The source of this uncorrected OCR text may be viewed in the DjVu format at: http://fax.libs.uga.edu/QK488xE4/3tgbi or http://purl.galileo.usg.edu/ugafax/QK488xE4/3tgbi VOLUME III V> . » Ireland Henry John Elwes, F.R.S. AND Augustine Henry, M.A. Edinburgh: Privately Printed x THE TREES OF GREAT BRITAIN AND IRELAND v t, V LEBANON CEDAR AT HIGHCLERE CASTLE From a Drawing by Charlotte Lady Philllmore. ;A *' ï .* J5.- - 'ft/ Trees Great Britain Ireland Henry John Elwes, F.R.S. AND Augustine Henry, M.A. VOLUME III Edinburgh : Privately Printed MCMVIII CONTENTS LIST OF ILLUSTRATIONS CEDRUS .... CEDRUS LIBANI, LEBANON CEDAR . CEDRUS BREVIFOLIA, CYPRUS CEDAR CEDRUS ATLANTICA, ATLAS OR ALGERIAN CEDAR CEDRUS DEODARA, DEODAR . LIBOCEDRUS . LlBOCEDRUS TETRAGONA LIBOCEDRUS CHILENSIS LIBOCEDRUS DONIANA LIBOCEDRUS BIDWILLI LIBOCEDRUS MACROLEPIS LIBOCEDRUS DECURRENS, INCENSE CEDAR CUNNINGHAMIA .... CUNNINGHAMIA SINENSIS LIQUIDAMBAR .... LIQUIDAMBAR STYRACIFLUA, SWEET GUM LIQUIDAMBAR ORIENTALIS LIQUIDAMBAR FORMOSANA NYSSA ..... NYSSA SYLVATICA, TUPELO . NYSSA AQUATICA, COTTON GUM NYSSA SINENSIS, CHINESE TUPELO . SASSAFRAS ..... SASSAFRAS TZUMU, CHINESE SASSAFRAS SASSAFRAS OFFICINALE, SASSAFRAS . CORYLUS ..... CORYLUS COLURNA, TURKISH HAZEL CARPINUS ..... CARPINUS ORIENTALIS CARPINUS POLYNEURA CARPINUS JAPONICA .... CARPINUS YEDOENSIS CARPINUS CORDATA .... CARPINUS LAXIFLORA CARPINUS CAROLINIANA, AMERICAN HORNBEAM CARPINUS BETULUS, COMMON HORNBEAM . OSTRYA ..... ill 4SI 453 467 469 476 484 485 486 487 488 488 489 494 494 499 500 5°S 506 508 5°9 512 514 515 515 520 520 525 527 527 528 529 529 53° 531 532 540 iv The Trees of Great Britain and Ireland OSTRYA CARPINIFOLIA, HOP HORNBEAM OSTRYA VIRGINICA, IRONWOOD OSTRYA JAPONICA, JAPANESE HOP HORNBEAM NOTHOFAGUS .... NOTHOFAGUS CLIFFORTIOIDES NOTHOFAGUS MENZIESII NOTHOFAGUS FUSCA . NOTHOFAGUS MOOREI NOTHOFAGUS CUNNINGHAMI . NOTHOFAGUS BETULOIDES NOTHOFAGUS OBLIQUA NOTHOFAGUS ANTARCTICA ARBUTUS ..... ARBUTUS UNEDO, STRAWBERRY TREE ARBUTUS HYBRIDA .... ARBUTUS ANDRACHNE ARBUTUS MENZIESII, MADRONA SCIADOPITYS ..... SCIADOPITYS VERTICILLATA, UMBRELLA PlNE PlNUS SYLVESTRIS, SCOTS PlNE CARYA ..... CARYA OLIVJEFORMIS, PECAN NUT . CARYA AMARA, BITTERNUT . CARYA ALBA, SHAGBARK HICKOKY CARYA PORCINA, PIGNUT CARYA SULCATA, BIG SHELLBARK CARYA TOMENTOSA, MOCKERNUT PLATANUS ..... PLATANUS ORIENTALIS, ORIENTAL PLANE PLATANUS OCCIDENTALS, WESTERN PLANE . ACER ..... ACER PSEUDOPLATANUS, SYCAMORE . ACER CAMPESTRE, COMMON MAPLE . ACER PLATANOIDES, NORWAY MAPLE ACER LOBELII .... ACER PICTUM .... ACER OPALUS, ITALIAN MAPLE ACER MONSPESSULANUM, MONTPELLIER MAPLE ACER INSIGNE .... ACER VOLXEMI .... ACER TRAUTVETTERI ACER RUBRUM, RED MAPLE ACER DASYCARPUM, SILVER MAPLE . ACER SACCHARUM, SUGAR MAPLE . ACER MACROPHYLLUM, OREGON MAPLE ACER NEGUNDO, ASH-LEAVED MAPLE SEQUOIA . ... SEQUOIA SEMPERVIRENS, REDWOOD . SEQUOIA GIGANTEA, WELLINGTONS . 540 543 544 545 547 548 549 549 55° 551 552 553 558 559 563 564 565 567 568 571 597 599 600 601 604 605 606 611 614 627 630 641 651 656 659 660 663 665 667 668 669 671 674 677 681 684 687 688 699 ILLUSTRATIONS Lebanon Cedar at Highclere Castle (from a drawing by Charlotte Lady Phillimore) Cedar on Mount Lebanon ....... Lebanon Cedar at Painshill ....... Lebanon Cedar at Goodwood Lebanon Cedar at Strathfieldsaye ...... Lebanon Cedar at Petworth .... Lebanon Cedar at Blenheim ....... Lebanon Cedar at Stratton Strawless ...... Lebanon Cedar at Birchanger ...... Cedar Avenue at Dropmore ....... Algerian Cedars at Téniet-el-Hâad ...... Algerian Cedar at Ashampstead ..... Algerian Cedar at Fota ....... Deodars in the Himalaya ....... Deodar at Bicton ........ Libocedrus chilensis in Chile - .... Libocedrus decurrens at Frogmore ...... Cunninghamia sinensis at Bagshot Park ..... Liquidambar in America (A), Nyssa sylvatica in America (B) . Nyssa at Strathfieldsaye ....... Sassafras at Claremont ....... Corylus Colurna at Wollaton Hall ...... Hornbeam at Cornbury Park ...... Pollard Hornbeams at Bayfordbury ...... Hornbeam at Easton Lodge ....... Hornbeams at Weald Park . . .... Hornbeam at Gordon Castle ...... Hop-hornbeam at Langley Park, Norfolk ..... 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 ....... v Frontispiece PLATE No. 127 128 129 130 IS' I32 '33 134 !35 136 137 138 !39 140 141 142 143 144 145 146 147 148 149 150 'Si 152 '53 154 155 156 '57 158 '59 160 161 The Trees of Great Britain and Ireland Scots Pine at Inveraray Castle Scots Pine at Dunkeld ...... 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 ...... Hickory at Kew . ..... Carya alba at Brocklesby Park ..... Oriental Plane at Ely Oriental Plane at Corsham Court ..... Western Plane in America (A), Oriental Plane in Syria (B) Red Maple at Bagshot Park ...... London Plane at Albury ...... 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 .... Sugar Maple at Park Place ...... Red Maple at Whitton ...... Western Maple in America ...... Redwood at Claremont ...... Redwood Forest in California ..... 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 Carpinus and Ostrya ; leaves, etc. ..... Nothofagus, Fagus, Castanopsis, and Castanea; leaves, etc. Carya ; leaves, etc. ....... Platanus, Catalpa, and Liriodendron ; leaves, etc. Acer; leaves, etc. ....... Acer ; leaves, etc. ... . . Acer ; leaves, etc. ....... PLATE No. 102 163 164 165 166 167 1 68 169 170 171 172 173 174 175 176 177 178 179 1 80 181 182 183 184 185 186 187 188 189 190 191 192 194 195 196 197 198 199 200 201 202 203 204 205 206 207 CEDRUS Cedrus, Lawson, Agric. Man. 379 (1836); Loudon, Arb. et Frut. Brit. iv. 2402 (1838); Bentham et Hooker, Gen. PL iii. 439 (1880); Masters, Journ. Linn. Soc. (Bot.') xxx. 30 (1893). Larix, Miller, Diet. No. 3 (1724) (ex farté). Pinus, Linnœus, Sp. PI. 1001 (1753) (exfarte). Pmus, section Cedrus, Parlatore in DC. Prod. xvi. 2, p. 407 (1864). Abies, Poiret in Lamarck, Did. vi. 510 (1804) (ex parte). TREES belonging to the tribe Abietinese of the order Coniferae, 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, monoecious, 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 bract ; ovules, two on each scale, inverted. m 45 * B 45 2 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, i.e. 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,1 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, if 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, f 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 \\ 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.—(II. 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, LEBANON CEDAR Cedrus Libani, Barrelier, Plante, Icon. 499 (1714); Loudon, Arb. et Frut. Brit. iv. 2402 (1838); Ravenscroft, Pinet. Brit. iii. 247 (1884); Kent, Veitch's Man. Coniferœ, 415 (1900). Cedruspatula, Koch, Dendrol. ii. 268 (1873). Pinus Cedrus, Linnaeus, Sf. PI. 1001 (1753). Larix Cedrus, Miller, Gard. Diet. ed. viii. No. 3 (1768). Larix patula, Salisbury, Trans. Linn. Soc. viii. 314 (1807). Abies Cedrus, Poiret in Lamarck, Diet. 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 \\ inch in length, broader than thick. Cones large and broad, ellipsoid, 3 to 4^ inches long, if to 2^ inches wide; scales 2 inches or more in width, with the claw inflected almost at a right angle. VARIETIES 1. Var. argentea, Antoine et Kotschy, /ter. Cille. 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. atlantica. 2. Var. decidua, Carrière, Conif. 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.1 Webster reports2 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. tortuosa. 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 nana? a dwarf form ; stricta, 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 nie 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.—(II. J. E.) " Hardy Coniferotis 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 Kernel Hempstead. The branches are flattened, horizontal, and very close together, giving the plant a dense, stiff appearance.—Gard. Chrcn. xix. 563 (1896). 454 The Trees of Great Britain and Ireland Supposed hybrids between the Lebanon and Atlantic cedars have been recorded,1 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 History 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 {Juniperus excelsa) would have been not only much easier to procure, but far more prized on every account.2 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. argentea, of our gardens. Northern Syria and Asia Minor form one botanical province, so that the Lebanon groves, 1 Beissner, Nadelholzkunde, 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 atlantica, 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 Libani as another. The hitherto adopted and only alternative of regarding Libani 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, says1 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 400. 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 i foot per annum for trees up to sixty years 1 Card. 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 3^ to 9^ lines, and in trees from 60 to 200 years as 3^ 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 (a8th 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 loo 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 a small 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.1 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- Général 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, hut at a later period this was hroken down. 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 Bârûk, and the third is above Maâsir, 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 Maâsir. The Bârûk 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 Bârûk 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 Bârûk 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 Caria1 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 Abies cilicica, Juniperus excelsa, and J. fœtidissima ; and is occasionally mixed with Pinus Laricio. In Lycia, dense woods of cedar were observed by Luschan2 in the Baba Dagh and between Zumuru and the Bulanik Dagh. The tree, however, appears to attain its maximum development in the Cilician Taurus, where there are fine forests of great extent in the Bulgar Dagh, which have been visited by Tchihatcheff,8 Kotschy,4 and W. Siehe.6 The latter states that the climate in which the cedar grows is a severe one, the snow lying several feet deep on the ground for about five months of the year. He describes 1 Collected in Caria by Pinard, according to Boissier, Flora Orientait;, v. 699 (iSSi). Dr. Stapf informs us that Luschan also saw the cedar in this province. 2 Cf. Stapf, Betträge Flora Lycien, Carien, u. Mesopotamien, 2 (1885). 3 Asie Mineure, ii. 496 (1860). * Reise Cilicischen Taurus, 58, 370 (1858). 6 Gartenflora, 1897, pp. 182, 206. Siehe has sent seed from the Cilician Taurus to vaiious places, and I have two vigorous young trees raised from them. 458 The Trees of Great Britain and Ireland the forest as an open one, the trees standing isolated and attaining about 130 feet in height and 10 feet in girth, and none of a larger size were noticed. Haussknecht found the cedar in the Berytdagh in Cataonia ; Heldreich collected it in the Davros Dagh in Pisidia ; and two or three other localities, where the tree is apparently neither abundant nor remarkable for size, are mentioned by Boissier. I saw the cedar in the Ak Dagh, on the road between Makri and Cassaba, in 1874, where the trees were growing in open woods at about 5000 to 6000 feet elevation, and were not anything like as large as those in the Lebanon. INTRODUCTION We have no certain evidence as to the earliest introduction of the cedar into England; but Loudon, p. 2412, considers that Evelyn was most probably the introducer of the tree, as he states in the third edition of the Sylva (p. 125), published in 1679, that he had received seeds from Mount Libanus. It has been supposed that Dr. Uvedale got the seeds which were planted by him at Enfield between 1665 and 1670 from Evelyn, who, however, does not mention this in the second edition of the Sylva, published in 1670; and until this tree is dead or cut down we shall not know its age for certain.' The oldest specimen2 of cedar in the British Museum is in a volume of Herb. Sloane, ix. fol. 90, the title-page of which bears the following inscription :—" Plants gathered about London about the year 1682 for my own (i.e. Sir Hans Sloane's) collection." Sir Stephen Fox was credited by his descendant, Lord Holland, with having introduced and planted the cedar at Parley, near Salisbury (cf. Loudon, p. 2413), which was cut down in 1813, when it weighed over 13 tons. Quenby Hall, Leicestershire, is also mentioned as having the oldest cedar in England, but this rests on family tradition only, and the tree at Quenby in 1837 was only 47^ feet by 7 feet 9 inches in girth. In Country Life, May 2, 1903, the late Mr. C. J. Cornish gives an account of a cedar at Childrey Rectory, near Wantage, which, " according to unbroken tradition," was planted by Dr. Edward Pocock, who was chaplain to the Turkey Company at Aleppo in 1629, and afterwards chaplain to the Embassy at Constantinople. He returned home in 1641 and was appointed to the living of Childrey in 1642. In 1903 it was still growing vigorously and increasing rapidly in size, and measured 25 feet in girth at five feet from the ground, and covered an area of about 1600 square yards. Though it has suffered much from the loss of branches broken by the weight of snow about twenty years ago it now presents a very handsome appearance as shown by the photograph which is given on p. 567 of Country Life, No. 330. Lord Savile informs me that a cedar, which he remembers as being the tallest that he ever saw, grew at Rufford Abbey. This is believed to have been planted 1 Boulger, in his biographical sketch of Uvedale \T\Journ. Bot. xxix. 13 (1891), gives some details of the Enfield cedar, but has not been able to verify the statement that it dates from 1670. The Enfield cedar is figured in Card, Chron. xxxii. 31, f. 12 (1902). Cf. also Card. Chron. viii. 505 (1890). 2 The statement in Gard. Chron. ii. 194 (1887), that there is mention in Belon's works, which were published in 1553 and 1558, that the cedar of Lebanon existed in France before 1558 is erroneous; and it is probable that the tree was not introduced into France till 1734- Cf. Loudon, p. 2414. Cedrus 459 by Charles II., who used to visit and stay at Rufford, where his rooms are now known as " the King's rooms." Its stump is now surrounded by iron railings and labelled "Cedrus Libani, planted by King Charles II." Loudon considered that the cedars at Chelsea1 mentioned by Sir Hans Sloane in 1685 as then existing (Ray's Letters, p. 176), but now dead, and those at Chiswick House, which are still flourishing, were the oldest in England. But I am informed by Mr. Challis, gardener to the Earl of Pembroke at Wilton House, that "in the year 1874 a very large cedar was cut down there, whose stem up to about 18 feet from the ground was nearly uniform in size, and then divided into twelve distinct branches, each nearly equal in size to a good-sized tree, some of them extending horizontally 70 feet from the trunk. The circumference of the bole five feet from the ground was 36 feet, a transverse section measured when down 11 feet 9 inches, and the number of concentric rings, after several careful counts, some of the rings being somewhat indistinct, was 236. A section of this stem was sent to the South Kensington Museum." If this is correct, and it seems to me that the exact statement of so experienced a gardener as Mr. Challis cannot be questioned, the tree must have been introduced in 1638, before Evelyn's time, and was not only the oldest but also the largest cedar on record in England. I have taken great pains to verify this statement by seeing the section mentioned ; but though careful search has been made in the Records of the British Museum (Natural History), as well as at the Victoria and Albert Museum at South Kensington, and in the letter books of the Royal Horti cultural Society, this wonderful specimen cannot now be traced or discovered. CULTIVATION The seeds of the cedar, whether imported or home grown, should be sown under glass in the spring ; for though they will germinate in the open air, their ^growth is so slow for the first three or four years that much time and loss will be saved by protecting them with a frame. If sown in pots they should be planted out in a frame at a year old, as the roots soon become cramped and pot-bound, and the young plants do not make good roots for some time if they have once been so checked. At two or three years old they may be planted in rich soil about the beginning of May when the buds are starting, and will require some years more in the nursery before finally planting them out. The Lebanon cedar requires a warm, deep, well-drained soil to bring it to perfection, and does not grow so well in the colder and moister parts of England. When once established it will endure our most severe winters without much injury, though it often suffers from heavy snowstorms, which break the branches. The seedlings vary considerably in habit, in vigour, and in colour, and as they do not bear pruning well when the branches become large, it is best to cut off the lower ones when quite small, so as to encourage an upright growth. 1 The last of the cedars in the Physic Garden at Chelsea, which had been dead for some years, was removed in 1904. In 1882 it was 60 feet high and 13 feet 9 inches in girth at 3 feet from the ground. Card. Chron. xxxv. pp. 185, 224 (1904). Cf. also ibid. xxvi. 336, f. 70 (1886), where a figure of the tree is given. Ill C 460 The Trees of Great Britain and Ireland As most people prefer the spreading forms of cedar for lawns or parks, the Lebanon cedar is probably the best for such places ; but when surrounded by other trees it may be drawn up to a great height with few side branches, though I should prefer the Algerian cedar for planting in such situations. Generally it may be said that the Lebanon cedar is the best for the hotter and drier parts of England, and the deodar for the moister and milder districts. The Algerian cedar seems to be hardier, and according to Sargent this is also the case in the United States ; but none of the cedars succeed in New England, though near Philadelphia, Washington, and at Biltmore, North Carolina, there are fine specimens of the Algerian form. The transplantation of large cedars is rarely desirable, but has been sometimes effected with success. A case is recorded1 in which a cedar at Southsea 30 feet high, with a spread of 36 feet, was transplanted at a cost of about ^"100. Mr. J. W. Odell, gardener at the Grove, Stanmore, in a communication to the Royal Horticultural Society on r4th February 1899, states that during a recent gale a large branch was broken off a cedar there, which showed that a great mass of adventitious roots had started from the seat of a previous injury and grown down wards towards the base of the tree, between the splintered portions of the wood. I observed a precisely similar occurrence in a cedar which was partly blown down at Stoke Hall, Notts, in October 1903. The roots were bright reddish-brown in colour, and the thicker ones, an inch in diameter, were covered with rough pustules. Some of these were sent to the Museum at Kew. REMARKABLE TREES Among the existing trees in England it is difficult to say which is the finest. If height and girth combined are taken there is none to equal the splendid tree at Pains Hill, near Cobham, now the property of C. Combe, Esq., of Cobham Park. An account of this place, published in Country Life for March 19, 1904, states that these cedars were probably planted between 1750 and 1760 by the Hon. Charles Hamilton. In 1781 Sir Joseph Banks visited Pains Hill with the younger Linnaeus, who said that he saw there a greater variety of fir trees than he had seen anywhere else. Curiously enough, Loudon, though he often mentions Pains Hill, gives no measurements, and neither Strutt, Lambert, nor Lawson alludes to the cedars there ; but when I saw them in 1904 I measured the largest (Plate 128) to be from 115 to 120 feet high by 26 feet 5 inches in girth. It grows on sandy soil near the lake and divides into several tall, straight stems, which form a spreading crown, and seems to be in perfect health. The next finest of this type that I have seen is perhaps a tree standing in Good wood Park, near the kitchen garden, which, when I measured it in 1906, was about 95 feet high, though its flat top makes the exact height difficult to ascertain, and 26^ feet in girth, the branches spreading over an area of 133 paces in circumference (Plate 129). Goodwood2 is perhaps more celebrated for its cedars than any other 1 Card. Chron. xxv. 42 (1899). 3 Cf. Card. Chron. xxvii. 124 (1900), where the finest cedar at Goodwood was reported to be 29^ feet in girth in 1900. Cedrus 461 place in England, as in 1761 many hundreds were planted by Peter Collinson for the Duke of Richmond. Loudon, on page 2414, quotes a MS. memorandum of Collinson's as follows :—" I paid John Clark, a butcher of Barnes, who was very successful in raising cedars, for 1000 plants of Cedar of Lebanon, 8th June 1761, ^79 : 6s., on behalf of the Duke of Richmond. These 1000 cedars were planted at five years old, in my sixty-seventh year, in March and April 1761 ; in September 1761 I was at Goodwood and saw these cedars in a thriving state. This day, 2Oth October 1762, I paid Mr. Clark for another large portion of cedars for the Duke of Richmond. The duke's father was a great planter, but the young duke much exceeds him, for he intends to clothe all the naked hills above him with evergreen woods." Of the cedars at Goodwood, Loudon goes on to say that 139 remained in 1837. According to Kent (op. cit. 419, note *), eleven fine cedars were uprooted in Goodwood Park by the fierce gale of ßrd March 1897. There are some splendid cedars at Wilton of which Lambert1 writes as follows :—" I am indebted to the Hon. and Rev. W. Herbert (author of the cele brated work on Amaryllidacées) for the following interesting particulars respecting the cedars at Highclere, the seat of the Earl of Carnarvon : ' The two oldest cedars at Highclere were raised in 1739 from a cone brought from Lebanon by Dr. Pococke2 in 1738. They were stunted plants for some time, and removed to their present situation in 1767. The largest of the two measured, in 1829, 9 feet in circumference, having grown only an inch in the last two years, the chalk being unfavourable to its growth. The largest cedar at Highclere, though much younger, measured in 1830, at three feet from the ground, 10 feet i inch in circumference ; it was reared from a cone, which came from the Wilton cedars in 1772, and was about 48 years old before it bore. It was known to the late Earl of Carnarvon that the cedars at Wilton were kept by his grandmother, the Countess of Pembroke, in pots at her window, till growing too large, they were planted upon the lawn, between the house and the water, a situation very favourable to their growth. Supposing them to have been 48 years old, when the cone was gathered from them in 1772, they must have been raised as early as 1724. It is most probable they were between 1710 and 1720; for the Countess of Pembroke who cultivated them died before her husband, who married again after her death, and died in 1733. The oldest cedars at Highclere are, therefore, now (in 1831) 92 years old; those at Wilton at least 106, probably between no and 120. Dr. Pococke found the cir cumference of the largest cedar with a round or single stem to be 20 feet ; but he does not state how near the ground he measured it."' I saw these trees in 1903 and measured them carefully; the best was then about 108 feet high and 21 in girth, with a spread of 109 feet. This tree has lost a large limb, the hollow caused by which has been carefully filled with cement. At Strathfieldsaye there are also some splendid cedars, a group of which on strongish clay soil have the same upright, small-branched character as the Windsor trees. The best of these is 110 feet high by 11 feet 9 inches in girth, with a clean 1 Genus Pinus, ii. 91 (1832). 2 This is not confirmed by Mr. Challis's statement on p. 459 ; and probably all the Wilton cedars were not of the same age. Dr. Richard Pococke travelled in the East during 1737 to 1742. 462 The Trees of Great Britain and Ireland bole of about 40 feet (Plate 130). When mentioned by Loudon it was considered the tallest in England, being then 108 feet high by 9 feet in girth. At Combe Bank, near Sevenoaks, there is a magnificent cedar, which Henry in 1904 measured about 105 feet high with a girth of 20 feet. There are no really large cedars at Syon, Kew, or Woburn. There is a magnificent tree at Blenheim, 28 feet in girth, but of no great height, and having the spreading habit which we usually see in this tree in England (Plate 132). Probably the tallest cedar in England is one in the pleasure grounds of Petworth Park, which I measured carefully in 1905, and could not make it less than about 125 feet high by 14^ feet in girth. It is remarkable for having a trunk clear of branches for no less than 56 feet, where a small limb comes off, but, with this ex ception, it is clean up to about 80 feet. Probably this habit is due to its growing in a sheltered position, more or less shut in by other trees, on a deep bed of sandy loam. Owing to its position this tree is very difficult to photograph (Plate 131). The next finest tree of this type which I have seen is one in the Belvedere Plantation at Windsor. This, according to Menzies, who says that the ground is marked in a map of 1750 as open, cannot be more than 150 years old; and it is at least 115 feet high with a girth of 16 feet. It is without any large branch until it reaches a height of 60 feet or more, and carries nearly the same girth to this elevation ; so that a plank 60 feet long and 3 or 4 feet wide at the top end could probably be cut from it. Menzies figures this tree1 and gives the dimensions in 1864 as only 75 feet by 12 feet 10 inches, which was probably less than its actual size at that time. There are several other fine trees in the same drive, but none equal to this ; and a young one close by, which was planted by Mr. Simmonds, Deputy Surveyor of Windsor Forest, about thirty-five years ago, is now about 40 feet high, and has the same straight-growing upright habit which cedars seem to develop best on deep sandy soil. In Hertfordshire there are many fine cedars, of which the most notable are growing on a lawn at Bayfordbury.2 Mr. Clinton Baker tells me that they were raised from seeds of the Enfield Cedar, and planted in 1765. They have been measured at various intervals by his grandfather, father, and himself, as follows :— l822 1837 I865 l880 1895 1900 1904 4 < No. I at 3 ft. Ft. 10 12 15 16 16 16 17 In. O 2 0 O 7 9 o o No. I at S ft. Ft. 10 ... 16 17 ... 18 18 In. 5 o 3 o 2 No. 3 at 3 ft.2 Ft. IO .. 15 17 18 19 19 In. IO . II 3 9 5 5 No. 3 at 5 ft. No. 7 at I ft. Ft. In. 10 8 13 o 16 5 18 o !9 5 19 9 Ft. In. 14 5 17 6 22 I 20 O 27 I 27 3 At Langleybury, Herts, a large cedar3 was growing in the grounds of 1 History of Windsor Great Park and Windsor Forest, Plate 14. 2 Figured in Card. Chron. xxvi. 521, f. IO2, and 553, f. 109 (1886). s Card. Chron. xiv. 392 (1880). 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. Barren, 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 i£ feet high, in 1747. When described by Grigor1 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 16^ 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 90 feet high, by 14^ 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 Gary 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 21^ 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 i6£ 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 I saw them on i6th 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. Libani. In Cumberland there are two splendid cedars at Eden Hall, the seat of Sir R. Musgrave, Bart., which, according to a paper1 by Mr. Clark of Carlisle, were supposed to be 270 years old, and one of them measured 86 feet by 22^ 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 said2 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,8 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). 3 Essays on Natural History, 69 (1808). Cedrus stated that it was 5 feet and i 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 10 o 13 i|- 14 o Second cedar ... 86 10 9^ n 4 Third cedar ... 7 10 99^ 10 8 Sweet chestnut . . . roi 117 120 Beech .... 94 9 u 10 3 Sycamore . . . . 811 9 7~k 9 II " 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. Ross1 it was, in 1883, 75 feet high by 17^ feet in girth; but I am informed by Mr. T. Muir that it is now 85 feet high by 19 feet 9 inches at i£ feet from the ground, with a spread of 101 feet. At Moncreiffe House near Perth, the seat of Sir R. Moncreiffe, there is a well-shaped tree, which Hunter2 mentioned as bearing many cones and measuring 66 feet by n feet. In 1907, when I saw it, it was about 80 feet by 14^ 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 10 feet at i|- 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. i Woods and Forests, Dec. 26, 1883, P- 59- 3 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 trees1 at Anneville near Dundrum, Co. Dublin, the largest of which was 14^ 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. Carrière2 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. Carrière 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 cedar8 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 Verrières, 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 Cedrela 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,4 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, Art. et 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 Ireland. 3 Traité Canif. 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. 4 In the Eastern States it is known as red cedar, but this term is applied to Thuya 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. (H. J. E.) CEDRUS BREVIFOLIA, CYPRUS CEDAR Cedrus brevifolia. Cedrus Libani, Barrelier, var. brevifolia, Hooker, Journ. Linn. Soc. (£of.), xvii. 517 (1879); Beissner, Nadelholzkunde, 300, fig. 75 J (1891). Resembling C. Libani in characters of leading shoot and branchlets, but with very short leaves, not exceeding \ 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 Libani, var. brevifolia, 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 account2 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 1 5 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 Nature, xxix. 597 (1884). Cf. also Proc. R. Ceog. 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 I ever had before." Madon, who wrote for the Government in 1881 a report1 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 Mâchera 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, reports2 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 1 Parly. Paper: End. V. in Cyprus, No. 366, 0/1881, 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. (H. J. E.) CEDRUS ATLANTICA, ATLAS OR ALGERIAN CEDAR Cedrus atlantica, Manetti,1 Cat. Plant. Hort, prope Modiciam, Suppl. Secundum, 9 (1845) ; Ravenscroft, Pinet. Brit. iii. 217 (1884); Kent, Veitch's Man. Conifères, 409 (1900); Masters, Gard. Chron. x. 425, f- 53 (1891). Cedrus of ricana, Knight, Syn. Conif. 42 (1850). Cedrus Libani, Mathieu, Flore Forestière, 564 (1897). Pinus atlantica, Endlicher, Syn. Conif. 137 (1847). Pinus Cedrus, Linnseus, 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 \\ inch) long, usually as thick as or thicker than broad. Cones shorter and more cylindrical than in C. Libani; scales \\ 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,2 that the glaucous variety constitutes a distinct species.3 The glaucous tint is an essentially unstable character,4 trees occurring in the wild state in which glaucous leaves appear only on some of the branches. This variety often arises in cultivation. Beissner6 mentions several varieties, which have been obtained in cultivation, as pyramidalis, columnaris, and fastigiata* characterised by peculiarities of habit ; and a variegated form in which the foliage of the young shoots is yellowish,7 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 Dccaisne, Rev. Hort. ii. 41 (1853). Cf. Gard. Chron. 1853, p. 132. 3 Cedrus argentea, Renou, Ann. Forest, iii. 2 (1854). « Cf. Fliehe in Mathieu, Flore Forestière, 564, note 2 (1897). 6 Nadelhohkunde, 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. vii. 197 (1890). 1 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, loc. cit. 47° 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 visited1 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 article2 by De Candolle in 1837. Dr. Trabut3 states that the tree occurs in the mountains behind Tetuan ; and it is supposed4 to exist to the south-east of Fez, where the traveller Rohlfs states that he saw larch growing. In Algeria the cedar5 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 pamphlet6 concerning the cedar, published by order of Governor-General Cambon, and M. Lefebvre's excellent book7 on the forests of Algeria. The chief forests are those in the vicinity of Ouarsenis, Téniet-el-Hâad, and Blida, and in the Djurdjura range in the province of Algiers ; and those on Mt. Babor, in the Mâadid mountains south of Sétif, and in the Aurès and Belezma mountains near Batna. The forest8 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 40 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 Bibliothèque Universelle de Genève, 1837, pp. 439, 440. » Les Zones Botaniques de l'Algérie, ^ (1888). 4 Lefebvre, Les Forêts de Cidre, I (1894). * A fine picture of a forest in Algeria is given in Garden and Forest, viii. 335, £ 47 (1895). 6 Les Forêts de Cidre (Alger-Mustapha, 1894). 7 Les Forêts de l'Algérie, pp. 406-421 (Alger-Mustapha, 1900). 8 Hutchison, Trans. R. Scot. Art. Soc. xiii. an, 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 6J feet long, in order to fell the trees. Cedrus 471 between 4900 and 6500 feet ; but on the Häizer 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 Mirbeckii, 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 Mâadid 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-Hâad, and those in the east in the vicinity of Batna, visited by me last January. The cedar occurs around Batna, both on the Aurès 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 T88i 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, id. 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 153. 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 phœnicea. These young trees are narrowly 47 2- 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 :— AGE. 125 years 160 „ 200 „ 255 - 305 .. Diameter. Feet. Inches. 2 2 3 4 4 72 I^ 3 3 Height of Market able Timber. Feet. 46 59 59 59 Total Height. Feet. 98 105 125 125 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 cones1 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-Hâad 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. Cedrus 473 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 Mirbeckii, 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, B), 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 Cèdre 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, libanol, 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 iu Garden and Forest, viii. 335 (1895), shows the flat-topped habit of mature trees in their native forest. 2 The odour disappears after exposure to the air for a few years, and is riot noticeable in the cedar furniture which is so common in the houses at Batna and Téniet-eI-H.;ad. Cf. Lefebvre, Les Forêts de l'Algérie, 350 (1900). 474 The Trees of Great Britain and Ireland When staying at Heythrop Park, Oxfordshire, in March 1901, I went out on a 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 Horticole, 1867, p. 59, gives the following measurements :— LEBANON CEDAR AGE. 1 year 2 years 3 » 4 - 5 - 6 „ 7 - 6-8 12-15 Height. Metres. Centimetres. O O 18-25 36 50 75 o o o o o I ATLAS CEDAR AGE. 1 year 2 years 3 » 4 » 5 » 6 „ 7 „ Height. Metres. 0 O 0 I I 2 Centimetres. 10-15 20-30 40-50 O 75 So 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- 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, an<^ 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-Hâad. In December 1860 the tallest of these was 18^ feet; in December 1866, 31 feet. When I measured it in 1906, it was 77 feet by 8 feet i 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,1 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 6£ 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-Hâad ; 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 s4 feet. At Heanton Satchville, North Devon, I saw a healthy young tree in a 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 fastigiata. 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 the ground, which was said to be sixty-five years old, is prohably a Lebanon cedar. Cf. Card. Chron. xxviii. 210 (1900). } 111 E 476 The Trees of Great Britain and Ireland well at Murthly and other places. At Smeaton-Hepburn, a tree,1 planted in 1847, was, in 1902, 69 feet high and 6^ feet in girth. At Fordell, in Fifeshire, the property of Lord Buckinghamshire, I 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 3^ 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 138). 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 7^ feet in girth. It seemed to me likely to become a most valuable forest tree in this region. (H. J. E.) CEÜRUS DEODARA, DEODAR Cedrus Deodara, Lawson, Agric. Man. 381 (1836); Loudon, Arb. 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. Coniferœ, 411 (1900). Cedrus Libani, Barrelier, var. Deodara, Hooker, Himal. Journ. i. 257 (1854), Nat. Hist. Rev. ii. n, tt. 1-3 (1862), and Fl. Brit. Ind. v. 653 (1888); Collett, Flora Simlensis, 486 (1902) ; Gamble, Ind. Timbers, 710 (1902). Cedrus indica, Chambray, Arb. Res. Conif. 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 2^- 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), Cedrus 477 VARIETIES A considerable number of varieties have arisen in cultivation, ten being mentioned by Beissner.1 1. Var. albo-spica. Growing shoots during spring and early summer of a milky- white colour. Trees of this kind at Dropmorez are pyramidal in habit, and make splendid growth. At Grayswood,3 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. crassifolia. Branches short and stout ; branchlets not pendulous ; leaves short and thick. 4. Var. verticillata. Branchlets whorled. 5. Var.fastigiata. 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. nivea.* Trees with thin, shining, deep green foliage have been distinguished as var. viridis. (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 Picea Morinda and Pinus excelsa, and three species of oak in their various zones. Sometimes the silver fir (Abies Pindrow) accompanies it, but more rarely ; the cypress (Ctipressus torulosa) in its favourite localities joins it ; the yew is often found under it ; and at low elevations it mixes with Pinus longifolia. Among other trees commonly found with it may be mentioned Betula alnoides, Populus ciliata, j^Esculus indica, elm, hazel, hornbeam, maples, bird-cherry, holly (Ilex dipyrena), Pieris ovalifolia, and rhododendron ; while among the shrubs commonly found in deodar forests may specially be noted species of Berberis, Indigofera, Desmodium, Cotoneaster, Euonymus, Salix, especially Salix elegans, Viburnum, Lonicera, Parrotia, and rose, while Clematis montana, Vitis 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 Nadelholzkunde, 307, 308. 3 Gard. Chron. xxxvii. 59, 105 (1905). 2 Card. Chron. xxxvii. 44, 76 (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 Pinus Gerardiana and Querem Ilex. The deodar can attain a very great size.1 Thomson2 mentions one near Nachar, on the Sutlej, that was 35^ feet in girth. Dr. Stewart measured one at Kuarsi 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 loo 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 Dun 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 cornes 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, Life of Sir C. J. F. Bunbury, ii. 238 (1906). 2 Western Himalaya and Tibet, 64 (1852). 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 seeds1 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 Thirty-first Report of the Commissioners of Woods, pp. 168-172, and pp. 440-454 (1853), and further in the Thirty-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 Boldrewood2 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. Keia Hand List of Conifera, 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 Tsuga 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 7^ 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 record1 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,2 Sunninghill,2 Bishopsteignton2 near Teignmouth, Enys2 in Cornwall, and Fota2 in Ireland. Seedlings have been raised from home-grown seed at Rozel Bay2 in Jersey and at Bicton.3 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 Card. Chron. 1852, p. 582, and x. 279 (1891). * Ibid. x. 423, 435, 436, 492, 679 (1891). 3 Ibid. 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 larvae 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 SEcidium 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 evidence1 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 :— Cedrus Deodara Cedrus Libani Cedrus atlantica Total Places reported on. 211 8l 74 Not injured. 80 51 48 Injured. Much injured. Killed. 50 19 19 26 6 2 55 5 5 Proportion of Killed and Much injured. i in 2^ i in 7^ i in It may be interesting to notice in what proportion the three different parts of the kingdom suffered. It was as follows :— Scotland England Ireland Total Places reported on. 64 142 4 Not injured. 19 61 3 Injured. 26 24 I Much injured. H 13 Killed. 5 50 Proportion of Killed and Much injured. i in 3^ i in 2^ 1 Published in Ravenscroft, Pinet. Btit. 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 Pinet. Brit., as long ago as 1858. One of these on the lawn measured in 1902 was 80 feet by ii feet 8 inches (Plate 140). The other is near the ornamental water in a more sheltered situation, and was then 90 feet by 9 feet i 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 7^ 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 ID 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 planted1 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 9^ 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 atlantica ; 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 1 Card. C/ircti. xxv. 138 (1899). Cedrus 483 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 by 4 feet 8 inches in girth. The best that we know in this county is perhaps one at 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 tree1 planted in 1841, when it was 2^ 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 7^ 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. 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.D., 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,2 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). Ill 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, Bâillon, Hist. PI. xii. 34 (1892). EVERGREEN trees with aromatic odour, belonging to the tribe Cupressinese of the order Coniferse, 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 : monoecious with those of the two sexes on different branchlets, or rarely dioecious, 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 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 Bidwilli, 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 miich longer than the median leaves. 5. Libocedrus chilensis, Endlicher. Chile. Median leaves minute, rounded at the apex, with a conspicuous gland. 6. Libocedrus 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. Libocedrus papuana, F. v. Mueller.1 New Guinea. 8. Libocedrus austro-caledonica, Brongniart et Gris.2 New Caledonia. LIBOCEDRUS TETRAGONA LibocedriQ tetragona, Endlichler, Syn. Conif. 44 (1847); Lindley and Paxton, Flower Garden, i. 47, f. 32 (1850); Kent, Veitch's Man. Conifères, 256 (1900). Libocedrus cupressoides, Sargent, Situa JV. 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 tree8 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 ^ 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 \ inch long, brown. Scales four, minutely 1 Trans. Roy. Soc. Victoria, \. 32 (1889). z Bull. Soc. Bot. France, xviii. 140 (1871). 3 This tree has been confused by travellers witb Fitzroya 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,1 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,2 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 Nothofagîis betuloidcs and Drimys Wintert. It was introduced by W. Lobb3 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, Flower Garden, i. 48, f. 33 (1850); Kent, Veitch's Man. Coniferce, 252 (1900). Thuya chilensis, Don, in Lambert, Pimts, ii. 19 (1824); Hooker, London Journ. Bot. ü. 199, t. 4 (1843). Thuya andina, Poeppig et Endlicher, Nov. 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 El Ars, 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 Filzroya patagonica. Cf. Hortus Veitchii, 46 (1906). 2 Scott, Princetown Univ. Exped. Patagonia, viii. 6, 18, 142 (1903). 3 dard. Chron. 1849, p. 563. Libocedrus 487 Cones1 on short branchlets, J 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 5^ 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 L. 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. Coniferœ, 254 (1900); Cheeseman, New Zealand Flora, 646 (1906). Libocedrus plumosa, Sargent, Suva N. Amer. x. 134 (1896). Docrydium phtmosum, D. Don, in Lambert, Pimts, ed. 2, Appendix 143 (1828). Thuya Doniana, Hooker, London Journ. Bot. i. 571, t. 18 (1842). A tree, attaining in New Zealand 100 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. chilensis ; 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 \ inch long ; scales four, each with a lanceolate acuminate, erect, 1 Cones ripened on young trees at Les Barres in France in 1900. Fardé, Art. 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 Vitœ, 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 Liboeedrus Bidivilli, ]. D. Hooker, Flora New Zealand, i. 257 (1867); Kirk, forest flora New Zealand, 159, tt. SZA, 83 (1889); Cheeseman, New Zealand Flora, 647 (1906). A tree similar to L. Doniana, 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 L. Doniana, but more slender ; on old trees tetragonal, ^th to ^th 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 L. Doniana, but smaller, J 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 L. Doniana, it ought to be hardy in the milder parts of the British Isles. (A. H.) LIBOCEDRUS MACROLEPIS Liboeedrus macrolepis, Bentham et Hooker, Gen. PI. iii. 426 (1880); Kent, Veitch's Man. Coniferœ, 255 (1900); Masters, Gard. Chron. xxx. 467 (1901); Henry, Garden, Ixii. 183, with figure of tree (1902). Calocedrus macrolepis, Kurz,fourn. 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 L. 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, Liboeedrus 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 f 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 Liboeedrus 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,1 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 CEDAR Liboeedrus decurrens, Torrey, Smithsonian Contrit, vi. 7, t. 3 (1854); Sargent, Suva N. Amer. x. 135, t. 534 (1896), and Trees N. Amer. 73 (1905); Kent, Veitch's Man. Conifera, 253 (1900); Mayr, Fremdländ. Wald- u. Parkbäume, 315 (1906). T/iuya Craigana, Murray, Botan. Exped. Oregon, 2, t. 5 (1853). Thuya giganiea, Carrière, 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 \ 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 Ann. 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." 49° 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, l to \ 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, i£ inch long, terete, brownish, glabrous. Cotyledons, two, if 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, f 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 Thuya Craigana in honour of Sir W. Gibson Craig, one of the members of the association. Carrière confused the tree with TJmya gigantea ; and for some time there was great confusion in the nomenclature of the two species. Libocedrus decttrrens 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 49 the floors of open valleys, usually growing singly or in small groves, often mixed with Pinusponderosa 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 Finns 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 i 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, Abies coneolor, 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 90 to 100 by 8 or 9 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 plieata 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. Washington, 1900. Ill 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 i6th 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 i 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 4ij 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 i 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 n 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 6^ 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. A 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 01 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 Libocedrus 493 by Henry, was 42 feet by 4 feet 10 inches in [905. 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, Libocedrus decurrens is rare in cultivation. At Stradbally Hall, Queen's County, a fine tree measures 53 feet high by 5| 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 Madré 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).1 The largest I have seen in France is at Verrières, 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 Hortus Vihnorinianus (1906). (H. J. E.) 1 Pardé, Arb. Nat, dus Bai-res, 32 (1906). CUNNINGHAMIA Cnnninghamia, R. Brown, in Richard, Cot/if. So, t. 18 (1826); Kentham et Hooker, Gen. PI. iii. 435 (1880); Masters, Join n. Linn. Soc. (Bof.) xxvii. 304, fig. 18 (1889), and xxx. 25 (1892). Belis? Salisbury, Trans. Linn. Soc. viii. 315 (1807). Jacnlaria, Rafinesque, in London, Card. Mag. viii. 247 (1832). Raxopitys, Nelson (Senilis), Piiiaeece, 97 (1866). A GENUS, belonging to the Coniferœ, with only one known living species,2 and doubtfully represented in the fossil state.8 Cunninghamia is considered by Bentham and Hooker, and by Masters, to be a member of the family Araucariese ; but it is placed by Eichler8 in Taxodinese. Seward and Ford, who have lately published an exhaustive monograph4 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 Araucarieae and the Taxodinese ; 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 Cunningliamia sinensis, R. Brown, loc. cit. (1826); Lambert, Genus Finns, ed. 2, t. 53 (1832); Loudon, Arb. et Frut. Brit. iv. 2445 (1838); Murray, Pines and Firs of Japan, 116, figs. 216- 224 (1863) ; Kent, Veitch's Ma». Coniferce, 292 (1900); Shirasawa, Icon. Ess. Forest. Japon, text 23, t. 9, fi". 1-24 (1900). CuHninghamia lanceolata, W. J. Hooker, Bot. Mag. t. 2743 (1827). Finns lanceolata, Lambert, Genus Finns, ed. i, t. 34 (1803). Belis jaculifolia, Salisbury, loc. cit. 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 Kellis, used for the daisies. 2 While the above was passing through ihe press, there has been received at Kew a specimen of a new species of Cunninghamia, lately discovered in the mountains of Formosa al 7000 feet altitude. This species, which will shortly be published by Mr. Hayata, differs from C. sineusis in having shorter leaves, acute aiid not acuminate at the apex. Mr. Hemsley is inclined to thinl< that a specimen, preserve«! in the Herbarium, which was collected on Mt. Omei, in Western China, by Faber, is possibly a third distinct species. 3 Engler u. Prantl, Natur. Pßanicnfamil. ii. 85 (1889). Cnnniiiglianiites, an allied fossil genus, has been found in the Keuper and Chalk deposits in Saxony, Bohemia, Westphali.i, Southern France, and Greenland. Cf. Schimper u. Schenck, Palicoatologie, 283 (1890). 1 The Araucarieœ: Phil. Tians. Key. Soc., vol. cviii. 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, i 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 \\ 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.1 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, i 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 i£ 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, rooi-suckers are often produced, which grow 1 Cf. Woods and Forests, 1884, p. 593, and Garden, xxix. 173 41886). 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. Cunninghamial 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.2 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 introduced3 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,4 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 TOO to 150 feet in height, and from 15 to 18 feet in girth ; and when growing in close stands, with straight stems clean to 40 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 4«slialia, ii. 564 (1826), R. Brown states that he requested Richard to change the name Belts, given by Salisbury, into Cunninghamia, in honour of both J. Cunningham, the discoverer of the tree, and of the collector Allan Cunningham. 2 Amallheum Bolanicum, i. t. 351, f. 2. (1705). 3 Aiton, llmtus Kewensis, \. 320 (1813). 1 Wanderings in China, 379 (1847); Tea Countries, ii. 178, 212 (1853); Residence among the Chinese, 189, 277 (I8S7). 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, hsiang-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,1 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 planted2 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 10 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 [905 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 II.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 Fremdland. Ifald- u. Parkbäume, 285 (1906). - Cf. Hayata, in Tokyo 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 tree1 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 i 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,2 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 al 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,3 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. VVicklow, 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 \\ 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,5 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,6 on the northern end of Lake Maggiore, a tree planted fifteen years is 23 feet in height. (H. J. E.) 1 John Smith, in Records of Krw 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. 4 Mouillefett, Traite des Arbres, ii. 1336 (1898). 5 Partie, Arb. Nat. des Barres, 57 (1906). e Christ, Flore de la Suisse, 77 (1907). LIQUIDAMBAR Liquidambar, Linnseus, Gen. PL 463 (1742) ; Bentham et Hooker, Gen. PL i. 669 (1865); Engler u. Prantl, Pflanzenfam. iii. pt. 2, 123 (1891). DECIDUOUS trees belonging to the order Hamamelideœ. 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 monoecious, 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,1 wild in China, which are imperfectly known and not introduced. 1 These are :— 1. Liquidambar Rosthontii, Diels, Flora von Central China, 380 (1901), a small tree occurring hi Szechwan; flowers and fruit unknown. It resembles in foliage L. orientalis. 2. Liquidambar sp., Hemsley, Journ. Linn. Soc. (Bot.) xxiii. 292 (1887). Specimens, consisting of detached leaves and 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. Card. Chron. xlii. 344 (1907). ni 499 H 500 The Trees of Great Britain and Ireland The species in cultivation are :— 1. Liquidambar styraciflua, Linnaeus. 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. Liquidambar orientalis, Miller. Asia Minor. Shoots glabrous. Leaves small, five-lobed, margin with large lobules ; under surface quite glabrous. 3. Liquidambar formosana, Hance. China, Formosa, Tonking. Shoots pilose. Leaves large, usually three-lobed ; under surface pilose, without conspicuous axil-tufts. LIQUIDAMBAR STYRACIFLUA, SWEET GUM Liquidambar styraciflua, Linnaeus, Sf. PI. 999 (1753); Loudon, Arb. et Frut. Brit. iv. 2049 (1838); Oliver, in Hooker, Icon. Plant, xi. 13 (1867); Sargent, Si/va N. Amer. v. io. t. 199 (1893), and Trees N. Amer. 340 (1905). Liquidambar macrophylla. Oersted, Am. Cent. xvi. t. io (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 \\ 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. Trelease2 observed in the case of Liquidambar trees 1 See Miss Gregory in Botanical Gazette, xiii. 282 (1888). 2 Garden and Forest, 1890, p. 195. Liquidambar 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,1 as it is usually called in the United States, has a very wide range of distribution. Its most northerly station is, according to Sargent,2 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 » Also known as Red Gum. 2 Garden and Forest, ii. p. 232. 5O2, 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 nj feet in girth, which was 94 feet to the first branch. Plate 144 A, 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 i2\ 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 Madré 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. Liquidambar 5°3 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,1 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 reported2 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,3 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, Arboretum Notes, 28. 3 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. i 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 g-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 (i) 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 25. 3d. per cube foot, and "hazel pine" only is. 3d. to is. 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.1 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. Deft. Agric. Forest Service Circular, No. 51 (1906). Liquidambar 5°S LIQUIDAMBAR ORIENTALIS Liqtddambar orientalis, Miller, Gard. Diet. No. 2 (1768); Oliver, in Hooker, Icon. Plant, xi. 13, t. 1019(1867); Hanbury, Science Papers, 139, with figure (1876); Bentley and Trimen, Medicinal Plants, ii. No. 107, t. 107 (1880). Liquidamlar imberbe, Alton, Hort. Kew. iii. 365 (1789); Loudon, Arb. et 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 z\ 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 Liquidambar styraciflua, but smaller. Fruiting head about i 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 Liquidambar 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 Signala, near Mêlasse, 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.1 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.2 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 Altingia excelsa, Noronha, a tree allied to Liquidamlar, which yields by incisions in the bark a sweet-scented resin. Cf. Bretschneider, Bot. Sinicum, 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 10 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 Liquidambar styraciflua 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 Baleine1 (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. 215 (1866), and Jour». Bot. 1870, p. 274; Oliver, in Hooker, Icon. Plant, xi. 14, t. 1020 (1867); Hemsley,y0w?7z. Linn. Soc. (£ot.) xxiii. 291. Liquidambar acerifolia, Maximovvicz, Mél. Biol. vi. 21 (1866) and viii. 419 (1871). Liquidambar Maximmuiczii, Miquel, Ann. Mus. Bot. Lugd. Bat. Hi. 200 (1867). A tree2 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 Partie, Arbor. Nat. des Barres, 205, note I (1906). 2 The peculiarities of the buds, leaves, and stipules have been fully described by Lubbock, mjotirn. Linn. Soc. (Bet.) xxx. 495 (1894). Liquidambar 5°7 with dense long pubescence. Petiole pilose, with two subulate, persistent, pubescent, glandular stipules. Fruiting heads spiny, i^ 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,1 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, Joiirn. Bot. i. 257, says it is the commonest tree in the mountain forests of Hainan. Hance, loc. cit., says that at Canton old stumps buried beneath the soil sucker freely. 2 It yields a resin, Ftn»-hsiang ; and a caterpillar, which feeds on its leaves, produces a coarse kind of silk, used for fishing-lines. Ill NYSSA ATyssa, Linnaeus, Gen. PI. 308 (1737); Bentham et Hooker, Gen. PL i. 952 (1867); Harms in Engler u. Prantl, Pflanzenfam. iii. 8, 257 (1898). Tupelo, Adanson, Fain. PI. ii. 80 (1763). Ceratostachys, Blume, Bijdr. Fl. Ned. Ind. 644 (1825). Agathisanthes, Blume, loc. cit. 645. Daphniphyllopsis, Kurz, Journ. Asiat. Soc. 1875, "• 2O1- DECIDUOUS trees or shrubs belonging to the order Cornacese. Leaves alternate simple, stalked, with margin entire or remotely one- to four-toothed, without stipules. Branchlets with discoid pith. Flowers small, diœcious 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 :—Nyssa sessiliflora, Hooker, a tree attaining 60 feet in the Himalayas and Java ; has not been introduced and would probably not be hardy in England. Nyssa sinensis, Oliver, has recently been intro duced from Central China. The remaining five species are natives of Eastern North America Nyssa acttminata, Small, a species imperfectly known, is a small shrub growing in pineland swamps in Georgia. Nyssa 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 Nyssa sylvatica, Marshall ; and no trees referable without doubt to it are known to us in England. Nyssa sylvatica, Marshall, and Nyssa aquatica, Marshall, occur rarely in cultivation in England. NYSSA SYLVATICA, TUPELO, PEPPERIDGE, BLACK GUM JVyssa sylvatica, Marshall, Arlust. Am. 97 (1785); Sargent, Suva N. Amer. v. 75, t. 217 (1893), and Trees N. Amer. 707 (1905). Nyssa multiflora, Wangenheim, Nordam. Holz. 46, t. 16, f. 39 (1787). Nyssa villosa, Michaux, Fl. Bor. Am. ii. 258 (1803); Loudon, Arb. et 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 répand 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, ^ to i inch long. Flowers on pubescent peduncles, appearing after the leaves ; staminate flowers numerous, stalked and in crowded clusters ; pistillate flowers sessile, two to fourteen in a head. Fruit ovoid, bluish-black, -\ to f 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 i^ inch long by f inch broad, on petioles ^ 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 Nyssa aqiiatica, with leaves grey and pubescent all over the under surface, and with one or two teeth often on the margin. Nyssa sinensis, which resembles in foliage Nyssa sylvatica, is dis tinguished by the appressed pubescence of the shoots. In winter Nyssa 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. biflora, Sargent, Suva N. Amer. v. 76, t. 218 (1893). Nyssa biflora, Walter, FL Carol. 253 (1788); Loudon, Arb. et 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 Nyssa bißora were all probably Nyssa 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 loo feet with a maximum girth of about 15 feet.1 It is found generally in wet soil on 1 But Ridgway measured a Mack gum hi Wabash Valley, 125 feet high Ijy 13 feet in girth, and 64 feet to the first limb. Nyssa 511 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 144 n 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, iii. 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 tree1 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 100 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 4^ 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 Card. Chron. xxvi. 162 (1899), is evidently eironeous, 14 feet Ioj inches being a misprint for 4 feet loj inches. 2 Schenck, in Biltmore 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 27^ 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. (H. J. E.) NYSSA AQUATIC A, COTTON GUM, TUPELO GUM Nyssa aquatica, Marshall, Arbust. Am. 96 (1785) ; Linnaeus, Sp. PI. 1058 (exparte) (1753) ; Sargent, Silva N. Amer. v. 83, t. 210 (1893), and Trees N. Amer. 711 (1905). Nyssa unifiera, Wangenheim, Nordam. Holz. 83, t. 27, f. 57 (1787). Nyssa denticulata, Aiton, Hort. Kew. iii. 446 (1789). Nyssa tomentosa, Michaux, FI. Bor. Am. ii. 259 (1803). Nyssa aiigulisans, Michaux, loc. dt. Nyssa grandidentata, Michaux f., Hist. Art. Am. ii. 252, t. 19 (1812); Loudon, Arb. 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. 10) elliptical or ovate-oblong, base rounded or tapering, apex long-acuminate ; margin entire or répand, 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, i to i^ 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 Nyssa 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 Nyssa 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 Illinois. An interesting account of the peculiar habit of this tree, as observed in the swamps of Arkansas, is given by Coulter.1 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. Wilson2 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 cultivated3 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 specimen6 of Nyssa aquatica, perhaps identical with the tree now living. 1 Report Missouri Bot. Garden, 1904, xv. 56, plates 18, 19. 2 Prof. Philadelphia Acad. Nat. Sc. 1889, P- <>9- 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 Holroycl,1 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 (Liriodendron). 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. gd. 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, CHINESE TUPELO Nyssa sinensis, Oliver, in Hooker, Icon. Plant, t. 1964 (1891). A tree, attaining in China 40 feet in height. Young shoots covered with a dense appressed white short pubescence, retained in the second year. Leaves (Plate 199, Fig. i) 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, £ to f 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, f 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.2 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.) 1 U.S. Dept. Agric., Forest Service Circular, No. 40 (1906). a E. II. Wilson in Card. Chron. xlii. 344 (1907). SASSAFRAS Sassafras, Nées ab Esenbeck u. Ebermaier, Handb. Med. Pharm. Bot. i. 418 (1830); Bentham et Hooker, Gen. Pl. iii. 160 (1880). DECIDUOUS trees belonging to the order Lauracese, with alternate pinnately-veined simple leaves without stipules. Flowers dioecious 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 thicke'ned 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, CHINESE SASSAFRAS Sassafras Tzumu, Hemsley, in Kew Bull. 1907, p. 55, and in Hooker, Icon. Plant, t. 2833 (1907). Litsea laxiflora, Hemsley, Journ. Linn. Soc. (Bot.) xxvi. 383, t. 8 (1891). Lindera Tzumu, 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 tzu-mu or /tuang- ch'iu tree. Resembling very closely the American species in the characters of the foliage and inflorescence, it was considered by Prof. Sargent1 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). in 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, TO feet high, growing in the Coombe Wood nursery, which was raised from seed sent by Wilson in IQOO. 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, SASSAFRAS Sassafras officinale, Nées ab Esenbeck u. Ebermaier, loc. dt. ; Bentley and Trimen, Medicinal Plants, iii. 220 (1880). Sassafras Sassafras, Rarsten, Pharm. Med. Bot. 505 (1882); Sargent, Suva N. Amer. vii. 17, tt. 304, 305 (1895), and Trees N. Amer. 337 (1905). Sassafras variifolium, O. Kuntze, Rev. Gen. PL ii. 574 (1891); Sargent, in Bot. Gazette, xliv. 226 (1907). Laurus Sassafras, Linnaeus, Sp. PL 371 (1753); Loudon, Art. et Frut. Brit. iii. 1301 (1838). A tree, attaining in America 90 feet in height and 18 feet in girth. Bark,1 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 répand, 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, i 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 i^ 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, f 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, i^ to 2^ 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 ;J 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 m a garden near London.1 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,2 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 i 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 1898. 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 is a younger tree in Mrs. Robb's grounds at Goldenfielcl, Liphook, and a small one in Kew Gardens planted by Sir W. T. Thiselton-Dyer. There is also a healthy young tree at Tortworth. 1 Gerard, Herball (ed. Johnson), 1524 (1633). 2 Woodlands, Nos. 489 seq. (1825). 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, vii. 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 1569. 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, Linnasus, Sf. PI. 998 (1753); Bentham et Hooker, Gen. PI. iii. 406 (1880); Winkler, in Engler, Pflanzenreich, iv. 61, Betulacece, 44 (1904). DECIDUOUS trees or shrubs, belonging to the order Betulaceae. Leaves alternate, distichous on the branchlets, stalked, simple, penninerved, doubly serrate ; stipules two, caducous. Buds composed of numerous imbricated scales, corresponding to stipules. Flowers monoecious, 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, Corylus Colurna, attains the dimensions of a timber tree, and comes within the scope of our work. CORYLUS COLURNA, CONSTANTINOPLE OR TURKISH HAZEL Corylus Colnrna, Linnaeus, Sp. PL 999 (1753); Loudon, Arb. et Frut. Brit. iii. 2029 (1838); Willkomm, Forstliche Flora, 377 (1887); Winkler, op. 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 521 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 ^ to i inch long, glandular pubescent or glabrescent. Catkins1 i^ 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 f 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. lacera, A. de Candolle, op. cit. 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 arc figured in Card. 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 40 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. :— Coryhis intermedia, Loddiges, Catalogue (1836) (Coryhis 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, i.e. 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 Göttingen and in the Forestry Garden at Münden, 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,1 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.2 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.3 It occurs in Asia Minor in Bithynia, Paphlagonia, and Anatolia. According to Radde,4 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-scat aie very irregular in number and shape, being circular dots or curved lines. 2 In Thessaly and Acarnania, according to Halacsy, Consp. Fl. Gratcce, iii. 135 (1904). 3 Beck, Veg. Illyrischen Länder, 300 (1901). 4 Pßanzenverb. 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. Linnaeus 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,1 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.2 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. Flora, 225 (1665). Ill 2 British Trees with Illusltalions, 9 (1901). L 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 London ; 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. Cohirna. 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. Gamble1 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. (H. J. E.) 1 Man. Indian Timbers, 684 (1902). CARPINUS Carpinus, Linnseus, Gen. PI. 292 (exparte) (1737); Bentham et Hooker, Gen. PL iii. 405 (1880); Winkler, in Engler, Pflanzenreich, iv. 61, Befulaceœ, 24 (1904). Distegocarpiis, Siebold et Zuccarini, Flor. Jap. Farn. Nat. ii. 103, t. 3 (1846). DECIDUOUS trees belonging to the order Betulacese. 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, monoecious, 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. DISTEGOCARPUS, Sargent, Suva 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, loc. cit. 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 laxißora, 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 orientalis, Miller. South-eastern Europe, western Asia. Leaves i J inch long, acute, deeply plicate. 2. Carpinus polyneura, Franchet. Central China. Leaves 2 inches long, acute, smooth and only slightly plicate. II. 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. Carpinusyedoensis, Maximowicz. Central China. Cultivated in Japan. Leaves 2^ inches long, rounded at the base, with conspicuous bands of appressed pubescence on the upper surface. Branchlets pilose. C. Leaves ovate, acimiinate 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. t Under surface glabrous between the nerves. 6. Carpinus laxißora, 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, •fa inch long. 7. Carpinus Betiilus, Linnaeus. 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. Betulus, but turning red in autumn. Branchlets with scattered long hairs. Buds ovoid, \ inch long. ft Under stirf ace 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 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. (A. H.) CARPINUS ORIENTALIS Carpinus orientalis, Miller, Gard. Diet. ed. 7, No. 3 (1759); Loudon, Arb. et Frut. Brit. iii. 2014 (1838); Winkler, Betulaceœ, 37 (1904). Carpinus duinensis, Scopoli, Fl. Carniol. ii. 243, t. 60 (1772); Boissier, Fl. Orient, iv. 1177 (1879); Willkomm, Forstliche Flora, 366 (1887). Carpinus nigra, Moench, Verz. Ausland. Bäume H. Stand. 19 (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,1 strongly plicate, the nerves being deeply impressed above, about i^ inch long by -f 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, \ 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 (1899); Burkill, Journ. Linn. Soc. (J3ot.\ xxvi. 501 (1899). Carpinus Turczaninowii, Hance, var. polyneura, Winkler, Betulaceœ, 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 2j 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 £ 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. orientalis ; nerves nine to twelve pairs ; petiole, £ to f 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. Turczaninowii, 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, Icon. Ess. Forest. Japon., text 47, t. 24, ff. 1-17 (1900); Winkler, Betulacece, 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. Fnm. 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. i) ovate-oblong, up to 4 inches long by i| 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 \ inch long, linear - lanceolate, pubescent, persistent during summer. Fruit : strobiles 2^ inches long ; bracts densely imbricated, f 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 Young plants were sent by Prof. Sargent in 1895 and 1897 to Kew, which have now attained about 10 feet in height. At Tortworth a young tree has produced fruit. The foliage of this species is remarkably distinct and handsome. (A. H.) CARPINUS YEDOENSIS Carpinus yedoensis, Maximowicz, Mil. Biol. xi. 314 (1881); Burkill, Journ. Linn. Soc. (Bot.), xxvi. 502 (1899); Franchet, Journ. de Bot. xiii. 203 (1899); Winkler, Betulacece, 35 (1904). A small tree. Young branchlets densely covered with long hairs. Leaves (Plate 201, Fig. 3), 2,^ inches long by i£ 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, f to \ inch long, pilose ; stipules, linear-lanceolate, caducous. Fruit : strobiles, 2\ inches long ; bracts loosely imbricated, f 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 Père 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. \. 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, Icon. Ess. Forest. Japon, text 46, t. 24, ff. 18-32 (1900); Winkler, Betulacece, 26 (1904); J. H. Veitch, Hortus Veitchii, 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 53° The Trees of Great Britain and Ireland pairs, impressed above ; petiole, f- inch long, with scattered long hairs ; stipules caducous. Fruit : strobiles, 3 to 6 inches long, long-stalked ; bracts densely imbricated, membranous, i to i^ 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. japonica, 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, jovate-oblong, 3 inches long by if 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. 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,1 the variety chinensis 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 laxißora, Blume, Mus. Bot. Lugd. Bat. i. 309 (1850) ; Oliver, in Hooker, Icon. 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, Icon. Ess. Forest. Japon, text 48, t. 25, ff. 15-30 (1900); Winkler, Betulacecc, 33 (1904). Carpinus fargesü, 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), 2^ inches long by i^ 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, loi: cit. Carpinus 531 stipules caducous. Fruit : strobiles, up to 3 inches long ; bracts very loosely imbricated, about f 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.1 It has not yet, apparently, been introduced into cultivation. Plants at Kew, sent under the name of C. laxißora, from the Arnold Arboretum in 1895, are C. japonica. (A. H.) CARPINUS CAROLINIANA, AMERICAN HORNBEAM Carpinus caroliniana, Walter, Ft. Carol. 236 (1788); Sargent, Silva N. Amer. ix. 42, t. 447 (1896), and Trees N. Amer. 190 (1905); Winkler, fletulaceœ, 31 (1904). Carpinus americana, Michaux, Fl. Bor. Amer. ii. 201 (1803); Loudon. Art. et 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. Betuhts 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, f to i 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. Betulus 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, \ inch long, with glabrous ciliate scales ; those of C. Betulus being large, fusiform, £ to J 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 arc distinguished by Burkill, h(. cit. in M 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 l 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 not germinate. (A- H.) CARPI N US BETULUS, COMMON HORNBEAM Carpinus Betuhts, Linnaeus, Sp. PI. 998 (1753); Loudon, Arb. et Frut. Brit. iii. 2004 (1838); Willkomm, Forstliche Flora, 358 (1887); Mathieu, Flare Forestière, 396 (1897). Carpinus vulgaris, Miller, Gard. Diet. ed. 8, No. i (1768). Carpinus sepium, Lamarck, Fl. Franc, ii. 212 (1778). Carpinus compressa, Gilbert, Exerc. ii. 399 (1792). Carpinus ulmoides, Gray, Nat. Arrang. Brit. PI. ii. 2 ^5 (1821). Carpinus carpinizza, Host, Fl. Austr. ii. 626 (1831). Carpinus intermedia, Wierzbicki, in Reichenbach, Icon. Fl. Germ. xii. f. 1297 (1850). Carpinus nervata, Dulac, Fl. Haut. Pyré n. 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 if 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 \ to \ inch long, pubescent ; stipules narrow, lanceolate, \ inch long, caducous. Male catkins, about i^ inch long ; scales ovate, acute, entire, veined longi tudinally ; stamens, 4 to 12, with long yellow anthers. Female catkins, nearly i 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 i^ inch long, entire or minutely serrulate ; nutlet, l 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. tropicalis, 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, 5 to ^ inch long ; scales, ciliate and pubescent towards the tips, brownish. Seedling :1 Primary root tapering, wiry, flexuose ; caulicle terete, pubescent, ^- inch long ; cotyledons fleshy, rounded-obovate, 3- 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. carpinizza, 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,2 fastigiate, pendulous, and variegated forms have originated. In var. purpurea, the young leaves have a reddish tint. Var. incisa, Aiton,3 has leaves with large sharp serrated teeth. A wide-branching tree of this variety at Bcauport, Sussex, is 6 feet 3 inches in girth ; and there is also a fine specimen at Smeaton-Hepburn, East Lothian. In var. quercifolia, Desfontaines,4 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 reported5 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 recorded6 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,7 it is a very rare tree ; and Townsend8 considers it to be a doubtful native of Hamp shire. Druce9 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 Vie in France. Godron, Les Hêtres ToitUlards (1869). 3 Aiton, Hort. Kew, iii. 362 (1789). 4 Desfontaines, Tab. Écol. Bot. Mus. Hist. Nat. 212 (1824). 6 Rev. .T. F. Noott in letter to Kcw, March 1894. 6 Watson, Comp. Cybele Brit. 311 (1870) and Topo*. Bot. 355 (1873!. 7 Mansell-Pleydell, Flora of Dorsetshire, 246 (1895). 8 Flora of Hampshire, 313 (1883). 9 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 Hertfordshire1 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.2 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,3 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. H.) 1 Pryor, Flora of Hertfordshire, 373 (i»»7)- " c- Reid. Origin Brit. Flora, 144 (1899). 3 Pflanzern/erb. fCaukasusländ. 183 (1899). Carpinus 535 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 Estate 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 FREES 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 n 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 9 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 40 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 40 feet. Another, a pollard, hollow on one side, measured 13 feet 6 inches in girth. These grow on a 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 40 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 Carpinus 537 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 [891 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 17J 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 ID 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 10 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 to 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 l838 5° to 6o anfl one 73 53$ The Trees of Great Britain and Ireland I measured three from 70 to 80 feet with a girth of 6^ to 8^ feet, one being covered with dense tufts of twigs, a kind of witches' broom, caused by Exoascus Carpini. 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. Walker1 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. Ren wick sends me particulars of a very remarkable hornbeam at Douglas Support, in Lanarkshire, which, in 1900, measured 78 feet by 8 feet i 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.2 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. 9S,ßctt Loudon. 2 I am 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. lie 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. 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.) in N OSTRYA Ostrya, Scopoli, Fl. Carniol. 414 (1760); Bentham et Hooker, Gen. PI. iii. 406 (1880); Winkler, in Engler, Pflanzenreich, iv. 61, Betulacea, 24 (1904). Carpinus, Linnaeus, Gen. PI. 292 (exparte} (1737). SMALL deciduous trees, belonging to the order Betulaceae, 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 Knowltoni, 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 CARPI NI FOLIA, HOP HORNBEAM Ostrya carpinifolia, Scopoli, Fl. Carniol. ii. 244 (1772); Willkomm, Forstliche Flora, 368 (1887); Mathieu, Flore Forestière, 403 (1897). Ostrya vulgaris, Willdenow, Sp. PI. iv. 469 (1805); Loudon, Arb. et Frut. Brit. iii. 2015 (1838). Ostrya italica, Spach. Ann. Se. Nat. sér. 2, xvi. 246 (1841). Ostrya italica, sub-species carpinifolia, Winkler, Betulaceœ, 22 (1904). Ostrya Ostrya, Sargent, Silva N. Amer. ix. 32 (1896). Carpinus Ostrya, Linnœus, Sp. PI. 998 (1753). A tree attaining 6o feet in height and 10 feet in girth ; stem cylindrical, bark greyish, finely fissured, and scaly. Young branchlets with dense appressed 54° Ostrya 541 pubescence. Leaves (Plate 201, Fig. 11) about 3 inches long by if 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 § inch long, appressed pubescent ; stipules persistent during summer. Nutlet ovoid, \ 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, T3^ 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,1 in the Basses-Alpes, which is mainly composed of Pinus 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,2 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.2 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 Pardé8 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 Mouillefert4 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 Fliehe, Bull. Soc. Bot. France, xlvi. 8 (1899). Cf. also ibid. xxxv. 160 (1888). 2 Christ, Höre 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 miocène 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). * Principales Essences Forestières, 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 I metre above the ground ; and the Ostrya 11£ 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 2\ 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 Loudon1 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 i897,3 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 40 feet high by 2 feet 7 inches in girth. At Munden, Watford, a tree, 32 feet by 2 feet n 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,4 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 dimensions5 of the Ostrya 1 Op. cit. viïi. 244 a. 2 Card. 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. * Walker, Essays on Natural History and Rural Economy (1812). 6 Bargally is fully described by Loudon, op. cit. i. 95-99 (1838). Ostrya 543 from a letter of the then owner, J. Mackie, as 60 feet high and 4 feet i 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, IRONWOOD, AMERICAN HOP HORNBEAM Ostrya -virginica, Willdenow, Sp. PI. iv. 469 (1805); Loudon, Arb. et Frut. Brit. iii. 2015 (1838). Ostrya -virginiana, Koch, Dendrologie, ii. pt. ii. 6(1873); Sargent, Suva JV. Amer. ix. 34, t. 445 (1896), and Trees N. Amer. 192 (1905). Ostrya Ostrya, Macmillan, Metaspennœ Minnesota Valley, 187 (1892). Ostrya italica, sub-species virginiana, VVinkler, Eetulacece, 22 (1904). Carpinus Ostrya, Linnaeus, Sp. PI. 998 (1753) (in part). Carpinus virginiana, Miller, Diet. 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^r inches long, slightly cordate at the base, with fewer nerves, about twelve pairs. The nutlet in this species is larger, \ to £ 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,1 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 carpinifolia. (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 Carpimts caroliniana. It is known in America as 1 Ann. Sc. Nat. sér. 2, xvi. 246 (1841), and Hist. Wg. xi. 218 (1842). 544 The Trees of Great Britain and Ireland Iron wood, 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 carpinifolia. (H. J. E.) OSTRYA JAPONICA, JAPANESE HOP HORNBEAM Ostrya japonica, Sargent, Garden and Forest, vi. 383, f. 58 (1893), forest Flora Japan, 66, t. 22 (1894), and Suva N. Amer. ix. 32 (1896); Shirasawa, Icon. Ess. Forest. Japon, text 49, t. 25, ff. 1-14 (1900). Ostrya virginica, Maximowicz, Mil. Biol. xi. 317 (1881). Ostrya italica, sub-species vtrgitiiana, Winkler, Betulaceœ, 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, Ka'i, and Totomi, and also at Nikko ; and farther south, in the island of Shikoku. Ostrya iaponica 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 Père Farges and by myself. Ostrya mandschurica, Budischtschew,1 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. NOTHOFAGUS Nothofagus, Blume, Mus. Bot. Lugd. Bat. \. 307 (1850); Oerstedt, Vidensk. Selsk. Skrift. V. ix. 331 (1873); Solereder, System. Werth Holzstruclur, 253 (1885); Krasser, Ann. K.-K. Naturhist. Hofmuseums, Wien, xi. 149 (1896). Calucechinus and Calusparassus, Hombron et Jacquinot, Voy. Pôle. Sud. Atlas, tt. 6-8 (1853). Lophozonia, Turczaninow, Bull. Soc. Imp. Nat. Mosc. xxxi. 396 (1858). Fagus, section Nothofagus, Bentham et Hooker, Gen. PI. iii. 410 (1880). THIS genus comprises the beeches inhabiting extra-tropical South America, Australia, Tasmania, and New Zealand, and was formerly considered to be a section of the genus Fagus, which, however, as now limited, includes only the species of the northern hemisphere. The two genera are distinguished as follows :— NOTHOFAGUS.—Trees or shrubs, with deciduous or evergreen foliage.1 Flowers monœcious or rarely diœcious, either solitary or in groups of threes. Fruit : involucre, two-, three- or four-valved, usually bearing externally transverse entire, toothed or lobed lamellae, with or without gland-tipped processes ; or in rare cases the valves are smooth and without appendages ; nuts, solitary or three in each involucre. FAGUS.—Trees with deciduous foliage. Flowers monœcious ; the staminate numerous in globose heads, the pistillate in pairs. Fruit : involucre, covered externally with bristly, deltoid or foliaceous processes ; nuts, two in each involucre. About seventeen2 distinct species of Nothofagus are known, constituting three natural sections, based on the characters of the foliage :— I. Leaves deciduous, soft in texture, folded in bud along the lateral nerves, crenate or serrate in margin. 1. Nothofagus antarctica, Oerstedt. Large tree, S. America. Introduced into cultivation. Leaves ovate, f to i inch long ; lateral nerves three to five pairs ; margin slightly lobed, unequally crenate, with three to five teeth between the ends of each adjacent pair of nerves. 2. Nothofagus Montagnei? Reiche. Shrub or low tree. Chonos Archipelago. Not introduced. A little-known species, of which I have seen no specimen ; leaves \ inch long, firmer in texture and more conspicuously veined above than those of the preceding species, from which it is also distinguished by the yellow-coloured pubescence on the branchlets. 1 Bunbury, in Bot. Fragments, 322 (1883), writes an interesting article on the different types of foliage which are met with in this genus. * N. alpitta, Reiche (Fagus alfina, Poeppig et Endlicher), is a doubtful species. s Calucechinus Mcntagiiei, Hombron et Jacquinot, lac. cit. t. 7 (1853). Fagus Moiitagnei, Philippi, Linnaa, xxix. 45 (l8S7). 545 546 The Trees of Great Britain and Ireland 3. Notkofagus pumilio? Krasser. Shrub or low tree, S. America. Not introduced. Leaves ovate, i to i£ inch long ; lateral nerves five to seven pairs ; regularly bicrenate in margin, with two teeth between the ends of each adjacent pair of nerves. Young branchlets clothed with dense yellow pubescence. 4. Nothofagus obliqua, Blume. Large tree, S. America. Introduced. Leaves, i to 2^ inches long, ovate-oblong, glaucous beneath ; lateral nerves seven to eleven pairs ; lobulate and serrate in margin. 5. Nothofagus procera? Oerstedt. Large tree, S. America. Not introduced. Leaves, 2 to 3^ inches long, oblong ; lateral nerves fifteen to sixteen pairs ; doubly serrate in margin. 6. Nothofagiis Gunnii? Oerstedt. Shrub, Tasmania. Not introduced. Leaves, ^ inch long, ovate ; lateral nerves six pairs, prominent beneath ; margin regularly crenate, with one tooth between the ends of each adjacent pair of nerves. II. Leaves evergreen, margin not entire. In this section, the leaves are usually very coriaceous, and glabrous, or with only slight pubescence. A. Leaves minutely crenulate in margin. 7. Nothofagus apicttlata? Krasser. Tree attaining 40 feet in height, New Zealand. Not introduced. Leaves i inch long, elliptical, conspicuously mucronate at the apex. B. Leaves crenate in margin. 8. Nothofagus Menziesii, Oerstedt. Large tree, New Zealand. Introduced. Leaves, ^ inch long, ovate, rhomboid or orbicular, doubly crenate, with two pubescent pits on each side of the midrib near the base on the lower surface. 9. Nothofagus Ciinninghami, Oerstedt. Large tree, Australia, Tasmania. Introduced. Leaves, \ inch long, ovate, deltoid or rhomboid, simply crenate ; lower surface without pits or resinous papillae. ID. Nothofagus betuloides, Blume. Large tree, S. America. Introduced. Leaves, J to i inch long, ovate, crenate or obtusely dentate ; lower surface dotted with resinous papillae. C. Leaves serrate in margin. \ i. Nothofagiis Dombeyi? Blume. Large tree, S. America. Not introduced. Leaves, i inch long, lanceolate ; nerves six pairs ; lower surface dotted with resinous papillae. 12. Nothofagiis nitida? Krasser. Large tree, S. America. Not introduced. Leaves, i to i^ inch long, trapezoid-ovate, acuminate ; nerves six pairs ; lower surface not dotted. 13. Nothofagus fusca, Oerstedt. Large tree, New Zealand. Introduced. 1 Fagus pumilio, Poeppig et Endlicher, Nov. Gen. et Sf. ii. 68, t. 195 (1835). Fagas antarctica, var. bicretiata, DC. Prod. xvi. 2, p. 120 (1864). 2 Fagus procera, Poeppig et Endlicher, Nov. Gen. et Sf. ii. 69, t. 197 (1835). 3 Fagus Gunnii, Hooker, Icon. Plant, t. 88l (1852). 4 Fagus apiculata, Colenso, Trans. AT. Zeal. Instil, xvi. 335 (1884). 6 Fagus Dombtyi, Mirbel, Mém. Mus. Paris, xiv. 467, t. 24 (1827). • Fagus nitida, Philippi, Linnxa, \xix. 44 (1857). Nothofagus 547 Leaves, i to \\ inch long, thin in texture, ovate, rounded at the apex ; serrations large, few, irregular ; nerves four to seven pairs. 14. Nothofagus Moorei, Krasser. Large tree, Australia. Introduced. Leaves, 2 to 3 inches long, ovate-lanceolate, acuminate ; sharply and finely serrate ; nerves ten to fifteen pairs. III. Leaves evergreen, entire in margin. In this section, the leaves on young trees are glabrous ; but on older trees they become densely tomentose on the under surface. They resemble considerably the leaves of certain species of Vaccinium. 15. Nothofagus cliffortioides, Oerstedt. Large tree, New Zealand. Introduced. Leaves, \ to ^ inch long, ovate, rounded at the base ; tomentum whitish. 16. Notliofagus Solandri,1 Oerstedt. Large tree, New Zealand. Not intro duced. Leaves ^ to \ inch long, oblong, cuneate at the base ; tomentum whitish. 17. Nothofagus Blairii? Krasser. Large tree, New Zealand. Not introduced. Leaves f inch long, ovate, rounded at the base, apiculate at the apex ; tomentum yellowish. (A. H.) NOTHOFAGUS CLIFFORTIOIDES Nothofagus diffortioides, Oerstedt, Vidensk. Selsk. Shrift. V. ix. 355 (1873). Fagus diffortioides, J. D. Hooker, in Hook. Icon. Plant, tt. 673 (1844) and SIÖB (1852), Flora Neiv Zealand, i. 230 (1854), and Handb. New Zealand Flora, 250 (1864); Kirk, Forest Flora New Zealand, 201, tt. 101, 101 A (1889); Cheeseman, New Zealand Flora, 643 (1906). An evergreen tree, attaining, in New Zealand, about 50 feet in height and 6 feet in girth. Young branchlets pubescent ; buds minute, ovoid, shining, brown. Leaves (Plate 202, Fig. 4) persistent for two or three years, distichous and crowded on the branchlets, coriaceous, minute, \ to \ inch in length, entire in margin ; on young plants ovate, rounded at both base and apex, green and glabrous on both surfaces ; on adult trees, ovate or ovate-oblong, rounded and unequal at the base, subacute at the apex, minutely punctate above, greyish-white with dense appressed pubescence beneath ; petioles short and pubescent. Male flowers solitary ; stamens eight to twelve. Fruit : involucre \ to \ inch long, three-lobed, each lobe with two or three entire transverse lamellae ; nuts one to three, winged, one or two triquetrous, the third flattened. This tree is known in New Zealand as the "mountain birch," and is confined to mountainous regions except in the south-western corner of the South Island, where it descends to sea-level. It is not found in the northern part of the North Island ; but elsewhere is very common in the forests3 between 2000 and 4000 feet elevation, 1 Fagus Solandri, Hooker, Icon. Plant, t. 639 (1844). 2 Fagus Blairii, Kirk, Trans. N. Zeal. Inst. xvii. 297, 306 (1885). 3 A view of a forest of this species in the South Island at 3000 feet, showing a dense undergrowth of young beech and tall smooth stems of older trees, is given in Schimper, Plant Geography, 760, f. 460 (1904). Ill O 548 The Trees of Great Britain and Ireland often forming the timber line, when it becomes a mere bush. The wood, though of no great size, is used for telegraph-poles, fencing-posts, railway sleepers, and wharf- piles, the heartwood being very durable in situations where it is exposed to alter nations of dryness and moisture. N. cliffortioides is extremely rare in cultivation. There are two specimens at Enys, Cornwall, which, according to Mr. John D. Enys, were 35 feet and 28^ feet high respectively in 1905 ; but when Elwes saw them in that year they were very slender and not thriving. These trees are semi-deciduous, most of the leaves, after turning brilliant red in autumn, falling off during winter ; whereas, in New Zealand, the foliage is strictly evergreen. Another tree is growing at Messrs. Veitch's nursery at Coombe Wood, where it stands out of doors without any protection. It is very slow, however, in growth, and is only about 12 feet in height. (A. H.) NOTHOFAGUS MENZIESII Nothofagus Menziesii, Oerstedt, Vidensk. Selsk. Skr ift. V. ix. 355 (1873). Fagus Menziesii, J. D. Hooker, Icon. Plant, t. 652 (1844), and Flora New Zealand, i. 229 (1854); Kirk, Forest Flora New Zealand, 175, t. 89 (1889)5 Cheeseman, New Zealand Flora, 640 (1906). An evergreen tree, attaining in New Zealand 100 feet in height and 15 to 25 feet in girth. Bark silvery-white, resembling that of the common English birch. Young branchlets covered with dense erect brown pubescence. Leaves (Plate 202, Fig. 9) persistent for two or three years, distichous on the branchlets, about \ inch long, coriaceous, deltoid, ovate or rhombic ; cuneate at the base, obtuse at the apex ; glabrous ; upper surface dark-green, shining ; lower surface pale-green, with usually two (occasionally only one or none) small pits fringed with brownish hairs near the base of the midrib ; lateral veins about three pairs ; margin irregularly and doubly crenate ; petioles short, pubescent. Male flowers solitary ; calyx four- to six-lobed ; stamens six to twelve. Fruit : involucre % to ^ inch long, cleft into four narrow lobes, each with five transverse scales, cut to the base into recurved linear gland- tipped processes ; nuts three, one two-winged, two three-winged, the wings produced upwards into sharp points. This species, which is known in New Zealand as the "silver birch "or "red birch," is common in the mountain forests of both the North and South Islands, ascending from sea-level to 3500 feet. The wood is dark-red, strong, and compact, and being easily worked, is suitable for making furniture. A small tree of this species is growing in the Temperate House at Kew ; and we are not aware that it has ever been tried in the open air. The tree has handsome foliage, and should be hardy in Cornwall and the south of Ireland. (A. H.) Nothofagus 549 NOTHOFAGUS FUSCA Nothofagus fusca, Oerstedt, Vidensk. Selsk. Skrift. V. ix. 355 (1873). Fagus fusca, J. D. Hooker, Icon. Plant, tt. 630, 631 (1844), and Flora New Zealand, i. 229 (1854); Kirk, Forest Flora New Zealand, 179, t. 90 (1889); Cheeseman, New Zealand Flora, 641 (1906). An evergeen tree, attaining in New Zealand 100 feet in height and 25 feet in girth. Bark of young trees white and smooth, becoming on old trees furrowed longitudinally and brown in colour. Young branchlets minutely pubescent. Leaves (Plate 202, Fig. 8) persistent for two years, distichous on the branchlets, f to i\ inch long, thin in texture, glabrous, broadly ovate, cuneate at the base, rounded at the apex, dark-green and shining above, pale-green beneath ; sharply serrate with a few large teeth in the upper two-thirds of the leaf ; lateral nerves four to five pairs ; petioles about -^ inch, pubescent. Male flowers solitary or in threes; calyx five- toothed ; stamens eight to sixteen. Fruit : involucre nearly ^ inch long, viscid- pubescent, four-lobed, each lobe with three to five transverse entire or fringed scales ; nuts three, as in N. Menziesii. This species is a native of New Zealand, where it is known as the " black birch " or "red birch." It grows in forests at elevations between sea-level and 3500 feet, being a splendid tree. Its distribution is North Island from Monguni and Kartaia southwards, but local to the north of the East Cape ; South Island from Nelson to Foveaux Straits, but rare in Canterbury and Eastern Otago. The wood is dark-red, strong and compact, and more durable than that of the other species ; it is frequently used for wharves, bridges, and fencing posts. A small tree of this species about 10 feet high, and said to be thirty years old, is growing in the Coombe Wood Nursery. At Castlewellan, Co. Down, there is a tree1 about 18 feet high, which was imported from New Zealand some years ago as a small plant in a Wardian case. It is growing rapidly, making an annual growth of a foot. The old leaves remain on the branches till the new ones appear, changing before they fall to a brilliant red, which contrasts well with the light green of the young growths. (A. H.) NOTHOFAGUS MOOREI Nothofagus Moorei, Krasser, Ann. K.-K. Naturhist. Hofmuseums, Wien, xi. 163 (1896). Fagus Moorei, F. v. Mueller, Fragm. Phyt. Austral, v. 109 (1865); Bentham, Fl. Australiensis, vi. 211 (1873). An evergreen tree, attaining in Australia 150 feet in height. Young branchlets pubescent ; buds ovoid, acute, brown. Leaves (Plate 202, Fig. 7) persistent for two or three years, distichous on the branchlets, coriaceous, glabrous; shining, dark- 1 This is figured as Fagus cliffortioides in Earl Annesley's Beautiful and fiare Trees, 71 (1903). 550 The Trees of Great Britain and Ireland green above ; pale-green beneath ; ovate or ovate-lanceolate ; 2 to 3 inches long on barren branchlets, about i inch long on flowering shoots ; unequally cuneate at the base, acuminate at the apex, finely and sharply serrate ; lateral nerves 8 to 12 pairs, prominent on the upper surface ; petioles very short, pubescent. Flowers unknown. Fruit : involucre about f inch long, four-lobed ; lobes lanceolate, acute with pubes cent scales terminating in glandular processes ; nuts three, two three-winged, the other two-winged. This tree was discovered by Mr. C. Moore, Curator of the Botanic Gardens, Sydney, in New South Wales, and is the only southern beech occurring in a sub tropical region. It forms dense forests at the head of Bellinger River and Bealsdown Creek, at about 4000 feet altitude ; and a few trees have also been seen near the source of the Macleay River. It was introduced into cultivation at Kew about fifteen years ago, and there is a small tree now growing there in the Temperate House. The only specimen living in the open air, so far as we know, is growing in the garden of Mr. Thomas Acton at Kilmacurragh, Co. Wicklow. It was 18 feet high in 1906, and had bark resembling that of Prumis avi^^m. (A. H.) NOTHOFAGUS CUNNINGHAMI Nothofagus Cunninghami, Oerstedt, Vidensk. Selsk. Skrift, V. ix. 355 (1873). Fagus Cunninghami, J. D. Hooker, Journ. Bot. ii. 152, t. 7 (1840), and Flora Tasmania, i. 346 (1860); F. v. Mueller, Fragm. Phyt. Austral, v. no (1865); Bentham, Flora Australiensis, vi. BIO (1873); Rodway, Tasmania ft Flora, 182 (1903). An evergreen tree, said to attain in Tasmania 200 feet in height and 40 feet in girth. Bark, as seen in cultivated trees, roughened by small scales and fissuring longitudinally. Young branchlets densely and minutely pubescent. Buds conical, sharp-pointed and curved at the apex, shining, brown. Leaves (Plate 202, Fig. 5) persistent for two or three years, distichous and crowded on the branchlets, coriaceous, about g- inch in length ; broadly ovate, deltoid or rhombic ; cuneate or cordate at the base, acute at the apex ; unequally crenate in margin ; both surfaces glabrous, veins inconspicuous and scarcely prominent beneath ; petioles short, pubescent. Male flowers solitary ; stamens eight. Fruit : involucre \ inch long, four-lobed, each lobe with five or six rows of dorsal transverse scales, split up into gland-tipped processes ; nuts usually three, two lateral triquetrous and three-winged, the other flattened and two-winged. This species is very common in Tasmania, where it is known as the " Tasmanian myrtle," and forms a large proportion of the forests in the mountainous and western humid districts. It ascends to 4000 feet, becoming at this elevation a mere shrub, a few feet in height. It also occurs on the mainland of Australia, in Victoria, in a few scattered localities, being most common according to F. v. Mueller on the Baw- baw mountains, and less common at Dandenong, Mount Juliet, Wilson's Promontory, Nothofagus 551 La Trobe, Tyre's and Thomson's Rivers, and in a few places in the cooler and moister parts of Gipp's Land. The woodl apparently varies : one kind, " red myrtle," being of a bright pink colour, with a grain like that of the English beech, and con sidered to be suitable for cabinetwork ; another kind, " white myrtle," brownish- grey in tint, is not so attractive in appearance. N. Cunninghami is very rare in cultivation. The finest tree (Plate 154), said by Lord Barrymore to have been planted about fifty years ago, is growing at Fota, Co. Cork, and measured, in 1904, 48 feet high by 3 feet 3 inches in girth. This tree has numerous branches, many of them ascending from near the base of the trunk. A tree at Kilmacurragh, Co. Wicklow, with branches ascending and curving at the tips, was 40 feet high by 3 feet 4 inches in 1906. This tree has excrescences on the trunk, similar to the so-called "wood-balls," which are often seen on the common beech. It flowered in 1906. At Osborne, Isle of Wight, there is a tree, 30 feet by 2 feet 2 inches, which when Elwes saw it in 1906 seemed thriving. It seems to be as hardy as any of the genus, and might be planted with good prospects of success in the extreme south-west of England near the sea. (A. H.) NOTHOFAGUS BETULOIDES Nothofagus betuloides, Blume, Mus. Bot. Lugd. Bat. i. 307 (1850); Reiche, Chil. Buch. 15 (1897); Wildeman, Voy. Belgica, 74 (1905); Macloskie, Princeton Univ. Expedit. Patagonia, Botany, 329 (1903-6). Fagus betuloides, Mirbel, Mêm. Mus. Pans, xiv. 465, t. 4 (1827); Loudon, Arb. et Frut. Brit. iii. 1982 (1838); Hooker, Journ. Bot. ii. 155 (1840), and Fl. Antarct. ii. 349, t. 124 (1847). Fagus Forsteri, Hooker, Jourti. Bot. ii. 156, t. 8 (1840). Hetula antarctica, Forster, Comm. Goett. ix. 45 (1789). Calusparassus betuloides and C. Forsteri, Hombron et Jacquinot, Voy. Pôle Sud, Atlas t. 7 (1853). A large evergreen tree. Bark smooth, grey ; scaling near the base in old trees. Young branchlets slender, viscid, covered with short pubescence. Buds minute, brown, ovoid. Leaves (Plate 202, Fig. 3) persistent for two or three years, distichous and crowded on the branchlets, rigid, coriaceous, f to i inch long by \ inch or slightly more in breadth, ovate, rounded at the base, acute at the apex, crenate or serrate in margin ; upper surface shining, glabrous, often viscid ; lower surface finely reticulate, glabrous, dotted with resinous papillae; petioles short. Male flowers solitary ; calyx funnel-shaped, four- to seven-lobed ; stamens ten to sixteen, with long and slender filaments. Fruit : involucre four-lobed, with erect filiform glandular processes. According to Loudon, both N. betuloides and N. antarctica were introduced in 1830, but he had not seen a specimen of either. Sir W. J. Hooker2 states that healthy young trees of both species, the first, as far as he knew, that ever had reached Europe, were sent in Wardian cases to Kew from Cape Horn in 1843, being 1 Report on Tasmanian Timbers by Mr. R. A. Kansome, of the Stanley Works, Chelsea, in Kew Bull. 1889, pp. 114, 115. 2 Notes Bet. Antarctic Voyage, 64 (1843) > cf- also Loudon, Card. Mag. 1843, p. 442. 552 The Trees of Great Britain and Ireland a consignment from his son, who was attached as botanist to the Antarctic expedition of the Erebiis and Terror. (A. H.) The largest and finest specimen which I have seen in cultivation is at Bicton (Plate 155). It measured in 1906 about 50 feet in height and 6| feet in girth: but bore no fruit on two occasions when I saw it. At Pencarrow, Cornwall, a tree,1 reported by Mr. Bartlett to have been obtained from Messrs. Veitch in 1847, was 36 feet by 4 feet 3 inches in 1903. One at Coldrinick in the same county was measured in 1905 by Mr. Bartlett, who gives its dimensions as 45 feet by 5 feet 5 inches. There is another specimen,2 about sixty years old, growing on Sir John Llewellyn's property at Caswell Bay, near Swansea, which he tells me measured 25 feet by 3 feet 2 inches in 1907. It is close to the sea and in con sequence has been shorn off by the sea wind to the same height as the Portugal laurels and poplars which grow beside it. At Grayswood, Haslemere, at 600 feet elevation, a tree, said by Mr. B. C. Chambers to have been planted in 1882, measured 34 feet by 2 feet 3 inches in 1906. At Hafodunos, Denbighshire, a tree, reported by Col. Sandbach to have been planted in 1855, was in 1904 36 feet high by 5 feet 2 inches in girth at 3 feet from the ground, dividing at 5 feet up into two stems. There is no tree growing now at Kew, one, a healthy specimen, having been killed3 by frost in January 1867. There is also a tree growing at Ashridge Park, Herts, which is about 30 feet by 3 feet, on which Miss Woolward has observed fruit, and a smaller one is in the Knap Hill Nursery, near Woking. A tree at Powerscourt, Co. Wicklow, was, in 1906, 33 feet high by 2 feet 10 inches in girth. I have a sample board from a tree of this species, which grew on the rockery at Lucombe and Pince's Nursery, Exeter. This was cut down when the nursery was cleared for building in March 1903. In 1886, it was 35 feet high and 2 feet 8 inches in girth at 3 feet from the ground,4 and had not grown much in the succeeding years. The timber was of poor quality, and had begun to decay. (H. J. E.) NOTHOFAGUS OBLIQUA Nothofagus obliqua, Blume, Mus. Bot. Lugd. Bat. i. 307 (1850); Reiche, Chil. Buch. 8 (1897); Wildeman, Voy. Belgica, 75 (1905). Fagus obliqua, Mirbel, Mein. Mus. Parts, xiv. 465, t. 23 (1827); Hooker, Jiourn. Bot. ii. 153 (1840). Fagusglauca, Philippi, Linnœa, xxix. 43 (1857). Lophozonia heterocarpa, Turczaninow, Bull. Soc. Imp. Nat. Mosc. xxxi. 396 (1858). A deciduous tree, attaining in Chile a height of over ioo feet. Bark, according to Reiche, dark in colour and fissured. Young branchlets glabrous ; buds small, 1 Figured in Card. Chron. xxxiii. 10, f. 5 (1903), where it is erroneously stated to have been introduced from New Zealand. 3 Figured in Card. Chron. 1872, p. 466, f. 136, and 1886, xxv. 104, f. 18. 3 J. Smith, Records of Kew Gardens, 277 (1880). 4 Gard. Chron. xxv. 104 (1886). Nothofagus 553 conical, sharp-pointed, glabrous, brown, few-scaled, and appressed to the branchlets. Leaves (Plate 202, Fig. 2) variable in size, i to 2^- inches in length, ^ to i inch in breadth, thin in texture, ovate-oblong, unequal at the rounded or cuneate base, sub-acute or obtuse at the apex ; dark-green above, very pale beneath, both surfaces glabrous except for slight pubescence on the midrib and nerves ; margin shallowly lobulate in the lower half, the lobules and upper part of the leaf serrate with minute triangular acute teeth ; nerves 8 to 11 pairs, prominent on the lower surface, running obliquely to the margin ; petiole ^ to ^ inch long. Male flowers solitary ; calyx irregularly lobed, stamens thirty to forty. Fruit : involucre four-valved, valves pubescent on the back with lobed appendages bearing stalked glands ; nuts three, two trigonous and three-winged, one flattened and two-winged. This species is very variable, especially as regards the size and pubescence of the leaf, and De Candollel distinguished three varieties :—Var. valdiviana : leaves small, glabrous, with cuneate base ; var. macranthera : stamens long, leaves pubes cent beneath ; and var. niacrocarpa, with the nuts longer than the valves of the involucre. N. obliqiia was introduced2 into England by Lobb in 1849, and in the following year it was said to have been growing freely in the open air in Messrs. Veitch's nursery at Exeter. None of the original plants appear, however, to have survived. Plants raised from seed, brought from Chile by Elwes in 1902, have grown with great vigour at Kew,3 being now about 8 feet in height. At Monreith, Sir Herbert Maxwell, who received a plant from Kew, reports that it has borne without injury 20° of frost, and may be assumed to be perfectly hardy. In Lord Ducie's garden at Tortworth, this tree has grown with astonishing vigour, being now 12 feet high and 8 inches in girth ; it endured the severe frost of May 1905 without any apparent injury. The seedlings which were raised at Colesborne, however, never throve, and died before attaining any size, which is possibly due to the presence of lime in the soil. (A. H.) NOTHOFAGUS ANTARCTICA Nothofagus antarctica, Oerstedt, Vidensk. Selsk. SAnfi.V. ix. 354 (1873); Reiche, CM. Buch, n (1897); Wildeman, Voy. Belgica, 73 (1904); Macloskie, Princeton Univ. Exfed, Patagonia, Botany, 326 (1903-1906). Fagus antarctica, Foster, Comm. Goett. ix. 24 (1789); Loudon, Arb. et Frut. Brit. iii. 1982 (1838)5 Hooker, Jmtrn. Bot. ii. 149, t. 6 (1840), and Fl. Antarct. ii. 345, t. 123 (1847). Cahicechinus antarctica, Hombron et Jacquinot, Voy. Pôle Sud, Atlas, tt. 6, 7 (1853). A deciduous tree, attaining in Terra del Fuego at low elevations a very large size. Young branchlets covered with dense erect pubescence, persistent in the second year. Buds, \ inch long, ovoid, slightly compressed, glabrous, few-scaled. Leaves (Plate 202, Fig. i) £ to i inch long, crumpled and uneven in surface, oblong- 1 Prod. xvi. 2, p. 119 (1864). 2 Card. Chron. 1849, p. 563 ; Lindley, Journ. Hort. Soc. vi. 265 (1851) ; Lindlcy and Paxton, Flower Garden, ii. 166 (1852). 3 Cf. Kew Bull. 1906, p. 379. 554 The Trees of Great Britain and Ireland ovate, unequal and usually cordate at the base, rounded at the apex, variable in pubescence, dark - green above, light - green beneath ; nerves usually four pairs ; margin with three or four pairs of shallow lobes, which are minutely and irregularly dentate, the teeth being rounded or acute ; petiole i to ^ inch, pubescent. Male flowers solitary; calyx five-partite; stamens ten, as long as the calyx. Fruit: involucre four-partite, each lobe with three to four transverse reddish scales ; nuts three, the central one two-winged, the lateral pair three-winged. Two distinct forms occur :— 1. Var. sublobata, DC.1—Petiole and upper surface of the leaf glabrescent ; lower surface glabrous except on the nerves, which are clothed with long appressed hairs. 2. Var. îtliginosa, DC.1—Leaves pubescent on both surfaces with minute erect hairs. This species was introduced in 1843, as mentioned in our account of N. betuloides, but it is doubtful if any of the original plants are still living. The only specimen which we have discovered is a bushy tree, about 15 feet high, which is growing alongside a fine tree of N. betuloides at Hafodunos, Denbighshire. Colonel Sandbach believes it to be about thirty years old. Plants raised from seed, collected by Elwes in Chile in 1902, are now in cultiva tion at Kew,2 in a peat-bed, and have attained about 6 feet in height. They are vigorous in growth, and have passed through the severe frosts of 1906-1907 without injury, and look as if they might grow to be trees of considerable size. (A. H.) DISTRIBUTION OF THE SOUTH AMERICAN BEECHES In extra-tropical South America, the beeches are the dominant trees, extending from a point on the west coast of Chile about lat. 33°, southward to Patagonia and Tierra del Fuego, and crossing the Andes into Argentina. The best account of the Chilean beeches is given by Reiche,8 from whom, supplemented by my own observations in 1901-1902, I take the following particulars. The most northerly species is N. obliqiia, which extends on the coast up to about lat. 33\ but in the extreme north does not form forest except in the interior valleys. About lat. 35" it is the principal tree in the forest which formerly clothed the lower slopes of the Andes, but which is now fast vanishing before the attacks of man. The tree is called Roble Pellin by the Spaniards, and grows to a large size with a tall straight trunk, attaining a height of 120 feet or more, and a girth of 20 to 30 feet. In the forest country, which commences south of the Maule River, it is mixed with N. Dombeyi ; and these two species form the principal timber supplies of Chile, and are largely cut for house-building, railway sleepers, and other purposes. Some cargoes of this timber have lately been imported into England, under the name of Chilean Oak ; and by the courtesy of the Great Western Railway Company, I have received one of these sleepers, which has a dense reddish wood, not at all resembling 1 Prod. xvi. 2, 120 (1864). De Candolle's var. bicrenata is Nothofagus fitmilio. 2 Cf. Kew Bull. 1906, p. 381. 3 Beiträge Kenntnis* Chilen. Buchen (Valparaiso, 1897). Nothofagus 555 that of the European beech, and apparently suitable to take the place of the lower grades of mahogany in the manufacture of furniture. It cracks, however, badly in drying, and will require very careful seasoning. In the forests of Chilian (lat. 36°), JV. obliqiia grows up to 4000 or 5000 feet, being replaced at higher elevations by N. antarctica and N. pumilio ; and reaches its southern limit in the region of Lake Llanquihue. A photograph by Mr. Bartlett Calvert, who accompanied me on my journey, shows the appearance of this tree in the forest at about 3000 feet, near the source of the Renaico River (Plate 156). Here the undergrowth is usually composed of a dense thicket of the Chilean bamboo (Chusqiiea sp.} ; and in the more open places the ground is often carpeted with a dense bed of Alstrœmeria aurantiaca, whose brilliant orange flowers produce a most lovely effect. In the wetter places it is associated with Drimys Winteri and many beautiful shrubs and herbaceous plants, of which Eiicryphiapinnatißda, Embothrium coccineum, Tropœohim speciosum, and several species of Fuchsia and Calceolaria are the choicest ornaments of our gardens in the warmer and damper parts of the south-west of England and Ireland. In many parts of the Chilean forests it is often covered with a lovely parasitic plant, Myzodendron linearifolium, DC.,1 which hangs in silvery masses from the branches. Nothofag^ls Dombeyi is known to the Chileans by its Indian name of Coigue, and is a large and common tree in Chile. It is usually associated with N. obliqua, but does not extend so far to the northward, not being found to the north of the Maule river. It is widely spread in Araucania, Valdivia, and Llanquihue, and occurs also on the Argentine side of the frontier. It grows on the island of Chiloe, and has been collected on the river Aysen (lat. 45°) ; but its extreme southern limit is not accurately known. Nothofagus nitida, which has been much confused with N. Dombeyi, is a common forest tree in the coast mountains of Valdivia, and grows on Chiloe and the Guaitecas Islands. The distribution of this species has not yet been satisfactorily determined. Nothofagus procera, known as Rauli, is less common than N. obliqua, to which it is allied, and usually grows scattered in the forest. Its northerly limit lies between 35° and 36° lat., and it does not occur farther south than the province of Valdivia, where it becomes a stately tree. It does not cross the frontier into Argentina. Nothofagus antarctica is widely distributed, extending from about lat. 38° to Tierra del Fuego. It is the commonest species which I found on my tour at high elevations, both on the Chilean and Argentine sides of the frontier. It is associated with Araucaria at 4000 feet, and is common also in the plain of Valdivia in marshy situations. In the mountains around the great lake of Nahuelhuapi, the leaves of this species had already assumed their autumnal tint in February. N. antarctica and N. betiiloides are the dominant trees in Patagonia and Tierra del Fuego ; and 1 I found this species in the low country about Temuco in Chile, and also on the Argentine side of the frontier in two or three localities. Two other species also occur :—j\f. oblongifolium, DC., which I found on Nothofagus antarctica, near the baths of Chilian at 5000 to 7000 feet elevation ; and M. punctulatiiin, DC., which I gathered on Nothofagus Dombeyi at Lake Meliquina, and in the dense evergreen forest which skirts the glaciers of the great Tronador mountain at 2000 feet in lat. 40°. Ill P 556 The Trees of Great Britain and Ireland according to Dusen,1 their distribution is regulated by the amount of rainfall. In the western parts of Tierra del Fuego, where the rainfall is heavy, the coast forest is evergreen and is mainly composed of N. behiloides and Drimys Winteri; and N. antarctica is only met with in the mountains. In the eastern part of Tierra del Fuego, where the rainfall is slight, the latter species descends to sea-level and grows in mixture with N. betuloidcs and N. pumilio. In Western Patagonia, the evergreen forest predominates in the Archipelago and on the western side of the mountain range, where much rain falls and the prevailing winds are south-westerly ; whereas, on the eastern slopes of the mountains, where the climate is comparatively dry, the forests are composed of deciduous trees. According to Dusen, the deciduous-leaved forest is well seen at a point 30 miles up the River Aysen. In the inland region the ground is covered by a thin park-like forest, which is almost exclusively composed of one species, N. antarctica. This tree does not grow in such close masses as the European beech, and, owing to the absence of dense shade, there is a luxuriant undergrowth of herbs and shrubs. These park-like forests prevail up to 2300 feet. Above this elevation steppes occur, which are studded with small groves of N. pumilio, the ground being covered with mosses. At 3000 feet N. pumilio is only a low tree, which gradually becomes smaller as it ascends, until at 4300 feet it forms a stunted forest of dwarfed trees, with their branches interlaced together. An earlier account of the Antarctic beeches is given by Sir J. D. Hooker,2 who states that N. antarctica strongly resembles the European beech in its deciduous leaves, form of trunk, and smooth bark. It ascends much higher at Cape Horn than N. betuloides, and is much the larger tree of the two when it is found growing at sea-level. N. betuloides, however, grows to a very large size about the Straits of Magellan, and being evergreen, is a marked feature of the scenery in winter, as its upper limit is sharply defined, and contrasts with the dazzling snow that covers the matted and naked branches of N. antarctica. Captain King3 observed many trees of N. betuloides 3 to 4 feet in diameter, one being as large as 7 feet. He describes the wood as heavy and far too brittle for masts or even boat-hooks, but cutting up into tolerable planks. Hooker considered the timber of the deciduous species to be superior. N. betriloides, while much commoner in the south, extends along the coast range as far north as Valdivia.4 It is replaced in the Guaitecas Islands by N. Dombeyi and N. nitida.6 Nothofagus pumilio has been much confused with N. antarctica, of which it was made var. bicrenata by De Candolle. It is very distinct in both foliage and fruit. It extends from Chilian and Nahuelbuta in Chile southward to the Straits of Magellan, and is usually a shrub, constituting the scrubby growth which prevails 1 Princeton Univ. Exped. Patagonia, Botany 2, 10, 26 (1903-1906) ; and Engler, Bot. Jahrbuch, xxiv. 179 (1897). 2 Fl. Antarct. ii. 345. 3 Voyage of the "Adventure" ami "Beagle," i. 576 (1839). Ball, Notes of a Naturalist in S. America, 225, says that N. betuloides " has a thick trunk, commonly three or four feet in diameter, but nowhere attains any great height. Forty feet appeared to me the outside limit attained by any that I saw at Eden Harbour or elsewhere." 4 Reiche, he. cit. <> Dusén, loc. cit. Nothofagus 557 above timber-line in the mountains. I saw this species near the baths of Chilian, lat. 37°, where it grew as a bush at 6000 to 8000 feet on the crests of the ridges in volcanic soil. According to Reiche, it occurs as undergrowth in the Araucaria forests at 4000 to 5000 feet elevation ; but in sheltered situations in the mountains of Valdivia and Llanquihue it occasionally becomes a tree 60 feet in height. (H. J. E.) ARBUTUS Arbutus, Linnœus, Gen. PI. 123 (1737); Bentham et Hooker, Gen. PI. ii. 581 (1876). Unedo, Hoffmannsegg et Link, Fl. Port. i. 415 (1809). EVERGREEN trees or shrubs, belonging to the order Ericaceae. Leaves simple, alternate, spirally arranged on the branchlets, coriaceous, persistent, stalked, pinnately-veined, entire or serrate, without stipules. Buds with spirally imbricated scales, within which the young leaves lie flat and are not rolled or folded. Flowers perfect, regular, in terminal compound racemes or panicles. Pedicel with two bracteoles, in the axil of an ovate bract ; bracts and bracteoles scarious, persistent. Calyx five-lobed, free, persistent, unaltered at the base of the fruit. Corolla gamopetalous, hypogynous, urceolate or globose, with five obtuse, recurved, imbricated teeth. Stamens ten, included ; filaments free, inserted on the base of the corolla, dilated and pilose at the base ; anthers deflexed, dorsifixed, two-celled, opening by two pores, each anther with two awns on the back, against which insects knock in their search for honey and scatter the pollen through the pores. Pollen- grains united in tetrahedral masses of four grains each. Disc annular. Ovary superior, five- or occasionally four-celled ; style columnar, stigmatose and obscurely five-lobed at the apex ; ovules numerous. Fruit a berry or drupe, the endocarp often being imperfectly developed. Seeds numerous, small, angled, with a coriaceous testa and a horny albumen. About twenty species are known, inhabiting the western and south-western parts of North America, Central America, Ireland, the countries in Europe bordering upon the Mediterranean, the Canary Islands, Morocco, Algeria, Asia Minor, the Crimea, and the Caucasus. Many of the species are only shrubs or very small trees, and others are not hardy or have not been introduced. Only four species1 attaining a considerable size in cultivation in the open air in England, one of which is a hybrid, will be dealt with :— A. Leaves serrate. Young branchlets glandular-pubescent. i. Arbutus Unedo, Linnaeus. Ireland, Southern Europe, Asia Minor, Morocco, and Algeria. Leaves green beneath ; petiole \ inch. Older branchlets dark brown, rough, and fissuring. 1 Arbutus canariensis, Lamarck, growing in the open air, is five feet high at Mount Usher in Wicklow ; but at Newry this species requires protection in winter. Artutus arizonica, Sargent, a native of the high mountains of Southern Arizona, if introduced, might be hardy. 558 Arbutus 559 2. Arbutus hybrida, Ker-Gawler. A hybrid. Leaves slightly glaucous beneath ; petiole f inch. Older branchlets fawn- coloured, smooth. B. Leaves entire. Young branchlets glabroiis. 3. Arbîdus Andrachne, Linnaeus. Albania, Greece, Asia Minor, Crimea, Caucasus. Leaves slightly glaucous beneath, contracted into short broad points at the apex, tapering at the base in cultivated trees ; petiole ^ inch. 4. Arbutus Menziesii, Pursh. Western N. America, from British Columbia to California. Leaves glaucous, almost white, beneath ; rounded or with a minute sharp point at the apex ; sub-cordate or rounded at the base ; petiole i inch. ARBUTUS UNEDO, STRAWBERRY TREE Arbutus Unedo, Linnœus, Sp. PI. 395 (1753); Loudon, Art. et. Frut. Brit. ii. 1117 (1838); Boswell- Syme, Eng. Bot. vi. 28, t 882 (1866); Hooker, Stud. Fl. Brit. Islands, 243 (1878); Mathieu, Flore Forestière, 225 (1897). Unedo edulis, Hoffmannsegg et Link, Fl. Port. i. 415 (1809). A small tree, attaining in Ireland 40 feet in height and 10 feet or more in girth, usually a shrub in the Mediterranean region. Bark rough, brownish-red, more or less fissured, and only rarely scaling off in part. Young branchlets reddish or green, covered with gland-tipped hairs, which persist in the second year ; older branchlets brown, rough, slightly fissuring on the surface. Buds minute, reddish ; scales imbricated, ovate, acute, ciliate. Leaves 2 to 4 inches long by i to 2 inches broad, very variable in shape, oblong, oblong-lanceolate, elliptic, or ovate, acute at the apex, tapering at the base ; upper surface dark-green, glabrous and shining ; lower surface pale-green, glabrous, with prominent midrib and inconspicuous pinnate-reticulate venation ; margin serrate or biserrate, the serrations acute or rounded. Petioles short, about \ inch long, glandular-pubescent. Flowers appearing in autumn, inodorous, in short drooping glabrous terminal panicles. Calyx - lobes minute, triangular. Corolla usually white, rarely pinkish, urceolate, with rounded ciliated teeth ; ovary glabrous. Fruit ripening in the follow ing autumn, at the same time as the appearance of the flowers of the succeeding year ; a stalked berry, pendulous, sub-globose, f inch in diameter, orange-scarlet, densely covered with minute pyramidal spine-like excrescences, edible, superficially resembling a strawberry, but entirely different in structure. Seedling.—Cotyledons two, raised above ground on a short caulicle, oval, rounded at the apex, abruptly narrowed at the base into a flat petiole, entire, £ inch long, dull-green above, pale-green beneath. Young stem reddish, with short glandular hairs ; primary leaves alternate, minute, oval or obovate, serrate and minutely glandular-pubescent in margin ; tap-root about 2 or 3 inches long. 560 The Trees of Great Britain and Ireland VARIETIES In the wild state there is considerable variation in the size and shape of the leaves, dependent upon conditions of soil, shade, and climate. Fliehe1 describes two distinct forms in France. In the hot and dry region of the Esterel, the leaves are small in size, not exceeding 2^- inches in length by f inch in breadth, and are very coriaceous, spathulate, with feebly serrated and revolute margins. In the forest of La Pinouse, near Quillan in Aude, which is mainly composed of Finns sylvestris with a slight mixture of beech and silver fir, the climate being cool and the altitude con siderable, the Arbutus has very large leaves, often 5 inches long by 2 inches broad, which are lanceolate with sharply serrate and non-revolute margins. The following varieties are often cultivated :— 1. Var. rubra, Aiton, Hort. Kew, ii. 71 (1789). (Var. Croomei? Hort.).—Flowers pink or reddish. Mackay3 noticed a single plant of this variety, growing on red slate near Glengariff. 2. Var. integerrima, Sims, Bot. Mag. t. 2319 (1822) (vars. integrifolia and rotundifolia, Hort.).—Leaves entire and smaller than in the type. This is said to have been raised by Loddiges from seed of the ordinary form. The leaves vary in shape, often being obovate or almost orbicular. 3. Var. quercifolia, Hort.—Leaves obovate-lanceolate, with a few irregular teeth in the upper half, about 2 inches long by f inch broad. In cultivation at Kew. 4. Var. turbinata, Persoon.—This variety occurs wild in Greece, and is remarkable for its large top-shaped fruit, more than an inch in length. 5. Other varieties have been noted, which I have not seen, as salicifolia with narrow leaves, crispa with crumpled leaves, ana. plena, with semi-double flowers. DISTRIBUTION This species is widely spread throughout the maritime regions of the countries bordering on the Mediterranean, occurring in Spain, France, Corsica, Sardinia, Italy, Istria, Herzegovina, Dalmatia, Greece, Turkey, Syria, Algeria, and Morocco. It is also met with in the maritime belt along the Atlantic from Portugal to Kerry in Ireland. It occurs either as undergrowth in the forests, where in favoured situations it reaches the dimensions of a small tree, or is one of the shrubs composing the maquis or heaths, which spread over large tracts of siliceous soil that have been denuded of trees in past ages. It is apparently only in Ireland that the Arbutus grows to be a forest tree, moderate in size, but equalling in height and girth the trees of other species, with which it is associated. In France, the Arbutus is common in the departments whose shores are bathed by the Mediterranean and extends inland as far as Drome and Lozère ; it is not un- frequent along the west coast from Bayonne to La Rochelle, and is recorded4 from 1 Butt. Soc. des Sciences, 1886, p. 26. 2 Figured in Garden, xxxiii. 320 (1888). 3 Fl. Hibernica, 182 (1836). 4 Arbutus is very abundant, in company with oak and mountain ash, in a wood, about li mile in length, on the abrupt and rocky slope of the cliff of Trieux, near Paimpol, in Côtes-du-Nord. Cf. Dr. Avice, in Bull. Soc. Bot. France, xliii. 123 (1896), and Coste, Flore de la France, ii. 506 (1903). In a note on the occurrence of this species in the Landes, in Bull. Soc. Bet. France, xlix. p. Ivii. (1902), it is stated that wherever the Arbutus grows, in that region, holly is absent, the two species seeming to exclude each other. Arbutus an outlying station in Brittany. In Corsica, it is very common as a shrub in the maquis ; but in some of the forests grows to be a considerable size, as in that of Bonifatte near Calvi, where I measured trees 25 feet in height and i foot in diameter, which were growing at 2000 feet altitude. In Corsica, a liqueur, called acqua vida de bagui, is made from the berries. In Spain and Algeria, I noticed it as a shrub, growing in ravines in the forests; but in Italy it sometimes attains a considerable size. The Arbutus is unquestionably wild1 in the south-west of Ireland, where it is associated with other plants, which like it are Mediterranean in type and not indigenous to other parts of the British Isles. It has been known to the Irish since early times, and is called caithne (pronounced cahney) in Kerry and cuince in Clare. The former name occurs in several place-names in Kerry, as Derrynacahney, the "oak-wood of the Arbutus," two miles south-east of Crusheen ; Cahnicaun wood, near the Eagle's nest, Killarney, which is coill caithneacan, the " wood of the little Arbutus," in Irish; Ishnagahiny Lake, five miles south-east of Waterville, which is nisge-na-geaithne, "Arbutus water," in Irish. The Clare name, cuince, is supposed to occur in several place-names, anglicised as quin, which, however, often represents a family name of another signification. Cappoquin, in Waterford, means the field of the Arbutus, and Feaquin, in Clare, the wood of the Arbutus. The occurrence of names like Quin, a parish in Clare, and Quinsheen, one of the islands in Clew Bay, Mayo, may point to an extension of the distribution of this plant far to the north in ancient times. At present, Arbutiis Unedo is restricted to Co. Kerry and the extreme south western part of Co. Cork. In the latter county it is thinly scattered through the woods in the vicinity of Glengariff, growing in company with oak, birch, holly, hazel, and mountain ash, and attaining about 25 feet in height and 3 feet in girth. It is said to grow here and there among the mountains to the west of Glengariff, and was seen by R. A. Phillips at Adrigole, ten miles to the west, high up in the mountains amongst rocks, and without the shelter of other trees. Phillips believes that it does not now grow to the eastward of Glengariff; and he could not find it in its former station, Ballyrizzard, near Crookhaven. The Arbutus has its head-quarters in Co. Kerry, in the Killarney district, being particularly abundant and luxuriant on the islands and shores of the lakes generally, where it forms a considerable part of the natural forest. At the base of Cromaglaun mountain, near the tunnel on the Kenmare road, there is a wood composed almost exclusively of Arbutus ; and it is also met with on the Cloonee lakes south of the Kenmare River.2 About Killarney the tree is indifferent as regards soil, as it grows on limestone on Ross island, on sandstone on Dinis island, and on slate, grit, and conglomerate 1 Its right to be considered an indigenous plant was contested by Smith, who, in his History of the County of Kerry (1756), states that it was introduced by the monks of St. Finnian, who founded the Abbey of that name on the banks of the lake, in the sixth century. Babington, in Afaç. Nat. Hist. ix. 245 (1836), says this idea is inconceivable as the tree grows in isolated spots far up in the mountains, and is truly an aboriginal. AH Irish botanists, and they are supported by authorities like Sir J. D. Hooker and Trof. Fliehe of Nancy, are agreed as to the tree being an undoubted native of the south-west of Ireland. 8 There are six trees on the islands in Glenmore Lake near Dereen, and a few on the mountains beside the lake, accord- ng to information I received when visiting Dereen in July 1907. 562 The Trees of Great Britain and Ireland elsewhere. It is much more affected by climate and aspect than by soil, and seeks the most humid and mildest situations. In the Killarney basin it occupies practically the whole northern shore of the northern lake, but does not grow on the exposed islands of this lake. It is absent from the shore itself, when this is marshy or composed of shingle or sand, and grows on the rocky headlands, where it forms a natural wood with oak, holly, and mountain ash. It is very common on the long indented promontory of Muckross, and reaches its greatest dimensions on Dinis Island, which is perhaps the dampest and most sheltered spot in the whole district, protected by high mountains on the east and west, but open to the south. It usually does not extend far from the lake shore, but in the very humid and shaded Tore ravine it recedes into the general woodland along the rocky banks of the torrent, and ascends to an elevation of several hundred feet. It flourishes also on the rocky and sheltered islands of the southern lake. In dense woods it has a fairly straight and single trunk ; but in the open it usually divides at a short distance from the ground into two or more stems, which tend to be spirally twisted and are often curved, each of them terminating in a much-branched wide crown of foliage. The bark of old trees scales off in longitudinal strips and becomes purplish-grey in colour, assuming in the sunlight a reddish tinge, resembling in this respect the branchlets, which are pale-green on the shaded side and crimson on the sunny side. The largest trees seen by me were about 40 feet in height ; one had three stems 4 feet 10 inches, 4 feet 3 inches, and 3 feet 2 inches in girth respectively, the butt measuring close to the ground 17 feet round. Plate 157 represents one of the finest of these trees on Dinis island. Major Waldron has recently found trees up to 5 feet 7 inches in girth. Much larger trees existed formerly, as one measured by Mackay in 1805, which was 9^ feet in girth. The Arbutus woods, like those of Kerry generally, suffered much from the ironworks, which were established in the eighteenth century, and the largest trees were cut down at this period. INTRODUCTION The date of the introduction of the Arbutus into English gardens is unknown ; but Mrs. J. R. Green has kindly sent me the following extract from the State Papers,1 showing that its existence in Kerry attracted in the sixteenth century the attention of the English settlers, who called it •wollaghan, a corruption of ubhla caithne (pronounced oolacahney), or "arbutus apples," a name used for the edible fruit :— "You shall receive herewith a bundle of trees called wollaghan tree, whereof my Lord of Leicester and Mr. Secretary Walsingham are both very desirous to have some, as well for the fruit as the rareness of the manner of bearing, which is after the kind of the orange, to have blossoms and fruit green or ripe all the year long, and the same of a very pleasant taste, and growing nowhere else but in one part of Munster, from whence I have caused them to be transported immediately unto you, 1 Cal. State Papers, Ireland, A.D. 1586, p. 240. Arbutus 563 praying you to see them safely delivered and divided between my said Lord and Mr. Secretary, directing that they may be planted near some ponds or with a great deal of black moory earth, which kind of soil I take will best like them, for that they grow best in Munster about loughs and prove to the bigness of cherry trees or (A. H.) more and continue long." CULTIVATION Though the Arbutus can hardly be called a tree in most parts of England, because it is rarely planted in situations which will enable it to assume a tree-like habit, yet it is so beautiful as a shrub, that no garden should be without it in districts which are warm enough in winter and damp enough in summer to allow it to thrive. It is easily raised from seed, and I have found little difference between the growth of seedlings raised from English and from French seed. Both suffer severely from frosts exceeding about twenty degrees, and from cold dry winds, and should therefore be kept under glass in winter till they are 2 or 3 feet high, when they should be planted out in a well-drained but not dry or heavy soil, in a place well sheltered from the north-east, but not overhung by other trees. Severe winters injure and often kill Arbutus in the eastern and midland counties, and large specimens are rarely seen except on the west and south-west coasts. Even there I have never seen one rivalling what Henry describes in Ireland, and it does not seem to be a long-lived tree in England. The best I have seen, perhaps, is on Sir E. Loder's beautiful grounds at Leonardslee, which is about 30 feet high, with a clean stem 8 or 10 feet high and 3 feet 4 inches in girth. The largest tree on record1 was one growing at Mount Kennedy, Wicklow, which in 1773 was 13 feet 9 inches in girth. It was supposed then to be somewhat more than loo years old. In 1794 it was still living, though it had been split by the wind, and torn up by the roots ; and fresh healthy shoots were springing up from some branches which had layered. The wood, which is of a reddish-brown colour, is hard and takes a good polish, but is very liable to split in drying, and so far as I know is not used for anything but small ornamental work, though it seems very suitable for inlaying or parquet. (H. J. E.) ARBUTUS HYBRIDA2 Arbutus hybrida, Ker-Gawler, Bot. Reg. t. 619 (1822); Louden, Arl. et Frut. Brit. 1119 (1838); Gard. Chron. ix. 211, f. 37 (1878). Arbutus andrachnoides, Link, Enum. Hort. Berol. i. 395 (1821). Arbutus serratifolia, Loddiges, Bot. Cab. t. 580 (1821). Arbutus intermedia, Heldreich, Flora, 1844, p. 14. Arbutus Unedo-Andrachne, Boissier, Fl. Orient, iii. 966 (1875). Arbutus hybrida, being a cross between A. Unedo and A. Andrachm, is variable in the wild state, sometimes being exactly intermediate between the two 1 Hayes, Practical Treatise on Planting, 128 (1794). 2 This name, though not the oldest, is the one by which the species has been usually known, and is adopted by us. HI n 564 The Trees of Great Britain and Ireland parents, and sometimes more closely resembling one of them. As seen in cultiva tion, the bark is smooth, like that of A. Andrachne. The branchlets have the glandular pubescence of A. Unedo, and the leaves are serrate, as in that species ; but have the slightly glaucous tint and conspicuous veins of the other species ; petioles glandular-pubescent. The flowers are borne in spring in large drooping panicles, which are usually glandular-pubescent. The fruit is of moderate size, and slightly tubercular on the surface. According to Loudon, var. Milleri, with large leaves and pink flowers, was raised in the Bristol nursery, being a cross between the red-flowered variety of A. Unedo and A. Andrachne. This seems to be rare in cultivation. Arbutus hybrida originated in the Fulham nursery early in the nineteenth century. It is, however, known in the wild state, being recorded by Heldreich and Halacsy for several localities in Greece. It is also reported to have been found by Albow1 at Pizunda, on the north-eastern shore of the Black Sea, which is re markable, as A. Unedo does not occur wild in this district, and the identification was possibly erroneous. A tree growing at Sedbury Park, near Chepstow, the residence of Colonel Marling, V.C., is by far the finest we have seen of this hybrid. It measures 39 feet high by 5 feet 10 inches at 5 feet, and 7 feet 4 inches at 3 feet from the ground. It is grafted on a stock of A. Unedo, but shows more of the character of A. Andrachne in its habit and bark. It has been propagated by inarching, and seems to be a hardier tree than A. Unedo (Plate 158). There are fair-sized trees at Kew. (A. H.) ARBUTUS ANDRACHNE Arbutus Andrachne, Linnaeus, Sp. PI. 566 (1762); Bot. Reg. ii. t. 113 (1813); Bot. Mag. t. 2024 (1819); Loudon, Art. et Fmt. Brit. ii. 1120 (1838). Arbutus integrifolia, Salisbury, Prod. 288 (1796). Arhitus Sieberi, Klotzch, Linnaa, xxiv. 71 (1851). A large shrub or small tree, attaining 30 to 40 feet in height. Bark peeling off in thin papery layers, smooth, thin, and reddish brown. Young branchlets reddish or green, glabrous ; older branchlets olive-green or brownish, smooth. Buds minute, reddish. Leaves, larger usually in cultivated trees than those of Arbutus Unedo, oval-oblong, contracted into short blunt points at the apex, tapering at the base ; upper surface dark green, glabrous, shining ; lower surface glaucescent, glabrous, with prominent midrib and distinct lateral veins ; margin entire. Petiole glabrous, about \ inch long. Flowers in erect viscid glandular-pubescent panicles, yellowish white, appearing in spring. Calyx-lobes deep, ovate, acute. Corolla contracted at the apex, with five reflexed short rounded ciliate lobes. Ovary pubescent. Fruit small, about \ inch, 1 Radde, Pflanzenverb. Kaukasuslaiid., 127, note (1899). Arbutus 565 rarely ^ inch in diameter, globose, orange coloured, smooth, hard, glandular on the surface. Arbutus Andrachne is a small tree or large shrub, resembling A. Unedo in habit, and like it occurring often in heaths and occasionally in the forests ; and only rarely forming small pure woods. It occurs in Albania, Greece, Cephalonia in the Ionian Islands, Crete, Rhodes, Cyprus, in the maritime regions of Asia Minor, Syria, and Palestine, in the Crimea and in the district of the Caucasus bordering upon the Black Sea. (A. H.) It was introduced into England from Smyrna in 1724, and cultivated at Eltham by Dr. Sherard. This tree though rarely planted in modern gardens * is, on account of its superior hardiness and its extremely beautiful bark, a more ornamental tree than the native species. Though I have never seen or heard of its producing ripe fruit in England, seedlings may be obtained from Continental nurseries, and some that I brought from Pallanza, in October 1906, have survived the journey without injury. The tree seems to enjoy lime in the soil. The bark is like smooth reddish-brown leather, covered with a thin silvery paper-like skin which peels off annually, and for this alone it is well worth growing. There was a very fine though not tall tree of this species on the lawn at Williamstrip Park, Gloucestershire, on rather heavy soil, which endured the inclement season and severe winters of 1879-80-81 without much injury, but is now dead. I saw in 1903 another which was 36 feet high and 4 feet in girth lying on the ground at Haldon near Exeter, which had been blown down some years before but was still living. The best that I know now living is in the Botanic Garden at Bath, and measures 27 feet by 6 feet 3 inches at i foot from the ground, shortly above which it divides into several stems. There is also a handsome tree about 25 feet high at Westonbirt, and one at Mamhead, 30 feet high, which is decaying at the butt. (H. J. E.) ARBUTUS MENZIESII, MADRO»A Arbutus Menziesii, Pursh, FI. Amer. Sept. \. 282 (1814); Sargent, Suva N. Amer. v. 123, t. 231 (1893), and Trees N. Amer. 728 (1905). Arbutus procera, Lindley, Bot. Reg. xxi. t. 1753 (1836); Loudon, Art. et Frut. Brit. ii. 1121 (1838). Arbutus laurifolia, Hooker, FI. Bor. Amer. ii. 36 (1840). (Not Lindley.) A tree attaining in America 100 feet in height and 20 feet in girth, but usually much smaller. Bark of branches and young stems thin, smooth, reddish, peeling off in large thin scales ; of older trunks dark reddish brown and covered with small thick scales. Young branchlets glabrous ; older branchlets reddish brown, smooth. Buds stouter than in A. Unedo, ^ inch long ; scales ovate, acute, apiculate. Leaves oval or oblong, larger than in A. Andrachne or A. Unedo, up to 5 inches long by 3 inches broad, rounded or contracted into minute sharp points at the apex, 1 A tree in Kew Gardens, 20 feet high, is figured in Card. Chron. iv. 724, f. 100 (1888). 566 The Trees of Great Britain and Ireland subcordate or rounded (rarely tapering) at the base ; upper surface dark green, shining, glabrous ; lower surface glaucous, almost white in colour, glabrous, with prominent midrib and conspicuous lateral veins ; margin entire, occasionally serrate on young plants. Petiole stout, ^ to i inch long, glabrous, usually winged on one or both sides for some distance by the decurrent base of the leaf. Flowers appearing in spring, in erect pubescent panicles, about 5 or 6 inches long and broad. Calyx-lobes scarious, white. Corolla white, urceolate. Ovary glabrous. Fruit ripening in autumn, sub-globose, ^ inch in diameter, bright orange- red, glandular on the surface, with a thin flesh and a five-celled thin-walled cartila ginous stone. Arbutus Menziesii occurs in the Pacific coast region from Southern British Columbia, where it grows on Vancouver Island and the islands at Seymour Narrows, through Washington and Oregon to California, reaching its most southerly point in the Santa Lucia Mountains. In Washington it is not uncommon on the cliffs along Puget Sound, and on high slopes, where it receives plenty of light. It usually grows on rich soil and, according to Sargent, is common and attains its largest size in the redwood forest of Northern California, becoming smaller to the north and south, and only growing as a shrub to the south of the bay of San Francisco. I did not observe it in the dense redwood forest near Crescent City ; but found it common inland to the east of the coast range in South-Western Oregon. Here it grew on dry hills at 2000 to 3000 feet altitude, in mixture with Pinus ponderosa, Libocedrus decurrens, and oak, in thinly forested country ; and resembled very much in habit, with its short trunk and broad branching crown, the Arbutus of Killarney. In a ravine near Kerby I measured a tree, 99 feet high by 5 feet i inch in girth, with a straight stem, clear of branches to 40 feet ; but this grew in exceptionally good soil, and was crowded by other trees — Lawson Cypress, Sugar Pine, Quercus densißora, Acer macropkyllum, etc. (A. H.) The largest tree of this species known, which has been figured by Sargent,1 is growing in the grounds of the reservoir at San Rafaël in Marin county, California. It measures 100 feet in height and 23 feet in girth at 3 feet from the ground. Arbutus Menziesii was introduced by Douglas in 1827; but is rather a rare tree in cultivation in England. It appears to be less hardy than the other species now described, and at Kew makes slow growth and often has its leaves and shoots injured by frost. It is found in gardens usually under the name of A. procera, and commonly attains the size of 20 to 30 feet. The largest that I have seen is at Bassetwood, near Southampton, the residence of J. R. Anderson, Esq. This tree is no less than 50 feet high, with a stem clear for about 20 feet, and 3 feet 2 inches in girth. A tree at Tortworth is 35 feet high by 4 feet 4 inches.2 In Scotland, at Castle Menzies, I measured one in 1907 which was 37 feet by 5 feet 2 inches, and did not seem to have suffered much from the severe frost of the previous winter, though the flower buds were killed. (H. J. E.) 1 Garden and Forest, v. 146, f. 23 (1892). In the same journal, iii. 509, f. 515 (1890), the tree is figured in its native forest. 2 Mr. Clinton Baker informs me that there is a tree at Bayfordbury which was sent to his grandfather about twenty-five years ago from America, as a very small plant. It is now 30 feet high, and bears fruit every year. SCIADOPITYS Sdadopitys, Siebold et Zuccarini, Fl. Jap. ii. i, tt. 101, 102 (1844); Bentham et Hooker, Gen. PI. iii. 437 (1880); Masters, Journ. Linn. Soc. (£ot.) xviii. 502 (1881), xxvii. 276, 320 (1889), xxx. 2i (1893), and Journ. Bot. xxii. 97 (1884). AN evergreen tree, belonging to the tribe Taxodinese of the order Coniferse, attaining in Japan a height of 120 feet and a girth of 12 feet. Bark reddish brown, scaling off in long strips. Branches sub-verticillate. Branchlets brown, glabrous, bearing minute scales, which represent true leaves, and cladodes, which are long, green, and leaf-like, performing the functions of true leaves, but differing from them in structure. The scales are borne spirally on the internodes, and are dry, brown, membranous, ovate-lanceolate, and decurrent. At the apex of the shoot there is a ring of similar scales, deltoid in shape, and densely pubescent on their inner surface, out of the axils of which arise a whorl of cladodes, ten to thirty in number, spreading all round the branchlet. These are 2 to 5 inches long, averaging \ inch in width, linear, rigid, narrowed towards the base, obtuse and minutely rigid at the apex ; upper surface dark green, shining, with a median groove ; lower surface green on each side of a deep white stomatiferous central furrow. Buds globose, composed of numerous spirally imbricated greenish scales ; terminal bud, at the apex of the shoot, in the centre of the whorl of cladodes, continuing the growth of the main axis in the following year ; a smaller bud, often present at the side of the terminal bud, developing into a lateral branch in the next season. As a rule, the main axis is bare, except for the scales, below the apex, which bears the whorl of cladodes and the buds ; but on strong-growing shoots a lateral branch is occasionally developed half-way up the internode. The cladodes are leaf-like shoots, and not true leaves, each representing an axillary branch with two coherent leaves ; but their true nature has given rise to a great deal of discussion ; and the elaborate papers of Dr. Masters cited above may be consulted on this subject. Male flowers in a terminal compact raceme, about an inch in length ; each flower f inch long, subsessile ; anthers numerous, spirally arranged, short-stalked, with an acute and reflexed crest and two pendulous cells, opening by a vertical slit ; pollen-grains globular, minutely tuberculate. Female flowers, terminal small cones composed of spirally arranged lanceolate bracts, which are serially continuous with the true leaves, empty at the base of the cone, higher up with fleshy semi-lunar ovular scales in their axils, half the size of the bracts and bearing one to nine ovules in a transverse series on their inner surface. As the cones increase in size, the 567 568 The Trees of Great Britain and Ireland ovular scales outgrow the bracts, and in the mature cone are much larger than and almost entirely coalesced with them. The cones, which are borne on short stout stalks, clothed with a few membranous bracts, either remain terminal and erect or are pushed aside by the growth of a lateral branch. They take two years to ripen, and remain persistent on the tree for some months after the dehiscence of the seeds. Ripe cones, about 3 inches long by 12- inch in diameter, oblong-ovoid, obtuse at the apex, composed of woody scales, which result from the coalescence of the ovular scales and bracts of the flower. The scales are fan-shaped, about f inch wide ; upper margin rounded and reflexed ; outer surface convex, marked by a transverse rugged irregular ridge ; inner surface concave, with slight depressions for the seeds. Seeds, five to nine on each scale, reversed, oval, compressed, dark brown, surrounded by a narrow membranous reddish-brown wing, notched at the base and marked at the apex by the white hilum ; seed with wing, about f inch long by £ inch wide. The seedling has a long slender tap root, and a terete green glabrous caulicle about an inch in length, which bears two cotyledons. These are sessile, linear, tapering to an obtuse apex, a little more than ^ inch long, dark green above, paler below with indistinct lines of stomata. Primary leaves like the cotyledons, but longer. Sciadopitys is a monotypic genus, only one species being known, which is a native of Japan. SCIADOPITYS VERTICILLATA, UMBRELLA PiNE1 Sciadopitys -verticillata, Siebold et Zuccarini, loc. cit. 1844; Kent, Veitch's Man. Conifères, 287 (1900); Shirasawa, Icon. Ess. forest. Japon, text 22, t. 8, ff. 15-36 (1900); Thiselton-Dyer, Hot. Mag. t. 8050 (1905)5 Mayr, Fremdländ. Wald- u. Parkbäume, 407 (1906). Taxus -verticittata, Thunberg, Fl. Jap. 276 (1784). The species has been described above. The tree is known in Japan as Koya- maki, or pine of Mt. Koya, one of the localities where it is found growing wild. Thomas Lobb sent a living plant in 1853 from the Botanic Garden at Buitenzorg in Java to Veitch's nursery at Exeter; but it soon died. It was afterwards intro duced by seeds brought from Japan by J. Gould Veitch in 1861, some being also sent about the same time by Fortune to Standish at Ascot. A variety in which the leaves are striped with yellow was introduced by Fortune ; but this seems to be now unknown in cultivation. Plants only 3 feet high produced cones in 1876 in the nursery of Messrs. Thibaut and Keteleer at Sceaux.2 The tree appears to have first borne fruit in Scotland3 at Ardkinglas, in 1878, and in England4 at Kew and Coombe Wood, in 1884. Proliferous cones,5 which bear cladodes at their apex, are of frequent occurrence in Japan, and have also been borne by trees cultivated in Europe. In 1 This is a translation of Sciadopitys, a name given on account of the leaf-like cladodes spreading out from the apex of the shoot, like the ribs of an umbrella. 2 Card. Ckron. v. 827 (1876). *Jcm: of Forestry, 1879, P. 5o8. * Card. Chron. i. 80 (1884). 5 Masters, Jottrn. Bot. loc. cit. f. 4. Sciadopitys 569 these the bracts, which are t ordinarily completely coalesced with the fruit-scales, become detached from them towards the apex of the cone, and are scale-like in character, producing cladodes in their axils. (A. H.) Shirasawa states that the tree grows wild in mixture with Abies ßrma and Cupressus pisifera at 600 to 5000 feet in the forests of Kiso and Shinano. Matsumura adds Mt. Hoonokawa in Tosa. According to Mayr, the tree is similar to the silver fir in its capacity for bearing shade ; but is extremely slow in growth, only attaining in Kiso, where the climate is favourable to it, a height of 30 feet in fifty years. Trees no feet in height and 2 feet in diameter average about 250 years old. Mayr gives a figure of two old trees, growing at Agematsu, which were nearly 120 feet in height and 4 feet in diameter; and this shows how the tree, even at an advanced age, preserves a narrow pyramidal form with an upright leader, without any sign of flattening of the crown. I saw this tree in its native forest to the best advantage in the lovely valley of Atera, on the west side of the Kisogawa, below Agematsu, at from 2000 to 3000 feet elevation. Here it was scattered in a forest of mixed conifers and hardwoods, and seemed to grow only on rocky slopes and ridges, where its narrow-pointed top made it conspicuous. The seedlings were numerous in dense shade growing on a bed of humus, and those that I took up had long but scanty roots, running deep, but spreading little. Their growth was very slow, not more than 3 to 6 inches annually for the first twenty years at least. On a steep rocky hill above the forester's house, at the end of the tramway which has been made up this valley, the largest trees were growing, mixed with Thujopsis, the undergrowth being very dense, and composed of Rhododendron, with Shortia uniflora spreading over the ground in great sheets. The largest that I was able to measure were 90 to 100 feet high and 9 to 10 feet in girth, one being n feet 9 inches at 5 feet from the ground. Plate I59A fairly represents the appearance of the tree here. In the forest near Koyasan I saw it again, mixed with Cupressus obtusa, but not attaining so large a size, though it seemed that in the dense shade the seedlings of Sciadopitys were more numerous and vigorous. Though often planted in parks and temple gardens, I never saw any trees as fine as those figured by Mayr at Agematsu, and it is clear that shade, perfect drainage, and a rich forest soil are essential to this species. According to Mayr, the wood is white in colour, the sapwood only f inch thick being like the heartwood. The wood is comparable to the best kind of spruce, and is soft and elastic. It is used in Japan for boat-building, making bath-tubs and casks, planking, etc. CULTIVATION Though this interesting tree has been planted in many places, yet it usually grows very slowly and seems to require a high summer temperature, with a warm and sheltered situation. Ripe seed was produced in Ireland at Castlewellan in 1900, 57° The Trees of Great Britain and Ireland but the seedlings which I raised from this source were always weakly, and notwith standing every care died after two or three years. Seed sent from Japan in 1906 also germinated weakly, and many of the seedlings damped off in the winter without making roots, though very carefully watered. They seem to require a very light, sandy peat when young, and refuse to grow in soil which contains lime, or where moisture is deficient during the summer. By far the finest tree that I have seen in England is at Hemsted, in Kent, where a tree was in 1905 no less than 38 feet high by 2 feet in girth, and showed its true habit very well. Owing to its being rather crowded by other trees, a photograph of this was difficult to take, but after several attempts had been made, Mr. Edwards was able to get the one reproduced in Plate 15915. The next largest I have seen is at Coombe Royal, in South Devon, where a tree about 25 feet high is growing, with a forked stem. As usually seen in gardens in England, it forms a shrubby pyramid, and I have seen no others over 15 to 18 feet high. At Castlewellan, however, it seems to thrive very well, and should do well in the south and west of Ireland and Wales. At the Villa Trubetskoi, near Intra, on Lake Maggiore, I saw a vigorous tree about 45 feet high, which had divided into three stems, and in 1906 bore no cones. Sciadopitys is perfectly hardy, and bears without injury the severe winter climate of Boston in New England, and of Grafrath in Bavaria, the thermometer descending in the latter locality to — 18° Fahr. ; but in severe winters the foliage turns brown. (H. J. E.) PINUS SYLVESTRIS, SCOTS PINE Pinus sylvestris, Linnaeus, Sp. PI. 1000 (exd. var.) (1753); Lambert, Genus Pinus, i. tab. I. (1803); Loudon, Arb. et Frut. Brit. iv. 2153 (1838) ; Willkomm, Forstliche Flora, 193 (1887) ; Mathieu, Flore Forestière, 579 (1897); Kent, in Veitch's Man. Conifera, 379 (1900); Kirchner, Loew u. Schröter, Lebengeschich. Blütenpfl. Mitteleuropas, \. 175 (1904); Mayr, Fremdländ. Wald- u. Parkbäume, 347 (1906); Borthwick, in Trans. R. Eng. Arb. Soc. vi. 205 (1906). A TREE commonly ioo feet, rarely attaining 150 feet in height, with a girth of ID to 15 feet. Stem usually straight and cylindrical, with the branches regularly whorled in young trees, forming a pyramidal crown ; in older and isolated trees, branching irregular, with a flattened crown. Bark different in the lower and upper parts of the trunk ; towards the base thick, fissured into irregular longitudinal plates, scaly, and reddish brown or greyish brown in colour ; on the upper part of the stem,1 owing to the outer portion continually falling off in thin papery scales, the bark remains very thin, smooth, shining and bright red. Young shoots greenish, smooth and shining ; becoming greyish brown in the second year ; marked with the pulvini of the scale-leaves, which are early deciduous. Buds long-oval, pointed, usually non- resinous, covered by lanceolate acuminate scales, fimbriated on their edges, the upper ones with their tips free and not recurved. Leaves two in a bundle ; sheaths at first white, ^ inch long, speedily becoming shrivelled, brown, and short ; the pair of leaves close together, but not appressed, usually about 2 inches long but varying under different conditions from i to 4 inches, dark green with interrupted lines of stomata on the convex side, glaucous with many well-defined lines of stomata on the flat inner side, plano-convex in cross-section, linear, stiff, acute at the apex, somewhat bent, smooth, finely serrate in margin ; resin-canals marginal. The leaves persist usually three years. Male flowers in dense clusters at the lower part of the current year's shoot, £ inch long, oval, short-stalked, surrounded at the base by four yellowish bracts ; anther with small rounded upright connective. Female flowers, solitary, opposite or occasionally whorled, apparently terminating the young shoot, erect at first, but becoming pendant immediately after pollination, stalked, globose, reddish, composed of rounded bracts and almost circular ovular scales, the latter having a beak-like process on the upper side and bearing two minute ovules. Cones shortly stalked, variable in shape, usually ovoid-conic with an acute apex, oblique or nearly symmetrical at the base, greyish or dull brown in colour, i to 3 1 According to Shaw of Boston, who is the greatest living authority on the genus Pinus, this peculiarity of the bark of the upper part of the tree being thin and reddish, owing to the constant shedding of scales, occurs only in three pines, viz. P. sylvestris, P. densiflora, and P. patula. i" 571 a 572 The Trees of Great Britain and Ireland inches long. Scales dark brown on the inner surface, oblong, ending in a rhomboidal apophysis, which is variable in form in different varieties and even in the same cone ; flattened, with a transverse keel and an elevated or depressed umbo, or raised and pyramidal, with four to five concave sides ; occasionally the apophysis ends in a hooked process. The cones open in spring to let out the seeds, which may be carried by strong winds to an immense distance ; and the empty cones usually remain on the tree till the following autumn. Seeds long-oval, ^ to \ inch, some blackish, others grey in colour, surmounted by a wing half-oval in shape, which is three times as long as the body of the seed. Seedling.—Cotyledons, four to seven, triangular in section, linear, slightly curved upwards, about | inch long ; stomata absent on the outer surface, present on the inner two surfaces, without wax, so that the cotyledons are green in colour and not glaucous. Primary leaves elliptic in section with hairs on their edges. The seedling grows about 2 to 4 inches high in the first year, ordinary needles being produced in the second year ; branches usually appear in the third year. The primary root is long, attaining about 8 inches in the first year, and giving off many lateral fibres. VARIETIES The common pine, spread over an immense geographical area and growing in the most diverse conditions of soil and climate, exhibits considerable variations in most of its characters. The stem may be straight and cylindrical with a single leader, only branching at the top and giving rise to a flattened crown of foliage in old age ; or it may be dwarf, branched from the base and crooked, simulating the smaller forms of Pinus montana. The young cones are usually reversed immediately after flowering ; but in certain regions they remain erect. The adult cones vary in size and shape and in the form of the apophyses, which may be flat or raised, pyramidal or hooked ; but all these variations in the apophysis may occur on the same cone. The male flowers may be yellow or reddish in colour. The leaves vary in length from one to four inches, and may be broad or narrow, stiff and sharp-pointed or soft in texture ; and in some cases they are much more glaucous than in others. They vary in duration from two to five years. Many of these varieties occur in individual trees in the same forest ; and in many cases, when the condition of the soil is changed as by draining, pines which have been small and stunted assume the ordinary tall form, and the shape of the cones probably does not remain constant, when the seedlings are raised in a new locality. It is difficult on this account to establish clearly marked geographical varieties. The experiments, which have been carried out at Les Barres,1 over a long term of years, show, however, that there are races of pines, which preserve their characters of straightness of stem, quickness of growth, or the reverse ; but these races cannot be distinguished by characters of cones or leaves ; and are the result of the selection of seed from vigorous or weak individuals. 1 Cf. Pardé, Arboret. Nat. des Barres, 71 (1906), where a full account is given of the plots of Riga, Haguenau, Scotch, and certain French varieties of Pinus sylvestris, which were mostly planted between 1823 and 1835. Firms Sylvestris 573 The following varieties, occurring in the wild state, have been distinguished, though they are not so clearly defined in nature, as they seem to be from their description. 1. Vzr.genuina, Heer. Cones usually solitary, long-stalked, symmetrical, acute at the apex ; apophysis flat or convex, not hooked. Needles about 2 inches long, persistent three years. This is the common pine, growing on good soil in Germany, Southern Scandinavia, Poland, and North-Western Russia. Two races have been distinguished on the Continent in cultivation :— (a) rigensis (Pinus rigensis, Desf.). Riga pine, raised from seeds collected near Riga. At Les Barres, this is the best race of P. sylvestris, the stem being very straight and cylindrical, rising to a great height, and with few lateral branches ; bark very red, stripping off above in very thin papery scales. Von Sievers states that the form native to the Baltic provinces of Russia is superior in growth and timber to that introduced there by seed from Germany. Willkomm, however, is of opinion that the so-called Riga pine is only a fine tall-growing form, and occurs in North Germany and Poland, as well as in Russia. (b) Hagmnensis, Loudon. Haguenau pine, raised from seed obtained in the forest of Haguenau in Alsace. At Les Barres, this form, though vigorous in growth, is defective, on account of its tendency to form numerous irregular branches, so that the stem is not so clean and does not reach the same height as the Riga variety. The bark is not so red, and is not so fine-scaled as in that variety. Two trees of the Haguenau variety, raised from seed, procured by Loudon in 1828, are growing at Seggieden in Perthshire, and are now about 65 feet high by 8 feet in girth. According to the forester, they are distinguishable in bark, buds, shoots, and leaves from the Scots pine growing near them. 2. Var. scotica.—This variety, which grows wild in the Highlands of Scotland, differs in the redder bark of the stem; in the shorter more glaucous leaves (i^inch long), often persistent four years; and in the shorter cones (i^ inch long), which are symmetrical, with apophyses usually flat near the base, tending to be pyramidal in the upper part of the cone. 3. Var. engadinensis, Heer.—Bark reddish ; needles short, i to \\ inch long, thick and stiff, persistent for five years ; buds resinous. Cones ovoid-conic, 2 inches long, oblique at the base ; apophyses convex on the outer side of the cone, umbo large and blunt. A small tree, rarely 30 feet high, growing in the Engadine Alps.1 It is perhaps a hybrid between P. sylvestris and P. montana. 4. Var. lapponica.—Pinus lapponica, Mayr, Fremdländ. Park. ». Waldbäume, 348 (1906). This variety, which grows in the north of Norway and Sweden and in Finland, is considered by Willkomm and Christ1 to be identical with var. engadhiensis, with which it agrees in the short, straight stiff leaves, persistent for 1 Dr. Christ, in Flore de la Suisse, 197, and Suffi. 31 (1907), considers the Engadine pine to be precisely the same as specimens he examined, which were collected at Quickjock in Lapland (lat. 67°). He also mentions (p. 285) a curious form of the common pine, slender and tall in habit, with very short green needles, which grows at Flims in Switzerland, and also in one or two places in Silesia. 574 The Trees of Great Britain and Ireland five years, in the resinous buds, and in the small cones with hook-like apophyses. Mayr, however, considers it to be a distinct species, and gives the characters which distinguish it from the common form of P. sylvestris, without pointing out in what respect it differs clearly from var. engadinensis. 5. Var. nevadensis, Christ.—Needles broad, short and stiff, very white on their Hat surfaces. Cones nearly sessile, oblique, with very pyramidal apophyses. Occurs in the Sierra Nevada in the south of Spain. 6. Var. reßexa, Heer.—Needles as in the common form. Cones long and slender, conic, with long hooks to the apophyses. This variety has been found growing on high peat-mosses in Switzerland and on poor sandy soil in Prussia, and occurs sporadically elsewhere. In the Caucasus and Asia Minor, P. sylvestris differs from the European form, in having very long and broad needles (3^ inches long), and very oblique cones with hooks on their outer side directed downwards. Specimens from the Amur have very long leaves (4 inches or more) with cones of the ordinary form. The pine of the Ural and Altai mountains (var. uralensis, Fischer) is only distinguished by having short and stiff leaves. In the dry climate of the south of France, in the Cevennes and in Provence, the needles of this species become short and are often disposed in slender tufts at the ends of the branchlets. In the French Alps near Modane, P. sylvestris grows in mixture with P. montana, var. uncinata ; and it is difficult to distinguish between these trees in this locality,— the branches densely covered with short leaves, persistent for four or five years, being alike in both species ; and the cones of both have hooked apophyses.1 How ever, at Modane, as elsewhere, the reddish bark and the dull colour of the cones will distinguish P. sylvestris ; while in the other species the bark is never red and cones are shining brown. Similar forests occur in Switzerland, where P. sylvestris and P. montana appear to pass one into the other ; and the occurrence of these apparently transitional forms has given rise to the belief that they are hybrids between the two species ; but this is not established beyond doubt. Willkomm describes two interesting varieties, due to poverty of soil and exposure. One is the shore-pine of the Baltic provinces of Prussia, which has a short bent stem with an irregular crown of foliage or is a mere bush ; the cones are very oblique and hooked. Another form is peculiar to the peat-mosses in Austria and Germany ; the stems are rarely more than 6 feet high, very' slender, and branched to the base ; needles very stiff and short (about i inch long), persistent for two years ; cones very small, with hooked apophyses. Several varieties have arisen in nurseries or as sports in the wild state. Var. virgata, Caspary.—Main branches irregularly whorled, arising from the stem at an angle of 30° to 60°, elongated and giving off a few twig-like branchlets, only the outermost of which are furnished with leaves. This curious variety2 was first 1 The young cones of P. sylvestris at Modane remain erect (like those of P. montana) and are not reversed immediately after pollination, as is usually the case elsewhere. 2 Willkomm, forstliche Flora, 199 (1887). Pinus Sylvestris 575 noticed in France; and some years later, in 1881, was found in the forest of Wandsburg in Prussia. Var. argentea, Steven.—Cones and leaves with a silvery tint. Found in the Caucasus. Var. monophylla, Hodgins.—A shrub, with the needles in each sheath attached to each other throughout their length, apparently forming one needle, but easily separated. Originated at Dunganstown, near Wicklow, about 1830. Var. microphylla, von Schwerin. —Needles thin, sharply pointed, only £ inch long. Originated as a seedling in 1883 at Wendisch-Wilmersdorf. Var. aurea.1—A low tree of dense habit, with leaves of a golden yellow colour usually in spring, the foliage becoming green in summer. Var. variegata.—Leaves variegated. This form has arisen several times in cultivation ; but was once found wild in Prussia by Caspary. Var. pyramidalis.—Fastigiate in habit. Schübeler says that trees of this kind are common in the forests of Norway and Finland. Var. pendula, Caspary.2—A weeping form, found in a wood near Tilsit, in East Prussia. Various dwarf forms are known, as pumila, nana, globosa. DISTRIBUTION The common pine has an extraordinarily wide distribution, occurring in regions of the most diverse climates and on almost all soils, and in the mountains as well as in the plains. It grows in Eastern Siberia, where the temperature falls to - 40° Fahr., and the period of vegetation hardly lasts for three months ; and is met with in Southern Spain, where the summer heat reaches 95° Fahr., and the period of vegetation lasts for nine months of the year. It occurs in dry regions like Provence, where there is little humidity in the air, and in the west of Scotland, where the air is laden with moisture all the year round. It is by preference a tree of siliceous soils, but occurs on almost all geological formations ; and in Scotland, Norway, and Sweden grows on peat-bogs too wet for the spruce to exist on. The area of distribution includes almost all Europe and the greater part of Northern Asia. The northerly limit, commencing on the north-west coast of Norway at Alten (70° N. lat.), passes through Lapland, south of the Enara lake (68° 50'), and touches Pasvig Fjord on the Arctic Sea at 69° 30'. Extending through the Kola peninsula from Kola bay, it crosses the White Sea at 66° 45' and in the Petchora territory goes as far north as 67° 15' ; and crosses the Ural at about 64°. In Siberia it never reaches quite as far north as the Arctic circle, though it nearly touches it on the Ob and the Yenisei rivers ; east of the Lena river it descends to about 64 . It reaches its extreme easterly point (about 150° E. long.) in the Werchojansk Mountains. The eastern limit descends from there through the Stanovoi Mountains 1 There is a useful note on the propagation of this variety in Gard. Chron. xi. 405 (1892). 2 Schrift. Fliys. Oekoiiom. Gesell. Königsberg, 1866, p. 49, fig. I. 576 The Trees of Great Britain and Ireland and the Seja territory to the Upper Amur. According to Komarov,1 it is a scarce tree on the banks of rivers in Manchuria. Its southerly limit in Siberia is not well known ; but it is known to occur in the mountains of Dahuria, in the territory around Lake Baikal, and in the Altai Mountains. Its southern limit in European Russia is a very irregular line, which begins in the Ural south of Orenburg at about lat. 52°, is most to the north in the government of Tula (lat. 54° 30'), and descends from there to Kharkof (lat. 49°), passing into Galicia about lat. 50°. Far south of this line, and separated from it by the Russian Steppes, on which no pine trees grow, occurs an area of distribution, not yet well made out, which includes the Caucasus, the mountains of the Crimea, Asia Minor,2 and North-Western Persia. There is also an isolated area, in which the pine is found growing wild, in Macedonia, on Mount Nidjé. From Galicia the southern limit in Europe (exclusive of the last-mentioned area) passes southwards to the Transylvanian Alps ; thence it extends along the mountains to Servia, where the tree grows on the Kopavnik mountain (about lat. 43"), continues through the mountains of Bosnia, Dalmatia, Illyria, Venetia, and through Lombardy to the Ligurian Apennines (about lat. 44"). It passes into France, across the Maritime Alps, into the Cevennes, and reaches the Eastern Pyrenees ; in Spain it descends through the mountains of Catalonia, Aragon, and Valencia to the Sierra Nevada in Andalusia, which is its extreme southerly point in Europe (lat. 37°). The westerly limit beginning here, stretches north-west through the mountains of Avila to those of Leon in North Spain ; and is continued through the mountains of Scotland to the north-west coast of Norway. In this vast area the pine is very irregularly distributed. The largest forests occur in the Baltic provinces of Russia, in Scandinavia, in Northern Germany, and in Poland. Towards the south it only occurs in mountains, and rarely forms pure forests of considerable extent. According to Huffel,3 it is rare in Roumania, where he saw it at the confluence of the Lotru and Oltu rivers at 1700 feet altitude, and in the valley of Bistritza.4 In the British Isles, the common pine is found wild at the present day only in the Highlands of Scotland, where a few forests still remain. These occur in the valley of the Spey at Rothiemurchus, Duthill, Abernethy, and Glenmore, and in the valley of the Dee at Invercauld, Braemar, and Glen Tanar. There is also a fine wild forest, the " Black Wood," on the south side of Loch Rannoch in Perthshire.5 That of Ballochbuie near Invercauld is probably the finest now existing. The pine was widely spread over the British Isles in ancient times, as is evidenced by the occurrence of remains of logs, stumps of trees, and cones in the 1 Flora Manshuria, \. 175 (1901). 2 Finns sylvestris grows on the Armenian plateau, and has been described in Linnaa, xxii. 296 (1849), as P. armena, Koch; P. Kochiana, Klotzsch ; and P. pontica, Koch. Cf. Moniteur Jardin Botanique Tiflis, ii. 26(1906). 3 Forêts de la Roumanie, 6 (1890). 4 M. B. Golesco, in an article on the forests of Roumania, in Bull. Soc. Dendr. France, ï. 171 (1907), states that in the Muscel district P. sylvestris is only found on calcareous soils ; and in a letter to Elwes confirms this statement, adding that it attains a diameter of one metre, ancl does not grow on the adjoining schist. 6 Buchanan White, Flora of Perthshire, 282 (1898), gives as additional localities for wild trees in Perthshire, Breadalbane, in Glen Lyon and near Killin and Tyndrum ; and mentions one or two other places where the pine is doubtfully nati\-c. According to the Rev. E. G. Marshall, Journ. Bot. xliv. 160 (1906), it is certainly native in the forest of Glenavon, but quite scarce, and the seedlings appear to be destroyed by deer hrowsing on them. Pinus Sylvestris 577 peat-mosses and submerged forests.1 In the south of England extensive forests occurred in Neolithic times, when the existing peat-mosses began to form ; but in other parts of the three kingdoms it is probable that the pine existed in many places in historic times.2 Of its existence in a wild state until lately in England, the evidence is very meagre. Holinshed,8 writing in 1586, says: "The firre, frankincense, and pine we do not altogether want, especiallie the firre, whereof we haue some store in Chatleie Moore in Darbishire, Shropshire, Andernesse, and a mosse neere Manchester, not far from Leircesters house ; although that in time past not onelie all Lancastershire, but a great part of the coast betweene Chester and the Solme were well stored." According to the Rev. Abraham de la Pryme4 there was a wood of wild pine on a hill at Wareton in Staffordshire in his day, the beginning of the eighteenth century ; and, in an old deed, fir trees were mentioned as growing scattered in Hatfield Chase in Yorkshire about the year 1400, the last surviving aboriginal pine here being cut down about 1670. The Wareton pines were described by Ray in a note6 dated Oct. 14, 1669: "We rode to see the famous fir-trees, some 2^ miles distant from Newport, in a village called Wareton in Shropshire,0 on the land of Mr. Skrimshaw. There are of them thirty-five in number, very tall and straight, without a bough till towards the top. The greatest, and which seems to be the mother of the rest, we found by measure to be 14^ feet round the body, and they say 56 yards high, which to me seemed incredible. The tenant's name of the house close by these fir-trees is Firchild, whose ancestors have been tenants to it for many generations." These trees, according to Dr. Higgins5 of Newport, are mentioned in an old book, Historia Vegetabfaim Sacra, published in 1694 by Westmacott, who says there were thirty-six of them, one of them being 47! yards high. Withering,7 writing in 1776, states that the trees at Wareton were no longer existing in his time. Pine forests apparently occurred in Roman times in the north of England, and remnants of these may have existed down till a recent period, concerning which the late Professor Newton told me of some very old Scots pines that used to grow about forty-five years ago on Wretham Heath, Norfolk, which local tradition said had never been planted, but grew there wild. They were always spoken of as the " Deal8 Trees," all other trees of this species that were planted being named Scotch firs. Whether there is any real foundation for this tradition is very hard to say, but it is possible that the seed > Cf. Clement Reid, Origin British Flora, pp. 16, 152 (1899) :_« Remains of this tree are found in Neolithic deposits, in 'submerged forests' and at the base of peat-mosses, nearly throughout Britain and in Ireland. In late Glacial times at Bovey Tracey, Devon, and at Hoxne, Suffolk (in bed C ?). Abundantly in the preglacial strata of Norfolk, but not in any of the mterglacial deposits in Britain. During the Neolithic period it seems to have been one of our commonest trees • but afterwards disappeared from the southern half of England. " 2 The orchid, Goodyera refens, which was formerly supposed to grow only in wild coniferous forests, as in the Highlands of Scotland, has begun to appear, of late years, in various localities, where the Scots Pine has been planted, both in England and m France ; and the problem as to how the seeds of the orchid reach these plantations is still unsolved. Cf. Kew Bulletin 1906 p 293 ; Actes Premier Congrès Internat. Bot. Paris, 382 (1900) ; and Fliehe, in Mém. Acad. Sta»u/as, 1878 3 Holmshed's Chronicles, i. 358 (1807), reprint of the edition published in 1586. 4 Phil. Trans. No. 275, p. 980 (1701). 6 Derham, Memorials of John Ray, 25 (1846). 8 Wareton, now usually written Warton, is in Staffordshire, not far from the Shropshire boundary ï Botany, ii. 593 (1776). e According to Britten and Holland, Diet. Eng. Plant-Names, ,46 (1886), deal** is used for Pin* sylvestris in East Anglia and Northamptonshire, the cone being commonly called deal-apple. 578 The Trees of Great Britain and Ireland from which these trees grew might have been brought from Norway in early times ; and Sir H. Howorth suggests that the existence of the Capercaillie, whose bones have been found in Tertiary deposits in the eastern counties, would have been impossible unless either pines or spruce existed to feed them in winter. The Rev. Leonard Blomfield read a paper before the Bath Antiquarian Field Club on December 9, 1885, in which he tried to prove that the numerous Scots pines, now growing in the neighbourhood of Bournemouth,1 are descended from aboriginal trees ; and gave the following list of names of places in England in which the word ßr occurs, indicating that these localities were in early days probably noted for woods of Pinus sylvestris :—Firbank in Westmoreland ; Furbecke or Firbeck, and Firbie or Firby, in Yorkshire ; Furbie, Firby, or Firsby in Lincolnshire ; Furcombe in the parish of Farnborough, Berkshire ; Furle or Furleigh in Pevensey Rape, Sussex ; Furland, a tithing of Crewkerne in Somerset ; and Furland Hill, between Brixham and Dartmouth. Loudon, p. 2167, says that the tree only began to be planted in Britain about the end of the seventeenth century ; but the following extract from a letter2 of James I. to the Earl of Mar, dated Oct. 30, 1621, shows that the introduction of the Scots pine into England was earlier. "The Marquis of Buckinghame, being desirous to have firre trees planted aboute his house at Burleigh on the Hille, hath earnestlie requested us to cause him to be furnished as well with the seede as with young trees, which his desire wee willinglie wold have performed with all expedition. And because wee know none who so readilie can give us satisfaction in this pointe as your selfe, we have thoughte good by these présentes to require you with all expedition to cause some store of seede to be gathered eyther in your owne boundes or in those of the Marquis of Huntlie, where it may be soonest had, and so soone as possiblie may be, sende a man of purpos to Burleigh on the Hille with so much of the freshest and fairest thereof as convenientlie may be caried. And that yee cause sette downe in writing at what time and in what kinde of grounde the same is to be sowed, and with the maner of sowing thereof; also when the time of year is fitting for removing and setting of plantes and young trees. Yee shall likewise sende one to Burleigh with four or five thousand of them, with the like instructions of time, place, and maner of setting and preserving." There is no reference to these trees in the History of Burley on the Hill, published in 1901 ; and enquiries have elicited no information, except that there are now on the estate six or eight Scots firs, which are not more than 25 feet high. A local woodman, about 60 years of age, whose father was woodman before him, never heard of the existence of old pines at Burley. The common Gaelic word for Pine vs, gius. It occurs in a few Scottish names of places, as Craiggush, Kingussie Altnaguish, Dalguise. This word is commonly used for pine also in Ireland, and ochtach occurs in books. In spite of the wide prevalence in ancient times of pine in Ireland, place-names with either of these words 1 The submerged pine forest on the sea-coast at Bournemouth is described by Sir C. Lyell in Principles of Geology, ii. 536 (1872). 2 Historical MSS. Commission, Reporten MSS. of Earl of Mar, p. 103 (1904). Pinus Sylvestris 579 are rare. Mr. T. P. O'Nowlan, a competent Gaelic scholar, has given me the following list : *— GOOSE ISLAND, oilean guts, in Lough Derg, Co. Tipperary, "island of pines." CLONYGOOSE, cluain giiis, parish in Carlow, north of Borris, " meadow or plain of pines." MULLAGHANUISH, mullach an gius, near Ashford, Limerick, "hill-top of the pine." GARROOSE, gardhagius, near Bruree, Limerick, "garden of pines." KNOCKNAGUISH, cnocan an giuis, about three miles north of Kenmare, "little hill of pine." KNOCKNAGUSSY, similar in meaning, is situated about three miles south-west of Lough Mask in Co. Galway, "hill of the pine." KNOCKHOUSE, cnoc gheainas, three miles south-west of Mullinavat, Co. Kilkenny, " hill of pines " ; the Gaelic word used here being a local variation of the common form. OGHTY ISLAND, oilean ochtaigh, near Roundstone, Co. Galway, "pine island." DROMOGHTY, drom ochtaigh, about three miles north of the tunnel on the Kenmare road, Co. Kerry, "pine ridge." Apparently, though the pine tree was centuries ago well known in Ireland, there is very scanty evidence as to its existence as an indigenous tree in modern times. Everywhere in Ireland the roots of pine trees are often found in situ in the upper layers of the peat-mosses, showing that forests of pine grew in the peat and attained a considerable size. These peat-mosses are probably of late formation.2 Ray3 quotes Mr. Harrison as an authority for pine "growing wild in the mountainous parts of Kerry where the Arbutus grows," about the beginning of the eighteenth century. Smith,4 writing in 1761, says that "these trees have been much destroyed in recent years ; for, except a small shrub here and there among the rocks, there are none standing at present of any large size." Mackay5 mentions, in 1825, a solitary pine tree standing near the foot of Mount Nephin in Mayo, which was supposed to be the last remnant of the pine forest of that county. This tree,6 very large and very old, was living in 1866, the exact locality being an open bog at Deal Castle, near Crossmolina, at the head of Lough Conn, and had been fenced in by the Earl of Arran. H ayes,7 writing in 1794, speaks throughout his valuable book of Pinus sylvestris as Scots fir; and evidently in his day all the pines in Leinster at least were the product of Scotch seed. In France the common pine is never met with growing wild in the plains. It is confined in the wild state to the Alps of Savoy, of Dauphiné, and of Provence, the 1 While the above was passing through the press, Mr. O'Nowlan sent me a further list, as follows :—Lough Aguse, .._ of two lakes, one near Fettigo, Donegal, and another in Fermanagh ; Lough Ayoosy and Aghoos, in Mayo ; Cappay'use Roscommon ; Meenaguse in Donegal ; Drumgoose and Derrynoose in Armagh ; and Annagoose Lake in Monaghan. 2 The evidence for this is too large a subject to be entered upon here. In certain peat-mosses no less than three distinct forests are discernible, occupying different depths ; and the uppermost forest, always of Finns sylvestris, probably dates from historic times. 3 Synopsis Methodica, 442 (1724). 4 State of the County Kerry, 372 (1761). 6 Catalogue of Plants in Ireland, 83 (1825) « Cybele Hibernica, 277 (1866). 7 Practical Treatise on Planting, 133, 167 (1794). Ill name in 580 The Trees of Great Britain and Ireland mountains of Auvergne, the Cevennes and Pyrenees, and is specially noted as occurring chiefly on the slopes with a southerly aspect.1 It is common in the eastern part of the Pyrenees, between 4000 and 6600 feet altitude. In Spain the common pine is also restricted to the mountains, only forming woods on northern slopes ; and in the Sierra Nevada, forms the large and splendid forest of La Granja on the north side of the Sierra de Guadarrama, where it ascends to 6500 feet.2 In Germany very extensive and pure forests of pine occur in the north-east, always on sandy soil in the plain. These forests are called heaths, as they contain wide stretches covered only with heather and many swampy areas. Such forests are common in the provinces of East and West Prussia, Pomerania, Brandenburg, Posen, upper Silesia, Saxony, and in the kingdom of Saxony. Other large forests of the same kind occur more isolatedly in north Schleswig, Hanover, Jutland, and Holland. In the valley of the Rhine, both in Alsace and Baden, very fine forests of pine are also met with, as at Haguenau, likewise on sandy soil. In the mountains of Middle and Southern Germany, the pine only grows in small groves or as isolated trees. Similarly throughout the Alps and Carpathians, in Hungary, and on both shores of the Adriatic the pine is rare, only occurring in small woods. The pine does not occur wild in the islands belonging to Denmark, and is totally absent from the Hungarian plain, the Bakony forest in Hungary, the Central Car pathians, Banat, and Slavonia, and is not met with in the alpine and subalpine regions of the high mountains of Central Europe. (A. H.) In Switzerland this tree does not seem to attain such a large size as in Scandinavia. A tree at Campodials in the Grisons, figured on Plate xi. of Les Arbres de la Suisse, is said to be 80 feet high by 10 feet 6 inches in girth at 4 feet from the ground. It grows on the edge of a forest at about 3000 feet elevation, on crystalline rocks. In Norway and Sweden the common pine constitutes by far the largest portion of the forest, and flourishes farther north than any other tree except the birch. Though truly virgin forest is now becoming a rarity in Norway, and in the more accessible parts of Sweden and Finland, yet the area of land covered with pine and spruce is still so large and so much better suited than England for the production of commercial timber that we shall, in my opinion, never be able to produce it of such good quality and at so low a price. I have seen Pimis sylvestris at its best in the forests of South-Eastern Norway in the valley of the Glommen ; where the bright yellow bark of the upper part of the tall trunks on the banks of the river is a marked feature of the scenery ; and in the far north in upper Saitenfjord where the oldest pines known to me now exist ; though here, as elsewhere, they are rapidly being felled to supply the great demand for building and mining timber. In the more central provinces of North and South Trondhjem, and on the coast, the pine does not seem to grow to such a great size, probably because the soil and climate are too wet to suit 1 It is not a native of the French side of the Vosges, but occurs on the German side at 1300 to 3000 feet altitude. It is not wild in the Ardennes or in the Jura. 2 Mentioned by Christ in Flore de la Suisse, 198 (1907). Cf. supra, p. 574, var. nevadensis. Pinus Sylvestris it as well as the spruce, for the pine is a lover of a sandy soil and a dry long winter, with a hot sunny summer. Dr. Schübeler, in his Viridarium Norvegicîtm, i. 375, gives many details about the pine, from which I gather that its range extends from the south, where it reaches an elevation of 3500 feet above the sea, to the inner valleys of Finmark, where in lat. 70° N. it attains in Alten and Porsanger fjords as much as 60 feet high and 7^ feet in girth. He tells us that formerly there were pines on the Dovrefjeld, near Jerkin, at an elevation of 3200 feet, as much as i foot in diameter, where no trees now exist; and that near Roros, now one of the bleakest and coldest towns in Norway, the forest was, in 1773, so dense as to be almost impassable. The tallest pines in Norway that he mentions were near Holden in Lower Thelemarken, where one was measured 104 feet high, with a diameter at the ground of z feet IQ inches, and at 70 feet high of 9^ inches. Another at the same place was 105 feet high, and 5 inches in diameter at 96 feet up. At Klosterskogen in Skien, one was measured 108 feet by 6 feet 5 inches at breast height. The greatest girth that he mentions is about 15^- feet. I have myself measured at Graddis in Junkersdal, within the Arctic Circle, and at an elevation of at least 1200 feet, pines of over 50 feet high and 12 to 13 feet in girth. One of these, which was cut down, was 34 inches in diameter and about 240 years old, but the outer rings were so close that I could not count them accurately, the first loo years'growth being over 26 inches in diameter, showing that" in this latitude at least the increase after this time is very slow. The tallest that I saw in this valley was 84 feet high near the Government Forest Nursery at Storjold. I observed that in Junkersdal the natural regeneration from seed was poor, and that in the upper parts of the valley the young seedlings were very small and stunted, and birch seemed to be taking their place. In this valley on July 10, 1904, vegetation had only just commenced, and the pines had not pushed their young growth, though Cypripedium Calceolus was in flower. A severe frost which took place in April, - 14° to - 16° Reaumur, after warm weather in March, had killed most of the young shoots where not protected by snow. Schübeler gives several illustrations of the curious forms which this tree some times assumes. His Fig. 59 shows a tree in which the branches are very short and which has the shape of a northern spruce rather than that of a pine. Fig. 60 shows a branch with a great bunch of forty to fifty closely packed cones surrounding it. Figs. 61 and 62 show the power which the tree possesses of sending out upright stems of considerable size from a fallen trunk whose roots still retain their hold on the ground. Fig. 63 shows an immense witches' broom, forming a dense mass of living twigs in a ball 10 feet in diameter, which surrounds the trunk of a pine growing at Aaseböstäl in Nordfjord. It is occasionally planted in Iceland,1 but does not long survive the severe climate, though Hooker was told that a single dwarf tree grew on an island in a lake between the head of Borgarfjord and Reyholt. As little is known with regard to the so-called Riga pine, which was for long 1 Babington, mjourn. Linn. Soc. (Bot.) xi. 50 (1870)." 582 The Trees of Great Britain and Ireland the most celebrated for masts and shipbuilding purposes, and has been found in France to be the best variety in cultivation, we may refer our readers to a recent publication by Von Sivers,1 with a map of the distribution of pine and spruce, which shows a comparatively small area of the former. The author states that though the pine is everywhere at home, it grows best on sand, especially where that is underlaid by good soil, and that in favourable places it reaches often a height of 150 feet. The area which is occupied by pine plantations in Estland, Livland, and Kurland is estimated at 638 square kilometres. It would, therefore, seem that the production of pine timber is not sufficient to continue the large export upon which in the past reliance could be placed. And though there are still large reserves of pine forest in Northern Sweden and Finland, yet it was stated by Mr. A. Howard at a recent meeting of the Society of Arts, in a discussion on Sir Herbert Maxwell's paper on Forestry, that the size of the deals imported from the Baltic is steadily diminishing, and that a much smaller proportion of n-inch boards is now sent than was formerly the case. In the forests of the lower valleys of the Altai Mountains in Siberia I have seen the pine attain a greater size than anywhere in Europe, some trees in the valley of the Biya river, a tributary of the Ob, which I observed in 1899, being estimated at 150 to 160 feet in height, and clean to 100 feet, at which height they looked as if they were 5 or 6 feet in girth.2 CULTIVATION 3 Of all the many species of pine, none is so widely distributed in Europe, so common all over Great Britain, so easy to grow as the Scots pine, or Scotch fir, as it is often incorrectly called. Its vigorous constitution and rapid growth when young enables it to exist and even to thrive in almost all situations, and though the variations which it has produced in a wild as well as in a cultivated state are innumerable, yet the most casual observer can hardly fail to distinguish it from any other species which is likely to be seen in cultivation. I have seen the tree in the greatest perfection on the sandy soils of Surrey, Sussex, Bedfordshire, and Notts, on the rich loams of the south-western and midland counties, on the dry sandy glacial deposits and heath-clad hills of the Highlands, and in many parts of Europe. Whether the Scots pine was at first principally propagated in England from native Scotch seed or from German seed is doubtful, and probably the earliest planted trees came from various sources ; but so far as my experiments have gone, it seems as though the seedlings grown from acclimatised trees are now more flourishing, and grow faster in the south of England than those from German, Highland, or Scandinavian seed. I have tried plants of the same age from all these sources in Gloucestershire, and have found those sent me from the New Forest the most promising in their younger stages. If rapidity of growth at first is any indication 1 Die Forstlichen Verhältnisse der Baltischen Provinzen, Riga, 1903. 2 Farther east, near Krasnoyarsk, a pine has been measured, which at 200 years old was 40 archines (93 feet) high, and II verschoks (ig| inches) in diameter ; but this is far surpassed by the pines found near Bélovège, where trees 150 years old are said to measure 60 archines (140 feet) high by 12 verschoks (21 inches) in diameter, and contain as much as 100 sagènes (about 250 cubic feet) of timber. Cf. Les Forêts de la Russie, Paris Exp. 1900. 3 Loudon's excellent account of the culture should also be referred to, pp. 2178-2183. Pinus Sylvestris 583 of vigour, I should prefer them, though I would not plant Scots pine as a forest tree on any soil where I could get larch to grow even fairly well ; and on dry chalk and limestone soils it never grows with the vigour that it does on sandy soils. Large parts of the open heath of the New Forest, though constantly pastured by horses, are becoming overgrown with Scots pine to such an extent that if they escape fire it seems as though they would eventually turn those open wastes into a more or less dense pine wood.1 But on clay soils, and wherever a rank growth of grass, ferns, or briars is found, natural reproduction is comparatively rare, and over the whole of the Cotswold Hills I only know of a few places where self-sown pines can be seen. If natural reproduction is desired, the best way of encouraging it is to uncover lines or patches of soil in the winter, on which the seed falling in April can germinate ; but the growth of these self-sown plants is, as usual with almost all natural seedlings, at first much slower than that of planted trees. In very old pine woods of loo to 150 years' growth, such as are found in Strathspey and in a few parts of England, the accumulated carpet of dead pine needles seems to prevent the young seedlings from establishing themselves ; and in the Belvidere plantation at Windsor Park, which is one of the finest in England, I saw no self-sown seedlings under the fine old trees, many of which are 100 feet and more in height. In such cases it is best to burn the heather or to graze it closely with sheep and cattle, and in many cases this is a necessary preliminary to preparing the ground for natural reproduction in Scots pine woods; but if the soil produces grass rather than heather, the regeneration is always less successful and requires more assistance. I shall not attempt to give any estimate of the financial results of planting Scots pine as an unmixed plantation, because the conditions of soil and climate are so varied that any estimates, such as we see commonly given in books on forestry, are usually misleading. On very sandy, dry soil it will probably pay as well or better than any other tree, because it can be planted so cheaply, and will regenerate itself so easily.2 But it must be kept thick enough to clean its stem before the branches get large, and in fact it may be better not to thin at all until 20 to 30 years old, when the weaker stems which will hardly pay to cut and carry out will be killed by their stronger neighbours. On high moorlands also it may be, and now often is, as profitable a crop as larch, because it grows well in windy and exposed situations ; but I would not plant it, except as a nurse to other trees, on any soil where experience has shown that a more valuable tree will grow to fair timber size, and the plan often adopted of mixing it in larch plantations on calcareous soil has led in many places to absolute failure. With regard to the possible yield of Scots pine in England, I have heard of nothing better than a part of the Dipton Woods near Hexham, Northumberland, the property of Lord Allendale. This was described in Trans. Scott. Arb. Soc. xx. 1 I was informed during a recent visit to the New Forest that the commoners already complain that the pasture is deteriorating from this cause. 2 I have seen no better example of natural regeneration than on the Duke of Bedford's property at Old Wavendon heath, near Woburn. 584 The Trees of Great Britain and Ireland 84 (1907), as containing something like 9000 cubic feet per acre, which at 4d. per foot works out at ^150 per acre. Mr. Gillanders informs me that the soil is fine fresh sand, the elevation 550 feet, and the aspect north-east. Some of the best foresters consider it an excellent nurse for oak, but beech is now usually preferred for this purpose, on any soil where the latter will thrive. From 70 to 100 years, or in the Highlands 120 to 150 years, is about the age at which the tree is usually mature for felling ; as, when younger than this, the timber is comparatively soft and inferior, though after creosoting it may be utilised for many purposes where strength is not important. Where pit-props are saleable, it is more profitable to cut the crop as soon as large enough for that purpose. PROPAGATION No coniferous tree is more easy to raise from seed, or easier to transplant. In Scotland the cones are usually gathered in autumn and the seeds extracted by kiln drying when required for sowing ; but they are better left on the trees till spring, and the seeds may then be easily extracted by damping the cones and exposing them to the sun till they open. The seeds will keep good for several years, and if not wanted to sow at once, are better extracted in the summer after they are ripe and kept until the following March, when they will germinate as readily as, and perhaps produce stronger plants than, those extracted by artificial heat in winter and sown the next spring. The best nursery practice is to sow them broadcast on slightly raised beds of sandy soil about three feet wide, and cover with about half an inch of fine earth, some of which may be raked off just before they begin to germinate, leaving a fresh surface uncaked by the rain and sun. If sown too thickly, the plants will be drawn up closely, and will not remain two years in the seed-bed without becoming crowded. Some people advise transplant ing at one year old, but in my experience two years is better, and, if carefully handled, the percentage of loss caused by transplanting is very small. If the plants are to be put out on heath or sandy land, the stronger ones may be permanently planted out from the seed-bed ; but in all soils which are grassy and weedy, it is better to keep them one or two years in nursery lines, which should be about i to i^ feet apart, and the plants 3 to 6 inches apart in the rows, according to whether they are intended to remain one or two years in lines. It is rarely desirable or necessary to allow them to remain more than four years in the nursery ; but if plants larger than i £ to 2 feet are required for special purposes, they must be transplanted when four years old and put in rows about one foot apart and two feet apart in the rows. The best time for planting out large trees is in the autumn, as soon as the terminal buds become hard ; but small plants should not be transplanted till after the period of severe frost has passed, or they will in most soils be lifted by frost. If, however, it is necessary to do so, stones should be put round the collar of the tree, not only to keep them fast in the ground, but also to keep out the drought during Pinus Sylvestris 585 the first year. I have found that this is a very successful method to adopt with all small trees on stony soil liable to drought. An account of the best way of growing the Scots pine from seed was written by the Earl of Haddington in 1760 to his grandson, and is quoted in the Highland and Agricultural Society's volume on the Old and Remarkabk Trees of Scotland, published in 1864. This account is very practical and based on personal experience, and interesting as showing how much care was taken by the planters of those days to ensure good results. REMARKABLE TREES As to the height the Scots pine attains in Great Britain, many particulars have been given by Loudon, which in most cases cannot be relied on for accuracy, but we have reliable measurements which show that the tree rarely exceeds no feet, and more usually is not over 100 feet. In the Vûtoria County History of Hants, it is stated in vol. ii. p. 469, that trees 130 feet high were growing at Beaulieu, but Lord Montagu tells me that he has never actually measured one over 116 feet, of which height one was blown down some years ago. I saw these trees in June 1906, and though many exceed 100 feet, and are clean to 70 or 80 feet, with a girth of about 7 feet, I could find none over no feet. At Rooksbury Park, near Wickham, Hants, there are some which, I think, are taller, growing, mixed with beech and oak, in a dense thicket of rhododendron. The largest I measured here was about 115 feet by 10 feet 4 inches, dividing at about 17 feet into three tall, clean stems. At Carclew, Cornwall, the seat of Colonel Tremayne, there is a fine avenue of pines, the tallest of which I found to be about no feet (Plate 160). At Pain's Hill, Surrey, Henry measured a tree of 106 feet. In the Belvidere plantation in Windsor Park, one of the finest old pine woods in England, planted about 1760, there are many trees of 100 feet and some perhaps a little more. There were some very tall trees at Hursley Park, Hants, of which I have no exact measurement, but I hear that few, if any, of them remain. At Buckling Hall, Norfolk, the property of the Marchioness of Lothian, there is an immense tree, perhaps one of the oldest in England, which, when described by Grigor in 1841,' was 70 feet high and 16 feet in girth at 6 feet from the ground. He thought it the largest tree of the kind in Norfolk. When I saw it in April 1907 it was 96 feet by 17 feet i inch, dividing at about 10 feet into two main trunks, which were chained together 40 feet up. It had a large burr at the base. At Stratton Strawless there are some fine trees planted about 1740 by Robert Marsham, measuring about 100 feet by 9 feet. A tree was reported by Loudon at Castle Howard, Yorkshire, as being 120 feet high, with a bole 100 feet long, but I could not identify this tree as still living in 1905. At Cocklode House, in Sherwood Forest, there is a fine avenue of Scots pines about 160 years old, which are 90 to 100 feet high, and 9 to 10 feet in girth, but many of them have been blown down. The tallest that I have ever seen or heard of is in the grounds at Petworth, 1 The Eastern Arboretum, p. loo. 586 The Trees of Great Britain and Ireland Sussex, the seat of Lord Leçonfield, where the soil seems particularly favourable to very tall trees. A careful measurement of this in 1905 gave the height as 120 feet by 11 feet in girth, and a bole of 35 to 40 feet, where it divides into two stems. Sir Hugh Beevor, who saw it in 1904, did not make it quite so tall. This tree appears in the foreground of Plate 162. No park in England contains a greater number of fine and picturesque old pines than Bramshill Park, Hants, the seat of Sir Anthony Cope, who tells me that he believes them to have been planted about the year 1600, and to be some of the oldest in England. The soil here is very light and sandy, and the oldest pines are in avenues, which have become rather irregular in course of time. The tallest trees that I measured here were not over about 80 feet high by 10 to 12 feet in girth, but there is one splendid tree in the Gravel Pit drive which is about 80 feet by 16 feet, of which I give an illustration (Plate 161). There are many self- sown seedlings of various ages in this park, but no other trees of remarkable size.1, At Dyrham Park, Gloucestershire, the seat of the Rev. W. T. Blathwayt, there is a very fine tree, of which I am indebted to the owner for a photograph, which measures about 73^ feet high by 14 feet 9 inches in girth, dividing at 4 feet into three trunks. There are some very fine clean Scots pines in Stowe Park, near Buckingham, one in the Queen's Quarter being over 100 feet high, with a clean bole over 60 feet long and n feet 3 inches in girth. In the Fir Grove, at Bayfordbury, Herts, there is a tree,2 with a clear stem of over 50 feet, which measured in 1905, 95 feet high by 9 feet 7 inches in girth. In Wales I have heard of no Scots pine of greater size than one at Penrhyn Castle, which Henry measured in 1904, and found to be no feet by 7^ feet, and about 70 feet to the first branch. At Gwydyr Castle he measured one about 85 feet by 11 feet 2 inches on which a mountain ash seedling was growing. In Scotland there are so many fine old trees that it is impossible to mention more than a few of them. Perhaps the finest, if not the tallest, is a tree at Inveraray, of which a beautiful photograph by the late Vernon Heath is in the museum at Kew. I measured this tree in September 1905, when it seemed to have changed very little in appearance, and though supported by chains above the fork, is very sound and healthy. It measures no feet by 14 feet, forking at about 35 feet, and leaning considerably to one side. The Duke of Argyll informs me that it was probably planted about 1620. Plate 162 shows the present appearance of this tree. There are many other very fine Scots pines at Inveraray on the lower slopes of Dun-i-cuach, but none equal to this in height or girth. On the banks of the Tay, near Dunkeld, there is a very graceful tree of weeping habit though of no great size (Plate 163), which measures 77 feet by n feet 6 inches in girth ; and at Blair Atholl there are some curious old pines in a row 1 Bunbury, who visited Bramshill Park in 1859, mentions the tradition that these trees were introduced by James I. from Scotland at the same time that he began building Bramshill, and states that there were three magnificent Scots pines at Eversley Rectory, which were coeval with those of Bramshill. Cf. Lyell, Life of Sir C. J. F, Bunbury, ii. 138, 139 (1906). 2 This tree measured, in 1816, 5 feet 8 inches in girth, according to an entry in an old note-book, now in the possession of Mr. H. Clinton Baker. Pinus Sylvestris 587 by the Inverness road, which the Duke of Atholl informs me are probably part of the booty carried off by his ancestors in 1684 from Inveraray, as described in Chronicles of the Atholl Family, by the present Duke.1 Nothing can better illus trate the importance which was paid to trees and planting even at this early period, when the Highlands were hardly civilised ; than that so many exotic trees should have existed at Inveraray, and that it should have been thought worth while to carry them to such a distance when wheeled carriages could not have traversed the country. An immense Scots pine, which I have not yet been able to visit, grows at Guisachan, Inverness-shire, now the property of Lord Portsmouth, whose forester, Mr. Davidson, informs me that in February 1907 it was 53 feet 10 inches high and 16 feet 8 inches in girth at the ground, and 15 feet 7 inches at 5 feet. At n feet from the ground, below the first branch, it is 16 feet 10 inches in girth. The trunk has been cut into at the base, which is believed by old people living near to have been done by smugglers, as an illicit whisky-still once existed near it. A drawing of this tree was made for the late Sir Dudley Marjoribanks, of which I have a copy. Mr. E. Ellice tells me that there are a number of very large old pines in the Guisachan Woods, girthing over 14 feet. Mr. E. Ellice of Invergarry informs me that there are a considerable number of old native Scots pines at that place, among them one which attracted the late Mr. Gladstone's attention, and of which he sends me a sketch, with the following measurements :—Height, 70 to 80 feet ; girth at the ground, 20^- feet ; at 5 feet, 16 feet 3 inches; at 10 feet, 15 feet 9 inches ; at 13 feet, 17 feet. A figure of this will appear in Vol. IV. Other trees near it measure 14 feet 3 inches, 12 feet 10 inches, and 12 feet 9 inches ; and these appear to be the parents of many more which may be divided, according to their age, into three classes : those of 120 to 150 years, of which there are some hundreds; those of from 80 to 100, of which there may be 15,000 to 20,000 ; and younger trees. The finest forest in this locality is in Glen Malie, on Lochiel's property, 1 " In 1684 or 1685 the Marquis of Atholl did carry out of the orchard enclosures and shrubberies at Inveraray— £ s. ci. Scots 600 Silver and Spanish fir trees, 6 years' growth . . ... 1800 500 Pinaster trees, 12 years' growth . ... 500 500 266 13 4 iSoo 600 400 120 200 266 13 4 200 500 Pine trees, 10 years' growth 400 Yew trees, 16 years' growth 6000 Holland trees (holly) 600 Beech trees 2000 Lime trees, 4 years 400 Buckthorn, 8 years 600 Black and White Poplars, 13 years 400 Chestnut 200 Horse Chestnut 300 Walnut 200 Fir trees, 5 years 20,000 Ash, Plane, and Elm trees 200 Pc.ir and Apple trees 200 Plum trees 300 Cherry. looo Apple and Tear stocks This claim was settled for £13,000 Scots or £1333, 6s. 8d."—Chronicles of AthoU Family, i. 265. Ill zoo 400 2400 400 20O 300 3000 j£i3.553 6 8 Scots. 588 The Trees of Great Britain and Ireland running up from the shore of Loch Arkaig for five miles, and in it there is one tree, even larger than that last mentioned, of which Mr. Ellice has sent me a sketch. At the narrowest part of the trunk, three feet from the ground, this tree measures 18 feet 8 inches in girth, and at about 10 feet divides into three tall trunks, each of which girths between n and 12 feet. Just below this fork it girths about 30 feet, and appears to be sound throughout. At Novar, in Ross-shire, there is an old plantation containing a number of very fine Scots pines, one of which measures 105 feet by 10 feet 3 inches, and larger ones can probably be found in this district, as well as in the sandy district which extends east from Inverness, where many large plantations of this tree flourish exceedingly. The finest individual trees and the finest Scots pine plantation that I have seen is in a place called Wishart's Burn, near Gordon Castle, Banffshire, on red sandstone soil. Though supposed to be about 180 years old, most of the trees are still in good health and quite sound, though wind has made some gaps in the plantation. When I visited them in April 1904 the tallest tree was about 117 feet high by IG feet n inches in girth. It forks at about 45 feet, but carries its girth so well that the bole would, I think, measure 45 feet by 28^ inches quarter girth, about 245 cubic feet, and the tops might contain 50 feet each, making a total of 345 cubic feet (Plate 164). Another tree standing near it was 114 feet by 8 feet 10 inches, and I estimated that the older trees here average over 100 feet high by 8 feet in girth. Mr. Webster, gardener to the Duke of Richmond, who showed me this beautiful spot, agreed with me that the average number of trees to the acre here was about sixty, and their average contents about 100 cubic feet ; but many have been cut and sold at as much as £j : los. each, to make masts for large herring boats. One of these trees probably is the one figured by Loudon (p. 2162) as a model of a fine Scots pine clear of branches to 50 feet, and containing 260 feet of timber. There are also very fine plantations of Scots pine in the neighbourhood of Castle Grant, the seat of the Dowager Countess of Seafield, in Inverness-shire, a place celebrated for its good forestry, and where better examples of thickly grown self-sown pine may be seen than anywhere else in Scotland. Mr. Grant Thomson, who has had charge of the extensive woods here for forty-five years, told me that the oldest planted trees are about 180 years old ; many of these are 80 to 90 feet high and 8 feet in girth, and number sixty to seventy per acre. Some have already begun to decay at the heart, and it was noticeable that on the thick bed of decayed pine needles under them seedlings would not grow. This has been referred to by Prof. A. Schwappach in a paper on the " Forests of Scotland " in Trans. Roy. Scot. Arbor. Soc. xv. 13 (1898). In this neighbourhood are the most celebrated and extensive natural forests of Scots pine in Great Britain, which I visited in April 1904. Glenmore forest, the property of the Duke of Richmond, was perhaps the best of these until 1783, when a great part of the mature timber was sold to an English merchant named Osbourne, who cut it down in twenty-two years and floated the timber to Spey- Pinus Sylvestris 589 mouth, where forty-seven ships of upwards of 19,000 tons burden were built from them at an expense for labour only of .£70,000 (cf. Loudon, p. 2161). Glenmore Lodge lies at an elevation of 1050 feet on the shore of Loch Morlich, where some of the finest pines still stand. One of them (Plate 165) is interesting as having for many years been the eyrie of an osprey whose nest is visible in the photograph, which was taken in the interval between two snowstorms ; but the birds, though carefully protected, have not bred there since 1900.* This tree measures 56 feet by 13 feet, and is very characteristic of the native Scots pine in its habit. But perhaps the most interesting tree in this forest is one from which a plank, now preserved at Gordon Castle, was cut and presented by Mr. Osbourne to the Duke, in 1806, as a memento of the forest. I could not count the rings of wood in it exactly, but the Duke of Richmond informs me that there are about 236. I measured the plank 5 feet 5 inches wide at the butt end and 4 feet 4 inches at the top. The sapwood is worm-eaten, and the colour of the wood has become very dark. I saw still lying on the hillside above Glenmore Lodge, near the upper limit of the Scots pine, at an elevation of about 1400 feet, a huge top, over three feet in diameter where it was cut off, and was assured by Francis M'Pherson, an old woodman, who showed it to me, that it was the top of the identical tree from which the above-mentioned plank was sawn. Though overgrown with moss and heather, much of the wood appeared to be still sound, after lying for nearly a hundred years. In confirmation of this I may state that Mr. J. Michie showed me, in Ballochbuie Forest, the remains of a pine which was sawn up and found sound after lying seventy years on the ground. In Abernethy Forest there are also many fine old pines, one of which, Mr. Grant Thomson's favourite, is shown in Plate 166. It measures about 60 feet high by 14 feet 3 inches in girth, and, though it divides into five tops, is a most graceful tree. Much of this forest was burnt down many years ago, but has become self-sown with young trees, and is now open wood covered with long heather, and a favourite wintering ground for deer. We measured a group of the best clean self- sown trees supposed to be about 120 years old, and estimated them at about 120 per acre, with an average timber height of 40 to 50 feet and a cubic content of about 25 feet (Plate 167). Such trees, where they stand, are worth about 6d. per foot. Many cones are gathered in this forest for seed, of which about 8 ounces from a bushel is the average produce ; and there is a large nursery where they are raised, the growth being very slow as compared to what one sees in England, on account of the cold and damp situation. I next visited Ballochbuie Forest, by the kind permission of His Majesty, who preserves this beautiful forest with great care. It is now perhaps the largest area of natural forest in Scotland, extending for several miles along the south side of the upper valley of the Dee. The photographs, Plates 168 and 169, give a good idea of the picturesque scenery of this forest and of the fine trees in it, many of which are 80 to 90, and some as much as 100 feet high, by 7 to 8 feet in girth. The 1 Mr. S. R. Clarke has sent me a photograph of a Scots pine at Fasnakyle, which is used annually as an eyrie by the Golden Eagle. 590 The Trees of Great Britain and Ireland best of those which I saw might measure 60 feet timber height by 18 inches quarter- girth or 135 cubic feet. In the best stocked areas the trees might average 120 per acre at 120 years old, of which forty trees of the first class would average perhaps 40 feet, forty second class 25 feet, and forty of the third class perhaps 12 feet each, or about 3000 feet to the acre. But these figures are only a rough estimate, as the King's trees are not cut for sale, and in consequence Mr. Michie could not give me exact figures, but thinks they are worth from 6d. to 8d. per foot standing. The Black Wood of Rannoch is an ancient natural pine wood extending for about three miles along the south shore of Loch Rannoch, and though there are no trees of exceptional size, there are many very picturesque ones, which are protected by the owner, B. C. Vernon-Wentworth, Esq., whose residence at Dell is near the east end of the wood. The largest which I measured was 91 feet by 11 feet 7 inches, with a very spreading base 15 feet 5 inches round at one foot from the ground. The greater part of the wood is open and covered with long heather, among which seedlings were fairly numerous wherever the soil was exposed. Many of the large trees were blown down thirteen years ago, and their timber, which was of very fine quality, was used by Sir J. Stirling Maxwell in the interior work of his house at Corrour. There is an excellent account of the Black Wood of Rannoch and its history in chap, xxxiv. of that admirable book, Hunter's Woods and Forests of Perthshire, which, though now getting out of date, as it was published in 1883, gives a better account of the great estates and their trees than exists for any other county in Great Britain. In this work dimensions are given of the finest trees then existing in the Black Wood, which are remarkable more for their great girth, spreading and massive branches, and picturesque appearance than for their height. On the shores of Loch Hourn, on the west coast of Inverness-shire, there are many native pines scattered among the birches, but none of large size, a few of those near the sea resembling the stone pine of Italy in habit. Henry observed that many of these trees do not ripen seed. With regard to the elevation at which the Scots pine grows in Scotland, we have various somewhat conflicting estimates. Mr. Michie tells me that Craig Doin (1900 feet) is about the highest level he knows it to reach in Ballochbuie. Mr. Seton P. Gordon, however, says1 that he has seen a young Scots pine growing at a height of about 2700 feet not far from the source of the river Dee on the south slopes of Brae Riach, though he regards this as very exceptional. Mr. Hugh Boyd Watt2 also considers 2700 feet quite an unusual altitude, and says, " From personal observations made on and around the Cairngorm mountains (and in no other district in this country do forest trees attain higher levels) I can say that even at 2000 feet above sea-level the Scots pine has difficulty in holding its ground. . . . On the southern slopes of Beinn a' Bhuird (Glen Quoich) considerable numbers of fairly well-grown Scots pines reach up from 2000 feet to 2100 feet, and I know no other place where what may be called the forest line is so high. ... In other localities, apparently favourable to their growth, the pines do not in any numbers exceed an 1 Country Life, i;th Aug. 1907, p. 245 ; 3ist Aug. 1907, p. 322. * Of. cit. 7th Sept. 1907, p. 359. Pinus Sylvestris 591 altitude of 1500 to 1700 feet. This is approximately the level at which they die out in Abernethy, Rothiemurchus, Glen Feshie, Glenavon, Invercauld, Birse, and Glen Tanar." When stalking on Ben Avon I saw with the telescope some pines in the upper part of Glen Derry which I supposed to be at an elevation of about 2000 feet, and Mr. Michie, who has seen these trees, thinks that this estimate is not far from the mark. (H. J. E.) In Ireland the common pine grows with great vigour and beauty, the bark becoming bright red in colour and the leaves very glaucous. The tallest trees, which I have seen, are at Curraghmore, the seat of the Marquess of Waterford, where, near a stream, I measured one no feet high and 7 feet in girth ; some, but difficult to measure accurately on account of their position in a dense wood, were probably 120 to 125 feet in height, the largest of these having a girth of 9 feet. At Doneraile Court, Co. Cork, there are some fine pines, growing scattered in an oak wood, the largest of which I made 97 feet by 11^ feet, with a clean stem to 50 feet. These trees are supposed by local tradition to be of native origin, and are called Irish pines; but they have evidently been planted, and there is no means of determining whether they originated from seed collected in Kerry from aboriginal pines still existing there in the i8th century, or, as is more probable, from Scotch seed, as they are probably about the same age as the famous larches at this place, which are reputed to have been sent to Doneraile by the Duke of Atholl. At Emo Park, Portarlington, there are many fine trees, the largest seen measuring 91 feet by 7^ feet and 88 feet by 9 feet i inch. There is also a splendid tree, growing near the gate of Mr. Walpole's beautiful garden at Mount Usher, on the Rossanagh property, which is 11 feet 9 inches in girth, and probably 80 feet in height. At Castledawson, Co. Derry, an old tree measures 80 feet in height by 11 feet 4 inches in girth. There are many fine trees scattered through Coollattin in Wicklow. These grow on moist boggy soil ; and I measured two clean of branches to 60 feet, which were 87 feet in height, and 9 feet 5 inches and 8 feet i inch respectively in girth. At Luttrelstown, near Dublin, Hayes1 measured a "Scots fir, eighty-five years' growth from the seed, of 11 feet 6 inches in circumference, and another of very great height ii feet 10 inches round." He gives several other instances of the rapid growth of the tree in Ireland.2 Mr. T. W. Webber, late Deputy Conservator of Forests in India, in the appen dix to his book on the Forests of Upper India, gives an interesting account of the growth of Scots pine in Ireland, the planting of which he strongly advocates. To the objection that home-grown timber is of inferior quality, he replies that the wood of Pinus sylvestris found in bogs in Ireland is often of great length and thickness, sound, fine-grained, solid and straight, and so excellent that it has been used by coach-builders as superior to Memel timber. Where such timber grew ages ago, 1 Practical Essay on Planting, 133, 167 (1794). 2 At Powerscourt an immense Scots pine was blown down by the great gale of February 1903, which I saw on the ground soon afterwards and which measured about 12 feet in girth. Some boards cut from the tree were kindly sent me by the late Lord Powerscourt, which show its growth to have been very rapid." (H. J. E.) 592. The Trees of Great Britain and Ireland similar material might be produced to-day, if close planting and slow growth were the rule. To prove this, he gives the actual dimensions of Scots fir grown under two different conditions in Ireland. Grown thirty to the acre, with spreading crowns. Girth Height Age. Diameter . Heart wood Sapwood . Rings per inch 5 feet 50 ,, 40 years 20 inches 12 8 „ 4, uneven These trees are quickly grown on deep soft soil, and are liable to be blown over. Timber, coarse, knotty, light, and perishable ; large amount of sapwood. Grown 200 to the acre, with small crowns. 5 feet. 75 » loo years. 20 inches. 19 i inch. io, regular. These trees were slowly grown on a hill-side on poor and stony soil ; standing close they resist storms. Timber fine grained, hard, heavy, durable, and equal to best Memel. Scarcely any sapwood. Mr. Webber has kindly written to me that the trees just mentioned grow on his own property at Kellyville, near Athy, in Co. Kildare. A beam, made out of the fine pine timber grown on the hill-side, placed in the front of a conservatory twenty-five years ago, is still sound and good. Mr. Webber has Scots pine thriving on pure rock, where there is little or no soil. He states that at Emo Park near Portarlington and on the road to Maryborough there are striking instances of pine succeeding on pure black bog, and self-sown seedlings maybe seen spreading all over the turf-moss. He reiterates the conclusions given above, namely, that the pine should be planted densely on poor soils, where it will resist the wind and yield timber without any appreciable sapwood, whereas on deep soft soils it is easily blown over and yields coarse and valueless timber. In the bog in Emo Park, Mr. Webber found great bases of Scots pines with their roots in the boulder clay, of gigantic size, showing that the tree was indigenous before the bog began to grow ages ago. In some parts of Ireland, Scots pine may be seen thriving on deep peat-moss, the condition necessary for success being judicious preliminary drainage. In mosses soaking with water, trees languish and die on account of the lack of air at their roots. On the other hand, if the drainage is too deep, the upper layer of the peat becomes so dry, that the trees suffer from want of water. Near Castledawson in Co. Derry, a considerable area of undrained peat-moss is covered by healthy and vigorous pine trees, which are natural seedlings, the product of seeds blown from an adjoining plantation. Here, however, the peat-moss rests on the side of a sloping sandhill and is not waterlogged. Natural pine seedlings are often seen on peat-mosses, struggling for life in the wettest situations ; and doubtless, if cattle and rabbits were excluded, these would in time take possession. At Churchill in Co. Armagh, the property of Harry Verner, Esq., considerable plantations of Scots pine, intermixed with a small proportion of larch, were made in 1861 on deep peat-moss, which had been thoroughly Pinus Sylvestris 593 drained. These trees, planted 3^ feet apart, are now forty-four years old and average 44 feet in height by 3 feet in girth. Two-year-old seedlings, one year transplanted, were used and a system of pitting was adopted. The holes were made about a foot deep, and were filled in with a mixture of clay and peat. The clay was brought from a distance, and no doubt its use added considerably to the cost of planting. Possibly peat-ashes, obtained by burning peat, heather, etc. on the spot would have answered equally well, and been less costly. The Scots pine succeeds better than any other tree on pure peat-moss, though alder and larch may be added in a certain proportion. At Clonbrock, in Co. Mayo, on an overcut bog, where the peat left uncut was 3 to 4 feet deep, Scots pine eighty years old averages only 47 feet in height by 4 feet in girth. Probably the lesser growth in this case is due to insufficient drainage. As there are immense areas of peat-mosses in Ireland, now yielding no return whatever, the possibility of afforesting them with Scots pine, or with a mixture of Scots pine and larch, is an important question ; and the success of the Churchill plantations is encouraging.1 Throughout Ireland there are extensive mountain tracts of barren land, covered with stones and rocks, which are of merely nominal value for grazing and are im possible to reclaim for agricultural purposes except at a ruinous expenditure. The Scots pine renders excellent service in turning these wastes to account. The late Lord Powerscourt made extensive plantations on the hill-sides of Co. Wicklow at 500 to 900 feet above sea-level, which paid handsomely. These plantations consisted in the main of a mixture as follows :—200 larch, 1500 Scots fir, and 500 spruce per acre, the plants being notched in, as, in Lord Powerscourt's opinion, they came on eventually as well as those which had been pitted at a much greater expense. The Scots fir have been gradually thinned out, the larch being left as the final crop. Lord Powerscourt was favoured by ready access to the sea, and by proximity to Wales, where his thinnings were readily sold as pit-props. He estimated that the initial cost of planting and fencing is /4 to ^5 per acre, and that, during the first twenty to twenty-five years, the thinnings pay for the expense of cutting and the interest on the first cost. After that the thinnings should bring in annually eight shillings an acre ; the final crop of larch at fifty years being probably worth about ^50 an acre. (A. H.) In the United States the Scots pine has been planted with more or less success, but does not seem likely to be as valuable for timber .as the native pines. The largest I saw was in the Wellesley Arboretum, near Boston, which was 49 feet high in 1904. In Professor Sargent's grounds it seems to be short-lived, only living for thirty to forty years. Ten miles from Boston, however, near Ponkapoag, it succeeds better on dry sandy soil, and I found some self-sown seedlings. At the Central Experimental Farm, near Ottawa, trees planted in 1888 were about 30 feet high in 1906, but Mr. W. T. Macoun2 reports that it suffers much from shade, and does not grow so fast as Norway spruce or European larch ; though he recommends it for nurses to other trees, and for producing fuel. n Q M^,pIantati°nS °n bog land at Knockboy, Co. Galway, were badly made, and turned out a disastrous failure Cf Dr. Schlich s report in KeW Bull. ,903, p. 22 . ana in Tra,,s. Roy. Scot. Arbor. Soc. xvi. pt. Ü. 249 (,901) Canadian Forestry Journal, iii. 77 (1907). 594 The Trees of Great Britain and Ireland Cases of inosculation are rare among pines, but a remarkable instance of this was pointed out to me by Mr. Savile Foljambe in the Catwhins, near the lodge leading out of Thoresby park into the Retford road. It seems probable that when the trees were young they had come in contact, and eventually fused ; the iron bands were put on afterwards, but the trees are now dead. Another case, somewhat similar, occurs in a pine tree growing on the estate of Chenevières, near Montbour (Loiret), France, which a photograph kindly sent me by M. Maurice de Vilmorin illustrates. Here it seems that one tree had forked at or close to the ground, and become connected by a thick branch at a much later period. A third instance of natural inarching in the Scots pine is described and figured by Count von Schwerin1 from a tree near Teltow in Germany. Tn this instance a branch of one tree grew into the bark of another and broke off, eventually forming just such a living connection between the two trees as is shown in Vol. I. Plate 4, but much thicker in proportion. The sap of the left hand tree appears to pass through this branch to the other, as the stem is thicker above the junction, and the branch has assumed the yellow bark of the upper part of the trunk. The large, usually globular masses of dense shoots which sometimes appear on this species, and more rarely on larch and spruce, are not caused by a parasitic fungus. Prof, von Tubeuf2 says that their origin is unknown, no insect or fungus having yet been discovered which might have caused the growth, which is composed of a mass of small buds, producing densely crowded tufts of short leaves. A specimen which was found at Schwarzenraben in Germany measured 53 centi metres in height and about the same in diameter, the weight of this mass being over eleven pounds.8 Such growths are not uncommon4 in England, and I have a photograph of one on a tree at Colesborne, which was about a foot in diameter. TIMBER On the timber of the Scots pine so much has been written that I will refer specialists to Laslett,5 who gives a long account, mostly from a shipbuilder's point of view, of the various foreign varieties known to him as Dantzig, Memel, Riga, and Swedish fir; but'makes no reference to the quality of native-grown timber, which, though men-of-war were built from it by Osbourne in the last century, seems to have been unused by the Admiralty since then, as it is now by the Post Office authorities in England, and by architects and builders generally. The reason of this is, no doubt, that the rapid growth of the tree in this country, in our mild climate, causes the wood to be much softer, 1 Mitt. D. D. Gesell. 194 (1906). 7 Ibid. 222, fig. 13 (1905). 3 Count von Schwerin, Mitt. D. D. Gesell. 222 (1905), says that in Bavarian Allgau, between Oberstaufen and Weiler, he has seen a forest of sixty-year-old spruce in which almost every tree was more or less affected by these growths, and supposes that the cause, whatever it is, must be contagious. He has seen similar growths on Picea Orientalin and suggests that some of the horticultural monstrosities such as Picea excelsa echiniformis and C. Lawsoiiiana forsteckensis have originated from a similar cause. 4 A specimen from a tree growing at Hunstanton Hall, Norfolk, was shown at a meeting of the Scientific Committee of the Royal Horticultural Society in April 1899. Cf. Card. Chron. xxv. 270 (1899). D Timber and Timber Trees, ed. 2 (London, 1894). Pinus Sylvestris 595 coarser, and less durable than that from North Germany, Russia, and Scandinavia ; whilst over-thinning causes it to be much more knotty. It is hardly possible to believe that the same tree can produce timber so different as examples which I showed at a lecture on English timber at the Surveyors' Institute, on 22nd February 1904, taken from an immense tree grown at Powerscourt, of about 12 feet in girth; and the beautiful fine-grained wood which I brought from Northern Norway, and which when well planed shines with a silky gloss. Every intermediate form may be found in this country ; but, as a rule, it is little valued in England except for mining timber, for cheap fencing, packing cases, and other uses. In Scotland, where it is as a rule slower grown, and better in quality, its value is kept down by foreign importations, though it is very largely used in making staves for herring barrels and many other purposes. But when old enough to have produced a large proportion of red heartwood, and free from knots, I have sold Scots pine for as much as 8d. per foot, and have found it very useful and durable timber for roofing and many other estate purposes ; and I am informed by Mr. Mitchell, forester to the Duke of Bedford at Woburn, that in that neighbourhood, where it is of good quality, it is largely used for rending lath, and that the buyers will give a high price for it when suitable for that purpose. Loudon gives many interesting particulars of the uses and quality of pine timber at home and abroad, and quotesl Mr. T. Davis who, in 1798, was the Marquess of Bath's forester at Longleat, to the effect that a cart-house, built from it on that estate, remained perfectly sound after eighty years' use. And Pontey, in his Forest Primer, published in 1805, als° defends the Scots pine against the "almost universally prevalent prejudice against it, which is no doubt based partly on ignorance and partly on the fact that it is often used when too young and unseasoned." But, whatever may be said against the wood when grown in the south of England, there can be no question that the Highland native pine timber, when clean, is a valuable, and in some cases also a very ornamental timber. I have seen at Castle Grant a very beautiful sideboard made on the estate, which showed the curiously twisted red and yellow grain which Mr. Grant Thomson tells me is only produced by the self-sown native trees ; and I am indebted to the kindness of Lady Seafield for some of the same wood, which was cut in the Forest of Abernethy. The entrance hall and a room at Balmoral Castle have been recently panelled with the same sort of wood, which has a very ornamental effect, and Mr. Michie tells me that it has also been largely used for internal decoration in Mar Lodge. I saw in the house of the postmaster of Bodö, in Norway, a very handsome table made from a variety of the wood, which is there called " Rie," and which seems to be caused by a disease in the tree producing excrescences and distortion in a part of its trunk, and I possess a piece of this wood which is so unlike pine wood that no one could recognise it as such. But these trees seem to be as rare in Norway as in Scotland, and command a high price locally. The oldest example of this wood in the form of panelling that I have seen or heard of is in the room known as Queen Mary's room in Castle Menzies, which was 1 Of. cit. 2168. Ill 596 The Trees of Great Britain and Ireland occupied by Mary Queen of Scots in 1577. This is now perfectly sound, and much darkened in colour by age. The width and somewhat knotty character of the boards tend to confirm what the late Sir R. Menzies told me, that it was made from locally grown Scots pine, which may have come from the Black Wood of Rannoch, or from other native forests that have now disappeared. It is remarkable that, notwithstanding all that has been written since the Earl of Haddington first raised the question as to the existence of different varieties of this tree in his Treatise on Forest Trees, in 1760, there seems to be as yet no exact knowledge as to whether the different kinds of timber produced by different trees are, as I believe, individual variations, largely due to soil, or whether, by sowing seeds from trees possessing superior qualities, they may be reproduced in other soils and situations. The best and most exact records of such experiments that I know of are given by M. M. de Vilmorin in his account of the varieties of Pinus sylvestris collected by his grandfather at Les Barres in France, and published in the Catalogue des végétaux Ligneux sur le Domaine des Barres (Paris, 1878), which show how much one family have done for the better knowledge of the economic value of trees, and for the benefit of their country. Briefly, this trial, extending over a period of over sixty years, shows that, in the soil and climate of Central France, the Riga variety of P. sylvestris has, on its first introduction as well as in the second generation, preserved its superiority over other varieties of the same tree—from the various parts of France, from Haguenau on the Rhine, from Switzerland, and from Scotland—by its good growth, freedom from branches, quality of timber, and facility of transplantation. Though this superiority might not be as marked in England, it points to the necessity of careful trials of seed from Riga which, so far as I know, have not yet been made in this country. In Scott. Arb. Trans, ix. 176(1881), there is a very valuable paper by J. M'Laren and W. M'Corquodale, on "The Supposed Deterioration of the Scots Pine"; it having been stated by George Don and other writers that there were two or more varieties, one of which was very inferior to the other. They conclude, after reviewing the experience of many competent foresters, that the quality of the timber depends on the subsoil and the climate more than on the variety, and that the rich red resinous timber, for which the Highland pine is distinguished, is not to be expected in the south. They say that, since the days of the fine old Memel pine, there is no pine timber imported to our country equal to the old Highland pine, and that what has caused it to fall into disrepute is that it is grown too fast and cut too young, coupled with the fact that it is more difficult to manufacture and dress than foreign timber. (H. J. E.) CARYA Carya,1 Nuttall, Gen. Am. ii. 220 (1818); Eentham et Hooker, Gen. PI. iii. 398 (1880). Scoria, Rafinesque, Med. Repos. New York, \. 352 (1808). Jficorius, Rafinesque, Fl. Ludov. 109 (1817). Hicoria, Rafinesque, Alsog. Am. 65 (1838). DECIDUOUS trees belonging to the order Juglandacese. Branchlets with solid continuous pith, not chambered as in Juglans and Pterocarya. Leaves alternate, compound, unequally pinnate, without stipules ; leaflets sessile or sub-sessile, serrate, penninerved. Flowers monoecious, without petals. Staminate catkins slender, drooping, in threes on a common penduncle or clustered and sessile or subsessile, arising either from buds in the axils of the leaf-scars of the previous year's branchlets or from the base of the current year's shoot, and appearing after the unfolding of the leaves ; flowers numerous in the catkin ; calyx two- to three-lobed, subtended by an ovate bract ; stamens three to ten, filaments short. Pistillate flowers, two to ten, in a cluster or spike, terminal on a leafy branchlet of the year ; ovary superior, one-celled, surrounded by a four-lobed cup-shaped involucre, formed by the union of a bract and two bracteoles ; calyx one-lobed ; stigmas two, sessile ; ovule solitary. Fruit, a nut, enclosed in a four-valved, thickened, hard and woody involucre, four-celled at the base, two-celled at the apex, tipped by the remains of the style ; seed solitary, without albumen, filling the cavity of the nut. The cotyledons remain underground in germination, the plumule being carried by the lengthening of their petioles out of the nut, which splits into two valves. The germination resembles that of the oak, the young stem bearing at first three to eight alternate minute lanceolate scales, above which the leaves are developed. The first leaf is simple, tri-lobed, or trifoliolate ; those succeeding, about five or six in the first year, being trifoliolate ; all are serrate and stalked. The difference observed in the length of the stems, in two or three seedlings, seen at Colesborne, may not be constant for each species.2 Twelve species of hickory are distinguished by Sargent,8 all natives of North 1 The generic name, Carya, though not the first one published, has always been used in England, and is now sanctioned by the regulations drawn up by the International Congress of Botany at Vienna in 1905. Cf. Verhand. Internat. Bot. Kongress. Wien, 1905, p. 239. With regard to the specific names, I have not altered those of Nuttall, which have been long in use. 2 Rowlee and Hastings, Bot. Gaz. xxvi. 349, pi. xxix. figs. 9, 10, 12 (1898), give figures of the seedlings of C. alba and C. porcina. 3 In Trees N. Amer. 132 (1905). Ashe, in Flora South-Eastern United States, 333 (1903), raised two varieties to the rank of species, making fourteen species in all. 597 598 The Trees of Great Britain and Ireland America, four species extending as far north as Canada, and one species confined to the high lands of Mexico. Six species1 (Plate 203) are in cultivation in the British Isles, and will be dealt with in the following account. The genus is divided into two sections :— I. Apocarya, De Candolle, Prod. xvi. 2, p. 144. Buds, with four to six valvate scales, which are often obscurely pinnatifid at the apex ; axillary buds, often two to three superposed, the uppermost one stalked. Husk of the fruit thin, and prominently ridged at the sutures. 1. Carya olivœformis, Nuttall. Buds greyish, densely pubescent, without glands. Leaflets, eleven to thirteen, rarely nine ; margin densely ciliate. 2. Carya amara, Nuttall. Buds yellowish, slightly pubescent, glandular. Leaflets, seven to nine; margin irregularly ciliate. II. Eucarya, De Candolle, Prod. xvi. 2, p. 142. Buds with ten to twelve imbricated scales, the outer falling early, the inner accrescent and becoming much enlarged and reflexed at maturity. Husk of the fruit thick, not ridged at the sutures. * Leaflets five, pubescent. 3. Carya alba, Nuttall. Young branchlets with brown stellate hairs ; base of the shoot marked with a dense pubescent ring. Leaflets, stellate-pubescent beneath, ciliate in margin. ** Leaflets ßve or seven, glabrous. 4. Carya porcina, Nuttall. Young branchlets glabrous or with only an occasional hair ; base of shoot without pubescent ring. Leaflets glabrous beneath, except for axil-tufts, non- ciliate in margin. *** Leaflets, seven or nine, pubescent. 5. Carya sulcata, Nuttall. Branchlets reddish-brown, glabrous towards the tip. Leaves not fragrant ; rachis nearly glabrous ; nerves in upper lateral pair of leaflets more than twenty pairs. 6. Carya tomentosa, Nuttall. Branchlets purplish-grey, pubescent and glandular. Leaves fragrant ; rachis pubescent and glandular; nerves in upper lateral pair of leaflets less than twenty pairs. (A. H.) 1 Carya aquatica, Nuttall, the water hickory, a native of river swamps in the southern parts of the United States, is not likely to succeed in any part of the British Isles. London, op. cit. 1444, mentions a tree of this species 40 feet high growing at Milford near Godalming ; but his identification was probably incorrect. Carya 599 CARYA OLIWEFORMIS, PECAN NUT Carya olivaformis, Nuttall, Gen. Am. ii. 221 (1818); Loudon, Arb. et Frut. Brit. iii. 1441 (1838). Carya angustifolia, Sweet, Hort. Brit. 97 (1827). Carya illinoinensis, Koch, Dendrol. i. 593 (1869). Carya Pecan, Schneider, ex Sargent in Bot. Gazette, xliv. 226 (1907). Juglans Pecan, Marshall, Arbust. Am. 69 (1785). Juglans illinoinensis, Wangenheim, Nordam. Holz. 54 (1787). Juglans angustifolia, Aiton, Hort. Kew, iii. 361 (1789). Juglans cylindrica, Poiret, in Lamarck, Diet. iv. 505 (1797). Juglans clivaformis, Michaux, FI. Bar. Am. ii. 192 (1803). Hicoria Pecan, Britton, Bull. Torrey Bot. Club, xv. 282 (1888); Sargent, Situa N. Amer. vu. 137, tt. 338, 339 (i89S). a"d Trees N. Amer. 133 (1905). A tree, attaining in America 170 feet in height and 18 feet in girth. Bark, brownish, deeply and irregularly divided into narrow forked ridges. Buds, similar to those of C. amara, but greyish-white in colour, densely pubescent and without glands ; lateral buds ovoid, pointed. Young branchlets densely pubescent, especially towards the tip, where no glands are present, but with minute glands at the base of the shoot. Leaves (Plate 203, Fig. 6), 12 to 20 inches long. Leaflets, eleven to thirteen or more, rarely nine, lanceolate, falcate, subsessile, acuminate, very unequally divided by the midrib ; upper surface with stellate pubescence on the midrib and nerves ; lower surface covered throughout with fine scattered pubescence and numerous glands ; margin plainly ciliate ; rachis densely pubescent. Staminate flowers in sessile or subsessile clustered catkins, usually on the previous year's branchlets. Fruit in clusters of three to eleven, pointed, four-winged and -angled, i to z\ inches long, \ to r inch broad ; husk thin, brittle, dark-brown, coated with yellow pubescence, splitting when ripe nearly to the base ; nut, thin- shelled, reddish-brown with irregular black markings ; seed sweet, reddish-brown. (A. H.) The Pecan is a native of the Middle States, occurring from Indiana, Southern Illinois and Iowa, southward through Western Kentucky and Tennessee to Alabama and Mississippi, and extending westward through Missouri, South-Eastern Kansas, Arkansas, Indian Territory and Louisiana to the valley of the Concho River in Texas. It is also met with in the mountains of Mexico. It chiefly grows on rich alluvial soil, along the banks of streams, and attains a greater size than any other hickory. Ridgway says it is one of the very largest trees of the forest, being only exceeded in height by the tulip tree and the scarlet oak. He records one tree, measured by Dr. Schneck, which was 175 feet high by 16 feet in girth, and another 30 feet in girth at the ground. It is largely cultivated in the Southern States for its fruit, which has been improved by selection and grafting, and is considered the best of the nuts of North America. It requires a much longer and hotter summer than any part of Great Britain affords ; and even in the south of Europe we have not heard of its being successfully 6oo The Trees of Great Britain and Ireland grown. Pardé1 has not heard of its having ever produced fruit in France, where it is hardy as far north as Grignon,2 but grows very slowly. It was introduced, according to Loudon, in 1766 into England, and he mentions, with some doubt as to the species, trees growing in the Horticultural Society's garden and other places near London ; but it is probable that this tree has never attained a considerable size in England. At Tortworth, a tree with the bark beginning to scale was, in 1905, 24 feet high and i^ foot in girth. Dr. Warre, the late headmaster at Eton, raised, from seed sent to him from New York, a tree, which is growing in the garden of Mr. E. C. Austen Leigh, at Eton, who informs us that it is about 11 feet high. Miss Woolward raised at Belton in Lincolnshire a plant, which is now about 4 feet high ; but it suffers severely from frost, the young growths being killed back every year. None of the seedlings which I have raised at Colesborne have thriven, as the summer here is evidently much too short and cold. (H. J. E.) CAR Y A AM ÄRA, BITTERNUT Carya amara, Nuttall, Gen. Am. ii. 222 (1818); Loudon, Are. et Frut. Brit. iii. 1443 (1838). Carya cordifolia, Schneider, ex Sargent in Bot. Gazette, xliv. 226 (1907). Jugions alba minima, Marshall, Arbust. Am. 68 (1785). Jugions cordiformis, Wangenheim, Nordam. Holz. 25 (1787). Jugions angiisti/olia, Poiret, in Lamarck, Diet. iv. 504 (1797)- Juglans sulcata, Willdenow, Berl. Baumz. 154 (1796). Juglans minima, Borkhausen, Handb. Forstbot, i. 760 (1800). Juglans mucronata, Michaux, Fl. Bor. Am. ii. 192 (1803). Juglans amara, Michaux f., Hist. Are. Am. i. 177 (1810). Hicoria minima, Britton, Bull. Torrey Bot. Club, xv. 284 (1888); Sargent, Silva N. Amer. vii. 141, tt. 340, 341 (1895), and Trees N. Amer. 135 (1905). A tree, attaining in America 100 feet in height and 9 feet in girth. Bark grey, smooth, ultimately separating on the surface into small thin scales. Buds bright- yellow, glandular ; terminal buds elongated, pointed and oblique at the apex, about \ inch long, with two pairs of valvate scales, often obscurely pinnatifid at the tip ; lateral buds, often two superposed, the uppermost stalked, four-angled, and pointed, the lowermost minute. Young branchlets, with scattered short pubescence, glandular towards the tip. Leaves (Plate 203, Fig. 5) 6 to 10 inches long. Leaflets, five to nine, variable in shape, lanceolate, ovate-lanceolate, or obovate-oblong, long- acuminate ; margin with occasional cilia ; lower surface with stellate hairs, especially along the midrib ; rachis minutely pubescent. Staminate catkins, pubescent, in threes on a slender peduncle, usually on the previous year's branchlet. Fruit, one, two, or three at the top of the branchlet, about an inch long ; husk thin, glandular, four-winged from the apex to about the middle ; nut thin-shelled ; seed bitter, reddish-brown. (A. H.) This species has a wide distribution, extending to the northward as far as Southern Maine, Ontario, Central Michigan, and Minnesota. It was the only hickory which I found near Ottawa, where it was common in the Gatineau Valley, 1 Arbor. Nat. des Barres, 253 (1906). * Seen by Henry in 1906. Gary 601 but not there a large tree. It extends westward to South-Eastern Nebraska, Eastern Kansas, Indian Territory, and Eastern Texas, and reaches its southern limit in North-Western Florida and Northern Alabama. It is usually found in lower and moister situations than the other species, and is one of the largest and commonest hickories in Southern New England, where it attains 70 to 80 feet in height and 10 to 12 feet in girth. It grows to its largest size on the alluvial lands of the lower Ohio basin ; and in Southern Indiana, Ridgway records a tree 113 feet in height and 6 feet 3 inches in girth. The bitternut is the commonest species of hickory in England, and grows to a considerable size. The finest tree is perhaps the one at Bute House, Petersham, which was, in 1903, 76 feet in height by 7 feet 5 inches in girth (Plate 170). At Barton, near Bury St. Edmund's, there are three trees, two in the arboretum, one of which I found in 1905 to be 80 feet by 5 feet 4 inches, and the third on the lawn, 74 feet by 7 feet 6 inches, but forked near the ground. These trees ripened fruit1 in 1864 ; but of late years do not appear to have borne any. At Syston Park, Lincolnshire, there are four trees of this species, one of which was flowering freely on loth May 1905, when it measured 71 feet by 5 feet 3 inches. Two of these are figured on Plate 171 from photographs taken by Miss F. Woolward. At Arley Castle there are five trees, the tallest of which measures 72 feet by 4 feet, the others being about 60 feet high with girths ranging from 3 feet 7 inches to 4 feet 7 inches. These are supposed to have been planted about 1820. At Bicton, a tree measured, in August 1906, 65 feet by 4 feet; and another growing at the Wilderness, White Knights, near Reading, is exactly the same size. At Devonshurst, Chiswick, a tree, now cut down, measured, in 1903, 68 feet by 5 feet. There is a tree, 58 feet by 6 feet 2 inches, in the Botanic Garden, Glasnevin ; but we have heard of no others in Ireland, where hickories seem to have been very little planted. (H. J. E.) CARYA ALBA, SHAGBARK HICKORY Carya alba, Nuttall, Gen. Am. ii. 221 (1818); Loudon, Arb. et Frut. Brit., iii. 1446 (1838). Carya ovata, Schneider, ex Sargent in Bot. Gazette, xliv. 226 (1907). Juglans ovata, Miller, Diet. ed. 8, No. 6 (1768). Juglans ovalis, Wangenheim, Nordam. Holz. 24 (1787). Juglans compressa, Gaertner, Fruct. ii. 51 (1791). Juglans alba, Michaux, Fl. Bor. Am. ii. 193 (1803). Juglans obcordata, Poiret, in Lamarck, Diet. iv. 504 (1797). Juglans squamosa, Michaux f., Hist. Arb. Amer. i. 190 (1810). Hicoria ovata, Britton, Bull. Torrey Bot. Club, xv. 283 (1888); Sargent, Silva If. Amer. vii. 153, tt. 346, 347 (1895), and Trees N. Amer. 139 (i9°5>- A tree, attaining in America usually 70 to 90 feet, rarely 150 feet in height and 15 feet in girth. Bark grey, ultimately separating into long strips, attached by the 1 Bunbury, Arboretum Notes, IOO (1889). 6o2 The Trees of Great Britain and Ireland middle and free at one or both ends, giving the trunk a characteristic appearance. Terminal buds ovoid, obtuse, \ to f inch long, with ten to twelve imbricated scales ; outer scales persistent during winter, falling a little before the unfolding of the leaves, triangular, keeled, apiculate or contracted into long points, dark-brown, pubescent ; inner scales downy, enlarging to 2 or 3 inches long, as the bud opens. Lateral buds about \ inch long, with four to five scales visible externally. Branchlets purplish-grey, covered with brown stellate hairs, scattered below, denser nearer the tip, and with a few yellow glands ; base of the shoot girt with a dense ring of pubescence. Leaves (Plate 203, Fig. i) 8 to 14 inches long. Leaflets, five, upper three obovate, lower pair ovate, all shortly acuminate ; margin ciliate with irregular tufted hairs, densest near the points of the serrations ; upper surface glabrous, except for stellate hairs on the midrib and nerves ; lower surface with a fine stellate pubescence, densest on the midrib and nerves ; rachis stellate-pubescent. Staminate catkins, glandular-hirsute, pedunculate in threes at the base of the year's shoot. Fruit solitary or in pairs, sub-globose, i to 2\ inches long, splitting freely to the base ; husk dark reddish-brown, glabrous, \ to ^ inch thick ; nut four- angled and -ridged, white ; seed large, lustrous, light-brown, sweet with an aromatic flavour. The above description is drawn from trees cultivated in England, which resemble in all essential characters, except the imperfect development of the fruit, specimens obtained by Elwes at the Arnold Arboretum. The size and shape of the leaflets, which are always five in number, and the amount of pubescence on the branchlets and on the rachis and surface of the leaves are very variable. Two or three trees1 of this species at Kew, which are about 25 feet high, have very large leaves which turn a brilliant yellow in autumn. Another tree at Kew, which was labelled Carya alba, differs from all other specimens which I have seen, as follows :— Branchlets and leaf-rachis almost glabrous, only showing when young a few stellate hairs. Leaflets, five, lanceolate, narrow, long-acuminate, nearly glabrous, with only a few stellate hairs, confined on the upper surface to the midrib, and scattered over the lower surface; margin non-ciliate. Buds, as in the typical form, but pointed and smaller. This tree, which is about 30 feet high, has very smooth bark, and is growing very vigorously. It is probably a glabrous form of Carya alba, and may possibly be Hicoria borealis, Ashe,2 which grows in Michigan and Southern Ontario. (A. H.) The Shagbark, according to Sargent, is widely distributed from the St. Lawrence valley near Montreal, along the northern shores of Lakes Erie and Ontario, west to South-Eastern. Nebraska, and south throughout the Middle States and along the Appalachian mountains to Western Florida, Northern Alabama, Eastern Texas, and Central Kansas. It attains its largest size in the Southern Alleghanies and in the lower Ohio basin, the largest trees recorded, one of which is figured in Garden and Forest, ii. 1 These have been labelled C. tomcntosa, from which they differ in having five-foliolate non-fragrant leaves ; and are the trees referred to in Card. Chron. xxviii. 295 (1900). 2 Ashe, Notes on Hickories, 1896; Britton and Brown, Illust. Flora N. United States, iii. $12, fig- 1156 b (1898). This form is not recognised by Sargent as a distinct species. Carya 603 463, being in Southern Indiana, where Ridgway1 says that trees of 130 feet high by 3 to 4 feet in diameter were not rare, and that some were certainly 150 feet, many trunks which seemed less than half the total height being 70 or 80 feet to the first limb. One of these in Wabash Co. measured 78 feet to the first limb, and was 14! feet in girth. Such giants, however, hardly exist now ; and during my travels in America I never saw a hickory more than about 100 feet high ; whilst those in New England and Canada are usually from 60 to 80 feet in height, and seldom exceed 6 feet in girth. The bark, which separates externally into long loose flakes, serves to distin guish this species readily from the others. The nuts, which vary much in size and shape and thickness, are superior in quality to any other native nut (except that of the Pecan), and are largely eaten. Some of the better varieties, which have thin shells and larger kernels, have been selected and propagated, so that the improve ment of this fruit in cultivation is likely to be as great as that of the walnut. This species was introduced in 1629, according to Loudon, who mentions large trees growing in 1838 at Syon, Fulham Palace, and other places near London. None of these are now living, and apparently the tree does not attain a great age in England. It succeeds about as well as the Bitternut. The largest tree that I have seen in England is hidden in a thick shrubbery at Botley Hill, Hants, the residence of Lady Jenkyns, and has an historic interest from the fact that it is almost certainly one of the trees planted by Cobbett, who lived there for some years about 1820. The old brew-house and oven, which, in his opinion, were two of the most necessary parts of an Englishman's house, still remain, as well as some rather stunted black walnuts. This hickory is about 75 feet high by 5 feet 4 inches in girth, with a bole of 30 feet. At the Wilderness, White Knights, near Reading, a tree measures 55 feet by 5 feet 8 inches. At Castle Howard, Yorkshire, there is a healthy symmetrical tree, growing on the site of an old nursery near the timber yard, which may have been sown in situ. I found it in 1905 to be 50 feet high by 3 feet 3 inches in girth. In the Pinetum, Brocklesby Park, Lincolnshire, there is a fine tree, which Mr. W. B. Havelock informs us is 79 feet high by 4 feet 10 inches in girth, with a straight stem running nearly to the top. It is growing in strong loam, sheltered by surrounding belts of trees, and is supposed to be about sixty years old. The bark is shaggy, and hangs in strips from the trunk (Plate 173). A fine tree of this species is growing in Syston Park, near Grantham, the seat of Sir John Thorold, who told me that his father planted it and the four bitternut trees mentioned above about fifty years ago. It measured in 1905 62 feet by 4 feet 4 inches, and has the characteristic scaly bark. At Boynton Hall, Bridlington, Yorkshire, there are three trees of this species, which Sir Charles Strickland informs us are respectively 50 feet by 7 feet, 40 feet by 4 feet, and 25 feet by 6 feet, the last being very bushy in habit, and growing in a very exposed position. These trees were raised from seed brought from America by 1 Prot. U.S. Nat. Mas. 1882, p. 77. Ill X 604 The Trees of Great Britain and Ireland Sir William Strickland, grandfather of the present owner ; and several other hickories planted at the same time have disappeared, one being blown down. At Golden Grove, Carmarthenshire, a seat of Earl Cawdor, a tree, which is supposed to have been planted in 1865, is 42 feet by 3 feet i inch. At Fonthill Abbey, Wilts, a tree, with very scaly bark, is 62 feet by 4 feet i inch. At Althorp there is a tree 75 feet by 3 feet 6 inches, growing in a dense thicket of laurels near the house. A hickory which was perhaps of this species grew close to the house at Moncreiffe near Perth, and is mentioned by Hunter as being the finest in Scotland. It had a bole 20 feet long by 5 feet 9 inches, and was cut down about six years ago. The timber was used for making gates. At Kinblethmont, Arbroath, H. Lindsay Carnegie, Esq., reports a tree 46 feet high by 2 feet n inches in girth. It was planted by his father about 1825. In Ireland, the only specimens which Henry has seen are two small trees about a foot in diameter, growing in the Botanic Gardens at Glasnevin, and two at Kilmacurragh, in Wicklow, about 35 feet high. (H. J. E.) CARYA PORCINA, PIGNUT Carya porcina, Nuttall, Gen. Am. ii. 222 (1818); Loudon, Arb. et Frut. Brit. iii. 1449 (1838). Carya olcordata, Sweet, Hort. Brit. 97 (1827). Carya glabra,1 Spach, Hist. Vég. ii. 179 (1834). Juglans glabra, Miller, Did. ed. 8, No. 5 (1768). Juglans squamosa, Poiret, in Lamarck, Diet. iv. 504 (1797). Juglans olcordata, Muehlenberg u. Willdenow, Neue Schrift. Ges. Nat. Fr. Berlin, iii. 392 (1801). Juglans porcina, Michaux f., Hist. Arb. Am. i. 206 (1810). Hicoria glabra, Britton, Bull. Torrey Bot. Club, xv. 284 (1888); Sargent, Suva N. Amer. vu. 165, ». 3S2-3SS (l895). and Trees N- Amer. 144 (1905). Hicorius glaber, Sargent, Garden and Forest, ii. 460 (1889). A tree attaining in America 90 feet in height and 12 feet in girth. Bark greyish, ultimately fissuring into narrow longitudinal ridges, occasionally on old trees breaking on the surface into loose thick scales. Terminal buds, ^ to ^ inch long, globose or ellipsoidal, with ten to twelve imbricated scales ; outer scales usually deciduous in winter, keeled, acute, or pointed, glabrous, ciliate, often glandular ; inner scales pubescent, enlarging to 2 inches long as the bud unfolds. Lateral buds small, ovoid, often glandular, with two scales visible externally. Branchlets glabrous, or rarely with a minute pubescence speedily disappearing. Leaves (Plate 203, Fig. 2) 8 to 12 inches long. Leaflets, five to seven, upper three obovate, lower one or two pairs oblong lanceolate, all acuminate ; margin without cilia ; upper surface glabrous, with numerous minute glands ; lower surface glabrous, except for slight tufts of pubescence in the axils of the midrib and lateral nerves, covered with numerous glands ; rachis glabrous. 1 This is given as Carya glabra, Schneider, by Sargent in Bot. Gazette, xliv. 226 (1907). Carya 605 Staminate catkins, scurfy-pubescent, pedunculate in threes at the base of the current year's shoot. Fruit, single or in pairs, very variable in size and shape ; husk thin, reddish brown, glandular, opening only at the apex or splitting to the middle or to near the base ; nut obscurely four-angled ; seed small, light brown, poor in flavour. Carya microcarpa, Nuttall * (Hicoria microcarpa, Britton2), appears to be only distinguished by the nuts, which are very small, compressed and globular. It is impossible to say whether it is in cultivation, as all the hickories in England produce smaller nuts than they do in America. This form is called by Sargent,3 var. odorata, a misleading name, as the tree has apparently no marked odour either in the foliage or in the fruit. (A. H.) The pignut, which grows usually on dry ridges and hillsides, has a most extensive distribution, ranging from Southern Ontario and Southern Maine in the north to Florida and Mississippi in the south, and extending westward to Nebraska, Kansas, and Texas. This species of hickory4 seems to succeed fairly well in England, though we know of very few trees. The best is at Kew, where there is a fine tree near the Temperate House (Plate 172), which is, however, becoming stag-headed. It measured, in 1907, 77 feet by 6 feet. It produces fruit abundantly, but the seed, so far as we have observed, is infertile. Another tree is growing at Leny, near Callander, Perthshire. Though in a somewhat exposed situation and at an elevation of several hundred feet, it had a stem 3 feet 4 inches in girth and might have been 50 feet high before the leading shoot was broken. (H. J. E.) CARYA SULCATA, BIG SHELLBARK, KINGNUT Carya sulcata, Nuttall, Gen. Am. ii. 221 (1818); Loudon, Arb. et Frut. Brit. iii. 1448 (1838). Caryapubescens, Sweet, Hort. Brit. 97 (1827). Carya cordiformis, Koch, Dendr. \. 597 (1869). Carya ladniosa, Schneider, ex Sargent in Bot. Gazette, xliv. 226 (1907). Juglans ladniosa, Michaux, f., Hist. Arb. Amer. \. 199 (1810). Juglans sulcata, Pursh, Fl. Am. Sept. ii. 637 (1814). Hicoria ladniosa, Sargent, Silva N. Amer. vii. 157, tt. 348, 349 (1895), and Trees N. Amer. 141 (1905). A tree attaining in America 120 feet in height and 9 feet in girth. Bark grey, ultimately separating in plates, which sometimes remain for many years hanging on the trunk. Terminal buds, ovoid, obtuse, f to i inch long, composed of ten to twelve imbricated scales ; outer scales not deciduous in winter, dark brown, ovate, keeled, pointed, tomentose, with scattered glands; inner scales silky pubescent, » Gen. Am. ii. 221 (1818). 2 Bull. Torrey Bol. Clut> xv. 283 (1888). 3 Silva, loc. cit. This varietal name is adopted by Sargent, as the tree appears to have been first described by Marshall, in Arb. Am. 68 (1785) as Juglans alba odorata. * The date of introduction is uncertain. Loudon states that in 1838 there were plants in the Hackney Arboretum. 606 The Trees of Great Britain and Ireland becoming 2 to 3 inches long at the opening of the bud. Axillary buds, J inch long, with four to five scales visible externally. Branchlets light reddish brown, glabrous towards the tip, covered below with a short dense pubescence. Leaves (Plate 203, Fig. 4) not fragrant, 15 to 22 inches long. Leaflets seven to nine, upper three or five obovate-oblong, lower one or two pairs lanceolate; margin ciliate, the cilia more numerous towards the base; upper surface glabrous, shining; lower surface with scattered short stellate tomentum ; nerves in upper pair of leaflets more than twenty pairs ; rachis with slight pubescence near the insertion of the leaflets, elsewhere glabrous. Staminate catkins, glabrous, pedunculate in threes at the base of the current year's shoot. Fruit, solitary or in pairs, 2 inches long ; husk downy or glabrous, hard and woody, ^ to ^ inch thick ; nut compressed, four- to six-ridged, with a hard bony shell ; seed light brown, sweet. (A. H.) This species usually grows on deep rich alluvial soil, which is inundated for several weeks in the year, and is one of the commonest trees in the great river swamps of Central Missouri and of the Ohio basin. It is rare and local east of the Alleghany Mountains, being occasionally met with in New York, Pennsylvania, and North Carolina ; and extends through the central states from South - Eastern Nebraska, Iowa, Illinois, and Indiana, southward to Kansas, Indian Territory, Arkansas, and Tennessee. It is exceedingly rare in cultivation1 in England, the largest tree we are acquainted with being one in Tortworth Churchyard, which bore a few nuts in 1905, when it was about 30 feet high by i foot 8 inches in girth. There are small plants growing at Hildenley Hall, Yorkshire, and at Grayswood, Haslemere. (H. J. E.) CARYA TOMENTOSA, MOCKERNUT Carya tomentosa, Nuttall, Gen. Am. ii. 221 (1818); Loudon, Arb. et Fnit. Brit. iii. 1444 (1838). Carya alba, Koch, Dendr. i. 596 (1869); Sargent, Bot. Gazette, xliv. 226 (1907). Juglans alba, Linnseus, Sp. PL 997 (in part) (1753). Juglans rubra, Gaertner, Fruct. ii. 51 (1791). Juglans tomentosa, Poiret, in Lamarck, Diet. iv. 504 (1797). Hicoria alba, Britton, Bull. Torrey Bot. Club, xv. 283 (1888); Sargent, Silva IV. Amer. vii. tt. 350, 351 (1895), and Trees N. Amer. 143 (1905). Hicorius albus, Sargent, Garden and Forest, ii. 460 (1889). A tree attaining in America 100 feet in height and 9 feet in girth, but usually much smaller. Bark grey, slightly ridged by shallow irregular interrupted fissures, and ultimately covered by closely appressed scales. Terminal buds broadly ovoid, ^ to f inch long, of 10 to 12 imbricated scales; outer scales, usually deciduous in winter, dark brown, keeled, apiculate, pubescent, glandular ; inner scales silky 1 Loudon states that it was introduced in 1804 ; but he appears to have been only acquainted in 1838 with small plants growing at the Horticultural Society's Garden, Loddiges' Nursery, and White Knights. None of these now survive. Carya 607 pubescent, enlarging to i to i^ inch long, as the bud unfolds. Lateral buds much smaller, with 4 scales visible externally. Branchlets densely covered with stellate hairs and shining glands, without any definite ring of pubescence at the base of the shoot. Leaves (Plate 203, Fig. 3) very fragrant, eight to twenty inches long. Leaflets seven or nine, upper three obovate, lower pairs ovate ; acuminate ; margin regularly ciliate ; upper surface glabrous, except for stellate hairs on the midrib, and with minute shining glands; lower surface covered with scattered stellate pubescence and numerous glands ; nerves in the upper pair of lateral leaflets usually less than 20 pairs ; rachis with stellate pubescence and shining glands. Staminate catkins, pubescent, pedunculate in threes at the base of the current year's shoot. Fruit, single or in pairs, globose, i to 2 inches long ; husk hard, thick, glandular, splitting to the middle or nearly to the base ; nut four-ridged near the top, thick-shelled ; seed small, dark brown, lustrous, sweet. (A. H.) This species has a distribution similar to that of the shag-bark and bitternut ; but is comparatively rare in the north, though it is found in Southern Ontario. It commonly grows on poor and sandy soil, and is the only hickory in the maritime pine-belt of the Southern States. It does not usually exceed 60 feet in height, except in the rich valleys of the Southern Alleghany Mountains and in Missouri and Arkansas. Ridgway records a tree in Southern Indiana 112 feet by loj feet, and says it is a common species in upland woods, being known as the black- or white-heart hickory. Its name of mockernut is derived from the thickness of the shell and the smallness of the kernel of the fruit, which makes it hardly worth eating. The fruit and leaves have a strong fragrant resinous smell, not present in the other species. Loudon states that this species was " introduced in ? 1766 " ; but he does not seem to have known of any trees in cultivation in 1838. This hickory is extremely rare, the finest specimen that we know of being one at Golden Grove, Carmarthen shire, which is 50 feet high by 4 feet in girth. It is supposed to have been planted in 1865. Near the Azalea Garden at Kew there is a fine healthy young tree, procured in 1872 from Booth's Nursery, near Hamburg, and now 46 feet high by 2^ feet in girth, remarkable for its fine large foliage, the fragrance of which is especially strong and can be perceived at a distance in the early morning. There are small trees growing at Tortworth and Hildenley. (H. J. E.) CULTIVATION OF THE HICKORIES Though so long introduced into cultivation, and at one time much more commonly planted than at present, no species of hickory has as yet established a reputation which justifies the hope that it may become a tree of real economic importance in this country. As ornamental trees the hickories are not equal either in size or in beauty to others which can be more easily grown, and though at least three of the species may be planted with good hopes of success, as an inter esting addition to parks and pleasure grounds, yet their cultural peculiarities must be carefully studied before doing so. 6o8 The Trees of Great Britain and Ireland Hickories though easily raised from imported nuts, require special treatment on account of their long thick tap-roots, which make them so difficult to transplant ; and as they grow slowly for several years and do not ripen their wood when young, in most places, their ultimate success must always be to some extent a matter of chance. Though they are found in America, in places where the soil is not specially deep or good, they require a hotter summer than we get to enable them to ripen fruit, and when a tree will not ripen seed, it can hardly be called acclimatised. I have made many experiments in raising them from seed, which at present have not given very good results, principally, I think, because of unfavourable local conditions ; but believe that if the following points, which are based on those adopted in the Arnold Arboretum,1 are attended to, the trouble will not be thrown away. The nuts must be procured from America as soon as ripe ; and if there is any influence in heredity, as I believe there is, from the Northern, rather than from the Southern or Western States ; but it is only fair to say that the seed which I collected myself in Canada and near Boston, did not produce such strong seedlings the first season as those which I procured from Philadelphia. The nuts should be sown at once in boxes of about 18 to 24 inches deep in rich sandy loam, about 2 inches apart, and covered with an inch or so of light soil. The boxes may be stored for the winter in a shed, and in spring brought into a frame or greenhouse to induce earlier germination. They should be kept under glass until all risk of spring frost has gone by, and perhaps are better kept in a frame the whole summer lightly shaded, and watered when necessary. The leaves will remain on throughout the autumn, when the box should be exposed to the full sun ; and as soon as the shoot, which does not exceed 4 to 8 inches in height the first year, is ripe, may again be put away for the winter in a dry place covered with leaves, and protected from mice. In the following May the seedlings may either be turned out and planted in a deep rich nursery bed, after cutting off the tap-root at about a foot, or if a warm sheltered spot can be found in a wood, where they can be cultivated and sheltered for some years, they may be planted out permanently without cutting the tap-root. But as the danger from vermin and early or late frosts will continue for some years, it may be better to keep them in the nursery till they are 3 to 5 feet high, provided that when transplanted a deep trench is first made on one side, so as to get up the whole of the root with as little injury as possible. Woods being their natural home they are more likely to grow into good trees when drawn up with others than when exposed in the open ; but we cannot point to an instance in Great Britain where they have been so treated, though some of the best trees we know are in dense shrubberies. As regards soil, it cannot be too deep, rich, or well drained, and a southern or western aspect is to be preferred. Under such conditions they may attain 50 to 60 feet in height in as many years, and in some parts of England even more. A certain amount of lime in the soil does not seem to be harmful. The hickories are not either in America or in England very long-lived trees, 1 Cf. Garden and Forest, x. 116 (1897). Carya 609 and none of great age are recorded here. When cut down, or when killed to the ground by frost in their young state, they push shoots freely from the stool, though they do not produce suckers. Dawson in Garden and Forest, ix. 77 (1896), gives an account of his method of grafting the cultivated varieties of hickory, and says that the best stock is the bitternut, which grows twice as fast at Boston as the common shagbark. He per forms the operation under glass in the month of January, by side-grafting close to the collar of the stock, and plunging the pots into sphagnum moss up to the top bud of the graft. TIMBER OF THE HICKORIES The best account of the wood is given by Michaux and Emerson. The timber of the different species is so similar in appearance, that I doubt if any one could identify without the names, the six species illustrated by Hough ; and as this author rarely mentions the age or origin of the trees from which his specimens were taken, or shows much personal knowledge of their peculiarities, his work is not of so much practical value as it might have been. Michaux specially commends the timber of the shellbark and the pignut. Emerson does not say which is best, but says that the most valuable is that which has been grown most rapidly, and places the pignut and shellbark first for weight. As fuel hickory is, or rather was in days when it was abundant, preferred to all other woods. But its greatest value is for carriage building, axe and tool handles, and especially for cask hoops, of which in Michaux's time large quantities were exported, as well as used at home. Now, however, it is superseded to a great extent for this purpose by iron. An article on hickory by Mr. J. F. Brown in Arboriculture, vi. No. 4, states that the great demand for hoops in the apple-growing districts of Virginia, is rapidly exhausting the local supply of young trees, which is now being filled from Southern Indiana, and that in consequence the supply of second-growth timber fit for wheels and carriage work is likely to become diminished, and in well-settled regions is already exhausted. He states that when the trees are cut and put on the market, no discrimination is made between the different species, though second-growth hickory is always preferred to the timber of old trees, because it is more elastic, tougher, and stronger. He quotes a report of a meeting of over 200 representatives of the carriage-building industry at Chicago, at which it was stated that the hickory trees have recently been attacked by insects to such an extent, that unless some means can be taken to check their ravages, there will be no more hickory available in ten years ; and though ash, maple, and other woods have been tried as a substitute, there is no other wood so suitable for this industry as hickory. It is imported to some extent to Europe, usually in the form of second-growth poles, which are produced from the stool and are used by carriage builders. Cobbett,1 with his usual enthusiasm for everything from America, urged that the hickory should be planted for coppice wood on account of the value of the hoops l Woodlands, arts. 295, 296 (1825). 6IQ The Trees of Great Britain and Ireland which are made from it, and stated that he had a young tree at Kensington which was seven feet high when only five years old from the seed. Mayr also thinks that the hickory might have some economic value in the warmer parts of Germany, as it has stood the hardest frosts at Munich without injury. But so far as we know, none of the trials which have been made in France, where the tree was introduced on a large scale by Michaux ioo years ago, have been successful, and I could not hear that any of the trees which he planted near Paris are now alive. (H. J. E.) PLATANUS Platanus, Linnaeus, Gen. PI. 358 (1737); Bentham et Hooker, Gen. PL iii. 396 (1880); Janko, in Engler, Bot. Jahrb. xi. 412 (1890); Usteri, Mém. Herb. Boissier, No. 20, p. 53 (1900); Schneider, Laubholzkunde, 435 (1905). TREES belonging to the order Platanacese, which consists of the single genus Platanus. Bark, at the base of old trunks, dark-coloured and scaly ; above on the stem and on the branches smooth, thin, light-grey or greenish, separating in large thin scales, which on falling expose large irregular surfaces of pale-yellow or whitish inner bark. Branchlets rounded, zigzag ; the apex withering and falling off in summer (no true terminal bud being formed), leaving an elevated orbicular scar close to the uppermost axillary bud, which prolongs the shoot in the following year. Buds in summer concealed in the funnel-like base of the leaf-stalk, after the fall of the leaf surrounded at the base by an incomplete narrow ring-like scar, sinuous in margin and divided into five parts, each marked by a group of bundle dots ;• a line extending round the twig from the sides of the leaf-scar indicating where the connate stipule fell off. Buds all axillary, uniform in size, conic, covered by a scale in the form of a cap, which contains immediately within two similar scales, and more internally, several scales open at the apex and each with a young leaf at its base. Flower-buds similar, but larger. The three outer cap-like scales split longitudinally as the bud expands, the second and third continuing to grow after the bud is unfolded, ultimately falling and marking the base of the branchlet with ring-like scars. Leaves deciduous, alternate, simple, stalked, palmately three- to seven-lobed ; lobes entire or dentate with minute or coarsely sinuate teeth ; venation pseudo- palmate, two strong lateral nerves diverging from the midrib a little above its base, and each often giving off on the outer side a basal nerve, thus forming with the midrib three to five main nerves, each of which ends in the apex of a lobe. Stipules two, lateral, united below into a tube embracing the branchlet above the insertion of the leaf, dilated above and more or less free, thin and scarious on flowering shoots, broad and leafy on vigorous barren branchlets, caducous or occasionally persistent. Flowers monoecious, fertilised by the wind, appearing with the leaves, the females developing first, minute, densely aggregated in unisexual heads, which are solitary or several in spikes or racemes ; the staminate heads on axillary peduncles, the pistillate heads on long terminal peduncles. Staminate flowers ; sepals, three to six, scale-like, half as long as the scarious, cuneiform, acute, three to six petals ; stamens, three to six, with very short filaments and clavate two-celled anthers, m 611 v 6i2 The Trees of Great Britain and Ireland crowned by a pilose capitate connective. Pistillate flowers ; sepals, four (three to six), rounded, short ; petals, four (three to six), long, acute ; staminodes pilose at the apex ; ovaries as numerous as the sepals, superior, sessile, surrounded at the base by long hairs, gradually narrowing above into long simple styles ; ovules one, rarely two. Head of fruit composed of numerous elongated obpyramidate achenes, surmounted by the persistent style, surrounded at the base by long rigid hairs. Seed solitary, oblong, suspended, containing a thin fleshy albumen and an axile erect embryo. The fruiting heads remain hanging on the tree during winter, the component achenes being ultimately dispersed by the wind. The dispersal of the pollen in the flowers of plane trees is effected by a peculiar mechanism, which bears some resemblance to that of the yew, and is well described by Kerner.1 The planes are readily distinguished by the simple alternate palmately-lobed leaves, the base of the stalks enclosing and concealing the buds. In winter, the conical buds, all lateral, with stipule-lines around the twig and the peculiar narrow sinuous leaf-scars are diagnostic. The genus is a very ancient one, fossil species2 having been found in North America in Cretaceous, Eocene, and Oligocène strata. In the Miocene and Tertiary epochs numerous species were spread throughout all Europe, Northern Asia, and North America as far north as the Arctic Circle. In the glacial period those became extinct in the northern parts of their area, and the existing species are confined to Canada, the United States, and Mexico in the New World, and to the Eastern Mediterranean region in the Old World. Their entire absence from Eastern Asia is remarkable, as tertiary plants of circumpolar distribution, which have survived to the present time, are usually found existing both in Eastern North America and in China and Japan. Six species8 are now living, which may be conveniently arranged as follows :— I. Adult leaves glabrous or nearly so, conspicuously toothed in margin. 1. Platanus orientalis, Linnaeus. Albania, Macedonia, Thrace, Greece, Crete, Cyprus, Rhodes, and Asia Minor. Leaves distinctly lobed, the sinuses extending at least one-third the length of the leaf. Fruiting heads bristly, several on the peduncle. Achenes, with long hairs arising not only at the base, but along the body of the achene ; apex pyramidal or conic, acute, passing gradually into the long style. 2. Platanus occidentalis, Linnaeus. Eastern North America from Ontario to Texas. Leaves indistinctly lobed, the sinuses not extending one-third the length of the leaf. Fruiting heads smooth, solitary, and terminal on the peduncle. Achenes with basal ring of long hairs, elsewhere glabrous ; apex truncate or rounded, with a depression, from which arises a very short style. 1 Nat. Hist. Plants, Eng. Transi, ii. 146 (1898). 2 Cf. L. F. Ward, in Proc. U.S. Nat. Museum, 1888, p. 39, who states that a prominent characteristic of these archaic forms is the presence of basal lobes on the leaves. These basal lobes are occasionally met with on the young shoots of the species now living. 3 Platanus glatrata, Fernald, Proc. Am. Acad. xxxvi. 493 (1901), is an imperfectly known species from Coahuila in Mexico. Platanus 613 II. Adult leaves densely tomentose beneath, margin entire or minutely and remotely toothed. * Lobes elongated. 3. Platanus racemosa, Nuttall, Sylva, i. 47 (1842). A tree, attaining 120 feet in height and 18 feet in girth, common on the banks of streams in California. Introduced on the Continent, unknown in cultivation in Britain. Leaves three- to five-lobed to below the middle ; lobes acuminate, tomentose on both surfaces ; occasionally with remote minute teeth ; base slightly cordate or truncate. Fruiting heads bristly, two to seven on the peduncle. Achenes with ring of basal hairs, elsewhere glabrous ; apex pyramidal or rounded, without a central depression ; style long. 4. Platanus Wrightii, Watson, Proc. Am. Acad. x. 349 (1875). A tree, attaining 80 feet in height and 15 feet in girth, common in mountain canons up to 6000 feet altitude in Southern New Mexico and Southern Arizona in the United States ; and in Sonora in Mexico. Not introduced. Leaves three- to five-lobed to below the middle or near the centre ; lobes acuminate, usually quite entire, tomentose on both surfaces, deeply cordate at the base. Fruiting heads smooth, two to four, racemose on the peduncle. Achenes with ring of basal hairs, elsewhere glabrous ; apex rounded, with a central depression, from which arises a short style. ** Lobes short and broadly triangular. 5. Platanus mexicana, Moricand, Bull. Ferr. Bot. 79 (1830). A large tree in Mexico, in Nuovo Leon and the provinces to the south of it ; frequently planted in the cities of North-Eastern Mexico ; the handsomest of all the plane trees.1 Not introduced. Leaves three- to five-lobed, densely white tomentose beneath, base truncate or cuneate. Fruiting heads bristly, solitary. Achenes with ring of basal hairs ; upper part of the body pubescent ; apex pyramidal, continued directly into the long style. 6. Platanus Lindeniana, Martens et Galeotti, Bull. Acad. Brux. x. 2, p. 343 (1843). Tree, 100 to 150 feet in height. South Mexico, near Jalapa, at 4000 feet altitude. Not introduced. Leaves with usually three very short lobes, ending in long bristle-like points, densely rusty tomentose beneath ; base truncate. Fruiting heads bristly, several on the peduncle. Achenes as in P. orientalis. 1 Garden and Forest, ix. 51 (1896). 614 The Trees of Great Britain and Ireland PLATANUS ORIENTALIS, ORIENTAL PLANE Platanus orientals, Linnaeus, Sp. PI. 999 (1753); Loudon, Arb. et Frut. Brit. iv. 2033 (1838); Boissier, Flora Orientate, iv. 1161 (1879); Gamble, Manual Indian Timbers, 661 (1902); Schneider, Laubholzkunde, 436 (1905). Platanuspalmata, Moench, Meth. 358 (1794). Platanus cuneata, Willdenow, Sp. PL iv. 473 (1805). Platanus acerifolia, Willdenow, Sp. PI. iv. 474 (1805). Platanus laciniata, Du Mont de Courset, Bot. Cult. vi. 436 (i8n). Platanus vulgaris, Spach, Ann. Sei. Nat. xv. 291 (excl. e angulosa) (1841). Platanus hispanica, Tenore, Cat. Ort. Nap. 1845, P- 91- Platanus digitata, Gordon, Garden, 1872, p. 572. A tree, with several forms distinct in foliage and habit, most of which appear to have arisen in cultivation. Bark scaling off in thin plates ; furrowed and thick at the base of old trunks, variable in the different varieties.1 Young branchlets at first densely stellate-pubescent, becoming green and glabrous in summer ; brown in the second year. Leaves at first covered with dense loose white stellate pubescence on both surfaces, later nearly glabrous, the pubescence being only retained here and there, mainly on the veins and midrib of the lower surface, which is paler than the dark-green shining upper surface. Petioles at first densely stellate, white pubescent, ultimately glabrescent. Fruiting heads several (two to seven) on the peduncle, bristly. Achenes with basal ring of rigid long hairs, similar hairs arising also along the body of the achene, the apex of which is more or less acute and ends in a long persistent style. Seedling:2 caulicle slender, about ^ inch long, surmounted by two narrow spathulate cotyledons, obtuse at the apex, tapering to a very narrow base, sessile, one-nerved, entire, glabrous, dark-green, about ^ inch long. First leaf resembling a petiole in shape, minute and glandular-pubescent. Second leaf spathulate-cuneate, with three teeth at the apex, alternately penninerved, two of the stronger nerves running into the teeth. Third leaf like the second, but larger. Succeeding leaves palmately five-lobed. In summer, the oriental plane and its varieties are readily recognisable by the leaves, bark, and habit. In winter, the twigs are rounded, striate, glabrous, with numerous inconspicuous lenticels, the apex ending in a short stump, bearing an orbicular scar, marking where the tip of the branchlet fell off in summer. Leaf-scars, on prominent pulvini, almost but not completely surrounding the bud as a narrow ring, sinuous in margin, and with five groups of bundle-dots. Stipular line surround ing the twig at the level of each leaf-scar. Base of the shoot ringed with scars, 1 Boissier states that the bark of the wild tree is rugose, and does not exfoliate, as is usually the case in cultivated trees, especially in var. acerifolia, of the origin of which he knew nothing. As there is considerable variation in the size of the scales of the bark on plane trees, it is probable that the difierence noted by Boissier is individual and not varietal or specific. Var. acerifolia grows usually very fast, and scales off in much larger plates as a rule than is the case in the other varieties, which are slower in growth. * Cf. Lubbock, Seedlings, ii. 505, f. 653 (1892). Platanus 615 marking the fall of the previous season's bud-scales. Buds distichous and alternate on the long shoots, arising at an angle of 45°, uniform in size, conical, smooth, lustrous, covered by a glabrous cap-like scale. VARIETIES 1. Var. typica.—The form in cultivation, known generally as P. orientalis, slightly different in foliage from the wild form, known as var. insularis. A tree, attaining enormous dimensions in South-Eastern Europe and Western Asia, with a short trunk, dividing into many wide-spreading branches. Leaves (Plate 204, Fig. 4) large, generally exceeding 6 inches broad by 5 inches long, usually five-lobed ; lobes extending about half-way to the base of the blade, oblong- triangular with an acuminate apex, entire ' in margin or with a few sinuate entire teeth ; sinuses deep, variable in shape. Each of the two basal lobes often gives off a short lobe below, making the leaf seven-lobed. Base of the leaf truncate or widely cordate, but usually extending along the midrib £ to \ inch below where the two main lateral nerves are given off. Upper surface dark-green, glabrous, shining ; lower surface paler, glabrous except along the nerves and midrib, and in their axils. 2. Var insîilaris, DC. Prod. xvi. 2, p. 159 (1864). The wild form, occurring in Albania, Greece, Cyprus, Crete, Rhodes, and Asia Minor. A tree, not reported to be very large in size, and said to have rough bark, with small scales, which fall off less readily than in the typical form. Leaves smaller than in the type, scarcely exceeding 5 inches broad and 4^ inches long, very variable in shape, usually five-lobed ; lobes oblong-triangular or triangular, coarsely three- to four-toothed, extending about halfway to the base of the blade, which is always cuneate, the lamina descending along the midrib J- to % inch below the insertion of the first pair of main nerves ; sinuses deep, variable in form. Mouillefert2 distinguishes two varieties of the wild form, one with narrowly lanceolate entire lobes and wide sinuses, the other with lanceolate sinuately-toothed lobes and very deep narrow sinuses. The range of variation in the shape of the leaf in the wild form is considerable. 3. Var. cuneata? Loudon, Arb. et Frut. Brit. iv. 2034 (1838). Platanus cuneata, Willdenow, Sp. PI. iv. 473 (1805). A tree, moderate in size, with bark resembling that of var. acerifolia. Leaves (Plate 204, Fig. 3) usually about 5 inches wide by 4^ inches long, occasionally 8 inches by 7 inches ; three- to five-lobed, the lobes as in var. typica ; base of the leaf broadly cuneate, the lamina extending along the petiole ^ to f inch below the insertion of the main lateral nerves ; petioles short. This variety only differs from var. typica in the markedly cuneate base, and approaches in character var. insularis. As seen in cultivation it usually forms imperfect small fruiting heads. 1 The form with entire lobes is sometimes distinguished as var. liguidambarifolia, Spach, loc. cit. 2 Essences Forestières, 221 (1903). 3 Probably var. undulala, Alton, Hort. Kew, iii. 364 (1789). 6i6 The Trees of Great Britain and Ireland According to Koch,1 this is a good species, growing as a shrub on the south eastern slope of the Caucasus, and having a very short stem from which spring many upright branches. Koch, however, did not find it in flower or fruit ; and as no plane has been observed growing wild in the Caucasus, it is probable that what he saw were stunted trees of ordinary PL orientait*, occurring as escapes from cultivation. At Grayswood, Haslemere, there are two plants, 4 feet high, with a fastigiate habit, which Mr. Chambers raised eight years ago, from seed sent from Kashmir. 4. Var. digitata, Janko, in Engler, Bot. Jahrb. xi. 412 (1890). Platanus digitata, Gordon, Garden, 1872, p. 572. This is a form of var. typica, in which the leaves are smaller than usual, with wider and deeper sinuses, the lobes extending three-fourths the depth of the blade and having large triangular toothed lobules. Gordon supposed this form to be a native of the Caucasus, and says that it was introduced by Messrs. Loddiges in 1842. He describes the fruiting heads as only half the size of those of the type ; but in the number and structure of the component achenes there is no difference. 5. Var. acerifolia, Aiton, Hort. Kew, iii. 304 (1799). Platanus acerifolia, Willdenow, Sf. PI. iv. 474 (1805). London Plane. Maple-leaved Plane. A tree, with a tall upright stem, giving off shorter branches than the typical form. Leaves (Plate 204, Fig. i) large, at least 8 inches wide by 7 inches long, with five short, broad, triangular lobes, separated by wide rounded or acute shallow sinuses, which only extend one-third the length of the blade; base truncate or widely cordate, the lamina often descending on the midrib a short distance below the insertion of the two main lateral nerves. Fruiting heads very variable in size and in number on the peduncle, often badly developed in English trees ; achenes similar in structure to those of the typical form, and never resembling those of P. occidentalis. Several forms of the London plane have been distinguished :— \yc.pyramidalis. Pyramidal in habit. Var. kelseyana. Leaves variegated with yellow. Var. Suttmri. Leaves creamy-white, more or less splashed or streaked with green, often very large, as much as 12 inches wide by 10 inches long. This is identical with var. argenteo-variegata, which was exhibited at the Royal Horticultural Society in July 1897, by Messrs. Russell of Richmond. It is one of the handsomest of variegated trees, the variegation usually lasting the whole season.2 A form with large leaves has been sent out under the misleading name of var. californica. Though var. acerifolia exhibits a wide range of variation in the cutting of the leaf, it always shows very distinct lobes, and cannot be confused with P. occidentalis, in which the lobes are indistinctly marked. 1 Dendrologie, II. Part i. p. 470 (1872). " Cf. Card. Chron. xxiv. 190 (1898). Platanus 617 DISTRIBUTION The oriental plane has been in cultivation from very early times in the Mediterranean region ; and the limits of its distribution in the wild state are difficult to determine accurately. It occurs