotyledons a pair of opposite leaves, which are about 2 inches long, broadly ovate, acute at the apex, cordate at the base, with three to five pairs of lateral lobes, unequal in size, toothed and ciliate in margin; petiole 2? inch, glandular-pubescent. Succeeding leaves are alternate and larger in size. VARIETIES In addition to the typical form described above, several geographical varieties occur, as the species is distributed over a wide area. 1. Var. glandulifera, A. de Candolle, Prod. xvi. 2, p. 132 (1864).—Occurs with the type in Europe and western Asia. In this variety the pubescence on the petioles, peduncles, and fruit-involucres is intermixed with glandular bristles; and the segments of the involucres are less acute and often dentate. 2. Var. dacera, A. de Candolle, of. cet. 131 (Corylus lacera, Wallich, List, 2798). —Leaves obovate, larger, up to 7 inches long, with ten to twelve pairs of nerves. Involucre-segments linear-lanceolate with glandular hairs. This variety occurs in the western Himalayas, from Kashmir to Nepal, at elevations of 6000 to 10,000 feet, and in many places is gregarious. Sir George Watt informs me that it is a handsome tree, usually growing in the mixed forests, and often attaining 80 feet in height. 3. Var. chinensis, Burkill, Journ. Linn. Soc. (Bot.) xxvi. 503 (1899) (Corylus chinensis, Franchet, Journ. de Bot. 1899, xiii. 197).—Leaves large, up to 7 inches long, with ten to twelve pairs of nerves, broadly ovate, unequal, acuminate ; petioles bristly. Involucres striate and constricted above the fruit, lobes forked, lobules 1 Abnormal male flowers with enlarged bracteoles are figured in Gard. Chron, xxvi. 691, fig. 135 (1886). 522 The Trees of Great Britain and Ireland acute and falcate. This variety occurs in China, and grows to about 4o feet high in mixed forests in Yunnan, Szechwan, and Hupeh. Apparently no varieties have originated in cultivation, but a hybrid has been obtained between this species and the common hazel, viz. :— Corylus intermedia, Loddiges, Catalogue (1836) (Corylus avellana x Corylus Colurna, Rehder, Mitth. Deuts. Dendrol. Gesell. 1894, p. 43).—This is a tall shrub or small tree with the bark of the common hazel, ze. darker and less scaly and fissured than that of C. Colurna. The fruit resembles that of the last species, but is shorter and scarcely glandular. Specimens of this are growing in the Botanic Gardens of Jena and Gdttingen and in the Forestry Garden at Miinden, but we know of none in England. IDENTIFICATION In summer the Turkish hazel is readily distinguishable by the scaly bark and the obovate leaves deeply cordate at the base and distichously placed on the branch- lets. In winter (Plate 126, Fig. 6) the following characters are available -—Twigs : brittle, shining, brownish-yellow, with few and inconspicuous lenticels and scattered glandular pubescence, usually, however, dense near the base of the shoot, which is ringed with the scars of the previous season’s bud-scales, one or two of the lowermost scales often persisting dry and darkened in colour; second year’s shoot with corky bark, which fissures and exfoliates slightly. True terminal bud absent, a small oval scar at the apex of the twig, on the side opposite to the highest leaf-scar, indicating where the tip of the shoot fell off in summer. Leaf-scars semicircular with three to six bundle-dots,’ somewhat obliquely set on prominent pulvini. Stipule-scars small, transverse, lunate, one on each side of the leaf-scar. Buds pretty uniform in size, alternate and distichous on the twig, from which they arise at a wide angle, ovoid, rounded at the apex; scales about ten, imbricated, pubescent, ciliate in margin. Pith small, circular. Male catkins present in winter on flower-bearing trees. DISTRIBUTION The Turkish hazel has a wide distribution, extending from south-eastern Europe, through Asia Minor and the Caucasus, to the Himalayas and Western China. In Europe it is found growing wild in Banat, Slavonia, Herzegovina, Bosnia, Servia, Roumania, and Greece.? In Banat, according to Willkomm, it sometimes forms pure woods in the mountains; and in Northern Albania it ascends as a bush to 3000 feet altitude.* It occurs in Asia Minor in Bithynia, Paphlagonia, and Anatolia. According to Radde,’ it grows in small groups on the south side of the main chain of the Caucasus and in many localities in Georgia, at 3500 to 5000 feet elevation, where it is a stately tree 50 to 70 feet in height, and with a 1 The cicatrices left by the leaf-bundles on the leaf-scar are very irregular in number and shape, being circular dots or curved lines. 2 In Thessaly and Acarnania, according to Halacsy, Consp. Fl. Grace, iii, 135 (1904). 3 Beck, Veg. Lllyrischen Lander, 300 (1901). 4 Phlanzenverb, Kaukasusland, 187 (1899). Corylus 523 stem diameter of 18 inches. The nuts of the wild tree are small, with a thick and hard shell. It also grows in the mountains of Karabagh, but does not occur in the Talysch district. (A. H.) CULTIVATION The Turkish hazel was first cultivated in western Europe by Clusius, who obtained it from Constantinople in 1582. Linnzeus states that in 1736 the finest specimen known was a tree in the Botanic Garden at Leyden, which had been planted by Clusius. It was apparently first cultivated in England about the year 1665 by John Rea,’ who states that he had then “many goodly trees of the filbeard of Constantinople.” He grafted these upon ordinary hazel stocks. The Turkish hazel is now a rare tree in England, seldom to be got from a nursery, though perfectly hardy and easy to grow from seed, which it ripens in most seasons in the southern half of England. I have raised many from a tree at Tortworth Court, and the Earl of Ducie has done the same. The seed usually germinates in the following spring if sown when ripe, but if kept till spring, some- times not until the next year. The seedlings, on my soil at least, have more inclination to become bushes than to make a single stem, but, if cut down two or three years after planting, will throw up strong suckers which may be trained into a tree, and should be planted in half-shady places or in an opening in a wood, as they are liable when young to be injured by spring frosts. REMARKABLE TREES No other place can show so many fine trees as Syon, where there are in the grounds at least five, all apparently of about the same age. The largest of these stands near the east bridge over the lake, and is about 75 feet high, with a bole about 30 feet long and 6 feet 9 inches in girth. Near the gardener’s house is another fine tree more spreading in habit, about 70 feet by 7 feet 6 inches, which is probably not the same as one figured by Loudon, which was then 61 feet high. This has been figured by the Hon, S. Tollemache as the Hazel. At Bute House, Petersham, Henry measured a well-shaped tree which, in 1904, was 56 feet by 6 feet 7 inches. At Corsham Court there is a remarkable tree about 50 feet high, which divides near the base into two stems, one of which is quite decayed, and the other, which has the appearance of having originated as a sucker from it, is quite sound and 6 feet 8 inches in girth. Lord Methuen tells me that he can remember this tree as formerly producing fruits which were sent up to table, but now it no longer bears any nuts. At White Knights I saw a grafted tree from which seedlings had sprung up in the shrubbery, and one of these, growing at the base of a stump, is 10 feet high at about ten years old. ; At Arley Castle there is a good tree which, in 1904, was by Mr. Woodward's measurement 60 feet by 5 feet 7 inches. 1 Flora, 225 (1665). 2 British Trees with Llustrations, 9 (1901). HI i. 524 The Trees of Great Britain and Ireland At Wollaton Hall, near Nottingham, the property of Lord Middleton, there is a tree 43 feet high which at 5 feet girths 7 feet 10 inches, and at 10 feet, where it forks, 8 feet. It has a spread of not less than 78 feet, which for this tree is very unusual (Plate 147). It is perhaps the most symmetrical of its kind that I have seen anywhere. In the Botanic Gardens at Oxford and Kew there are fair-sized specimens. In Scotland and Ireland we know of no trees of great size, and none were recorded by Loudon; but at Glasnevin there is one about 35 feet in height, which divides into three stems close to the ground, and has very pendent wide-spreading branches. TIMBER Little or nothing is known of the timber in England, but a wood has been imported to France under the name of “ Noisetier,” which I believe to belong to this species, and which, as exhibited by M. Hollande of Paris, is very handsome. I purchased some very handsome veneer from Mr. Witt of London, which he told me had come to him direct from Constantinople, and which I believe was cut from the root of C. Colurna. Two good-sized logs of this tree were in the collection of Servian timbers shown at the Balkan States Exhibition in London in 1907; one of them is now in the Kew Museum. Gamble? says that in the Himalaya it is a well- grained timber, which does not warp, of a pinkish-white colour, and often shows a fine shining grain resembling that of bird’s-eye maple. (EL. J. B.} 1 Man. Indian Timbers, 684 (1902). CARPINUS Carpinus, Linneus, Gen. Pi. 292 (ex parte) (1737); Bentham et Hooker, Gen. Pl. iii. 405 (1880) ; Winkler, in Engler, Pfanzenreich, iv. 61, Betulacee, 24 (1904). Distegocarpus, Siebold et Zuccarini, Flor. Jap. Fam. Nat. ii. 103, t. 3 (1846). Decipvous trees belonging to the order Betulacez. Leaves, alternate, distichous on the branchlets, stalked, ovate, doubly-serrate, penninerved, the nerves ending in the points of the teeth ; stipules scarious, caducous or persistent. Flowers appearing in early spring with the unfolding of the leaves, unisexual, moncecious, without petals. Staminate flowers in pendulous, cylindrical catkins, arising from buds produced near the ends of lateral branches of the previous year; stamens, three to twenty, crowded on a pilose receptacle adnate to the base of a concave scale; filaments short, two- branched, each branch bearing a one-celled anther, tipped with a cluster of long hairs. Pistillate flowers, in loose, semi-erect catkins, which are terminal on the branchlets of the year; in pairs at the base of an ovate, acute, deciduous scale ; each flower sub- tended by a small bract and two minute bracteoles, and consisting of a two-celled ovary, surmounted by a minute epigynous calyx and two elongated styles ; each cell containing one ovule. Fruit, in pendent, stalked strobiles, composed of imbricated, foliaceous or mem- branous involucres, resulting from the developed bract and bracteoles of the flower, each with a nutlet at its base. Nutlet, ovoid, compressed, longitudinally ribbed, crowned by the calyx and remains of the style, one-seeded, and falling from the involucre in autumn. Seed, filling the cavity of the nutlet, without albumen ; cotyledons fleshy, carried above ground in germination. The genus consists of about eighteen species inhabiting the temperate regions of North America, Europe, and Asia. Two sections are distinguished :— I. Distecocarpus, Sargent, Sz/va N. Amer. ix. 40 (1896). Scales of the staminate catkins lanceolate, stalked. Fruit-involucres, mem- branous, infolded below, completely covering the nutlet, closely imbricated in the strobile. Trees with scaly bark. Two species, C. japonica, Blume, and C. cordata, Blume. II. Eu-Carpinus, Sargent, oc. cz. Scales of the staminate catkins ovate, sub-sessile. Fruit-involucres, foliaceous, open or only slightly infolded over the nutlets, loosely imbricated in the strobile. Trees usually with smooth bark. This section includes the remaining species. 525 526 The Trees of Great Britain and Ireland Carpinus and Ostrya are very similar in foliage ; and the following key, based upon the characters of the leaves and branches (Plate 201), distinguishes all the species of both genera which are in cultivation in England. Carpinus laxiflora, though not yet introduced, has been included, as it has been much confused with the other Japanese hornbeams. Key to CARPINUS AND OSTRYA I. Leaves not exceeding 2 inches in length. 1. Carpinus orientals, Miller. South-eastern Europe, western Asia. Leaves 1} inch long, acute, deeply plicate. 2. Carpinus polyneura, Franchet. Central China. Leaves 2 inches long, acute, smooth and only slightly plicate. Il. Leaves exceeding 2 inches in length. A. Leaves lanceolate. 3. Carpinus japonica, Blume. Japan. Leaves about 4 inches long, much longer in proportion to their width than in the other species, with numerous (eighteen to twenty-four pairs) nerves. B. Leaves ovate, acute at the apex. 4. Carpinus yedoensis, Maximowicz. Central China. Cultivated in Japan. Leaves 24 inches long, rounded at the base, with conspicuous bands of appressed pubescence on the upper surface. Branchlets pilose. C. Leaves ovate, acuminate at the apex. * Leaves deeply cordate at the base. 5. Carpinus cordata, Blume. China, Korea, Manchuria, and Japan. Leaves 4 to 5 inches long, broad in proportion to their length, with fifteen to twenty pairs of nerves. ** Leaves rounded or only slightly cordate at the base. + Under surface glabrous between the nerves. 6. Carpinus laxiflora, Blume. China, Japan. Leaves 2} inches long, rounded at the base, abruptly contracted into a very long acuminate apex. Branchlets with scattered long hairs. Buds minute, ps inch long. 7. Carpinus Betulus, Linneus. Europe, western Asia. Leaves 3 inches long, slightly cordate at the base, turning yellow in autumn. Branchlets with scattered long hairs. Buds fusiform, $ to $ inch long. 8. Carpinus caroliniana, Walter. North America. Leaves as in C. Betudus, but turning red in autumn. Branchlets with scattered long hairs. Buds ovoid, } inch long. t+ Under surface pubescent between the nerves. 9. Ostrya carpinifolia, Scopoli. Southern Europe, Asia Minor, Syria. Leaves 3 inches long, not velvety to the touch above, rounded at the base. Branchlets with dense appressed pubescence. Carpinus 527 10. Ostrya japonica, Sargent. China, Japan. Leaves 3 to 4 inches long, velvety to the touch above, slightly cordate at the base. Branchlets with dense, scarcely appressed, pubescence. 11. Ostrya virginica, Willdenow. North America. Leaves 3 to 4 inches long, not velvety to the touch above, slightly cordate at the base. Branchlets glandular-pubescent. (6-H) CARPINUS ORIENTALIS Carpinus orientalis, Miller, Gard. Dict. ed. 7, No. 3 (1759); Loudon, 47d. e¢ Frut. Brit, iii. 2014 (1838); Winkler, Betwlacee, 37 (1904). Carpinus duinensis, Scopoli, FY. Carniol. ii. 243, t. 60 (1772); Boissier, AZ. Orient. iv. 1177 (1879); Willkomm, Forstliche Flora, 366 (1887). Carpinus nigra, Moench, Verz. Ausland. Béiume u. Staud. tg (1785). A small tree or large shrub, rarely attaining 50 feet in height ; bark smooth and greyish. Young branchlets covered with a very minute dense pubescence, with which are intermixed scattered long hairs. Leaves (Plate 201, Fig. 7) small,’ strongly plicate, the nerves being deeply impressed above, about 14 inch long by 3 inch wide, ovate or ovate-elliptical, acute at the apex, unequal and slightly cordate at the base ; margin sharply bi-serrate, ciliate ; upper surface dark green, shining, with scattered long hairs; lower surface light green, pilose on the midrib and nerves, glabrous between the nerves, with minute axil-tufts; nerves nine to thirteen pairs; petioles, } to % inch, pilose; stipules linear-lanceolate, pubescent at the apex, + inch long, often persistent during summer. Fruit: strobiles, up to 2 inches long; bracts densely imbricated, 2 inch long, obliquely ovate, not lobed, sharply and irregularly serrate. This species is a native of south-eastern Europe and western Asia. It occurs in Italy and Sicily, reaching its northern limit in Istria, Croatia, Slavonia, Banat, and Transylvania, and extending southwards through the Balkan States to Macedonia and Greece. It is also met with in the Crimea, Asia Minor, and the Caucasus. It was introduced into cultivation in England in 1739 by Miller. It appears to be exceedingly rare, the only specimens we have seen being at Kew, where there are several small trees, one of which, planted in 1878, is now about 20 feet high. (A. H.) CARPINUS POLYNEURA Carpinus polyneura, Franchet, Journ. de Bot. xiii. 202 (1 899); Burkill, Journ. Linn. Soc. (Bot.), xxvi. 501 (1899). Carpinus Turczaninowii, Hance, var. polyneura, Winkler, Betulacee, 38, f. 12 (1904). A small tree, attaining 30 feet in height; bark greyish, slightly fissuring and scaly. Young branchlets with scattered long hairs. Leaves (Plate 201, Fig. 5) 1 In wild specimens the leaves are often larger, 2 to 23 inches in length. 528 The Trees of Great Britain and Ireland small, weakly plicate, the nerves being only slightly impressed above, about 2 inches long by Z inch broad, ovate, acute at the apex, unequal and slightly cordate at the base; margin bi-serrate, ciliate ; upper surface dark green, shining, with scattered, appressed hairs; lower surface-as in C. orzentalzs; nerves nine to twelve pairs ; petiole, $ to 2 inch, pilose ; stipules linear, pubescent along the margins, } inch long, persistent during summer. Fruit : strobiles 2 inches long; bracts loosely imbricated, obliquely ovate, $ inch long, outer margin slightly serrate, inner margin sub-entire, not lobed, without a basal auricle. This species is a rare tree in the mountains of Eastern Szechwan and Western Hupeh in China ; and is closely allied to, if not a mere variety of, C. Zurczaninownt, Hance, which is common in Northern China. C. polyneura differs little in technical characters from C. orientalis, but is very distinct in appearance owing to the leaves being smooth and flat and not deeply plicate, as in the other species of hornbeam. It is only represented in cultivation by a single tree, about 15 feet high, in Kew Gardens, which was raised from seed sent by me in 1889. (A. H.) CARPINUS JAPONICA Carpinus japonica, Blume, Mus. Bat. Lugd. Bot. i. 308 (1850); Shirasawa, con. Ess. Forest. Japon, text 47, t. 24, ff. 1-17 (1900); Winkler, Betulacee, 25 (1904). Carpinus Carpinus, Sargent, Garden and Forest, vi. 364, f. 56 (1893); Forest Flora Japan, 64, t. 21 (1894). Distegocarpus Carpinus, Siebold et Zuccarini, Fl. Jap. Fam. Nat. ii. 103 (1846). A tree attaining.in Japan 50 feet in height and 5 feet in girth; bark furrowed and scaly. Young branchlets with scattered long hairs, which fall off in autumn. Leaves (Plate 201, Fig. 1) ovate-oblong, up to 4 inches long by 1% inch broad, acuminate at the apex, oblique at the base, which is rounded or slightly cordate ; margin finely bi-serrate, non-ciliate; upper surface dark green, pubescent on the midrib and nerves ; lower surface pale green, with scattered long hairs on the midrib and nerves and slight axil tufts; nerves, eighteen to twenty-four pairs, impressed above; petiole $ inch long, pubescent; stipules 4 inch long, linear - lanceolate, pubescent, persistent during summer. Fruit: strobiles 24 inches long ; bracts densely imbricated, 3 to % inch long, ovate, sharply serrate; nutlet covered by a minute orbicular lobe, attached merely by its base to the bract, the outer margin of the latter being slightly infolded below. This species is a native of central and southern Japan, and, according to Sargent, is common on the Hakone and Nikko Mountains between 2000 and 3000 feet elevation. It was collected near Nikko by Elwes, and at Nagasaki by Oldham. It was introduced by Maries in 1879; but no trees of this date are now to be found, there being only small plants about 3 feet high in the Coombe Wood Nursery. It is perfectly hardy in New England, where it produced fruit for the first time in 1891 in the Arnold Arboretum, where it had been introduced a few years previously. Carpinus 529 Young plants were sent by Prof. Sargent in 1895 and 1897 to Kew, which have now attained about 1o feet in height. At Tortworth a young tree has produced fruit. The foliage of this species is remarkably distinct and handsome. (its Elid CARPINUS YEDOENSIS Carpinus yedoensis, Maximowicz, Mél. Biol. xi. 314 (1881); Burkill, Journ. Linn. Soc. (Bot.), xxvi. 502 (1899); Franchet, Journ. de Bot. xiii. 203 (1899); Winkler, Betulacee, 35 (1904). A small tree. Young branchlets densely covered with long hairs. Leaves (Plate 201, Fig. 3), 24 inches long by 14 inch wide, ovate, acute at the apex, rounded at the base; margin biserrate and ciliate; upper surface with conspicuous bands of long appressed pubescence in the intervals between the lateral nerves; lower surface pilose on the midrib and nerves, glabrous or with scattered long hairs in the intervals between the nerves; nerves ten to twelve pairs; petiole, 2 to 4 inch long, pilose ; stipules, linear-lanceolate, caducous. Fruit: strobiles, 24 inches long; bracts loosely imbricated, 3 inch long, semi-ovate, coarsely serrate on the outer side, subentire on the inner side, which is slightly infolded at the base, forming a small auricle partly covering the nutlet. This species is only cultivated in Japan, where it was first seen by Maximowicz. It was discovered growing wild in the mountains of North-Eastern Szechwan in China by Pere Farges, and may have been brought to Japan by Buddhist monks in early days, like many other Chinese plants. Young plants were raised from Japanese seed in 1901 by Purpus, in the Botanic Garden at Darmstadt. In the nursery at Kew there are two or three plants, growing vigorously, and about 3 feet in height, which were obtained from Simon Louis in 1904. (A. H.) CARPINUS CORDATA Carpinus cordata, Blume, Mus. Bot. Lugd. Bat. i. 309 (1850); Sargent, Garden and Forest, vi. 364 (1893), viii. 294, f. 41 (1895), and Forest Flora Japan, 65 (1894); Burkill, Journ. Linn. Soc. (Bot.), xxvi. 501 (1899); Shirasawa, Jcon. Ess. Forest. Japon, text 46, t. 24, ff. 18-32 (1900); Winkler, Betulacez, 26 (1904); J. H. Veitch, Hortus Vettchit, 359 (1906). Distegocarpus (?) cordata, De Candolle, Prod. xvi. 2, p. 128 (1864). A tree, attaining in Japan and China a height of 50 feet and a girth of 6 feet ; bark, dark grey, deeply furrowed and scaly. Young branchlets covered with a very minute pubescence, intermixed with scattered long hairs. Leaves (Plate 201, Fig. 2), ovate, up to 5 inches long and 2 inches wide, acuminate at the apex, unequally and deeply cordate at the base; margin finely bi-serrate, non-ciliate ; upper surface dark green, with scattered long hairs; lower surface light green, pubescent between the nerves, pilose on the midrib and nerves, without axil tufts; nerves fifteen to twenty 530 The Trees of Great Britain and Ireland pairs, impressed above; petiole, 3 inch long, with scattered long hairs; stipules caducous. Fruit: strobiles, 3 to 6 inches long, long-stalked ; bracts densely imbricated, membranous, 1 to 1% inch long, irregularly serrate ; the inner margin furnished below with an orbicular lobe, infolding and concealing the nutlet ; the outer margin slightly inflected at the base. The basal lobe is much larger than in C. zaponzca, and is united to the bract, not only by its base, but also along one side. Var. chinensis, Franchet, Journ. de Bot. 1899, p. 202.—Leaves, ovate-oblong, 3 inches long by 12 inch broad, with eighteen to twenty pairs of nerves, slightly cordate and unequal at the base, shortly acuminate at the apex. This variety strongly resembles in the shape of the leaf certain forms of C. japonica, but has the fruit of C. cordata. It seems to be intermediate between the two species, and is found in the mountains of Eastern Szechwan in China. It was introduced into cultivation by Mr. E. H. Wilson in 1901, and young plants are growing in the Coombe Wood Nursery. i According to Sargent, Carpinus cordata is one of the largest and perhaps the most beautiful of the hornbeams. It grows on the main island of Japan only at high altitudes, its true home being in the deciduous-leaved forest of central and northern Yezo. It is also a native of Korea and Manchuria; and occurs in China, in the typical form, in the province of Shensi,’ the variety chzmensts growing more to the south. This species was introduced from Japan by Maries in 1879, and produced fruit in 1886 in the Coombe Wood Nursery, where the largest specimen now living is only 15 feet in height. A tree at Tortworth is about 20 feet, and has borne fruit, from which, however, Elwes did not succeed in raising seedlings. There is also a small tree at Grayswood, Haslemere. It seems to be very rare in cultivation, and there are no specimens growing in the Hornbeam Collection at Kew. (A. H.) CARPINUS LAXIFLORA Carpinus laxiflora, Blume, Aus. Bot. Lugd. Bat. i. 309 (1850); Oliver, in Hooker, zon. Plant, t. 1989 (1891); Sargent, Garden and Forest, vi. 364 (1893), and Forest Flora Japan, 64 (1894); Burkill, Journ. Linn. Soc. (Bot.), xxvi. 501 (1899); Shirasawa, Jcon. Ess. Forest. Japon, text 48, t. 25, ff. 15-30 (1900); Winkler, Betwlacee, 33 (1904). Carpinus Fargesii, Franchet, Journ. de Bot. 1899, p. 202. A tree, attaining in Japan 50 feet in height and 5 feet in girth; bark smooth, grey, sometimes almost white in colour. Young branchlets with scattered long hairs. Leaves (Plate 201, Fig. 8), 24 inches long by 14 inch broad, ovate or ovate-elliptical, contracted above into a long acuminate apex, rounded or slightly cuneate at the base; margin, bi-serrate, non-ciliate; upper surface with scattered long appressed hairs ; lower surface with long appressed hairs on the midrib and nerves, glabrous between the nerves; nerves thirteen to fifteen pairs ; petiole, $ inch long, pilose; 1 Burkill, doc. cit. Carpinus 531 a stipules caducous. Fruit: strobiles, up to 3 inches long; bracts very loosely imbricated, about % inch long, semi-ovate, outer side irregularly serrate, inner side sub-entire, with a lobe near the base, which is infolded, but does not conceal the nutlet. This species is a native of China and Japan. According to Sargent, it is very like the European hornbeam in habit, fluted stem, and smooth bark. It is common in all the mountain forests of Hondo, where it is most abundant at elevations between 2000 and 3000 feet. Near Agematsu, in Shinshu, at 2000 feet altitude, it was collected by Elwes, who saw no tree of any great size or beauty, though the leaves turn red and yellow in autumn. In Yezo, it descends to sea-level on the southern shores of Volcano Bay, where, near the town of Mori, it is common in oak forests, and grows to its largest size. In China, this species grows in the mountains of Hupeh, Eastern Szechwan, and Kiangsi; but is rare, displaying considerable variation in the character of the leaves and fruit.’ It has not yet, apparently, been introduced into cultivation. Plants at Kew, sent under the name of C. /axzflora, from the Arnold Arboretum in 1895, are C. japonica. (A. H.) CARPINUS CAROLINIANA, American HorNBEAM Carpinus caroliniana, Walter, Fv. Carol. 236 (1788); Sargent, Selva NV. Amer. ix. 42, t. 447 (1896), and Zrees N. Amer. 190 (1905); Winkler, Betulacee, 31 (1904). Carpinus americana, Michaux, Fl. Bor. Amer. ii. 201 (1803); Loudon, Av. e¢ Frut. Brit. iii. 2013 (1838). Carpinus Betulus, Koehne, Deutsche Dendrologie, 116 (1893).° A bushy tree, attaining, in America, rarely 40 feet in height and 6 feet in girth, with stem and bark like the common hornbeam. Young branchlets with a few scattered long hairs, the minute glandular pubescence often seen in C. Betulus never being present. Leaves (Plate 201, Fig. 6) as in the common hornbeam, but usually with fewer nerves, nine to ten pairs; and unequal, rounded, or slightly cuneate at the base. Stipules lanceolate, $ inch long, caducous. Fruits: strobiles, 2 to 3 inches long; bracts loosely imbricated, triangular-ovate, { to 1 inch long, with two short unequal lateral lobes, and a much longer middle lobe, which is usually serrate on only one margin ; pedicels of each pair of bracts united only at the base. In the absence of fruit, this species is difficult to distinguish from C. Beculus from which Koehne could not distinguish it even as a variety. In autumn, the beautiful red tint of the foliage of the American species is diagnostic. The best mark of distinction lies, however, in the buds, which are small, ovoid, acute, 4 inch long, with glabrous ciliate scales; those of C. Betulus being large, fusiform, 4 to 4 inch long, with pubescent ciliate scales. . This species, which is known in America as the blue beech or water beech, is found along the borders of streams and swamps, from Southern and Western Quebec 1 Three varieties are distinguished by Burkill, foc. c7t, M Ill 532 The Trees of Great Britain and Ireland to Florida, extending westward to Northern Minnesota, Eastern Nebraska, Kansas Indian Territory, and Eastern Texas. It is also met with in a slightly modified form? in the mountainous regions of Southern Mexico and Guatemala. It is most abundant and of its largest size in the southern Alleghany mountains and in Southern Arkansas and Texas. . It was introduced into England by Pursh in 1812; but is very rare in cultivation, the best specimen we have seen being at Arley Castle. It has no claim to be considered as a forest tree, its only merit being the scarlet colour of the foliage in autumn. Elwes gathered seeds of this species near Ottawa in 1904, which did (A. H.) not germinate. CARPINUS BETULUS, Common HornBeaM Carpinus Betulus, Linneus, Sp. Pl. 998 (1753); Loudon, 47d. e¢ Frut. Brit. iii. 2004 (1838) ; Willkomm, Forstliche Flora, 358 (1887); Mathieu, Flore Forestiére, 396 (189 7). Carpinus vulgaris, Miller, Gard. Dict. ed. 8, No. 1 (1768). Carpinus sepium, Lamarck, Fl. Frang. ii. 212 (1778). Carpinus compressa, Gilbert, Exerc. ii. 399 (1792). Carpinus ulmoides, Gray, Nat. Arrang. Brit. Pl. ii. 245 (1821). Carpinus carpinizza, Host, Fl. Austr. ii. 626 (1831). Carpinus intermedia, Wierzbicki, in Reichenbach, Jcon. Fl. Germ, xii. f. 1297 (1850). Carpinus nervata, Dulac, Fl. Haut. Pyvéin. 141 (1867). A tree, usually attaining only a moderate size, 60 or 70 feet in height and 8 feet in girth; but in England occasionally as large as 90 feet by 12 feet. Stem never perfectly circular in section, being more or less longitudinally fluted or ridged, with shallow rounded depressions between the ridges; bark smooth, thin, grey. Young branchlets with scattered long hairs, a very minute dense glandular pubescence being also often present. Leaves (Plate 201, Fig. 4) about 3 inches long by 1$ inch broad, oval or ovate, acuminate at the apex; broad, unequal, and rounded or slightly cordate at the base; margin bi-serrate, non-ciliate; upper surface dark green, glabrous, or rarely pilose on the midrib and nerves ; under surface light green, with appressed long hairs on the midrib and nerves and minute axil tufts ; lateral nerves, ten to fifteen pairs, impressed on the upper surface, prominent beneath ; petiole 4} to 3 inch long, pubescent ; stipules narrow, lanceolate, } inch long, caducous. Male catkins, about 14 inch long; scales ovate, acute, entire, veined longi- tudinally ; stamens, 4 to 12, with long yellow anthers. Female catkins, nearly 1 inch long; scales ovate, acuminate, ciliate. Fruit: strobiles up to 3 inches long; involucres loosely imbricated, in pairs, with their pedicels connate for the greater part of their length, three-lobed, the lateral lobes small and usually entire, the middle lobe, about 13 inch long, entire or minutely serrulate ; nutlet, 4 inch long, seven- to eleven-nerved, glabrous, with the apex umbonate and surrounded by a six-lobed calycine ring, within which are the remains of the style. 1 Var, ¢ropicalis, Donnell Smith, Bot, Gaz. xv. 28 (1890), ; Carpinus 533 In winter, the twigs are smooth, shining, glabrous, with five-angled pith, and are marked at the base of the year’s growth by ringlike scars, due to the fall of the accrescent scales of the bud of the previous season. Terminal bud not formed, the tip of the branchlet falling off in summer and leaving a small circular scar close to the uppermost axillary bud, the latter prolonging the shoot in the following season. Leaf-scars small, crescentic, three-dotted, with a short stipular scar on each side. Buds, distichous on the branchlets, unequal in size, on prominent leaf-cushions, appressed against the stem, fusiform, 4 to $ inch long; scales, ciliate and pubescent towards the tips, brownish. Seedling :' Primary root tapering, wiry, flexuose; caulicle terete, pubescent, % inch long; cotyledons fleshy, rounded-obovate, 4 inch long, auricled at the base, shortly stalked, glabrous, green above, whitish beneath; stem zigzag, pubescent, giving off alternate stalked bi-serrate leaves, which resemble those of the adult plant, but are smaller and occasionally lobulate in margin. VARIETIES The common hornbeam shows little variation in the wild state, the only form worth noticing being var. carpinzzza, which is found in Transylvania. In this variety the leaves are often distinctly cordate at the base with only seven to nine pairs of nerves ; and the fruit-involucre has very short lateral lobes. Under cultivation, pyramidal,’ fastigiate, pendulous, and variegated forms have originated. In var. purpurea, the young leaves have a reddish tint. Var. zzczsa, Aiton,’ has leaves with large sharp serrated teeth. A wide-branching tree of this variety at Beauport, Sussex, is 6 feet 3 inches in girth; and there is also a fine specimen at Smeaton-Hepburn, East Lothian. In var. guerctfolza, Desfontaines,* the leaves are smaller than in the type and are irregularly and deeply cut or lobed. In this variety, leaves of the ordinary form are often present on the same branch with those of the pinnatifid kind. Two remarkable trees of this variety are reported® to be growing on the bowling green of the Woodrow Inn, in Cawston Parish, near Aylsham, Norfolk. DISTRIBUTION The common hornbeam is indigenous in the south of England; but its true native limits cannot now be exactly determined, It is recorded® from Somerset, Wilts, Dorset, Hants, Berks, Oxford, Bucks, Herts, Surrey, Sussex, Kent, Essex, Cambridge, Suffolk, and Norfolk ; but in many cases, especially in the south-western counties, the records are probably of planted and not really wild trees. In Dorset,’ it is a very rare tree; and Townsend * considers it to be a doubtful native of Hamp- shire. Druce® considers it to be indigenous in Oxfordshire on the chalk, but always 1 Cf, Lubbock, Seedlings, ii. 532, f. 667 (1892). 2 A solitary wild specimen of the pyramidal hornbeam formerly grew in the forest of Gremsey, near Vicin France. Godron, Les Hétres Tortillards (1869). 3 Aiton, Hort, Kew, iii. 362 (1789). 4 Desfontaines, Zad. Ecol, Bot. Mus. Hist. Nat. 212 (1824). 5 Rev. J. F. Noott in letter to Kew, March 1894. 6 Watson, Comp. Cybele Brit. 311 (1870) and Topog. Bot. 355 (1873). 7 Mansell-Pleydell, Alora of Dorsetshire, 246 (1895). 8 Flora of Hampshire, 313 (1883). ® Flora of Oxfordshire, 268 (1886). 534 The Trees of Great Britain and Ireland planted on other formations. There is no doubt, however, that it formed a consider- able part of the ancient forests, which existed to the north and east of London ; and in the Lea division of Hertfordshire ' it still forms the chief portion of the underwood ; whilst it is common in Essex and Kent, where it is usually treated as coppice. The hornbeam has been found in the fossil state in Suffolk, in the interglacial strata at Hoxne, and in the preglacial strata at Pakefield.” Carpinus Betulus is widely distributed on the continent of Europe, and occurs also in Asia Minor, the Caucasus, and Persia. In Europe, its northern limit, beginning in Norfolk in England, crosses over to Denmark and South Sweden, where it ascends on the west coast to lat. 56° 30’, and on the east coast to 57° 13’, reaching its extreme northerly point on the island of Gothland in lat. 57° 20’. In Norway, Schubeler says, it is not wild; but he has seen a tree at Christiania, planted in 1818, which in 1885 measured 36 feet by 4 feet. In Russia, the hornbeam occurs as far north as lat. 56° 10’ on the coast of Courland, and is confined to the provinces which lie west of an irregular line drawn from near Riga to the Sea of Azov, its most easterly localities being in the governments of Vitebsk, Mohilef, Chernigof, and Poltava, and in the Crimea. South and west of the above limits, the hornbeam is spread through France, Belgium, Holland, Germany, Austrian Empire, Balkan Peninsula, Greece, Switzerland, and continental Italy; but is not found wild in Spain, Portugal, Corsica, Sardinia, and Sicily. In France, the hornbeam is most common in the north and east, where it forms a large part of the coppice forests, and also occurs as undergrowth in the high forests of beech and oak. Its southerly limit in France is a curved line extending from Grenoble through Toulouse to near Bordeaux. Towards the west and south, it becomes a rare tree, and is totally absent from Brittany. It is rather a tree of the plains and low hills than of the mountains; but ascends in the Vosges to 2000 feet, in the Jura to 2300 feet, and in the French Alps to 2800 feet. Treated as coppice, its growth is very rapid in France, where it takes the first rank as firewood. In Germany the hornbeam is widely spread in the plains and low hills, where it grows usually, as in France, in company with the beech and other deciduous trees, either as scattered individuals or in small groups. In east Prussia, where the beech does not occur, the hornbeam replaces it and grows to great perfection, often forming part of the spruce and pine forests. Pure woods are rare, though some of consider- able extent occur, according to Willkomm, in Alsace, Baden, and South Bavaria. In Austria, Hungary, the Balkan States, and Greece, the hornbeam is no longer a tree of the plain, but grows in the mountains in the beech forests. It ascends in the Harz mountains to 1250 feet, in the Bavarian Alps to 2900 feet, and in the Swiss Alps to 3000 feet. According to Radde,’ it is met with through the whole region of the Caucasus, at elevations ranging from sea-level to 5600 feet. It is also recorded from the northern provinces of Asia Minor, and from Ghilan in Northern Persia. (A. HL) 1 Pryor, Flora of Hertfordshire, 373 (1887). ? C, Reid, Origin Brit, Flora, 144 (1899). 3 Phanzenverb. Kaukasusland, 183 ( 1899). Carpinus Bea CULTIVATION The seeds of the hornbeam ripen in October, but with few exceptions do not germinate until after a second winter, and must be treated in the same manner as those of the ash. The seedlings though very hardy as regards spring frosts, grow slowly at first, and require about four years in the nursery before they are strong enough to plant out. Though on sandy soil the tree produces fruit freely and the seedlings bear shade as well as those of the beech, yet the hornbeam does not in England, as in some parts of France, tend to overpower the oak ; and its economic value was formerly much greater than it is now, on account of its being one of the very best trees for firewood, It may, however, be used for underplanting, and as a nurse for other trees on soils too wet for beech, and is admirably suited for making clipped hedges. When the shoots are interlaced they form an impassable barrier, and bear clipping as well as any tree. It also bears pollarding and coppicing extremely well, some of the old pollards which are seen in the eastern counties being of very great age; but when not so treated it does not appear to be a very long-lived tree, and rarely exceeds 200 years. In France, Mouillefert says, it lives 100 to 120 years, and rarely over 150 years, but I think it must considerably exceed this age in some parts of England. The hornbeam is more critical as to soil and climate than most of our native trees; and though Loudon says it is always found on stiff clay and on moist soils where scarcely any other timber tree will grow, this is hardly correct. I have never seen a really fine tree on any but fertile soils, and though it is the most abundant tree of Epping Forest, from which Loudon probably derived his idea; there is not, so far as I know, a really fine specimen in that district, though this may be partly due to their being nearly all pollards. I searched in vain for self-sown seedlings, with roots fit to transplant, and of fifty sent me by Mr. M‘Kenzie, superintendent of Epping Forest, only one survived. He tells me that though large numbers of seedlings may be seen after a good seed year, yet most of them very soon disappear, as the deer and cattle bite them off when not protected by bushes. As a wild tree it is principally found in the south-eastern and eastern counties where the lowest rainfall occurs, but it grows well in the west and in Ireland, and even as far north as Morayshire. Mouillefert says that in France fresh and permeable sandy soils suit it best; and that sandy, gravelly, and flinty clays also suit it well, even when calcareous, but that it languishes or perishes on those which are too stiff, marshy, peaty, or very dry ; and I think this is correct as regards England also. On account of its weak development of roots when young it requires shelter at first, and though it will stand shade fairly, it succeeds best as an isolated tree when adult. As a forest tree it can only be considered of secondary importance, and Forbes does not include it in his state Forestry. As as ornamental tree, it has great value, both on account of the graceful pendent branches, which when in flower and fruit are very beautiful, and for the brilliant yellow colour of the leaves in autumn. 536 The Trees of Great Britain and Ireland REMARKABLE TREES Large hornbeams are not at all common, and exist so far as I have seen in comparatively few places, mostly old parks. The largest and finest that I know of, though by no means the tallest, is near the reservoir at Cornbury Park, Oxford, where there is a tree whose height I could not measure exactly, though it probably exceeds 75 feet, with a bole 11 feet 10 inches in girth and 12 to 14 feet long, which spreads out at that height into an immense number of branches covering a circle of 95 paces (Plate 148). There are two other trees of nearly similar size and habit on the north side of the beech avenue, one of them leaning very much on one side with drooping branches. Sir Hugh Beevor has recently measured a tree, 100 feet in height and 9g feet 8 inches in girth, on Sir Robert Dashwood’s property near West Wycombe. But there is no place where I have seen hornbeams so tall or so numerous as at Cobham Park, Kent, where there must be hundreds of trees 70 to 80 feet high, and many with clean boles 20 to 4o feet long. Among so many it is hard to say which are the largest, but one which I measured near the old heronry, and not far from the ash grove, was over 90 feet high, dividing at about 7 feet into four stems, each of which ran up straight and clean for about 4o feet. Another, a pollard, hollow on one side, measured 13 feet 6 inches in girth, These grow ona soil which suits the ash perfectly. Four shoots from a stool in a wood here measured 76 feet high and 2 to 3 feet in girth, At Mersham-le-Hatch, Ashford, Kent, the seat of Sir Wyndham Knatchbull, Bart., I saw, in 1907, a remarkable wood called Bockhanger, composed of very old pollard hornbeams many of which are hollow and much decayed. They grow on a sandy loam, covered in spring with bluebells, and have for generations served to supply the mansion with firewood, of which the steward told me twelve to fourteen cords were annually consumed. The largest of these trees was about 16 feet to the crown, and had a very large kidney-shaped wen on one side, over which it measured 16 feet 4 inches in girth. Another tree here showed the remarkable power of the hornbeam in repairing wounds in its trunk. A large double-stemmed tree, widely split and hollow at the base, had higher up completely covered the open cleft with young healthy wood and bark in the same way that old yews often do. A most remarkable hornbeam, on account of its very wide-spreading branches, grows in Fredville Park, Kent, and though not over 35 to 4o feet high, covers an area of no less than 103 paces round. It has about fifteen main branches which show the characteristic irregularities that old hornbeams always have. The branches are so thick that foxes often choose the crown of this tree as a lair, and when covered with fruit, as it was when I saw it in June 1907, it is a most striking and beautiful tree. It grows in a deep fertile loam overlying chalk, but rather wet in winter. The hornbeam is, in Essex, especially in Epping Forest, most commonly seen as a pollard, the practice of lopping the branches for firewood having been very general in old times. A photograph showing the appearance of the tree when so PLATE aS HORNBEAM AT CORNBURY PARI AUNAAUOUAVA LV SWVAEDZNYOH AUVT1O0d ‘OF ALVIG PLATE 150. HORNBEAM AT EASTON LODGE MNVd GIVAM LY SINVAINYOH Carpinus 539 treated was taken for me by Mr. Elsden of Hertford, at Waterhall, a farm on Mr. H. Clinton Baker’s property near Bayfordbury, Herts, in January 1907 (Plate 149). At Essendon, Herts, Mr. Baker, in 1906 measured a tree, 81 feet by 11 feet 2 inches; a pollarded tree at the same place being 56 feet high by 18 feet in girth. Sir Hugh Beevor measured in 1891 a hornbeam in Hatfield Park, Herts, which was 17% feet in girth at about 4 feet from the ground. The finest and largest examples of pollard hornbeams that I have seen are in Easton Park, Essex, the seat of the Earl of Warwick. A group of these trees, growing near the park-keeper’s house, which was shown me by Mr. Rogers, agent for the Easton property, contains several trees of great beauty, which were in flower on 7th April. The largest of these measures no less than 28 feet round the head_at about 8 feet from the ground, and 12 feet 2 inches at 2 feet (Plate 150). Another near it, dividing into two stems which are united at the crown, was 25 feet in girth at 7 feet and 174 feet at 2 feet. A third, growing at some distance, has perhaps the finest head of all, and measures 26 feet round the head with a bole about 11 feet high. Mr. Shenstone tells me that the largest he has seen in Epping Forest is 27 feet in girth round the head, and he showed me another very old one in Braxted Park which was over 20 feet round. Mrs. Delves Broughton has sent me a photograph (Plate 151) of a very fine group of hornbeams in Weald Park, Essex, the seat of C. J. H. Tower, Esq., in which, according to the measurements sent me by Mr. T. W. Bacon, the two largest trees are 75 feet by 16 feet 9 inches, and 88 feet by 15 feet 4 inches. At Elveden, Suffolk, there is a very well-shaped and handsome tree in front of the house, which, as I was told by the late Prof. A. Newton, is probably not more than 140 years old, and measured, when I saw it in 1907, 75 feet by 10 feet. At Nibley, Gloucestershire, there is a tree, of which Col. Noel has been good enough to send me a photograph, which measures about 80 feet by 11 feet 6 inches with a bole of 8 feet and a spread of 80 feet diameter. In Bitton churchyard, Gloucestershire, there is a tree planted since 1817 by Canon Ellacombe’s father which is 65 feet by 8 feet 2 inches. At St. Pierre Park, near Chepstow, Major Stacey showed me a very fine hornbeam which, though not very tall, and with a bole only 10 feet high by 11 feet 7 inches in girth, spreads over an area 112 paces round. In the wooded part of Kew Gardens, there are several fine trees, the best of which is 70 feet high and 1o feet in girth, dividing into three stems at 7 feet from the ground. One tree, 5% feet in girth, has bark on the lower part of the trunk, divided into raised longitudinal ridges, which are covered with small scales. At Heron Court, Hants, there is a beautiful tree near the front entrance, 70 feet by 10 feet 5 inches with a spread of 25 yards. At Brocklesby, Lincolnshire, Lord Kesteven measured, in 1906, a tree 77 feet high by 9 feet 4 inches in girth, At Castle Howard the hornbeam grows well and there are several large trees, the tallest being about 80 feet high, the thickest 9 feet 3 inches in girth. At Studley Park, Yorkshire, in the valley below Fountains Abbey, there are several very fine hornbeams, probably the same as those figured by Loudon (ff. 1933, 1934, 1935), which were in 1838 50 to 60 and one 73 feet high. 538 The Trees of Great Britain and Ireland I measured three from 70 to 80 feet with a girth of 63 to 83 feet, one being covered with dense tufts of twigs, a kind of witches’ broom, caused by Exoascus Carpint. In Scotland the hornbeam is less common than in the south, but grows to a large size in the warmer districts ; though, as it is not mentioned either by Hunter, or in the Remarkable Trees of Scotland, it is evidently looked on as a rare tree in the north. Walker! speaks of one formerly growing at Bargally, which was 70 feet high, with a clear trunk of 20 feet. The finest I have seen is a tree at Gordon Castle, perhaps the one mentioned by Loudon as being then 54 feet high; in 1904, it was 68 feet by 8 feet (Plate 152). At Murthly, in the lower park near the Tay, there is an old tree measuring, in 1906, 65 feet by 9 feet 8 inches; and Henry measured one at Scone of the same dimensions. Mr. J. Renwick sends me particulars of a very remarkable hornbeam at Douglas Support, in Lanarkshire, which, in 1900, measured 78 feet by 8 feet 1 inch, with a bole of 17 feet long, and a spread of 60 feet, the branches having long pendulous twigs, which form a screen all round the tree and hang nearly to the ground.’ Another remarkable tree is at Eglinton Castle, Ayrshire, which separates into three stems near the ground, and measures at the narrowest point below the fork 14 feet in girth; its three stems girth 5 feet 9 inches, 5 feet 6 inches, and 4 feet 11 inches respectively. Mr. Renwick sends me particulars of other fine hornbeams as follows :—at Househill, Renfrewshire, 10 feet girth, 72 feet spread; at Tulliechewan Castle, Dumbartonshire, 60 feet by 8 feet 3 inches; at Gargunnock House, Stirlingshire, 8 feet 11 inches girth, 83 feet spread. ° The hornbeam is rarely planted in Ireland. The largest tree, which Henry has seen, is growing beside the Killarney Lake, at Mahony’s Point. It measured, in 1904, 15 feet 8 inches in girth, at 18 inches above the ground, giving off six great stems, the three largest of which were—8 feet 4 inches, 7 feet 7 inches, and 6 feet 3 inches in girth. This tree is about 70 feet in height, and the diameter of its spread is 80 feet. It is in perfect health and bears fruit regularly. At Adare, Co. Limerick, in 1903, Henry saw a fine tree, which measured 53 feet by 8 feet 8 inches, the spread of branches being 65 feet. At Glenstal, in the same county, there is a tree of exactly the same dimensions, as regards height and girth. At Kilrudderry, Co. Wicklow, a tree, which had been blown down, measured 8 feet 9 inches in girth; and here there is a very fine hornbeam hedge, about 15 feet in height. TIMBER The wood of the hornbeam is the hardest, heaviest, and toughest of our native woods, but though extremely strong, is not flexible ; and as it is seldom found large enough and clean enough to cut into planks, it is little used in England except for fuel, for which it is one of the best woods known, burning slowly with a 1 Essays, p. 95, fide Loudon. j ? Tam informed by Mr. Douglas that the peculiarity of this tree consists in the long drooping twigs, which are 20 to 30 feet in length, and hang like small cords to the ground on all sides, concealing the trunk, whilst the upper branches do not droop at all. He thinks that this is due to its being a grafted tree. A photograph, which he is good enough to promise me, will be given in a later volume, HORNBEAM AT GORDON CASTLE : Carpinus 539 bright flame, and making the best of charcoal, As it decays quickly when exposed to wet, it is of no use for outside work, and will not take creosote. The trunk of the tree is often very deeply furrowed, and the wood is said to be cross- grained and difficult to work. It is or was considered the best wood for cogs, mallets, and wooden screws for carpenters’ benches, also for pulleys and butchers’ blocks. Its value is uncertain, and depends largely on the locality, and on the size and age of the tree. With regard to the use of this wood by pianoforte manufacturers, Mr. J. Rose, of Messrs. Broadwood and Sons, to whom I am indebted for much information, writes me as follows :— “ Hornbeam is still used for piano action work in England, though American maple has replaced it to a considerable extent. French hornbeam, and, I believe, Dutch also, are used for the purpose, because of larger size and more freely grown than the British product, and also because, when all charges are included, it is probably cheaper. There is a marked difference in the English hornbeam and that grown in France and elsewhere on the Continent. This is perhaps hardly perceptible in a small sample, but the English wood is smaller and more irregular, but of a distinctly firmer texture, so hard and close as sometimes to resemble ivory. It works beautifully with fine saws and small drills; but the waste is serious. The foreign timber is larger and more freely grown, producing much larger boards, but the grain is coarser, and the texture of the wood less firm, and more liable to split when in small pieces, such as are used in action work.” (H. J. E.) III OSTRYA Ostrya, Scopoli, Fl. Carniol. 414 (1760); Bentham et Hooker, Gen. Pi. iii. 406 (1880) ; Winkler, in Engler, Pfanzenreich, iv. 61, Betulacee, 24 (1904). Carpinus, Linneeus, Gen. Pl. 292 (ex parte) (1737): Smati deciduous trees, belonging to the order Betulacez, agreeing with the genus Carpinus in the characters of the branchlets, buds, foliage, and staminate flowers. Pistillate flowers, in dense erect spikes, inserted in pairs on the base of ovate acute leafy scales, each flower enclosed in a sac-like involucre, formed by the union of a bract and two bracteoles, which is open at the apex at the time of flowering, after- wards becoming closed. Calyx dentate, adnate to the two-celled inferior ovary ; style short, divided into two linear subulate stigmatic branches ; ovules solitary in each cell. Fruits: disposed in stalked ovoid strobiles, composed of densely imbri- cated involucres, which are vesicular, closed, flattened, membranous, longitudinally nerved, reticulate, pubescent at the apex, and hirsute at the base with sharp, rigid, stinging hairs. Nutlet, sessile in the involucre, ovoid, compressed, longitudinally ribbed, crowned by the remains of the calyx; seed solitary, pendulous. Four species of Ostrya have been distinguished :—Ostrya Knowltonz, Coville, a rare tree in Arizona, not yet introduced, and three species, occurring in North America, Eastern Asia, and Europe and Asia Minor, which are so closely allied that they have been considered by most botanists to be only geographical races of one species. These three species are all in cultivation, and as they can be distinguished (see Key to Carpinus and Ostrya, p. 526), will be treated by us separately. OSTRYA CARPINIFOLIA, Hor Hornseam Ostrya carpinifolia, Scopoli, Fl. Carniol. ii. 244 (1772); Willkomm, Forstliche Flora, 368 (1887) ; Mathieu, Hore Forestiére, 403 (1897). Ostrya vulgaris, Willdenow, Sp. Pl. iv. 469 (1805); Loudon, Ard. et Frut. Brit. iii, 2015 (1838). Ostrya ttalica, Spach. Ann, Sc. Nat. sér. 2, xvi. 246 (1841). Ostrya ttalica, sub-species carpinifolia, Winkler, Betulacea, 22 (1904). Ostrya Ostrya, Sargent, Silva N. Amer. ix. 32 (1896). Carpinus Ostrya, Linneus, Sp. Pl. 998 (1753). A tree attaining 60 feet in height and 10 feet in girth; stem cylindrical, bark greyish, finely fissured, and scaly. Young branchlets with dense appressed 540 Ostrya 541 a pubescence. Leaves (Plate 201, Fig. 11) about 3 inches long by 1$ inch wide, ovate, shortly acuminate at the apex, unequal and rounded at the base; margin sharply bi-serrate and ciliate; covered above and below with appressed pubescence, spreading more or less over the whole surface, and not confined to the midrib and nerves, as in Carpinus Betulus, and with minute axil tufts on the lower surface ; nerves twelve to fifteen pairs ; petiole + to 3 inch long, appressed pubescent ; stipules persistent during summer. Nutlet ovoid, 4 inch long, crowned by a tuft of hairs ; calyx-limb obsolete. In winter the twigs are slender, zigzag, more or less pubescent. No true terminal bud is formed, the apex of the branchlet falling off in summer and leaving a minute circular scar at the side of the uppermost axillary bud. Buds small, 33, inch long, ovoid, viscid, set obliquely on prominent leaf-cushions; scales 6 to 9, imbricated, greenish with a dark brown margin, more or less pubescent. Leaf-scar semicircular, with two bundle-dots above and one group of three smaller dots below. Ostrya carpinifolia reaches its most westerly point in the extreme south-eastern corner of France, where it occupies a few isolated stations in the Basses-Alpes and Alpes-Maritimes Departments. In the forest of Miolans,’ in the Basses-Alpes, which is mainly composed of Pzxus sylvestris, it is found on a northern slope, over an area of about 400 acres, occurring chiefly as undergrowth and ascending to about 2700 feet altitude. In the Alpes-Maritimes it descends in some places to nearly sea- level. It extends eastward through Southern Switzerland, the Tyrol (where,’ near Botzen, it ascends to 3500 feet altitude), Carinthia, and Lower Styria to Southern Hungary, and spreads southwards through Carniola, Croatia, and the Balkan States to Greece, growing usually in rocky situations, more commonly on limestone than on other formations. It is common throughout Italy and Sicily in the oak and chestnut regions, ascending to 3800 feet elevation; and forms woods of con- siderable extent around Lake Como, especially above Lecco, on the shores of Lake Lugano, and at Gaudria and Salvatore.? It occurs as a rare tree in Corsica and Sardinia. It is also met with in Asia Minor and in the Lebanon. It attains about a hundred years of age; and according to Pardé* produces coppice shoots like the hornbeam. (A. H.) CULTIVATION It was introduced into cultivation in England some time before 1724, as it is mentioned in Furber’s Nursery Catalogue published in that year. Though an orna- mental tree which attains a good size and is perfectly hardy, it has always been very rare in this country. According to Mouillefert* its growth is about equal to that of the Hornbeam. I have raised plants from French seed which grow faster on my soil than those of the hornbeam, and seem at least as hardy, as they were 1 Fliche, Bull. Soc. Bot. France, xlvi. 8 (1899). Cf. also 2d¢d. xxxv. 160 (1888). 2 Christ, lore de la Suisse, 238 (1907). In the same work, p. 507, it is stated that this species has been found in the fossil state in miocene beds at Ardeche ; and another species, probably a mere variety, has been found in the same strata at Var. 3 Arb. Nat. des Barres, 281 (1906). 4 Principales Essences Forestiéres, 148, note (1903). At Grignon in France, planted together in the arboretum, on calcareous soil with a chalky subsoil, at thirty years old the Hornbeam is 11 metres high by 70 centimetres in girth at 1 metre above the ground ; and the Ostrya 114 metres by 73 centimetres in girth. It bore here without injury the severe winter of 1879. 542 The Trees of Great Britain and Ireland uninjured by the severe spring frost of May 21-23, 1905, and ripened their young wood well in October. They,may be distinguished by the larger leaves with a pair of persistent linear stipules at the base. REMARKABLE TREES From the dimensions given by foreign authors I doubt whether in its native country the Hop Hornbeam ever attains a much larger size than the one which I figure (Plate 153). This remarkable tree is at Langley Park, near Norwich, the seat of Sir Reginald Beauchamp, and cannot be of great age, as it is not mentioned in an account of this place in Grigor’s Eastern Arboretum, published in 1841. It is grafted on a stock of the hornbeam which measures 8 feet in girth below the graft, while the trunk above it is no less than 15 feet 8 inches. Its height is difficult to estimate, but may be about 50 feet. A large tree formerly grew at Kew, on which Mr. J. G. Jack, in Garden and Forest, v. 602, remarks as follows:—‘‘An unusually fine specimen of a hop hornbeam, 50 feet high, branching near the ground and spreading about 70 feet, with a trunk over 3 feet in diameter, was grafted on a stock of hornbeam at 23 feet from the ground, and is a good deal larger than its stock, with a swelling at the point of juncture. No one can help remarking the striking contrast between the rough bark of the Ostrya and the comparatively smooth bark of the Carpinus.” This tree was perhaps the one figured by Loudon’ in 1838, which was then said to be 60 feet high, with a trunk 3 feet in diameter, and the finest specimen in England at that time. In 1890 it was figured in the Gardeners’ Chronicle? as a handsome wide-spreading tree, but soon after began to decay, and was cut down in 1897,’ when it measured 59 feet high by 9 feet 4 inches in girth at 3 feet. Fruit was abundantly produced; but no perfect seeds were ever developed. A part of its trunk is preserved in the Museum at Kew, and I am indebted to the Director for a sample of the timber, which somewhat resembles that of the pear. According to Mouillefert it has all the qualities of hornbeam wood in a superior degree. There is a fine specimen in the Botanic Garden at Oxford, which measures about 40 feet by 4 feet. This tree, though quite healthy, is much infested by mistletoe. At Tortworth there is a tree about 4o feet high by 2 feet 7 inches in girth, At Munden; Watford, a tree, 32 feet by 2 feet 11 inches, is said to have been planted about 1830. In Scotland we know of no tree of this species of large size now existing, though a large one formerly grew at Bargally,t a place between Gatehouse and Newton-Stewart, once the property of Andrew Heron, a celebrated planter, who died in 1729. Loudon went there in 1831, and gives the dimensions® of the Ostrya 1 Op. cit. viii. 244 a. 2 Gard. Chron. viii. 275, Fig. 47 (1890). Also figured in Woods and Forests, 1884, p. 318. The shapes of the trees figured in Loudon and in the Gardeners’ Chronicle are very different. 3 Kew Bull. 1897, p. 404. 4 Walker, Zssays on Natural History and Rural Economy (1812). ® Bargally is fully described by Loudon, op. cit. i. 95-99 (1838). MIOANON ‘WUVd APTONVT LY NVAINYOH-dOH a a "4 kt ‘ Ostrya 543 from a letter of the then owner, J. Mackie, as 60 feet high and 4 feet 1 inch in girth at 4 feet in 1835. Henry could find no trace of this tree when he visited Bargally in 1904. At Glasnevin, Dublin, there are two trees, 30 and 25 feet high, narrowly pyramidal in habit. These are 34 years old and are growing on the bank of a stream. (H. J. E.) OSTRYA VIRGINICA, IRonwoop, American Hor HornBeamM Ostrya virginica, Willdenow, Sp. P2. iv. 469 (1805); Loudon, Ard. e¢ Frut. Brit. iii. 2015 (1838). Ostrya virginiana, Koch, Dendrologie, ii. pt. ii. 6 (1873); Sargent, Silva WM. Amer. ix. 34, t. 445 (1896), and Zrees NV. Amer. 192 (1905). Ostrya Ostrya, Macmillan, Metasperma Minnesota Valley, 187 (1892). Ostrya ttalica, sub-species virginiana, Winkler, Betulacee, 22 (1904). Carpinus Ostrya, Linnzeus, Sp. Pl. 998 (1753) (in part). Carpinus virginiana, Miller, Dict. ed. 8, No. 4 (1768). A tree attaining 60 feet in height and 6 feet in girth, but usually smaller. This species, as seen in cultivation, is mainly distinguished from Ostrya carpinifolia by the presence on the young branchlets, petioles, and midrib of the leaf beneath, of short, erect, gland-tipped hairs. The leaves (Plate 201, Fig. 9) are usually larger, 3% inches long, slightly cordate at the base, with fewer nerves, about twelve pairs. The nutlet in this species is larger, } to 4 inch long, fusiform, flattened, without a tuft of hairs at the apex, surmounted by a plainly visible calyx-limb. Two forms of this species occur in the wild state, which have been distinguished by Spach,’ as follows :— Var. glandulosa.—Young branchlets, petioles, and peduncles covered with gland-tipped short bristles. Specimens in the Kew herbarium from Ontario, Niagara Falls, and the Alleghany Mountains belong to this variety, which is the one known in cultivation in England. Var. eglandulosa.—Glandular bristles not present on any part of the plant. Young shoots pubescent. This variety appears to be common in the western and southern: parts of the United States, and does not appear to have been introduced into cultivation. In the absence of fruit, it would be difficult to distinguish this variety from Ostrya carpintfolra. (A. H.) The tree grows, according to Sargent, on dry gravelly slopes and ridges, often in the shade of oaks and other large trees ; and is a native of Canada and the United States, occurring on the northern shores of Lake Huron in western Ontario, eastward through the valley of the St. Lawrence to Chaleur Bay and Cape Breton Island ; extending southward to Northern Florida and Eastern Texas, and westward to Northern Minnesota, the Black Hills of Dakota, Nebraska, and Kansas. It is most abundant and of its largest size in Southern Arkansas and Texas. I saw it at Mt. Carmel in Illinois, and in the Arnold Arboretum, where it was a finer tree in size and habit than Carpznus caroliniana. It is known in America as 1 Ann. Sc, Nat. sér. 2, xvi. 246 (1841), and Hist. Vég. xi. 218 (1842). 544 The Trees of Great Britain and Ireland Ironwood, and is used for levers and tool handles, the wood being very tough and strong. Michaux states that on the estate of Duhamel du Monceau, in France, there were trees 20 feet high, from which self-sown plants had sprung up. It was introduced into England by Bishop Compton in 1692, but is rarely met with except in botanic gardens. At Kew there are four trees, 20 to 30 feet in height. Others are growing at Eastnor Castle and at Grayswood, near Haslemere, where, though not planted above twenty years, it is growing vigorously, and looks as if it would make a handsome tree. A tree in the Edinburgh Botanic Garden was, in 1905, 39 feet high by 3 feet 3 inches in girth. Seedlings raised in my garden grow more freely than those of the common hornbeam; but not so fast as those of Ostrya carpintfolra. (H. J. E.) OSTRYA JAPONICA, Japanese Hop Hornseeam Ostrya japonica, Sargent, Garden and Forest, vi. 383, f. 58 (1893), Forest Flora Japan, 66, t. 22 (1894), and Siva NV. Amer. ix, 32 (1896); Shirasawa, Leon. Ess. Forest. Japon, text 49, t. 25, ff. 1-14 (1900). Ostrya virginica, Maximowicz, Mé. Biol, xi. 317 (1881). Ostrya italica, sub-species virginiana, Winkler, Betulacez, 22 (1904). A tree attaining in Japan a height of 80 feet, with a tall straight stem, 5 feet in girth, but usually smaller. This species is considered by Maximowicz and Winkler to be identical with the American species, and there is said to be little or no difference in the fruit, which I have not seen. In cultivation, the Japanese tree is readily distinguished as follows :—Leaves (Plate 201, Fig. 10) velvety to the touch on the upper surface, which is covered with a dense erect pubescence ; nerves, ten to twelve pairs, fewer than in the other species; base slightly cordate. Young branchlets densely white pubescent, without glandular hairs, which are also absent from the petiole and midrib of the leaf. According to Sargent, this species is nowhere abundant in Japan, occurring only as scattered individuals in the forests of deciduous trees which cover Central and Southern Yezo, and growing also in the province of Nambu in Northern Hondo. Shirasawa, however, gives a more extensive distribution, stating that it is found also throughout the central chain of Hondo, in the provinces of Musahi, Kai, and Totomi, and also at Nikko; and farther south, in the island of Shikoku. Ostrya waponica is also a native of China, being an exceedingly rare tree in the mountain forests of Eastern Szechwan and Western Hupeh, where it was discovered by Pére Farges and by myself. Ostrya mandschurica, Budischtschew,' recorded from Manchuria, is probably identical with this species. The Japanese Hop Hornbeam was introduced in 1888 into the Arnold Arboretum by seed sent from Japan by Dr. Mayr, and has proved hardy in the climate of Eastern Massachusetts. There are two trees at Kew, sent by Prof. Sargent in 1897, which are now about 15 feet high and growing vigorously. There is also a healthy young tree at Grayswood, Haslemere. (A. H.) 1 In Trautvetter, Act. Hort, Petrop. ix. 166 (1884), I have seen no specimens of this. 4. Ag C.Betulus. O. virginica. O. japonica. Q.carpinifolia. Huitt del, Huth lith CARPINUS AND OSTRYA. PLATE.