The source of this uncorrected OCR text may be viewed in the DjVu format at: http://fax.libs.uga.edu/QK488xE4/6tgbi or http://purl.galileo.usg.edu/ugafax/QK488xE4/6tgbi Great Britain Ireland BY Henry John Elwes, F.R.S. AND Augustine Henry, M.A. Edinburgh : Privately Printed THE TREES OF GREAT BRITAIN AND IRELAND 4» * « y u t H' * *. i i • l t of Great Britain Ireland BY Henry John Elwes, F.R.S. AND Augustine Henry, M.A. VOLUMF VI Edinburgh: Privately Printed MCMXII I Y r À CONTENTS LIST OF ILLUSTRATIONS . . PICEA ....... PICEA EXCELSA, COMMON SPRUCE .... PICEA OBOVATA, SIBERIAN SPRUCE . PICEA ORIENTALIS, CAUCASIAN SPRUCE PICEA SCHRENKIANA, SCHRENK'S SPRUCE PICEA SMITHIANA, WESTERN HIMALAYAN OR MORINDA SPRUCE PICEA GLEHNII ...... PICEA POLITA ...... PICEA BICOLOR . . PICEA MAXIMOVVICZII ..... PICEA NIGRA, BLACK SPRUCE .... PICEA RUBRA, RED SPRUCE .... PICEA ALBA, WHITE SPRUCE .... PlCEA ALBERTIANA, ALBERTA WHITE SPRUCE PICEA ENGELMANNI, ENGELMANN'S SPRUCE PICEA PUNGENS, COLORADO SPRUCE PICEA SPINULOSA, SIKKIM SPRUCE .... JUNIPERUS ....... JUNIPERUS COMMUNIS, COMMON JUNIPER JUNIPERUS RIGIDA ...... JUNIPERUS OXYCEDRUS .... JUNIPERUS MACROCARPA ..... JUNIPERUS BREVIFOLIA ..... JUNIPERUS CEDRUS . . ... JUNIPERUS FORMOSANA ..... JUNIPERUS DRUPACEA ..... JUNIPERUS RECURVA ..... JUNIPERUS SQUAMATA ..... JUNIPERUS WALLICHIANA ..... JUNIPERUS PHŒNICEA ..... JUNIPERUS FLACCIDA ..... iii VI 1 1335 1337 1359 1362 1364 1366 1369 1370 1372 1374 1375 1377 1380 1385 1387 1389 1392 1400 1408 1409 1412 1413 1414 1417 1419 1420 1423 1424 1426 VI The Trees of Great Britain and Ireland MAGNOLIA HYPOLEUCA . ... MAGNOLIA Kouus ...... MAGNOLIA SALICIFOLIA ..... MAGNOLIA CONSPICUA . ... MAGNOLIA DENUDATA ..... MAGNOLIA PARVIFLORA ... MAGNOLIA WATSONI ..... MAGNOLIA STELLATA ... HALESIA ....... HALESIA DIPTERA .... HALESIA CAROLINA, SNOWDROP TREE MORUS ....... MORUS NIGRA, BLACK MULBERRY . MORUS RUBRA, RED MULBERRY .... MORUS ALBA, WHITE MULBERRY .... EUCALYPTUS ...... EUCALYPTUS CORDATA, TASMANIAN HEART-LEAVED GUM EUCALYPTUS PULVERULENTA, AUSTRALIAN HEART-LEAVED GUM EUCALYPTUS GLOBULUS, BLUE GUM .... EUCALYPTUS PAUCIFLORA, WEEPING GUM EUCALYPTUS VIMINALIS, MANNA GUM EUCALYPTUS COCCIFERA, MOUNTAIN PEPPERMINT EUCALYPTUS GUNNII, CIDER GUM .... EUCALYPTUS WHITTINGEHAMENSIS, WHITTINGEHAME GUM EUCALYPTUS ACERVULA, SWAMP GUM EUCALYPTUS VERNICOSA, DWARF GUM EUCALYPTUS MUELLERI, MUELLER'S RED GUM EUCALYPTUS URNIGERA, URN-BEARING GUM . PAGE 1592 1594 1595 1596 1597 1598 1598 '599 1600 1601 1602 1605 1606 1608 1609 1612 1620 1622 1623 1631 '635 1638 1642 1646 1647 1649 ILLUSTRATIONS Common Spruce in Massachusetts ..... Common Spruce at Studley ...... Spruce Avenue at Oakley Park ..... Common Spruce at Gwydyr Castle ..... Spruce at Kilworth, Ireland ...... Western Himalayan Spruce at Melbury .... Black Spruce at Coles borne ... . . Sikkim Spruce at Castlewellan Common Juniper at Colesborne ..... Juniperus recurva at Castlewellan . ... Catalpa at Ham Manor . . . Robinia at Frogmore . . ... Robinia at The Mote, Maidstone ..... Trunk of Robinia in America ; Evergreen Magnolia in North Carolina Wild Cherry in Savernake Forest ..... Cherry at George's Green, Slough ..... Pear at Lassington, Gloucester ..... Pear at Stockton, Worcester . . ... Magnolia at West Dean Park ..... Blue Gum at Torquay . . . . Blue Gum at Penmere ...... Eucalyptus coccifera at Povvderham ..... Eucalyptus Gunnii at Brightlingsea ..... Eucalyptus Whittingehamensis at Whittingehame Eucalyptus Muelleri at Derreen ..... Eucalyptus ; leaves and fruits .... Walnut at Cam-yr-AIyn Park .... Oriental Plane at Weston Park Scots Pine in Glen Maillie ..... Larch at Poltalloch ... Alpine Laburnum at Countesswells, Aberdeen Spruce Plantation at Rhindbuckie Hill, Durris PLATE No. 34° 341 342 343 344 345 346 347 348 349 35° 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 vu NOTICE TO SUBSCRIBERS THE favourable manner in which the Notice to Subscribers issued with the Fifth Volume of this work ( The Trees of Great Britain and Ireland) was received, leads us to hope for their approval of the course we are now about to adopt, after consultation with some of our best supporters. In consequence of the great additions to our knowledge, and the extreme difficulty of some of the genera described, the last part of the work has increased to over 500 pages, which, if published in one volume, would far exceed in size and cost those hitherto published. We have therefore decided to complete the work in two volumes. Vol. VII. will be issued, together with a general Index to the whole work, before the end of the year. In accordance with what was indicated in our last Notice to Subscribers the cost of these two volumes and the Index will be Five Guineas. H. J. ELWES. A. HENRY. IX PICEA THE characters of the Genus Picea and of the two sections into which it is divided have been given in Vol. I. pp. 75-76, with a description of the species belonging to the section Otnorica. At that time the Sikkim spruce (P. spinulosa), one of this section, was imperfectly known, and a full account of it is now given at the end of this article. See p. 1392. In the section Eu-picea, the leaves are quadrangular or rhombic in section, and bear stomatic lines on all their four sides. About fifteen species of quadrangular- leaved spruces are known,1 which may be readily distinguished by the following key, based on the characters of the branchlets, buds, and leaves. KEY TO SECTION EU-PICEA I. Branchlets quite glabroits. * Leaves on lateral branches radially arranged, spreading uniformly on all sides. 1. Picea Smithiana, Boissier. Western Himalayas. See p. 1366. Branchlets pendulous, grey. Buds large, resinous, pointed. Leaves slenden about i^ in. long. 2. Picea Maximowiczii, Regel. Japan. See p. 1374. Branchlets not pendulous, reddish brown. Buds small, resinous. Leaves, f to \ in. long. ** Leaves on lateral branches in an imperfect radial arrangement, not pectinate in two sets on the lower side of tlie branchlets, which are not pendulous. 3. Picea Schrenkiana, Fischer and Meyer. Central Asia, in the Alatau and Thianshan ranges. See p. 1364. Branchlets ashy grey. Terminal buds subglobose, girt with a ring of keeled pointed pubescent ciliate scales. Leaves rigid, sharp-pointed, f to \\ in. long. 4. Picea pungens, Engelmann. Wyoming, Colorado, Utah, New Mexico. See p. 1389- Branchlets at first glaucous, becoming reddish brown. Buds with the tips of their upper scales usually loose and reflexed. Leaves stout, rigid, with a hard sharp-pointed apex, f to ij in. long. 1 Not including the spruces of China, of which two or three species introduced by Wilson are in cultivation at Coombe Wood, but are too young to describe. vi 1335 « 1336 The Trees of Great Britain and Ireland 5. Picea polita, Carrière. Japan. See p. 1370. Branchlets yellow. Buds shining reddish brown, with closely imbricated scales. Leaves rigid, stout, curved, ending in a spine-like point. *** Leaves on lateral branches, imbricated on the ^lpper side of the branchlet; those below, pectinate and spreading laterally in two sets. 6. Picea alba, Link. North America. See p. 1380. Branchlets greyish or pale brown, usually glaucous. Buds with glabrous non- ciliate bifid scales. Leaves disagreeable in odour when bruised, about \ in. long. 7. Picea bicolor, Mayr. Japan. See p. 1372. Branchlets yellow, glabrous on lateral branches, pubescent in the furrows on leading shoots. Buds with scarious scales. Leaves, with two conspicuous white stomatic bands, each of five to six lines, on the two dorsal sides, and two bands of two lines on the two ventral sides. II. Branchlets^ variable, quite glabrous or with slight scattered pubescence. 8. Picea excelsa, Link. Europe. See p. 1337. Branchlets reddish, usually quite glabrous, or with slight pubescence often confined to the grooves between the pulvini. Terminal buds conical, acute, without resin, girt with a ring of keeled pubescent ciliate pointed scales. Leaves, usually f to i in. long, with two to three stomatic lines on each of the four sides. 9. Picea albertiana, Stewardson Brown. Alberta, British Columbia, Montana, Wyoming. See p. 1385. Branchlets greyish yellow, either glabrous or with minute pubescence usually confined to the pegs from which the leaves arise. Buds slightly resinous with rounded entire scales. Leaves, in an imperfect radial arrangement on the lateral branches, \ to i in. long. III. Branchlets always plainly pubescent. Leaves arranged on lateral branches, as in P. excelsa. * Terminal buds with a ring of conspicuous long subulate scales. ID. Picea nigra, Link. North America. See p. 1375. Branchlets covered with dense short glandular pubescence. Leaves bluish or glaucous green, about \ in. long. u. Picea rubra, Link. North America. Seep. 1377. Branchlets, as in P. nigra. Leaves yellowish green or dark green, not glaucous, ^ to f in. long. 12. Picea Gleknii, Masters. Saghalien, Yezo. See p. 1369. Branchlets reddish, with short non-glandular pubescence, confined to the furrows between the pulvini. Leaves slender, ^ to \ in. long. 1 Cf. P. bicolor, No. 7, which has pubescent leading shoots and glabrous lateral branches. Picea ** Terminal buds without long stibulate scales. 13. Picea orientalis, Carrière. Asia Minor, Caucasus. Seep. 1362. Branchlets slender, pale brown, covered with dense short non-glandular pubescence. Leaves, \ to f in. long, shining dark green, blunt and bevelled at the tip. 14. Picea Engelmanni, Engelmann. Western North America. Seep. 1387. Branchlets greyish yellow, with a sparse minute glandular pubescence. Leaves disagreeable in odour when bruised, bluish green, ^ to I in. long. 15. Picea obovata, Ledebour. Northern Scandinavia, Russia, Siberia; sporadic at high altitudes in the mountains of Central Europe. See p. 1359. Branchlets reddish brown, covered with a dense minute non - glandular pubescence. Leaves, if to % in. long, short-pointed, with three to four stomatic lines on each side. PICEA EXCELS A, COMMON SPRUCE Picea excelsa, Link, in Linnaa, xv. 517 (1841); Willkomm, Forstl. Flora, 67 (1887); Mathieu. Flore Forestière, 540 (1897); Ascherson and Graebner, Syn. Mittekurop. Flora, i. 196 (1897) ; Schröter, in Vierteljahrs. Naturf. Ges. Ziirich, xliii. 125-252 (1898); Kent, Veitch's Man.' Conif. 432 (1900); Kirchner, Loew and Schröter, Lebcngesch. Blutenpfl. Mitteleuropas, 99 (1904); Clinton-Baker, Illust. Conif. ii. 38 (1909). Picea rubra, Dietrich,1 Fl. Berol. 795(182 4). Picea vulgaris, Link, in Abhand. Akad. Berlin, 1827, p. 180 (1830). Picea Abies, Karsten, Pharm. Med. Bot. 324 (1881). Pinus Abies, Linnseus, Sp. PI. 1002 (1753). Pinus Picea, Du Roi, Obs. Bot. 37 (1771) (not Linnseus). Pinus excelsa, Lamarck, Fl. Franc, ii. 202 (1778). Abies Picea, Miller, Diet., 8th ed., No. 3 (1768). Abies excelsa, De Candolle, in Lamarck, Fl. Franc, iii. 275 (1805); Loudon, Arb. et Frtit. Brtt. iv. 2293 (1838). Abies carpatica, Lawson, Pinet. Brit. ii. 137, t. 20 (1867). A tree, often attaining in Britain 120 to 140 ft. in height and 10 to 12 ft. in girth, in central Europe attaining 200 ft. high and 15 to 20 ft. in girth. Bark on young stems brownish, thin, smooth ; on older trees thick, and scaling off on the surface in thin small scales. Young branchlets, reddish or yellowish brown, glabrous or with a minute scattered non-glandular pubescence, often confined to the furrows between the pulvini. Buds conical, acute, reddish brown, without resin, with rounded scarious scales ; terminal bud girt with a few acuminate keeled pubescent ciliate scales. Leaves on erect shoots radially spreading, more or less appressed to the twigs with their tips directed upwards : on lateral branches, pectinate below, the lower side of the twig being laid bare, most of the leaves being directed forwards and outwards ; while on the upper side of the twig, the leaves in the middle line are more or less 1 Dietrich's name and description apply to the common European spruce, and not to the American red spruce, as is often erroneously supposed. 1338 The Trees of Great Britain and Ireland appressed, with their tips directed forwards and slightly upwards. Leaves, variable in size, usually £ to f in., occasionally i in. long, rigid, straight or curved, ending in a short callous point, rhombic in section, with two or three stomatic lines on each of the four sides ; resin-canals variable, occasionally absent or only one present, usually two, one at each end of the transverse axis of the rhomb, close to the epidermis. Staminate flowers solitary in the axils of the leaves of the branchlets of the preceding year, rarely terminal on lateral branchlets, ovoid, about an inch long ; stamens numerous, spirally arranged, reddish, each with two pollen-sacs directed downwards and dehiscing longitudinally, and a prominent denticulate connective ; pollen grains, each with two air-vesicles. Pistillate flowers, appearing in summer as brown buds at the tips of the branchlets of the current year, developing in the following spring, about 2 in. long ; sessile, erect, cylindrical, purplish red ; scales carmine red, oval, with a truncate erose apex ; bracts about half the length of the scales, not increasing in size after the time of flowering, ovate - lanceolate, denticulate, with a long acuminate apex. After fertilisation the young cones leave the erect position, and gradually become pendulous, their scales becoming closely imbricated, and in the usual form of the species green in colour. Cones ripe in October, when they turn brownish ; cylindrical, pendulous, variable in size, about 4 to 6 in. in length ; usually opening in spring and letting the seeds escape when a dry east wind is blowing ; falling from the tree in the subsequent summer or autumn ; scales thin and flexible, rhombic, with a truncate emarginate or dentate apex, variable in size, f to f in. wide, i to i^ in. long, pale brown and glabrous on the exposed part, dark reddish brown and minutely pubescent on the concealed part ; bract about J in. long, lanceolate, denticulate at the acute or acuminate apex. Seed about J in. long, dark dull brown ; seed with wing about f in. long ; wing broadest near the obliquely rounded denticulate apex. Seedlitig.—Seeds sown in spring germinate in four or five weeks, the radicle first making its way out of the seed coats, and the caulicle carrying up the cotyledons, which are at first enveloped as with a cap by the albumen of the seed. The cap is soon cast off, and the cotyledons spread in a whorl. The cotyledons are six to ten in number, united at their base by a sheath, about |- in. long, triangular in section, with the upper edge faintly serrate, without resin-canals, stomatic on the two inner surfaces, deciduous at the end of the second year. The plant at the end of the first year is about 2 to 3 in. high, the young stem bearing, in addition to the whorl of cotyledons, spirally arranged primary needles, which are rhomboiclal in section, serrulate on the four angles, with two resin-canals, and inserted on raised pulvini. Branching occurs in the third or fourth year, when the leaves assume their adult form, being entire and not serrulate. No tap root is formed, the root dividing into numerous branches spreading in all directions. Throughout the life of the tree the absence of the tap root, seen in the seedling, persists ; and the roots of the spruce are usually spreading and do not penetrate the soil to any great depth. The spruce is normally monoecious, but instances have been known of Picea individuals which always bear staminate flowers ; and hermaphrodite flowers have been observed. The flowers are pollinated by the wind, the pollen being carried to an immense distance ; as far as'' eight miles in a case which was noticed near Munich. In the vicinity of spruce forests the pollen often descends in enormous quantity, covering the ground and the surface of lakes and rivers with yellow patches. I. The following variations occur in the form of the scales of the cone :— 1. Var. eurofiœa, Schroter, op. cit. 142 (1898). Var. montana, Ascherson and Graebner, op. cit. 198 (1897). Picea vttlgaris, Link, var. europœa, Teplouchoff, in Bull. Soc. Nat. Mosc., xli. pt. ii. 249 (1869). Cone-scales rhombic, gradually narrowing in the upper third to a truncate, slightly inflexed, emarginate or denticulate apex. This is the common form of P. excelsa, widely distributed throughout central Europe, and also occurring in southern Sweden. In the Alps it is rarely found over 5000 feet elevation. 2. Var. acuminata, Beck, in Ann. Nat. Hofm. Wien, ii. 39 (1887). Cone-scales, contracted suddenly into a long bifid recurved undulate apex. This variety is of rarer occurrence in central Europe than the preceding ; but is found in the Jura1 and the Alps, and is said to be common in eastern Prussia and in southern Sweden. 3. Var. triloba, Ascherson and Graebner, op. cit. 199 (1897). Scales of the cone trilobed at the apex. This is a much less common varia tion, which has been noticed in a few trees growing at Blankenburg2 in the Harz Mountains, at Soglio8 to the north of Lake Como, and in Moravia.4 II. There appear to be two races of the common spruce in the continental forests, which are mainly distinguishable by the colour assumed by the unripe cones in August. 4. Var. chlorocarpa, Purkyne, in Allg. Forst, ^l. Jagdzeit, lui. i (1877). Cones remaining green in August. 5. Var. erythrocarpa, Purkyne, loc. cit. Cones becoming dark violet in August. Purkyne considered that important differences in the growth of the tree, in the character of the wood, in the staminate and pistillate flowers, and in the soil occupied by each form, were correlated with the differences in the colour of the cones ; but Schroter considers that these are not established, and suggests further investigation. III. The spruce varies much in habit in the wild state, and several remarkable sports have been described. 6. Var. viminalis, Caspary, in Sehr. Phys. Oekon. Ges. Königsberg, xiv. 126 \ Finns viminalis, Sparrman, ex Alstroemer, in Vet. Ac. Handl. Stockh. xxxviii. 310 (1777). Pinus hybrida, Liljeblad, in Svensk FÎ. (1792). 1 Cf. Aubert, Flore de la Vallée âejoiix, 345 (1900). ., -—-. JTJ \-:/~~v A. Braun, in l'erh. Bot. Verein Prov. Brandenburg, xviii. Sitzb. 13 (1876). 3 Ascherson and Graebner, ex Schroter, pp. cit. 204, fig. 31 (1898). 4 Wilhelm, in Oesterr. Forstzeit. i8S8, p. 169. 134° The Trees of Great Britain and Ireland Branches in remote almost horizontal whorls, with very long and slender branchlets (often 10 ft., occasionally 20 ft. long) without or with very few lateral branchlets. Leaves radially spreading. This remarkable form of the weeping spruce was considered by Linnaeus J to be a hybrid between the spruce and Pinus sylvestris. It has been observed in about twenty places in Sweden, where it is vulgarly called Tysk gran or German spruce, in about the same number of localities in Norway, and in isolated cases in Livland, East Prussia, Poland, Thuringia, Tyrol, Styria, Carinthia, Carniola, and Switzerland.2 When sown, the peculiar habit is occasionally reproduced.3 7. Var. pendula, Jacques and Hérincq, Man. Gén. Plantes, iv. 340, 341 A remarkable form of the weeping spruce, narrow and columnar in habit, with pendulous branches almost appressed to the stem. Conwentz4 has described this form, known to him as a single tree8 in the Stellin forest near Elbing in West Prussia, another6 at Jegothen, near Heilsberg in East Prussia, and two others7 near Schierke in the Harz Mountains. Kraemer8 found another in a forest near Kreut in Bavaria. Solitary examples have also been found in Switzerland,9 in northern Hungary,10 and in the Bukowina.11 The seed of the weeping spruce near Jegothen, when sown by Conwentz,4 gave twelve trees, only one of which showed a tendency to the weeping habit. A similar tree with longer leaves, lighter in colour than the typical form, was discovered12 about the year 1860 by Mr. R. Smith Carrington in a plantation near Kinlet Hall, Shropshire, which was propagated by R. Smith and Company, Worcester, who sold it under the name Abies excelsa inverta™ Gordon, Pinet. Suppl. 4 (1862), a name scarcely worth keeping distinct from van pendula, Jacques and Hérincq, which antedated it a few years. A fine example, about 30 ft. high, was growing14 in 1897 at Ide Hill, Sevenoaks, Kent; and a good specimen exists at Murthly Castle. There is also a good example15 at Barbier's nursery, Orleans. Other kinds of weeping spruce, probably including Abies excelsa pendula, Loudon, a form introduced by Booth, are irregular in habit and much more spread ing. A very fine example occurs at Durris. 1 Linnœus refers to it as Abies procera •viminalis in Fl. Swc. 288 (1745). 2 Cf. Schröter, op. cit. 151, who draws attention to the fact that P. Breweriana, of the Siskiyou Mountains, has this habit as a constant specific character. 3 Cf. Wilhelm, in Verh. K. K. Zool. Bot. Ges. Wien, xxxvii. (1887). 4 Beob. Seltene Waldbäume W. Preussen, 135 (1895). 6 Figured in Gartenflora, 1899, p. 618, fig. 86; and by Conwentz, op. cit. 141, figs. 12, 13. 0 Figured by Conwentz, op. cit. 147, fig. 14. 7 Figured in Gartenflora, 1901, p. 315, figs. 48, 49 ; and by Conwentz, op. cit. 150, 152, figs. 15, 16. 8 In Flora, 1841, p. 700. 8 Schröter, op. cit. 156 (1898). 10 Schilberszky, in Kerteszeti Lapok, vii. (1892), describes a weeping spruce near Leutschau. 11 Cf. Oesterr. Forst, u. Jagdzeit. 1897, p. 356. 12 Nicholson, in Woods and Forests, 1884, p. 691 ; and The Garden, xxv. 229 (1884). 13 Picea excelsa inversa, Beissner, Nadelholzkunde, 361 (1891). 14 Gard. Chron. xxii. 368, fig. 109 (1897). Cf. also Gard. Chron. xxix. 263, fig. 98 (1901). V> Figured in Gartenflora, 1899, p. 617, fig. 87; and by Conwentz, op. cit. 163, fig. 17. Picea The "Cornish fir" which was mentioned by Hayes1 as growing in 1794 at Avondale in Co. Wicklow, was pendulous in habit and bore large cones, sometimes nearly a foot in length. The remarkable pendulous spruce2 at Shelsley Walsh, in the Teme Valley in Worcestershire, bears cones 9 in. in length, and appears to be identical with Hayes' variety. 8. Var. columnaris, Carrière, Conif. 248 (1855). Narrowly columnar in habit, with short horizontal branches, clothed with dense short branchlets and foliage. This form, which has been known a long time in cultivation, exists in the wild state in Switzerland, where six trees are known by Schröter in the five localities of Stanserhorn, Stockhorn, la Brévine, Chavannes, and la Berboleuse, all at high altitudes between 4000 and 5800 ft. The columnar spruces is to be carefully distinguished from the narrow spruce, known as the spitzfichte? in which the habit does not result as a sport, but is due to a severe climate, which checks the growth of the branches. The spitzßchte is similar to the columnar spruce in form, being narrowly cylindrical, but the stem is sparingly clad with short branches, wide apart, and forming a thin crown of foliage. The spitzfichte is never seen at low levels in the Alps and Jura, but occurs near the timber line, often forming small groves in exposed situations. This climatic form is much more common in P. obovata in Lapland, Finland, and northern Scandinavia. 9. \ix.pyramidata, Carrière, 247 (1855). Var. stricte, Schröter, op. cit. 158 (1898). Branches ascending at a narrow angle, forming a nearly fastigiate tree. Trees of this kind are occasionally seen in the forests of central Europe, and are rarely found in the seed bed in nurseries. ID. Var. strigosa, Christ, in Garden and Forest, ix. 252 (1896). A form with numerous slender horizontal branchlets, spreading from all sides of the branches, giving the tree the habit of the common larch. This variety occurs in one locality in the canton of St. Gall in Switzerland. il. Var. eremita, Carrière, in Jacques and Hérincq, Man. Gén. Plantes, iv. 341 (1857). A tree of slender pyramidal habit with numerous branches, directed upwards at a small angle with the stem, short stout branchlets, large buds, and distant short thick sharp-pointed needles. Var. Remonti, said by Kent5 to be a dwarf modification of this, is described by Masters6 as of dense compact pyramidal habit, recalling that of Cupressus Lawsoniana, var. erecta viridis. 1 Planting, 165 (1794). It is first mentioned apparently in London Catalogue of Trees (1730), as the long-coned Cornish fir, said to have heen "brought from America some years previously and planted in Devon and Cornwall." 2 Erroneously referred to P. Smithiana (as P. Morinda) in Card. Chron. 1869, p. 713, and xix. 132 (1896). 3 Dr. Christ, iu Garden and Forest, ix. 252 (1896), uses the term columnar spruce for the spitzfichte, which is not strictly accurate. 4 First named and described by Berg, in Jahrbuch K. Sachs. Akad. Forst. Tharand. xiii. 83 (1859). « Veitch's Man. Conif. 433 (1900). e In Card. Chron. vii. 578 (1890). 134* The Trees of Great Britain and Ireland 12. Var. virgata, Caspary, op. cit. xiv. 125 (1873). Abies excelsa, De Candolle, var. virgata, Jacques, in Anti. Soc. Hort. Paris, xliv. 653 (1853). Alles excelsa Cratistonii, Knight and Perry, Syn. Conif. 36 (1850). Picea excelsa, Link, var. denudata, Carrière, in Rev. Hort. iii. 102, fig. 7 (1854). Branches very few and usually not in whorls, elongated, straight or curved, with very few or without branchlets. Leaves radially arranged, either longer or shorter than in the common spruce, persistent ten or twelve years. This variety,1 which is known as the snake spruce, owes its peculiarities to the arrest of nearly all the buds, which do not develop. Most of the examples recorded are young trees, but one2 forty years old at Buttes, near Neuveville in the Swiss Jura, was 40 ft. high in 1898. The snake spruce is not uncommon in Norway, where Schübeler found it in seventeen localities between lat. 59^° and 6i^° ; and also occurs here and there in Sweden between lat. 58° and 63°. Isolated examples are reported from Finland, Livland, and Courland which are probably P. obovata ; and others occur in different parts of Germany. It is common in Bohemia; and one example is known in Moravia. Schröter mentions seventeen trees growing in ten localities in Switzerland. Carrière knew only one example, growing in Cochet's nursery at Suynes, near Brie-Comte-Robert, in Seine-et-Marne. Varieties intermediate between the snake spruce and vars. pendula, monstrosa, and viminalis also occur, but are very rare. 13. Var. monstrosa, Schröter, op. cit. 170 (1898). (Not Carrière.3) Abies excelsa, De Candolle, var. monstrosa, Loudon, Arb. et Fritt. Brit. iv. 2295 (1838). Abies aclada, Salvi, in Flora, 1844, p. 519. Picea excelsa, Link, var. motiocaulis, Nördlinger, ex Willkomm, Forstl. Flora, 76 (1887). This variety, which never develops any lateral branches, has a single thickened stem, bearing leaves near the apex, persistent for many years, and about i^ in. in length. This variety was first described by Loudon, who mentions a single specimen growing in the Chiswick garden, twelve years planted, and about 3 ft. in height. A specimen at High Canons, Hertford, produced cones of the ordinary form in4 1907. Salvi found in 1842 four specimens, growing wild in the Euganean Hills, west of Padua. One of these which was transplanted to Isola Bella in Lake Maggiore, where I saw it in 1909, is attached to a bamboo, and trained up the wall of the château ; it measures about 30 ft. in height and is nearly as thick (i-i£ inch) at the top as at the bottom, bearing leaves with very sharp points only on the upper two feet of the stem. Schröter records another specimen at Stockach in Baden, another in Bohemia, and another at Ansbach in Bavaria. A form of this variety is recorded 1 An analogous form of the common silver fir, Abiespectinata, var. virgata, Caspary, in Bot. Zeit. 778, t. ix. (1882), occurs ; but only four examples are known—two in Alsace, one in the Bohemian forest, and another in the Swiss Jura near Neuveville. The latter is described and figured by Schröter, op. cit. 168, fig. 15 (1898). Cf. vol. iv. p. 722. 2 The oldest known to Schröter was one near Dorpat, in Livland, said by Berg, in Schrf. Natutf. Ges. Univ. Dorpat, ii. t. 2 (1887), to be sixty years old. 3 Carrière, Canif. 248 (1855), wrongly applied the name monstrosa to var. virgata, Caspary. 4 According to Card. Chron. xxv. 146 (1886), var. monstrosa at Lucombe, Pince and Co.'s Nursery, Exeter, produced cones in 1886 which were similar to those of the ordinary spruce. Picea from Silesia and Thuringia, which bears a few undivided branches at the base, the upper part being without branches. 14. Var. globosa, Berg, in Schrift. Nattirf. Ges. Univ. Dorpat, ii. 19, 20 (1887). In this variety, normal growth is replaced by numerous close branches, irregularly dividing into a great number of branchlets, similar to a witches' broom, and forming either a globose bush without any leader, or a conical bush with a leader arising out of a globose base. I saw a remarkable example of the globose spruce in 1909 at the Forestry Experimental Station, Zurich. Seedlings had been raised, one quarter of which had reverted to the habit of the ordinary spruce, the others being very various in appearance and intermediate between the parent form and the normal habit of the species.1 In the true dwarf forms2 of the spruce, the branching is regular, but the growth of the shoots is very small, and the needles are very short. The most important are :— 15. Var. Clanbrassiliana, Carrière, in Jacques and Hérincq, Man. Gén. Plantes, iv. 341 (1857). Abies excelsa, De Candolle, var. Clanbrassiliana, Loudon, Arb. et. Frut. Brit. iv. 2294 (1838). A compact low dense globose bush, seldom higher than 5 or 6 ft. ; branches and branchlets, much shortened ; leaves about £ to \ in. long ; buds very red in colour. This is supposed to have been found on the Moira estate, near Belfast, about the end of the eighteenth century, when it was introduced into England by Lord Clan- brassil. This dwarf form has been found growing wild in Thuringia, and near Stock holm, and in Jemtland in Sweden. It is always sterile, and is propagated by cuttings. Elwes found at Tullymore Park, Co. Down, a large bush of this form measuring ID ft. high and 28 ft. in circumference, which he was informed was either the original or a part of it, and was supposed to be about one hundred and fifty years old. A specimen at Aldenham has reverted to the normal type, and is now growing rapidly into an erect tree. 16. Var. tabulœformis, Carrière, Product, et Fixât. Variétés, 52 (1865), Conif. 333 (1867). A prostrate form, with slender branchlets spreading horizontally over the ground. This is said by Carrière to have been taken, probably as a cutting, from a witches' broom, growing on an ordinary spruce in the Trianon. Torssander3 found a similar plant in Södermanland in Sweden, thirty years old, and only 20 in. high. 17. Other dwarf forms have been named, as vars. pumila*pygmœa? Gregoryana? Maxwelli? etc. 1 Cf. Engler, in Mitt. Schweiz. F'orst. Versuch, viii. pt. il 117, figs. 8, 9 (1904). 2 See under Witches' Brooms, p. 1345. 3 In SSdermanland Botan. Notiser, 1897, p. 169. 4 Beissner, Nadelhohkunde, 365 (1891). 6 Loudon, op. cit. 2295 O^S8)- 6 Said by Gordon, Pinettim, 9 (1875) to have been raised in the Cirencester Nursery. 7 Originated as a seedling in Messrs. Maxwell's nursery, Geneva, New York. Cf. Woods and Forests, 1884, p. 502, and Rehder, in Bailey, Cycl. Amer. Hort, üi. 1333, fig. 1798 (1901), VI C 1344 The Trees of Great Britain and Ireland IV. Several varieties are known in which the leaves are coloured :— 18. Var. aurea, Carrière, Conif. 246 (1855). Leaves yellowish white, shining. The golden spruce has been found wild in Carinthia. 19. Vzr.ßnedonensis, Gordon, Pinet. Supp. 4 (1862). Leaves pale yellow at first, changing to a bronze colour, and ultimately be coming green.1 This originated at Finedon Hall, Northamptonshire, where it came up accidentally in a bed of common spruce. It often loses its colour in cultivation, and at Colesborne has entirely reverted to the normal green. Var. mutabilisz has the young shoots creamy yellow in colour, changing to green by the end of the season. Mr. Bean3 saw in Hesse's nursery, Weener, Hanover, a very beautiful variety, with creamy white young shoots, which is called var. argenteo-spica. 20. Var. variegata. Carrière, Conif. 246(1855). Leaves variegated with pale yellow. A variegated form is mentioned by Loudon ; and Wittrock4 found a tree with leaves variegated white at Helsingfors. V. The colour of the bark of the common spruce varies from whitish grey to brown, probably due to influence of soil and climate. The following sports have been observed. 21. Var. corticata, Schröter, op. cit. 184 (1898). Bark thick, up to 3^ in., longitudinally fissured, and resembling that of a pine in external appearance, though in microscopical structure like the ordinary spruce. Schröter knew in 1898 only six spruces with thick bark, occurring in Austria, Bohemia, Hesse, Bavaria, and Switzerland ; but more than twenty are now known6 in the latter country alone. 22. Var. tuberculata, Schröter, op. cit. 190 (1898). Lower part of the stem covered with corky excrescences, projecting about an inch above the surface of the bark, where side branches are given off.6 Four examples only were known to Schröter in 1898, two in Austria, one in Bavaria, and one in Swit zerland ; but Badoux6 states that many more have since been found in Switzerland. VI. In addition to the varieties and sports just described, which are of unknown origin, there are many peculiar forms of the spruce which are due to external in fluences, and which cannot, properly speaking, be named varieties or sports. 1. The candelabra spruce is often produced, when the leading shoot is broken off by the force of the wind or by other causes. A whorl of secondary branches becomes erect below the broken part of the stem, and forming a series of leaders, grows up, giving the tree a candelabra-like appearance. 2. Dwarf spruces,7 which are mere bushes, with irregular branches, dense 1 Fowler, in Card. Chron. 1872, p. 76, speaks of the inconstancy of the colour in different parts of the tree. 2 Cf. Masters, in Card. Chron. vil 578 (1890). 3 Kew. Bull. 1908, p. 391. * In Hartman, Skand. Flora, 35 (1889). 6 Badoux, injourn. forest. Suisse, 1907, quoted by Beissner, in Mitt. Deut. Dend. Ges. 1910, p. 122. 6 Cf. Cieslar, in Centralblatt Gesamte Forstwesen, xx. Heft 4, pp. 145-149 (1894). Schröter compares these corky excrescences with those developed on the stems of Zanthoxylum, studied by Barber, in Ann. Jlot. vi. 155 (1892). 1 Ficea ellipsoconis, Borbas, Magyar Bot. Lapok, i. 26 (1902), a shrub-like spruce growing as scrub near tree-limit in " the western Carpathians, with short broad cones, is considered by Pax, in Pßanzcnverb. Karpathai, ii. 177 (1908), to owe its peculiarities to the high altitude, similar shrubs being recorded for the eastern Alps by Beck. Picea foliage, and numerous leaders occur in alpine regions, and are due either to the severe climate or to constant cropping by goats and sheep. 3. Witches' brooms on the spruce have hitherto been supposed to be due to the irritation of fungi, bacteria, or mites. Tubeuf,1 in January 1907, sowed seed which he obtained from a witches' broom that had borne cones. The greater part of the seedlings are normal, but a certain number are dwarf and bushy. Tubeuf supposes that some of the former will in time develop witches' brooms on some of their branches, and that the latter will probably remain dwarf, resembling the varieties2 Clanbras- siliana, pumila, etc., already referred to as of unknown origin. 4. Masters3 gives a figure of a remarkable branch of a spruce, in which the leading shoot had split into two portions for some distance, re-uniting above to form again one stem. DISTRIBUTION P. excelsa is a native of Europe, extending from the Pyrenees, Alps, and Balkans northward through south Germany and east Prussia to Scandinavia, and eastward through the Carpathians and Poland to western Russia. In France, the spruce occurs in the mountains, mixed with the silver fir, and in the zone above it, the lower limit in the Vosges and Jura being about 2000 ft. It is not at all abundant in the Vosges, where it ascends to 4300 ft. ; but in the Jura covers large areas, and reaches 5000 ft. altitude. It attains its greatest importance in Savoy, which is the only region in France where the spruce is the dominant tree, forming one half of the whole area occupied by forests. The forest of Thônes,4 near Annecy, which I visited in 1904, is one of the best examples of a spruce forest in France; and is treated on the selection system. It contains about 320 acres, lying between 2500 and 4300 ft. elevation on a steep slope, and is a mixture of two-thirds spruce and one-third silver fir. The standing timber is estimated at 7000 cubic ft. per acre, the annual felling averaging 53 cubic ft., with a revolution fixed at 144 years. The spruce is absent in the Cevennes, and is extremely rare on the north side of the Pyrenees, where it is replaced by fine forests of Pinus montana. Willkomm records it for the Pyrenees of Catalonia and Aragon, where it is not at all common. Its most southerly point in western Europe is the forest of La Cinca, south of Mt. Maladetta in lat. 42° 30'. In Germany, the northern limit of distribution, beginning in the Vosges, passes through the Pfalz, and after crossing the Rhine at lat. 50°, makes a bend to the west ward through Westphalia, and reaches the Weser Mountains, where, near Minden, the spruce attains its most northerly point as a wild tree in western Germany, lat. 52° 20'. From here the limit passes through Hildesheim, Wolfenbüttel, Walbeck near Magdeburg, and Halberstadt to Altenburg ; whence, taking a north-easterly direction, it is continued through Spremburg and Soran to Ostrowo, reaching the Russian frontier at lat. 52°. It then passes northward, parallel with the frontier, to 1 Cf. Prof. Somerville, in Quart. Journ. Forestry, iv. 309 (1910). 2 Cf. var. tabulceformis, Carrière, ante p. 1343. 3 In Card. Chron. xxiii. 274, fig. 52 (1885)- * Cf. Bitll. Soc. Forest. Franclie-ComU, vii. 630 (1904). 1346 The Trees of Great Britain and Ireland Gilgenberg in the province of East Prussia, and reaches the Gulf of Danzig at Elbing. In southern Germany it is scattered, or in small woods in the plains and valleys, being probably planted ;1 and as a wild tree is nearly confined to the mountains, where it occupies a distinct zone, with clearly defined upper and lower limits. The largest pure forests in Germany occur in the Harz mountains, which are almost entirely covered with spruce, ascending to 3300 ft., and in the Iser and Riesen mountains, up to 3900 ft. In the other great forest regions of Germany, as in the Thuringian, Bavarian, and Bohemian forests, the Fichtel and Erz mountains, central Saxony, etc., the spruce, in mixture with the silver fir, covers immense areas. In the province of East Prussia, there are very large forests on the plain, in which the spruce grows in company with the common pine, birch, alder, and willows ; but it is absent on pure sandy soils, where the common pine reigns supreme. The spruce is met with throughout the Alps in Switzerland and Italy, ranging in Tessin between 2700 and 6000 feet, and occupying a small outlying area in the Euganean hills in Lombardy. It is quite unknown in the Apennines. In Austro-Hungary, extensive forests of spruce, often almost pure, occur in the Carpathians from Silesia to Bukowina, and in the Transylvanian mountains. The largest spruce recorded2 is one which grew in the Carpathians, measuring 226 feet in height and 11^ feet in girth at breast height. In the Balkan peninsula8 the spruce reaches its most southerly limit, a line extending from the mountains of northern Albania to the Kopaonik mountain in Servia, whence it is prolonged eastward to the Rhodope mountains in Rumelia about lat. 42°. In Bosnia, Servia, Montenegro, Croatia, and Herzegovina, the spruce usually grows in mixture with the beech and silver fir, occupying a zone on the moun tains between 3000 and 6000 feet ; but in northern Albania the lower limit rises to 4000 ft. Huffel4 states that in Roumania, the spruce attains enormous dimensions, a tree, which was cut down in 1888 in the forest of Tarcau, measuring 195 ft. in height by 3 ft. 3 in. in diameter at breast height ; it was 392 years old. The spruce is much less common in Roumania than the silver fir, ascending to 6000 ft. in Wallachia and to 5000 feet in Moldavia ; while in Bukovina the spruce is more abundant than the silver fir, and occupies a zone between 2600 and 5200 ft. In Russia the southern limit of the spruce (including P. obovata and P. excelsa) extends from the frontier of Galicia, at lat. 50°, eastwards through northern Volhynia and Starodul in the government of Chernikof, crossing the river Oka at lat. 53° or 54°, to the southern boundary of the government of Kazan. From this northwards to the Arctic circle, the spruce is prevalent ; but the exact boundary of the two species is unknown. So far as I have seen specimens, the spruce in Finland and near St. Petersburg is P. obovata, which all authors agree is the only spruce found in north eastern Russia, as nothing like P. excelsa is seen to the eastward of the rivers Dwina 1 It is supposed never to be native in situations below 1300 ft., though it thrives when planted. Left to nature the beech speedily supplants it on all soils at low elevations in southern Germany. 2 Wessely, quoted by Mathieu, FI. Forestière, 541 (1897). I have not been able to verify this record; but Schröter and Kirchner, op. cit. 115 (1906) state that Enderlin measured in the Grisons two trees as follows :—one, 143 ft. high, 6 ft. 3 in. in diameter, with a volume of 1300 cubic feet ; the other, 152 ft. high, 4 ft. 11 in. in diameter, with a volume of 1150 cubic feet. 3 Beck von Mannagettn, Veg. Illyrisch. Land. 287 (1901). 4 Extrait Bull. Minist. Agric. Paris, 1890, p. 6. Picea 1347 and Viatka. Korskinsky states] that the typical form of P. excelsa occurs only in western Russia in the region adjoining the German plain ; and the varieties which he describes as occurring in central and northern Russia, and linking P. obovata with P. excelsa, seem to me to be simply P. obovata with cones slightly larger than those occurring in the Ural range. Von Sievers2 says that the spruce is the only shade-bearing tree in the Baltic provinces, and attains in favourable situations a height of 160 feet. It occurs naturally on better soil than the pine (which occupies poor sandy soil) and competes with the birch and alder on clay, thriving well on deep peat, if this is rich in mineral salts. The original conifer of Norway was Pinus sylvestris, the remains of which are found everywhere in peat mosses. The spruce is a late emigrant from Sweden and Lapland. It occupies in Norway three distinct regions :3— A. The spruce is found in the far north in isolated stations, as on the Varanger- fjord, lat. 69" 30', at Karasjok, lat. 68° 30', and at Saltdalen, on the west coast, lat. 67° io'.4 The spruce here is P. obovata, these stations being outposts of the north Russian spruce, which extends eastwards through Enara Lapland to the Kola peninsula, and through Swedish Lapland from Sulitjelma to Palojuensun on the Muonio river. B. In the Trondhjem district the spruce reaches the coast, and is connected with the northern Swedish spruce, through four passes in the range separating Norway from the Swedish province of Jemtland. This spruce is P. obovata.6 C. In southern Norway the spruce, which appears to be P. excelsa, occupies a distinct area, separated from the last by the Dovre-fjeld, and continuous with the spruce forests of south Sweden, there being no high mountains intervening between the two countries for a considerable distance north of Svinesund. Throughout this region, no remains of the spruce have been found in peat mosses, though those of the common pine are plentiful ; and the spruce is evidently a late emigrant, not having yet reached the west coast. Through Romsdal, Bergenhus, and Stavanger provinces, and the district of Lister, the area covered by forest is not extensive, the principal trees being pine and birch, while the spruce is rarely if ever found wild, except in the inland district of Voss, situated about 40 miles east of Bergen. According to Schubeler, the spruce ascends on the Jotunfjeld to 3250 feet, and in Hallingdal to 3400 feet. South and east of the mountains, the greater part of the very extensive forest area consists of spruce, mixed to some extent with pine and birch. The Norwegian spruce is said to contain a relatively small amount of resin, and is there fore largely used in the production of mechanical and chemical wood-pulp, an industry, which in some places has begun to threaten the continued existence of the spruce forests. The spruce bark is also used for tanning. 1 Tent. Fl. Koss. Orient. 494 (1898). * Forst. Verhält. Bait. Prov. 18 (1903). 3 Cf. Semander, in Engler, Bot. Jahrbuch, xv. 3 (1893). 4 Elwes found it here in 1903 only as a rare isolated tree, and was told that the Ranenfjonl, 50 miles south was its real northern limit. t 6 Specimens collected at Trondhjem and at Bracke in Sweden are identical, and are indistinguishable from specimens gathered in Perm in north-eastern Russia. 1348 The Trees of Great Britain and Ireland In Sweden the exact limits1 between P. obovata and P. excelsa are unknown to me ; but there is no doubt that in the northern part of the country, from Jemtland northwards, P. obovata is the sole species. In southern Sweden, the tree appears to be P. excelsa? and its distribution3 is peculiar, as it does not extend to the extreme south, not occurring in Skâne north of lat. 56° 10', and not extending to nearer the west coast, from Strömstad to Halmsted, than fifteen miles at any point. Its remains have not been found in the peat mosses south of Jönköping on Lake Wetter ; and it is supposed to have spread southward of this point in quite recent times ; and this is confirmed by ancient maps of North Skâne, which show that there were no spruce forests in this district at the beginning of the i7th century. (A. H.) In Scandanavia the spruce is called gran, or rödgran. From what I have seen in the forests of north and south Trondhjem it is usually found on the better class of land, and even there does not grow so large as farther south, ascending to about 2000 ft. in Tydal and Stordal, and attaining about 100 ft. in height by 8 to IQ ft. in girth. I have not noticed, even near the coast, that the trees are browned by the west wind, as they are sometimes even, far inland, in England, but the tree is rarely seen on the exposed parts of the coast, or on the islands, where the Scots pine grows alone. Schübeler, Viridarium Norvegictim, figs. 66, 68, 69, figures three trees remarkable for their habit, one having the branches very short and crowded on the upper part of the stem, and another a good example of the snake spruce, var. mrgata, Caspary. Figures 73, 74, and 75 show instances of natural layering ; and figures 76 and 77, trees grown from a fallen stem. Figure 78 shows a candelabra-shaped tree growing near Horten in the Christiana fjord. It is stated that the varieties known in cultivation as vars. nana, inverta, and Clanbrassiliana have all been found wild on the coast of Norway. The tallest spruce mentioned in Norway by Schübeler was in Hurdalen (lat. 60° 24'), and measured 130 ft. high by 3 ft. 5 in. in diameter; and I am informed of one recently cut in South Rendalen, which was 125 ft. high, and 15 in. in diameter at 80 ft. from ground, and 25 in. at 20 ft. Five logs over 20 ft. long were cut from this one tree. The largest spruce I have heard of in Sweden is mentioned by Schübeler (p. 409). It grew in Oster Gotland (lat. 58°) and measured 150 Swedish feet (44.54 m.), with a diameter of 6 ft. (1.78 m.). In Professor Göppert's memoir4 on the Primœval Forests of Silesia and Bohemia there are many illustrations of the remarkable forms which the spruce assumes when left absolutely in a state of nature, in regions where the snow lies long and deep. These forests are not described in detail, but are above the region of deciduous trees, and consist mainly of spruce and silver fir, with Sorbus Auciiparia, Salix silesiaca, and Lonicera nigra as underwood. Many of the fallen and rotting 1 Wittrock, in Act. Hort. Berg. iv. No. 7, p. 69 (1907), agrees with me in laying stress on the character of the twigs in the discrimination of the two species, P. obovata and P. excelsa, which he considers to exist in Sweden. Cf. his article in Krok, Hartman's Skand. Flora, 1889, p. 34. 2 Cf. Sylven, in Skog. Tidsk. 1909, Fack. pp. 201-261. 3 Cf. Hesselman and Schotte, in Mcdd. Slat. SkogsforsSksanstalt, Heft 3, pp. 1-52, with maps (1906). 4 Göppert, Skizzen zur Kenntniss der Urwälder Schlesiens und Böhmens, in Nova Ada Acad. Leap. Carol. Nat. Cur. xxxiv. (1868). Picea stems are covered with trees which have sprung from seeds germinating in the moss on these trunks. Göppert mentions one about 50 ft. long on which he counted thirty- six living trees of various ages from 4 ft. to 80 ft. high. On another 70 ft. long, there were thirty-two trees from eighty to one hundred years old, all of which had their roots resting on the fallen tree which had given them birth. Such examples are figured in his plates vii., via., and ix. Another form is shown in plate ii., figs. 7, 8, and 9, which illustrate trees which have grown from seeds falling on stumps of broken or dead trees at a considerable height from the ground, and which have forced their roots down through the decaying wood, in one case from a height of 16 feet, to the ground. When the stump decayed the roots were strong enough to support the young tree, which eventually was left standing like a Pandanus on a pyramid of its own roots. In some cases, as shown in plate iv., fig. n, two trees which had originated separately on the same stump became perfectly inarched at the root. Plate iv., fig. 12, shows a remarkable instance of a stump no less than 6 ft. in diameter, which had become covered with a thick layer of moss, and assumed the appearance of a gigantic mushroom, on the top of which no less than seven young spruces from 2 to 40 ft. high were growing without their roots having reached the ground at all. Plate ix., fig. 22, proves, according to Göppert, the immense period which may elapse in these forests before the fallen trees are absolutely decayed and resolved into humus. It shows, A, a fallen tree, of which the wood was nearly all dissolved into long brown pieces, only held together by the overgrowing thick moss, into something like the original shape of the trunk ; B is a tree which had fallen on the top of it at a later period, and was decayed about half through ; C is a living tree estimated at 300 years old, which had germinated on B, and buried its roots partly in and partly on one side of it. Göppert believes that from 1000 to 1200 years may have elapsed since the germination of the lowest tree, A ; but it seems to me that even if it was 400 years old when it fell, the second, B, may have fallen soon afterwards, and owe the slower decay of its wood to the comparative dryness of its position above A. Still it proves that the decay of such a comparatively soft wood as spruce or silver fir (the species is not in this case specified) is extraordinarily slow under the conditions prevalent in these forests. As a proof of its slowness of growth in some instances Forstmeister John remarked that the spruce in the densest parts of the forest attained an age of 120 to 160 years without exceeding 5 to 7 in. in diameter. I have myself cut in Norway a spruce which showed over forty annual rings, and was still thin enough to serve as a walking-stick, which I used through three seasons of elk-hunting before it broke under my weight. In the Böhmerwald the spruce comes to perfection at a higher level than the beech and silver fir, from 3000 to 3400 feet ; and in the Kubany forest there are thousands of trees from 120 to 150 ft. high, and 12 to 16 ft. in girth. It attains a greater age than the silver fir, some trees showing no less than 700 annual rings, though still quite sound. From 3600 ft. up to the highest peaks, which in this range of mountains attain little over 4500 ft., the spruce changes its habit, the stems I35° The Trees of Great Britain and Ireland becoming shorter and the branches more spreading and drooping on account of the heavier snowfall, so that when adult they assume a regular pyramidal or conical shape. At 3500 ft., a tree 3 ft. in diameter showed 420 rings ; and on the top of Kubany at 4100 ft., another with a diameter of 2 ft. had 235 rings. On the Arber mountain, the highest peak of all, at 4200 ft., Göppert saw a tree 3 ft. thick, but only 40 ft. in height ; but even at this elevation the majority of the trees are neither crippled nor diseased, as is often the case near the limit of trees in the Alps and Riesengebirge, where they are covered with lichens. Above these altitudes the lower branches often spread on the ground and form natural layers, which grow upright and make a colony of small trees around their parent. Such an instance is shown in plate i. fig. 2, and another even more curious on fig. 3, where the main trunk of a tree about 5 ft. in girth curved to one side and threw up a secondary straight stem from the nearly horizontal part of its bole. Figure 4 shows a fallen stem 32 ft. long, which remained living and bore no less than five erect trees from 10 to 37 ft. high, which apparently drew the whole of their nourishment from the original roots of the parent tree. Another peculiarity which occasionally appears in these forests are trees with immense swellings on their trunk, in the form of irregular burrs equally developed all round the trunk. Plate i. fig. 5, shows a spruce 18 in. in diameter at the ground, which has a regular swelling, shaped like a flattened orange, no less than 12 ft. in diameter, and from the centre of which the straight trunk again emerges with a diameter of 16 in. Göppert saw in Silesia an even more extraordinary tree (plate i. fig. 6), which was 45 ft. high and 2 ft. in diameter near the ground. At 7 ft. a regular swelling suddenly began (which is described as covered with many branches, but in the drawing shows none on one side) with a diameter of 10 to 12 ft. and 23 ft. high, above which it tapers off into a normal stem. No disease could be found in the bark or wood, which appeared completely sound, and the upper part of the tree is shown crowded with healthy branches. A remarkable group of four spruces growing at timber line (about 1850 metres altitude), on the north side of the Great Scheidegg, is figured by Dr. Klein,1 which from their position appear to have all sprung from seeds which have grown on a rotting trunk. Another remarkable illustration of the effect of wind on the growth of the spruce at high elevations is shown in plate 54 of the same work. A group of trees growing at about 4600 ft. on the Feldberg in the Black Forest, from 10 to 16 ft. high, have the branches cut off clean on the west side, which is attributed by Dr. Klein not alone to the drying effect of the wind in winter and spring, but also to the heating of the branches on one side only by the sun. In the same work are several illustrations of the dense spruce bushes,'called " feisstannli " by the Swiss, which are common in alpine regions, and are caused by the constant cropping of goats and sheep. In the virgin forests of the Capella Mountains in Croatia I saw, in 1910, some spruce of immense height;2 and measured one of about 170 ft. by 12^ ft. I was informed that, in this forest, spruce had been felled 190 ft. high and about 12 ft. in 1 Karsten and Schenck, Vegetationsbilder, \\, t. 38 (1905). 2 Cf. Quart. Journ. Forestry, v. 31 (1911). Picea Ï3S1 girth, and that about 300 years is the maximum age at which this tree remains sound. The spruce is not a native of Britain at the present epoch ; but remains of it have been found in pre-glacial beds at Cromer, Mundesley, Bacton, and Happisburgh in Norfolk.1 CULTIVATION It appears to have been introduced early in the sixteenth century, as Turner includes it in his Names of Herbes published in 1548 ; and both Gérard and Parkinson state that it was found in different parts of Britain. The spruce is easy to raise from seed, but the seedlings grow very slowly for the first three or four years, and are rarely large enough to plant out until they are four to six years old. I have noticed a great deal of variation in the time at which their new growth appears, and it is well to separate the earliest, which are very liable to be injured by spring frost, whilst those which do not start into growth till June remain uninjured. Few conifers are easier to transplant either in spring or autumn, provided the roots are not allowed to become dry ; but if exposed to the air in dry or cold weather a good many will die, or languish for two or three years after planting. The tree grows on almost any soil, but requires a sheltered situation to attain a great height and only comes to perfection where the soil is moist and not liable to dry up in summer. Grown in dense woods, the spruce is liable to be blown down by the wind ; but isolated trees make much stronger roots and are moderately storm-firm. In places near the sea the foliage is often injured by the salt contained in the air, and even as far as forty miles from the Bristol Channel I have seen the spruce completely browned on the side exposed to the wind in March. Though the seed ripens freely in most seasons and germinates readily, the spruce rarely reproduces itself from seed in England owing to its slow growth at first and the weak hold of its young roots on the soil, which cause the seedlings to wither up in summer or to be thrown out of the ground in winter. I only know a few places on my estate where self-sown spruce can be found ; and the seedlings have grown so slowly that I am convinced it is not an economic practice to reproduce spruce by seed, except in places where the ground is under snow for a long period. In the Highlands among heather self-sown seedlings are much commoner ; and on the shores of Loch Rannoch and in some of the old pine woods at Castle Grant there are considerable numbers of self-sown seedlings, but nothing like the number seen in Scandinavia or in the German forests, where they are protected by deep snow for a long period in winter. The spruce is a tree which has been planted more largely in England than it deserves to be ; for though it will, when established, grow on poor ill-drained soil faster than most conifers, yet the value of its timber when felled is less than that of almost any other tree ; and it is, on account of its shallow rooting habit, very likely to be blown down if the wind once gets into the plantation. VI C, Reid, Origin of British Flora, 151 (1899). D 1352 The Trees of Great Britain and Ireland As a nurse it is, on land unsuitable for Scots pine, one of the best we have, if not allowed to overcrowd the hardwoods planted with it; because its branches protect the ground from frost and drought, and its rapid growth acts as a wind break and draws up the other trees. Its roots are more superficial than those of larch or Scots pine ; and it is much cheaper to plant and less liable to injury from frost than silver fir. Loudon quotes (Arb. et Frut. Brit. p. 2305) the experience of W. Adam, Esq., of Blair, Kinross-shire, who was a great advocate of planting spruce as a nurse to oak and elm ; but it must be cut out or its lower branches lopped before it becomes large. In the east and north-east of Scotland the tree seems more promising as a forest tree ; and in the opinion of Mr. Crozier may produce a more valuable crop than either larch or Scots pine at considerable elevations. He gives me particulars of a plantation of 400 acres on the Durris estate, at an elevation of over 800 ft., which was marketed under his own supervision. The age of the trees was sixty years, the number per acre 560, averaging 10 cubic feet each. Sold standing at 5d. per foot they realised £i 16 per acre. Some parts of this plantation planted with Scots pine only made .£15 per acre, and the best of it under larch was estimated at £70 per acre. In this plantation the spruce was planted in patches, none over three or four acres, usually on sites unsuitable on account of excessive moisture for larch or Scots pine. The greater parts of the area, however, might well have been planted with spruce, as the locality is favourable to its growth, and similar results to the above are the rule rather than the exception on the whole estate. I have lately received from Mr. D. Munro of Banchory a photograph (Plate 371) of this plantation, most of the trees in which were blown down after it was sold. Mr. Crozier states that there is a large demand for home-grown spruce boards for box-making in Scotland, but that the wood must be fairly free from knots and discoloration ; and he considers it one of the most useful timbers for house-building. He adds that when planters realise that the limit of altitude for spruce planting lies above the pine belt, and not below it as seems to have been the general idea, and that it must be grown thicker than larch or Scots pine, its economic merits will become more evident than they are at present. In confirmation of Mr. Crozier's opinion that spruce is a valuable tree for planting for profit at a high elevation, I may refer to Messrs. Robinson and Watt's very full report1 on the Coombe Plantation, which lies between 900 and 1500 ft. altitude near Keswick in Cumberland. This plantation, which was sixty-one years old in 1910, consists mainly of larch, with a mixture of spruce, amounting to only one or two per cent at the lower levels, but to ten per cent at over 1250 ft. These authors state that here " spruce grows well at all elevations, and everywhere attains a greater volume than larch under the same conditions." At the higher altitudes it much exceeds the larch in volume. The influence of altitude and exposure on the spruce itself is shown in the following table :— 1 Journ. Board of Agric. xvii. 273, 360 (1910). Picea 1353 Elevation : feet. goo 1150 1250 MS0 1520 Height of tree in feet. 80 not stated Quarter-girth measurement, in cubic feet. 44 32 26 48 35 I have failed to obtain any other exact and reliable figures as to the value of a crop of spruce grown in England, except on such small and isolated patches of land, that they would give no fair criterion. As a shelter tree it makes a good edging to the roads in a plantation, and can be headed down or clipped when it has grown tall enough to keep the wind out. It bears clipping well and makes a good dense hedge on soils not liable to drought. Sargent1 states that as an ornamental tree in America, it loses vigour at twenty-five to thirty years old, except in the most favourable situations; and he only recommends it as a nurse for other trees, as it is very hardy and grows rapidly at first. As a proof, however, of the extremely vigorous growth of the spruce in America, I may say that the tallest tree in Mr. Hunnewell's Pinetum at Wellesley, Mass., which I had the pleasure of visiting with Professor Sargent in 1904, was a Norway spruce which was planted about 1852, and was in 1894 80 ft. high with branches spreading over a circle 60 ft. in diameter. When I saw it, it had increased little in height, but its lower branches had spread to 75 ft. diameter and some of them had rooted ; flowers were just showing on May 9, and cones were produced on branches close to the ground, which is rarely the case in Europe. This tree is figured on Plate 340. According to Pinchot,2 it thrives throughout the entire north-east of the United States and southward at the higher elevations ; but in the west, favourable results have been attained only as far as the eastern part of the prairie region, and then only in the more protected localities. He considers that it should be planted on a large scale in the cut-over land in the north, where the tree will provide a future supply of wood pulp, as it is in every way superior to the native spruces. REMARKABLE TREES If I could trust the measurements which have been given me I should say that the tallest spruce in this country is a tree at Rooksbury Park, Hants, the seat of J. C. Gamier, Esq. It is in a densely crowded thicket of rhododendron, surrounded by beech, and was said by Mr. A. Arnold to measure no less than 178 ft.; but after seeing it twice I could not believe that it was over 150 ft, and, owing to its position, could not measure it myself.3 In Oates Wood at the top of Cowdray Park, near the superb silver fir figured 1 Silva N. Amer. xii. 24, note (1898). 2 U.S. Forest Service, Planting Leaßet, No. 20 (1908). 3 At my request Mr. Arnold has recently re-measured this tree with a theodolite, and informs me that though he could not get a clear view of its top, he now estimates it at 149 feet. r354 The Trees of Great Britain and Ireland on Plate 208, there are two remarkably tall trees, which in 1903 I estimated at 140 to 150 ft., but owing to the steep slope on which they grow, and to the adjoining trees, I could not measure a base line. They were 11 ft. 6 in. and lo^ ft. in girth in 1906. Mr. Harold Pearson has recently had these measured by Mr. T. Roberts, forester at Cowdray, who informs me that he levelled a base line, and found the height in January 1911 to be 153 ft.; but this tree which has three leaders is not so handsome as the other, which he thought was about 148 ft. high. The tallest which I have myself measured are two trees growing on the edge of the lake near Fountains Abbey, Studley Royal, Yorkshire, which Loudon describes as the tallest spruces known to him, and says were 132 ft. high. When 1 saw them in 1905 I found one to be 140 ft. by 12 ft. 10 in., and the other 137 ft. by ii ft. They are free from branches for 30 to 40 ft, and seemed in excellent health, though probably over 150 years old (Plate 341). In Earl Bathurst's woods at Cirencester there are two narrow avenues of spruce known as the Cathedral firs, because they resemble the cross aisles of a cathedral. Of these Plate 342, from a negative taken by Mr. T. E. Gerald Strickland, gives an excellent picture as they were four years ago, but since then several have been blown down, one of which was over 100 ft. high, and showed on the stump 134 annual rings. Those standing average from 110 to 120 ft. high by 8 to 10 ft. in girth. On my own land at Lyde near Colesborne, in a deep sheltered valley, there is a tree about 125 ft. by 8 ft., but this is beginning to decay at the base, though not much over 100 years old. At Bowood, Wilts, I saw a very fine tree which measured, in 1908, 125 ft. by io ft. 8 in. ; but there may be better ones here. On the Earl of Powis's estate at Walcot, Shropshire, there is a wood of spruce about loo years old on a steep hill-side next to the Plassey plantation, a photograph of which has been reproduced in the Quarterly Journal of Forestry, iii. p. 358. I have seen no spruce plantation in England which equals this, and am indebted to Mr. R. H. Newill, agent for the estate, for the following account :— " When I came to measure up an area in the Spruce Plantation, near Plassey, I found it difficult to find a piece without any gaps in it, as the wind has been busy of late years, and has blown down many trees. Eventually I chose a piece near the top and squared i^ chain along the bank and i chain down it, an area of 0.15 acre. On this were twenty-two trees standing, of which I enclose measurements.1 No. i 2 3 4 5 6 7 8 Length, Feet. 80 76 79 80 7i 84 81 83 Quarter- girth, Inches. 9* 9* 12 8 9 10 9* 9i Cubic Feet. 50.1 47-7 79- 35-6 39-11 58-4 50.9 52- No. 9 10 n 12 13 H 15 Length, Feet. 80 82 75 80 80 35 (70 \I2 Quarter- girth, Inches. 12 «si n 14 12 9 13* 6* Cubic Feet. 80 103.9 63 loS.IO 80 19.8 88.7 3-6 No. 1 6 17 18 19 20 21 22 Length, Feet. 78 80 81 76 ( 39 21 ( 15 75 74 Quarter- girth, Inches. 12 II* 8* 13* '3i 7 5 9 10 Cubic Feet. 78 73-5 40.7 96.2 49-4 7-1 2.7 42.2 51-4 Total 22 trees, 1401 cubic feet. Picea These vary from 19 to 103 cubic feet, and average 63.7 cubic feet. I found stumps of seven more trees on the area, and taking them at the same average there would be about 1849 feet ; or about 193 trees measuring 12,326 cubic feet per acre. The age of the trees is about 100 years ; the rings are well marked to eighty- five years, afterwards so very close together that it is difficult to count them. All the trees are going back very fast, I believe every one is decayed at the butt ; and in the lower part of the plantation many are blown down or broken off each year. We could only obtain 3d. to 4d. per foot for this class of timber, and it was in order to turn it to a more profitable use, that I put down the creosoting plant." Assuming that this plantation had been clean felled at eighty years of age and that it had then contained 10,000 cubic feet per acre, the annual increment would have been 125 feet per acre; and taking the price at 4d. per foot standing the value of the crop would have been ,£166 per acre. The trees are facing north-east at an elevation of 600 to 900 ft., and the old red sandstone here seems to suit all kinds of trees, both hardwoods and conifers, as well as any soil in England. At Kyre Park there is a remarkable old spruce of the candelabra type which has an immense rugged bole broken off at about 30 ft., and 15 ft. 9 in. in girth. One of its upright branches is no less than io ft. 4 in. in girth, and twelve others have naturally layered themselves in a circle 64 ft. in diameter, and grown up into trees, two of which are 90 ft. high by 8 ft. in girth. Another remarkable instance of layering in the spruce is at Langley Park, Slough, where a tree on the lawn has been broken off at about 20 ft. and whose lower branches have formed a complete bower, resembling on a smaller scale that formed by the Whittingehame yew. Some of the small branches, only one to three inches thick, have formed a woody mass and thickened enormously at the point where they have taken root. In Wales the finest spruce we have seen are in a wood above Gwydyr Castle, where in 1905 I measured two trees in a grove round the bowling green, which were about 125 ft. high by 9 ft. 8 in. and 6 ft. io in. in girth. In this grove, which is shown in Plate 343, the spruce seems to clean itself better than in England, and I estimated that there might be 8000 to 10,000 cubic ft. per acre. Mr. Richards, forester to Lord Penrhyn, informed me that at Tyn-y-Coed in the same district of North Wales, a spruce plantation was felled in 1902 and sold to Mr. J. Jones of Liverpool, a tree in which is said to have been 149 ft. high, and that 158 trees in this wood contained 11,937 cubic ft., an average of over 75 cubic ft. ; two of them measuring respectively 80 ft. by 23 in. quarter-girth = 294 cubic ft., and 67 ft. by 27 in. = 338 cubic ft. It is evident from these figures that even if the value of the timber is low as compared with imported spruce, yet that it may pay well in this particular district, provided the trees are grown thickly enough. In Scotland the largest spruce of which we have any record grew at Blair Atholl, and was visited by the Scottish Arboricultural Society in 1879. It was then said to measure 142 ft. high, and to contain over 420 cubic ft. of timber.1 I 1 Hunter, Woods of Perthshire, 60 (1883). 1356 The Trees of Great Britain and Ireland was informed by the late Mr. Pitcaithley, forester to the late Earl of Mansfield, that this tree was blown down about 1893, when the height above given was verified. The tallest that we now know of, are probably some trees on the banks of a deep glen at Dupplin Castle, which I saw in 1907, and on account of their leaning to one side could not measure accurately, but thought must be from 130 ft. to 140 ft. high. A fine tree in the same place, dividing into two stems at thirty feet from the ground, measured about 100 ft. by 12 ft. 10 in. At Methven, Henry measured a tree in 1904 as 125 ft. by 8 ft. 3 inches. At Inveraray I saw trees over 120 ft. by 8 to 10 ft., and the forester, Mr. Campbell, told me that he had measured one blown down on Ben-y-Cuach 130 ft. long. In the woods of Glenaray the spruce seems to grow very well, being sheltered from the westerly gales ; but I do not remember to have seen such large or thriving trees elsewhere on the west coast of Scotland. In the east of Sutherlandshire, I am told by Mr. Gillanders that spruce grows well and cleans itself better than in the south. In Ireland we have not seen any trees of extraordinary size, and as a rule the climate is not adapted to the production of high-class spruce timber. But Mr. A. E. Forbes has sent me an account of a remarkable plantation near Fermoy, which I reproduce verbatim, and am indebted to Lieut. and Quartermaster T. Smith, R.E., for a negative which gives a good idea of this plantation (Plate 344). " A very fine clump of common spruce is growing in Glenshiskin Wood, which forms part of the property purchased by the War Department a few years ago near Kilworth. This wood occupies a valley running into the Kilworth mountains, a low range of hills formed from the Old Red Sandstone formation. A small mountain stream flows down the centre of this valley, and at one point, about 300 feet above sea-level, an alluvial deposit has been formed along its course of about an acre in extent. In this deposit a clump of almost pure spruce was planted about eighty years ago, and judging from appearances was never thinned or attended to in any way. From time to time poles were doubtless removed from it as required ; but no systematic thinning could have been carried out, as many of the trees still stand within four or five feet of each other. " This clump probably presents as fine an example of spruce growth in Britain as can be found anywhere. The trees vary in height from 90 to no ft., and form long clean poles with little taper, and ranging from 8£ to 18 in. quarter-girth at 4^ ft. from ground. The trees in two-thirds of an acre were carefully measured by Mr. M'Rae, forester at Dundrum, Co. Tipperary, in the spring of 1910, and the summary of the results obtained is given below :— Picea T357 Spruce Larch Stumps of Felled Trees No. 161 ii I 10 Age. 78 years » ... Average Height. 98 feet 99 » ... Average Quarter- girth at 4^ ft. 12^ in. 1 2 £- in. ... Total Cubic Contents. 8050 55° ... 8600 "By dividing the trees into three stem classes, a volume of 9600 cubic feet was obtained. Assuming the estimate of 8600 ft. to be correct, the average contents of the trees is about 50 ft., and the total yield per acre would be over 12,000 cubic ft., which for the period of eighty years is higher than anything I have heard of in Great Britain. " Within recent years, trees similar in size to those still standing have been removed, and the stumps still exist. It is quite possible, therefore, that the existing crop is smaller in volume than that which stood on the ground a few years ago. The high yield is, of course, largely due to the exceptionally favourable soil and situation of the site on which the trees were grown. A fine, rich, and well-drained soil, well provided with soil and atmospheric moisture, and a situation sheltered from all winds, provide ideal conditions for the growth of spruce or any other tree able to thrive with a moderate amount of summer heat. Oak and beech growing in the immediate vicinity of these trees are very poorly developed and covered with lichen and moss, indicating the cool and humid conditions which prevail." TIMBER Next to that of the Scots pine, the wood of the spruce is the largest import from the Baltic ; and from Norway the proportion of spruce timber is probably greater. On account of climatic and economic causes, it seems probable that this will always be the case, though in the west coast ports American spruce takes its place. For scaffold and ladder poles, small spars and masts, and oars, we cannot hope to compete with the north of Europe ; whilst for flooring, joists, and almost all purposes except those for which knotty boards are not objected to, it seems equally hopeless for British growers to attempt to compete with the well-known white deal of commerce. The reasons why the value of home-grown spruce timber is so low are, first, its very knotty character, caused by the persistence of the branches, which die more slowly than those of other conifers ; and, secondly, its want of strength and durability as compared with larch. Continental foresters tell us that the first defect may be obviated by close planting, and cite the large profit which is derived from this tree in Germany and Scandinavia. I have inquired of many of our best practical foresters ; but I have never been able to find any plantation in England, and only very small areas in Scotland, Wales, and Ireland, where spruce, which stood close 1358 The Trees of Great Britain and Ireland enough to kill the branches, have attained a considerable size, or where a spruce plantation has been a really profitable investment. My own experience is, that land where spruce may be well grown is fit to produce a much more valuable timber, and that on ordinary land it will starve to death before it will clean itself from branches. The late Mr. Philip Baylis, Deputy- Surveyor of Dean Forest, told me that spruces there, 50 to 60 ft. high, and so thickly planted that no vegetation would exist under them, still retained the dead branches to within 5 or 6 ft. from the ground ; and I think that this will apply to most places in England and Scotland. When the tree is of large size it usually becomes rotten at the heart near the ground ; and the top is often broken by the wind. Though the timber may be worth 4d. to 6d. per cubic foot for rough boarding or packing-cases, or for temporary sleepers and pit props in collieries or railways under construction, yet in quantity it is the most unsaleable wood we have. When, as often happens, large quantities are blown down by a heavy gale, I have known cases where no one would go to the expense of cutting up and removing the trees if they had them for nothing, and the proprietor has had considerable expense in doing so without any return whatever. When blown down, the shallow spreading roots tear up the ground for some distance round the tree and are very costly to get rid of, or if left leave the ground in a bad state for re-planting. Where, however, the soil and climate allow the spruce to be crowded closely enough to clean itself, before it becomes rotten at heart or is blown down, spruce timber may be used for estate building purposes, if not with actual economy, yet in many cases more advantageously than by selling it. Sixpence per foot is something like the average price, though 3d. to 4d. often has to be accepted. On shallow and dry soils the spruce often begins to decay at the heart for some feet from the ground at the age of fifty to seventy years, and on such soils should not be planted at all. Its spreading roots, which are extremely tough and elastic, are used in Scandinavia for the knees of boats, though rarely so utilised in England. The tough and durable branches made into a wattled fence will last for a long period, and are the common farm fence in many parts of Norway and in the Alps. When facilities exist for creosoting, spruce may be used for fencing and other outside work, such as sheds and outbuildings ; but unless treated with some preservative it soon decays when exposed to wet and dry. The spruce trees which produce the bois de resonance, used for sounding- boards in musical instruments, grow at high elevations in the Alps, the Jura, and in the Bohemian and Bavarian forests. These are very old trees, the growth of which has been extremely slow and very uniform, the annual rings not exceeding 1*2 in., and containing only a slight amount of autumn wood. These trees are usually covered with lichens, and their selected timber sells at very high prices, as much as 95. to 12s. per cubic foot. Burgundy pitch is a resinous product of the spruce, well known under the name of Burgony Pitch and Pix Burgundica as long ago as 1640. It was formerly Picea produced in the Vosges Mountains, but now, according to Flückiger and H anbury,1 mainly in Finland, the Black Forest, Austria, and Switzerland. Flückiger states that at Oppenau, in Baden, the principal place of its manufacture in Germany, it is mixed with French turpentine from Bordeaux and with rosin from N. America ; and the tapping of the trees in Government forests in Baden and Württemburg is now prohibited on account of the injury caused thereby to the timber. It is very generally adulterated in England, and is mainly used as an ingredient in plaisters.1 (H. J. E.) PICEA OBOVATA, SIBERIAN SPRUCE Picea obovata, Ledebour, Fl. Alt. iii. t. 499, iv. p. 201 (1833); Trautvetter, in Middendorf, Reise, i. pt. ii. 87, 170 (1847); Maximowicz, Prim. Fl. Amur. 261 (1859); Regel, Tent. Fl. Ussur. 137 (1861); Herder, in Hot. Jahrb. xiv. 160 (1891); Willkomm, Forstliche Flora, 93 (1887); Kent, Veitch's Man. Conif. 441 (1900); Komarov, Fl. Mansh. i. 197 (1901); Clinton-Baker, Illust. Conif. ii. 42 (1909). Picea vulgaris, Link, var. altaica, Teplouchoff, in Bull. Soc. Nat. Mosc. xli. pt. ii. 250 (1869). Picea excelsa, Link, var. obovata, Schröter, in Viertel/. Naturf. Ges. Zürich, xliii. 138 (1898). Pinus Abies, Pallas, Fl. Ross. i. 6 (1784) (not Linnaeus). Pinus obovata, Antoine, Conif. 96 (1840-1847). Pinus orientalis, Ledebour, Fl. Ross. iii. 671 (in part) (1847-1849) (not Linnaeus). Abies obovata, Don, ex Loudon, Arb. et Frut. Brit. iv. 2329 (1838). A tree, attaining in Russia and Siberia the dimensions of P. excelsa, which it resembles in habit of growth and in bark. Young branchlets reddish brown, covered with a dense minute pubescence, which is retained for several years, the older branchlets becoming greyish yellow. Buds, about £ in. long, conic, composed of closely appressed scales, rounded at their apices ; terminal bud girt with a ring of keeled acuminate ciliate scales, and closely surrounded at the base by the uppermost leaves. Leaves, arranged as in P. excelsa, deep green in colour, | to f in. long, ending in a short point, quadrangular in section, with three to four stomatic lines on each side. Cones i\ to 3^ in. long, \\ to i \ in. in diameter when open, shining brown when ripe ; scales numerous, thin, tough, flexible, longer than broad, T60 to & in. wide, and ^ to ^ in. long, fan-shaped, widest near the upper edge, tapering to the base on each side ; upper margin thin, undulate, rounded or with a slightly projecting occasionally bifid apex ; exposed part pale brown, glabrous ; concealed part reddish brown, minutely pubescent; flat or slightly concave internally from side to side; bract £ in. long, lanceolate, narrowing to an acute denticulate apex. Seed £ in., brownish black ; seed with narrow wing f to § in. long, broadest near the rounded denticulate apex. The description of P. obovata given above is drawn up from specimens procured from Siberia, and from Perm in Russia, by Mr. H. Clinton-Baker, from specimens collected in Finland by Mr. M. P. Price, and from specimens which I gathered in VI 1 Pliarmacographia, 616 (1879). E 1360 The Trees of Great Britain and Ireland northern Sweden, near Bracke, and in Norway, near Trondhjem ; all agreeing in the character of the cones, branchlets, buds, and leaves, and constituting, in my opinion, a species distinct from P. excelsa, of which P. obovata is generally considered to be a variety by Schröter and other modern botanists. These authorities have apparently paid no attention to the characteristic pubescence of P. obovata, a matter of importance, as in the genus Picea the presence or absence of pubescence on the branchlets is one of the most diagnostic features in the discrimination of the different species. The cones, moreover, are amply distinct in the two species. P. obovata varies somewhat in the size of the cones and in the shape of their scales ; and two main varieties have been distinguished, which are, however, connected by intermediate gradations. These varieties are : (a) the typical form described above, which is characterised by the scales of the cone being entire on margin ; and (b) var. fennica. 1. Var./enm'ca, Henry. Picea excelsa, Link, var. fennica, Schröter, in Viertelj. Naturf. Ges. Zürich, xliii. 138 (1898). Picea excelsa, Link, var. medioxima, Willkomm, Forst. Fl. 75 (1887). Picea vulgaris, Link, var. uralensis, Teplouchoff, in Bull. Soc. Nat. Mosc. xli. pt. ii. 250 (1869). Finns Abies, Linnaeus, \ax.fennica, Regel, in Gartenflora, xii. 95 (1863). Pinus Abies, Linnaeus, var. medioxima, Nylander, in Bull. Soc. Bot. France, x. 501 (1863). Pinus Picea medioxima, Christ, Flore de la Suisse, 254 (1883). Abies medioxima, Lawson, Pinet. Brit. ii. 159 (1867). Cone-scales, with their upper margins rounded and finely denticulate. Leaves dark green in colour. According to Schröter this variety occurs sporadically in Amurland and Siberia, and is the common form in the Ural range and throughout Russian Lapland, northern Sweden, and northern Norway, occurring with less frequency in Finland, Livland, Kazan, and Poland. Solitary trees with cones similar to this variety have also been recorded from numerous stations in the mountains1 of central Europe, from the Vosges and Jura throughout the Alps to the Carpathians and Bosnia. 2. Var. alpestris, Henry. Picea alpeslris, Stein, in Gartenflora, xxxvi. 346 (1887). Picea excelsa, Link, var. alpestris, Schröter, op. cit. 141 (1898). Abies excelsa alpestris, Briigger, in J. B. Naturf. Ges. Graubundens, xvii. 154 (1874), and xxix. 122 (1884). Abies excelsa medioxima, Heer, in Verh. Schw. Nat. Ges. 1869, p. 70 (not Nylander). Cones 3 to 5 in. long, with scales rounded and entire in upper margin. Leaves short, ^ to f in. stout, very glaucous. Trees with a whitish grey bark, and with remarkable bluish white foliage, which have been found at high elevations (between 4400 and 6400 ft.) in a few localities in the Swiss Alps, from Landbeck in the Tyrol2 to Engstelnalp in the Bernese Oberland, and from Lake Walen to Lake Como. These trees were first investigated by Heer and Briigger on account of the special name given to them by 1 Cf. Christ, in Garden and Forest, ix. 273 (1896). 2 Beissner, in Mitt. Dent. Dend. Ces. 1905, p. 143, describes trees like var. alpestris in the Engadinc. Picea 1361 the peasants, aviez selvadi, or wild silver fir, the common spruce being known as pign. I have seen no specimens, but apart from the glaucous foliage, which is a trivial and inconstant character in conifers, P. alpestris would seem to be identical with P. obovata. A vast amount of literature' has been written on the relationship of P. obovata to P. excelsa, the general result of which shows that a complete series of transitional forms connecting the two species may be found ; but these are only met with in the regions where the two spruces come in contact—elsewhere they are quite distinct and easily recognisable. It is possible that these transitional forms are due to hybridisation ; and further study by experimental sowings is needed to clear up the matter. P. obovata is the most widely distributed of all the spruces, extending over the vast northerly region of eastern Europe and Asia, where the climate is severe in winter and continental in character. It occurs in northern Scandinavia, Lapland, Finland, northern and eastern Russia, throughout Siberia to the Sea of Ochotsk and Kamtschatka,2 and in Manchuria. It extends far to the northward, reaching lat. 67° in the Kola peninsula, lat. 68° in the Ural range, attaining its most northerly point in Siberia on the Yenisei at lat. 69° 5', and crossing the Stanovoi mountains at lat. 64°, where it comes in contact with P. ajanensis. According to Komarov8 it is abundant throughout the wooded parts of Manchuria, where it grows along the banks of rivers, either forming pure woods or scattered amidst other trees. Its eastern and southern limits in Asia are imperfectly known, but it forms great forests in the mountains of Dahuria and in the Altai and Sayan ranges. Seebohm4 describes it as extending on the Yenisei " nearly as far north as the larch, where it is a very important tree for commercial purposes. Its wood is white, of very small specific gravity, extremely elastic ; and it is said not to lose its elasticity by age. It makes the best masts for ships, and is for oars the best substitute for ash. Snow-shoes are generally made of this wood. The quality is good down to the roots, and it makes the best knees for shipbuilding." In European Russia its southern limit is the northern edge of the Orenburg steppe ; and it forms vast forests in the governments of Perm, Vologda, Ekaterin burg, Ufa, Viatka, and Kama, that are either pure or mixed with larch, Pinus Cembra, Abies sibirica, and birch. It appears to be the spruce prevalent in Finland and in the Baltic provinces ; but in western Russia is mixed with P. excelsa, the limits between the two species being undefined, owing to the occurrence of transitional forms. Similarly in Scandinavia5 it is the common spruce in the north, while in the south P. excelsa appears to be the prevalent form. Its occurrence as a sporadic tree in the mountains of central Europe, under the form described as P. alpestris, 1 Cf. Teplouchoff, lee. cit. Korshinsky, in Tentamen Fl. Ross. Orient. 493 (1898), admits that cones like those of P. excelsa are never seen in eastern Russia. At the junction of the rivers Kama and Viatka the woods are said to be composed of both species. Cf. Kihlman, Pß. Stud. Rtiss. Lapland, 143 (1890), on the variation of the spruce in Finland, Lapland, and northern Scandinavia. Dammer, in Gard. Chron. iv. 480 (1888), may also be consulted, as well as the numerous authorities quoted by Schröter, op. (it. 240 (1898). 2 It is a doubtful native of the Kurile Isles, according to Miyabe in Mem. Boston Soc. Nat. Hist. iv. 261 (1894). 3 Flora Manshuriat, i. 197 (1901). * Siberia in Asia, 233 (1882). 6 Cf. under P. excelsa, pp. 1347, 1348. 1362, The Trees of Great Britain and Ireland is peculiar ; but may be explained as a remnant of the pre-glacial forests.1 In habit, P. obovata is usually more columnar than P. excelsa ; but little reliance can be placed on this character as a mark of distinction. This species had not been introduced into England in Loudon's2 time; and it is very doubtful if it occurs in cultivation in this country, except at Bayfordbury, where seedlings were raised in 1908 from seed brought from Siberia by Mr. C. F. H. Leslie. According to Kent,8 "the Siberian spruce soon perishes under the stimulus of the high temperature of this country." Small trees in botanic gardens reputed to be this species appear to me to belong to the transitional form between P. excelsa and P. obovata, which has less pubescence on the branchlets. Plants raised from Finnish seed, procured from Rafn, are much slower in growth at Colesborne than common spruce. In Germany, according to Mayr,4 it is slower in growth than the native spruce, and is not more hardy. It appearsB also to be equally slow in growth in the Arnold Arboretum, U.S.A. (A. H.) PICE A ORIENTALIS, CAUCASIAN SPRUCE Pkea orientalis, Carrière, Conif. 244 (1855); Boissier, Fl. Orient, v. 700 (1884); Masters, in Gard. Chron. xxv. 333, fig. 62 (1886), and Hi. 754, fig. 101 (1888); Kent, Veitch's Man. Conif. 443 (1900); Clinton-Baker, Illust. Conif. ii. 44 (1909). Pinus orientalis, Linnaeus, Sp. PI. 1421 (1763); Lambert,6 Genus Pinus, i. t. 39, fig. a (1803). Abies orientalis, Poiret, in Lamarck, Diet. vi. 518 (1804); Loudon, Arb. et Fruit. Brit. iv. 2318 (1838). A tree, attaining in the Caucasus 180 ft. in height and 12 ft. in girth. Bark brown, fissuring irregularly on old trees into thin scales. Young branchlets pale brown, slender, densely covered with a short pubescence, retained in the second and third years. Buds conical, acute, about \ in. long, brown ; terminal buds girt at the base with a few keeled acuminate scales. Leaves, on lateral branches arranged as in P. excelsa, very short, £ to f in. long, dark green, shining, bevelled and obtuse at the apex, quadrangular in section, with one to four lines of stomata on each of the four surfaces. Staminate flowers, cylindrical, J in. long, carmine red in colour; anther connective suborbicular, minutely denticulate. Cones, 3 to 4 in. long, f to i in. in diameter when closed, cylindrical but tapering to a narrow apex, violet coloured7 when growing, brown when ripe ; scales 1 Christ, Flore de la Suisse, 197 (1883), compares the distribution of this spruce with Pinus syhestris, var. etigadinensis, the pine on the Engadine, which he considers to be identical with Pinus lapponica, Mayr, the form of the common pine that occurs in northern Scandinavia and Lapland. Cf. ante, vol. iii. 573. 2 Cf. Loudon, Trees and Shrubs, 1030 (1842). * Veitch's Man. Conif. 442 (1900). « Fremdländ. Wald- u. Parkbäume, 333 (1906). 6 Sargent, in Garden and Forest, x. 481 (1897). 6 The cones figured by Lambert in this edition, t. 29, fig. b, were from China, and are possibly those of P. ajanensis. Lambert, in his second edition, t. 39 (1832), gives a new and coloured drawing of leaves and cones, collected by Sir Gore Ouseley near Tiflis, repeating also the figures of the cones from China. 7 The scales of young cones are green, with a narrow carmine-coloured margin. Picea obovate with a cuneate claw, £ to f in. wide, rounded entire and slightly bevelled in the upper margin ; bract £ in. long, with a narrow claw and a rectangular lamina, truncate at the apex. Seed dark coloured, £.in. long, with the wing J in. long ; wing broadest about the middle, upper margin rounded. This species is readily distinguished by its very short blunt leaves, and pale brown pubescent branchlets. DISTRIBUTION The oriental spruce is a native of Asia Minor and the Caucasus. It is widely spread in most of the mountain ranges of Asia Minor, being recorded for Troas, Mysia, Galatia, and Phrygia, where it generally occurs between 3000 and 7000 feet elevation. It is also met with in the valleys of the Antitaurus. It is, however, much more common, forming large forests, in the mountains between Trebizond and Erzerum, where it was discovered by Tournefort1 at the beginning of the eighteenth century. In the Caucasus it is generally associated with Abies Nordmanniana, and occurs in Georgia between 2500 and 7500 feet. In the Lesser Caucasus its eastern limit is the meridian of Tiflis, being totally absent to the eastward and in the province of Talysch. As a rule it ascends higher than Abies Nordmanniana, occasionally form ing the timber line at 7500 feet. The largest tree recorded by Radde,a measured, when felled, 184 ft. in height, with a diameter of stem of 4 ft. i in., and a cubic content of 925 ft.; it was 390 years old. CULTIVATION The species, according to Beissner,8 was introduced into Europe in 1837, but Loudon, writing in 1838, speaks of it as not in cultivation; and it appears4 to have come into this country in 1839. It has been in cultivation in the United States5 since about 1850, where it has proved hardy as far north as eastern Massachusetts, and is one of the most beautiful of all the exotic conifers that have been planted in the neighbourhood of Boston. (A. H.) REMARKABLE TREES None of the spruces seems more generally successful in cultivation than this ; and though it does not grow so fast as the common or the Sitka spruce, it is a really good ornamental tree, hardy in all parts of Great Britain, and ripening seed in most places. We have measured many specimens of from 60 to 70 ft. high and a few taller, among which the following may be mentioned :— At Dogmersfield Park, Hants, the seat of Sir H. Mildmay, a fine tree with many cones, 78 ft. by 7 ft. 8 in. in 1907. At Strathfieldsaye a handsome specimen 76 ft. by 7 ft. 8 in. At Highnam a tree about 67 ft. by 7 ft. in 1905. At Penrhyn a tree recorded6 as 58 ft. high in 1891, which was, when measured by me in 1906, 75 ft. by 5 ft. IG in. 1 Voyage mi Levant, 288 (1717). 2 JCaukasuslandern, 223 (1899). 3 Nadelhohhmde, 374(1891). 4 Lawson, Pinet. Brit. ii. 163 (1865). Loudon, in Trees and Shrubs, 1029 (1842) says: "Of late many plants have been raised in Knight's exotic nursery, from seeds received from Mingrelia and the neighbourhood of Tiflis." 6 Sargent, Silva N. Amer. xii. 22, note (1898), and in Garden and Forest, 1895, p. 55. 6 Journ. K. Hort. Soc. xiv. 485 (1892). 1364 The Trees of Great Britain and Ireland In Canon Ellacombe's garden at Bitton there is a dwarf bush of considerable age, which when covered with young cones is very ornamental. From it I have raised seedlings which grow very slowly. In Scotland the finest tree we know of is at Durris, which in 1904 was 61 ft. by 6 ft. 3 in. In Ireland Henry measured at Fota one which in 1903 was about 67 ft. by 6 ft. (H. J. E.) PICEA SCHRENKIANA, SCHRENK'S SPRUCE Picea Schrenkiana, Fischer and Meyer, in ßull. Acad. Sei. St. Petersb. x. 253 (1842); Regel, in Garten flora, xxvi. 69 (1877), and xxix. 49 (1880); Fedtschenko, in Bull. Herb. Boissier, vii. 189 (1899); Kent, Veitch's Man. Conif. 451 (1900); Clinton-Baker, Illust. Conif. ii. 48 (1909). Picea tianschanica, Ruprecht, in Mem. Acad. Sä. St. Petersb. xiv. No. 3, p. 72 (1870). Picea obovata, Ledebour, var. Schrenkiana, Masters, mjourn. Linn. Soc. (Bot.) xviii. 506 (1881). Pinus Schrenkiana, Antoine, Conif. 97 (1840-1847). Pinus obovata, Antoine, var. Schrenkiana, Pariatore, in De Candolle, Prod. xvi. 2, p. 415 (1868). Pinus orientalis, Linnseus, var. longifolia, Ledebour, FI. Ross. iii. 671 (1847). Abies Schrenkiana, Lindley and Gordon, in Journ. Hort. Soc. Land. v. 212 (1850). A large tree, attaining in Turkestan the dimensions of P. obowata. Young branchlets ashy grey, stout, glabrous. Buds dome-shaped or sub-globose, \ in. in length, rounded at the apex, light brown, with scarious scales ; terminal bud girt with a ring of acuminate keeled pubescent ciliate scales and closely surrounded at the base by the uppermost leaves of the branchlet. Leaves in an imperfect radial arrangement, dense and pointing forwards on the upper side of the branchlet, spreading with a few leaves pointing forwards and not truly pectinate on the lower side of the branchlet ; £ to \\ in. long, straight or curved, rigid, gradually tapering at the distal end to a long fine sharp-pointed apex1; obscurely quadrangular in section, with three to four lines of stomata on each of the four sides. Cones, 3 to 4 in. long, cylindrical, narrowing towards the obtuse apex, shining dark brown when ripe ; scales numerous, closely imbricated, longer than broad, about \ in. wide, obovate-cuneate, with the upper exposed part thin and glabrous, concealed part thicker and minutely pubescent ; upper margin rounded, entire, undulate ; bract \ in. long, ovate. Seed, light brown, \ in. long ; seed with wing \ in. long ; wing narrow, widest near the rounded apex. This species was discovered in 1840 by Schrenk, and is widely distributed in Central Asia, occurring mainly in the Alatau mountains and in the Thianshan2 range in Turkestan, where, according to Fedtschenko, it forms vast forests, now rapidly disappearing, as far south as lat. 41°, at 4000 to 8000 ft. altitude towards the north, and at 8000 to 10,000 ft. towards the south. It does not appear to 1 In wild specimens from old trees, the leaves end in a short acute callous tip. * Both Regel and Komarov agree that the spruce in the Thianshan range, considered by Ruprecht to be a distinct species (F. tianschanica), is identical with P. Schrenkiana in the Alatau range. Picea occur in the mountains uniting the Thianshan range with the Pamirs ; and its western limit is probably the Alexandrovoski mountains in Russian Turkestan. Its eastern limit is not as yet clearly known j1 but Przewalski found extensive woods of it, not only in the Thianshan range, but also in the upper course of the Yellow River in Mongolia, near Lake Kokonor, and in the adjoining Nan-Shan range.2 Mr. M. P. Price informs us that its most northerly point appears to be in the Barluk mountains, lat. 46°, where there are a few scattered forests in the higher valleys. He observed this tree at 9200 feet altitude in the pass between the valley of the river Baratala and the plateau of Lake Sairam. It bears the greatest extremes of heat and cold in these regions. Most of the trees, which he saw, scarcely exceeded 50 to 70 ft. in height and 7 to 8 ft. in girth. On a section 2 ft. ID in. in diameter from the base of a tree, which had grown in the vicinity of Lake Issik Kul and was preserved in the museum at Vernoe, he counted 296 annual rings. The wood is used for building houses in Russian Turkestan ; but is of little economic importance on account of the inaccessibility of the forests. In the eastern part of the Thianshan range, where the climate is very severe, and the thermometer sinks at least 7° F. below freezing every night during summer, P. Schrenkiana, nevertheless, forms open woods at about 8000 ft. elevation, which are remarkable for the peculiar narrow columnar form of the trees. This is well shown by two photographs, taken by Baron von Dungern, which are reproduced in Mitt, Deut. Dend. Ges. 1910, pp. 227, 229. He explains the cypress-like habit as due to the fact that the shoots of the lateral branches are almost invariably frozen, soon after their production in early summer ; whilst those of the leading branches, which are later in the season in emerging from the bud, escape destruction by the severe frosts. This species was distributed by the St. Petersburg Botanic Garden after its re-discovery in 1877 by Regel in Turkestan. It has never become common in cultivation. There are two trees at Kew, 8 and 10 ft. high, obtained from Messrs. Veitch in 1882; and smaller specimens at Bayfordbury and in other private collections. It appears to be hardy, though slow in growth, and is very distinct in appearance, most of the branches being rigid and ascending. (A. H.) 1 The spruce collected in Kansu, in north-western China, by Futterer and Holderer, identified with P. Schrenkiana by Diels, Flora von Central-China, 217 (1901) ; and another, collected by Bretschneider, near Peking, similarly identified by Masters, in Journ. Linn. Soc. (Bot.) xxvi. 554 (1902), appear to be identical with Picea Mastersü, Mayr, Fremdländ. Wald- u. Parkbäume, 328, figs. 105-107 (1906). 2 Cf. Koppen, Hohsewächse Europ. Russlands, ii. 538 (1889). Regel, in Act. Hott. Fetrop. vi. 485 (1880), states that it grows not only in the high mountains, but also along the rivers Baratala, Kash, and Vuldus. 1366 The Trees of Great Britain and Ireland PICE A SMITH I AN A, WESTERN HIMALAYAN OR MORINDA SPRUCE Picea Smithicma, Boissier, Fl. Orient, v. 700 (1884); Kent, Veitch's Man. Conif. 454 (1900). Picea Morinda, Link, in Linnœa, xv. 522 (1841); Masters, in Gard. Citron, xxiv. 393, fig. 85 (1885); Hooker, Flora Br. India, v. 653 (1888) (in part); Gamble, Indian Timbers, 716 (1902); Brandis, Indian Trees, 692 (1906); Clinton-Baker, Illust. Conif. ii. 40 (1909). Picea Khutrow, Carrière, Conif. 258 (1855). Pinus Smithiana, Wallich, PI. Asiat. Rar. iii. 24, t. 246 (1832). Pinus Khutrow, Royle, Illust. Him. Plants, 353, t. 84 (1839). Abies Smithiana, Lindley, in. Penny Cycl. i. 31 (1833) ; Loudon, Arb. et Frut. Brit. iv. 2317 (1838). Abies Khutrow, Loudon, Trees and Shrubs, 1032 (1842). Abies Morinda, Nelson (Senilis), Pinaceœ, 49 (1866). A tree, attaining in the Himalayas over 200 ft. in height and 20 ft. in girth. Bark greyish brown, divided by shallow fissures into small rounded or quadrangular scales. Young branchlets grey, shining, glabrous. Buds about £ in. long, spindle- shaped or ovoid, acute at the apex, brownish, resinous ; scales numerous, densely imbricated, rounded at the apex ; terminal bud girt at the base with a ring of acuminate keeled scales. Lateral branches always pendulous, with the leaves radially arranged and directed outwards and towards the apex of the branchlet at an acute angle. Leaves long and slender, about \\ in. long and -£% in. broad, incurved, tapering towards the apex, which ends in a slender cartilaginous point ; obscurely 4-angled, with about two lines of stomata on each of the four sides. Staminate flowers, about I in. long and \ in. in diameter, cylindrical, obtuse, light yellow ; anther connective orbicular, crenate. Cones, 4 to 6 in. long, \\ to 2 in. in diameter, cylindrical, narrowed towards the base, obtuse at the apex ; bright green and smooth when growing ; shining brown when mature; scales about an inch wide, broadly obovate from a cuneate base, smooth, convex, rounded and entire in margin ; bract obsolete. Seed dark brown, \ in. long, with the wing f in. long ; wing spatulate, broadest near the truncate denticulate apex. P. Smithiana occurs throughout the western Himalayas, between 7000 and 11,000 ft. elevation, being common from Garhwal to Kashmir, and also occurring in Gilgit, Chitral, and Kafiristan. It extends westwards to Afghanistan, where Aitchison found it in the Kuram and Hariab district, between 8000 and 11,000 ft., occasionally extending as high as 12,000 ft., where it struggles for existence with Pinus excelsa. According to Gamble, it is a very fine tree in the Himalayas, often attaining a greater height than the deodar, but probably never equalling the latter in girth. Large trees measured near Mundali in Jaunsar were 175 to 215 ft. in length and 19 to 23 ft. in girth.1 It forms mixed forests with Abies Pindrow, which cover mainly the northern and western slopes of the mountains, usually between 7500 and 8500 ft. In these forests the spruce is more common on the drier ridges, the silver fir growing in the moister ravines. P. Smithiana also forms mixed forests 1 I am informed by Sir G. Watt that a tree, recorded by Sir E. Buck, near Nagkunda, measured no less than 250 ft. high. Cf. Frontispiece of Vol. V.—H. J. E. Picea 1367 with the deodar. Grown in dense forest the stems are often free from branches to a great height, crowned by a conical pyramid of foliage with pendulous branches. In this condition, it produces seed at intervals of three or four years, and in small quantity. The rank undergrowth consists of Strobilanthes, small bamboos, rasp berries, balsams, and other plants, which render natural reproduction of seedlings rare and difficult. Clear cutting and artificial regeneration have been found to be the most successful modes of treating these forests. This spruce is attacked in the Himalayas by the aphis, Chermes abietis, which is common on the European spruce, and produces cone-like excrescences on the twigs. A fungus, Peridermium incarcerans, Cooke, often occurs as curious tassel-like orange bunches on the branchlets. The leaves are attacked by another fungus, sEc^d^^lm Thomsoni? (A. H.) CULTIVATION P. Smithiana was introduced into cultivation in 1818 by Dr. Govan of Cupar, who gave the seed to the Earl of Hopetoun, from which the first trees were raised at Hopetoun House, near Edinburgh. It is a thriving tree in many parts of the British Isles ; and though the young shoots are liable to be nipped by frost, this does not seem to do the tree permanent injury. It does not, however, seem to succeed on limestone soil. The tallest specimen2 I have seen in England is at Melbury, where, in 1906, I measured one 85 to 90 ft. high and 8 ft. 10 in. in girth. (Plate 345.) At Carclevv a tree was reported8 in 1891 as 80 ft. high, but when I measured it in 1905 it was 86 ft. by 7 ft. 9 in. At Pencarrow Mr. Bartlett measured a perfect specimen planted by Sir W. Molesworth about 1850, which in 1907 was 57 ft. by 6 ft. 7 in. At Bicton in 1902 I measured one 65 ft. by 6 ft. 9 in. At Redleaf, Kent, in 1907 a tree 75 ft. by 9 ft. had many cones on the lower branches, which rested on the ground. At Walcot there is a fine tree 60 ft. by 5^ ft. At Barton, Bury St. Edmunds, there are two fine trees, one of which was 84 ft. high by 7 ft. i in. in girth in 1904, the other 77 ft. by an inch less in girth. These trees were raised from seeds sent by Lady Napier, to whom they had been given by Wallich, and the seedlings were planted out in 1843. The trees were not injured in the least by the severe winter of 1860-1861, and commenced to bear cones for some years before 1869, having a very abundant crop in that year.4 A tree at Hardwicke House, Suffolk, planted later than those at Barton, was measured by Sir Hugh Beevor in 1904 as 73 ft. by 7 ft. In Wales the largest I have seen, a tree at Margam Park, was 81 ft. by 6^ ft. in 1907. In Scotland there are many good specimens, of which those at Hopetoun are the oldest, having been raised from seed sent to the Earl of Hopetoun by 1 Described and figured by Berkeley in Gard. Chron. 1852, p. 627. * The Picea Smithiana reported in Gard. Chron. 1869, p. 713, to be growing at Shelsley Walsh in the Teme Valley in Worcestershire, is P. excelsa. Cf. p. 1341. 3 Jonrn. R. Hort. Soc. xiv. 488 (1892). 4 Bunbury, Arboretum Notes, 134. VI F 1368 The Trees of Great Britain and Ireland Dr. Govan in 1818. When I saw them in 1904 the best of these was about 70 ft. by 8^ ft. Fowler * states that the two trees at Hopetoun House were planted in their present position in 1824, one being a seedling, the other a grafted plant worked on the common spruce, four feet above the ground. In 1871 the graft had outgrown the stock all round for 2 to 3 inches. The seedling tree in that year was 60 ft. high by 7 ft. in girth at four feet from the ground, the grafted tree being scarcely so tall. Mr. T. Hay, gardener at Hopetoun, remeasured these trees in January 1911, and informed me that the grafted tree is still in fair condition, and measures 70 ft. high. Its girth below the graft is 6 ft. 2 in., and above it 7 ft. 2 in. The seedling tree is more healthy and measures 75 ft. by 8 ft. 8 in. at 4 ft. from the ground. At Smeaton Hepburn, a tree, planted in 1840, was measured by Henry in 1905 as 67 ft. by 6 ft. 5 in. In Ireland this species thrives remarkably well, and there are many fine specimens. At Woodstock, Kilkenny, in 1909 I measured a tree 72 ft. by 8^ ft. At Mount Shannon near Limerick, a tree measured, in 1905, 69 ft. by 8^ ft. in girth. At Fota, Queenstown, there is a fine tree, which was, in 1903, 63 ft. by 8 ft. At Glenstal, Co. Limerick, in the same year, a tree was n^ ft. in girth, with an estimated height of 70 ft. At Bessborough in Co. Kilkenny, a tree, which was figured in the Gardeners Chronicle, May 21, 1904, is, we are informed by Viscount Duncannon, 60 ft. high by 6 ft. 9 in. in girth. Another at E mo Park, Portarlington, was 60 ft. by 8 ft. in 1907 ; and one at Coollattin was 59 ft. by 4 ft. 8 in. in 1906. In the United States,2 the tree is too tender for the climate of Boston, and does not do well even at Washington. There are no large trees of this species in the United States. TIMBER According to Gamble, the rate of growth in India is fairly fast, averaging about ii rings per inch of radius, or 125 years to a girth of 6 ft. The wood is similar to that of the European spruce, and affords excellent planking for floors, walls, and ceilings. It is used for shingles, for packing cases, for building huts, for water- troughs, etc. In some places it is utilised for making tea boxes. It averages in weight 30 Ibs. to 32 Ibs. per cubic ft. The bark was formerly used extensively by the shepherds for roofing their huts, but this practice has been stopped in the Government forests. On account of the expense of transport, it is never likely to be exported. (H. J. E.) 1 In Card. Chron. 1872, p. 76. 2 Garden ami Fotest, 1893, p. 14, and 1897, P- Picea 1369 PICEA GLEHNII Picea Glehnii, Masters, in Gard. Chron. xiii. 300, fig. 54 (1880), and in Journ. Linn. Soc. (Bot.\ xviii. 512, fig. 13 (1881); Mayr, Abiet. Jap. Reiches, 56, 102, t. 4, fig. n (1890), and Fremdländ. Wald- u. Parkbäume, 327 (1906); Kent, Veitch's Man. Conif. 437 (1900); Shirasawa, Icon. Ess. Forest. Japon, ii. t. 3, figs. 19-42 (1907); Clinton-Baker, Ilhest. Conif. ii. 39 (1909)- Abies Glehnii, Schmidt, in Mem. Acad. Imp. Se. St. Pétersb. xii. 176, t. 4 (1868). A tree, attaining in Yezo over 100 ft. in height. Bark different from any of the other spruces, reddish in colour, and fissuring into broad thin loose plates. Young branchlets slender, reddish, with dense short pubescence in the furrows between the pulvini, not spreading over the surface of the latter. Buds ovoid, brown, ^ in. long, composed of a few glabrous scales ; terminal buds girt with a ring of scales ending in long subulate points. Leaves arranged on the branchlets as in P, excelsa, £ to \ in. long, slender, ending in a short cartilaginous point ; rhombic in section, with about two stomatic lines on each of the two upper sides, and a single line on each of the lower sides. Cones, about 2 in. long by i in. in diameter when closed ; violet, with a red edge to the scales when growing, shining brown when ripe ; cylindrical, with an obtuse narrowed apex ; scales, when ripe, spreading from the axis at a right angle, suborbicular, with a cuneate claw, about ^ in. wide, with the thin upper margin entire, slightly erose, or faintly denticulate ; bract spatulate, £ in. long, denticulate at the apex. Seed blackish, £ in. long, with wing f in. long ; wing broadest about the middle, rounded at the apex, outer margin denticulate. This species is readily distinguishable from the other short-leaved spruces by the reddish branchlets, with the pubescence confined to the furrows. It resembles P. orientalis in the colour of the foliage, but is very distinct in the terminal buds, which have a ring of subulate scales, similar to P. nigra and P. rubra. DISTRIBUTION P. Glehnii* was discovered in Saghalien in 1861 by Glehn, the comrade of F. Schmidt on the expedition sent out by the Russian Geographical Society to Eastern Asia. It was subsequently found in Yezo by Maries in 1877. In Saghalien, this species is confined to the southern half of the island, where it grows on the plains and in the valleys, never attaining, according to Schmidt, a great size, being seldom over a foot in girth. According to Mayr, it is probably absent from the Kurile Isles,2 as it was not noticed by him on Shikotan ; according to Komarov,3 it does not occur in Russian Manchuria. (A. H.) 1 The Formosan spruce, identified with P. Glehnii by Matsumura in Tokyo Bot. Mag. xv. 141 (1901), is quite distinct, and has been named P. morrisonicola, Hayata, in Journ. Coll. Sei. Tokyo, xxv. 220 (1908). 2 Miyabe does not include it in his Flora of the Kuriles. 3 Flora Manshuriœ, i. 200 (1901). 137° The Trees of Great Britain and Ireland P. Glehnii attains its maximum development in Yezo, where, according to Mayr, it is much commoner in the west of the island than in the east. In western Yezo it forms mixed forests in company with P. ajanensis, chiefly on the cooler parts of the mountains, the trees reaching on an average nearly 120 feet in height. Mayr mentions a peculiar forest of this species, which occurs on the volcanic Iwo- san (1500 feet elevation) east of Lake Kucharro. In the eastern part of the island, it forms pure forests in the river valleys in swampy situations, which are often several hundreds of acres in extent ; but the trees are of no great size, averaging only 80 ft. in height. This species is known to the Japanese as Shinko matsu or Aka-eso. According to Miyabe, it is rare near Sapporo and only found at high elevations mixed with P. ajanensis. Near Lake Shikotsu at 1500 feet elevation, I found it much less abundant than P. ajanensis, and could not procure any fruiting specimens. A self-sown seedling which I brought from here is growing very slowly at Coles- borne and is now only i foot high. I could not learn whether the wood of the tree is distinguished from that of the common Yezo spruce. Some very broad clean pieces which I saw in the saw-mill at Sunagawa had a close grain and a shiny satiny surface when planed, making it suitable for interior work where strength is not required. CULTIVATION According to Beissner1 seeds of this species arrived in Germany before 1891, from which young plants were raised. It is scarcely known in cultivation in England. There are young plants at Kew, about 2 ft. high, which are thriving ; and small specimens at Bayfordbury and Brickendon Grange, Herts, and in the Cambridge Botanic Garden. It is too soon as yet to form any opinion as to the suitability of this species to our climate ; but I do not expect that it will attain any size. (H. J. E.) PICEA POLITA Picea polita, Carrière, Conif. 256 (1855); Masters in Gard. Chron. xiii. 233, fig. 44 (1880), and Journ. Linn. Soc. (Bot.) xviii. 507, pi. 19 (1881); Mayr, Abiet. Jap. Reiches, 46, t. 3, f. 7 (1890), and Fremdländ. Wald- u. Parkbäume, 335 (1906); Kent, Veitch's Man. Conif. 446 (1900); Shirasawa, Icon. Ess. Forest. Japon, ii, t. 2, figs. 18-29 (I9°7); Clinton-Baker, Illust. Conif. ii. 45 (1909)- Picea Torano, Koehne, Deutsche Dendrologie, 22 (1893). Abies Torano? Siebold, in Verhand. Batav. Genoot. Konst. Wet. xii. 12 (1830). Abiespolita, Siebold et Zuccarini, Flor. Jap. ii. 20, t. in (1842). Pinnspolita, Antoine, Conif. 95 (1840-1847). A tree, occasionally attaining in Japan 120 ft. in height, but usually considerably smaller. Bark fissuring into small scales, exposing the yellowish brown cortex 1 Nadelhohkunde, 377 (1891). 2 This specific name is uncertain, as it was unaccompanied by any description, and cannot be adopted. Picea beneath. Buds ovoid, acute at the apex, up to f- in. long, shining reddish brown; scales closely imbricated, ovate, rounded at the apex. Young branchlets stout, glabrous, shining, pale yellow. Leaves on lateral branchlets in an imperfect radial arrange ment, all directed outwards with their tips curving upwards ; about f to f in. long, ^ in. wide, very rigid, stout, curved, ending in a sharp spine-like point ; compressed rhomboidal in section, with 4 to 6 lines of stomata on each of the four surfaces. Cones, about 3 to 4 in. long, i \ in. in diameter when closed, yellowish green when growing, shining chestnut brown when mature ; ovoid-cylindrical, obtuse at the apex ; scales obovate, with a cuneate base, about ^ in. wide ; upper margin rounded, with a few irregular denticulations ; bract oblong, £ in. long, slightly narrowed at the denticulate apex. Seed mottled grey, about £ in. long, with wing f in. long ; wing broadest near the truncate denticulate apex. The very rigid sickle-shaped leaves, ending in prickly spines, and arranged radially on the branchlets, are unlike those of any other spruce. P. polita is confined to the main island of Japan, having nearly the same dis tribution as P. kondoensis and P. bicolor, extending from about lat. 355-° to lat. 38°, and not reaching the extreme north of the island. It is found in warmer situations than the other two spruces, and, unlike them, never forms pure woods. It always occurs as isolated trees or in small groups, scattered through the broad-leaved forest. It is the tallest of the three, the largest specimens seen by Mayr being nearly 120 ft. high ; and is a much rarer tree, of no economic importance in Japan, where it is known as hari-momi. This species was introduced into cultivation by J. Gould Veitch in 1861, and is perfectly hardy ; but it has nowhere attained considerable dimensions. Kent states that the best specimens occur in Devon and Cornwall ; but the largest which we have seen is one at Highnam, Gloucester, 30 ft. by 2 ft. in 1910. There is also a healthy specimen at Bayfordbury, planted in 1879, which has borne cones ; and another at Hatfield, very thriving. A tree at the Heatherside Nursery, Farn borough, about 20 ft. high, bore cones in 1909. There are two good young trees at Castle Kennedy. A fine specimen at Castlewellan, planted in 1884, was about 25 ft. high in 1907. According to Mayr this species, with P. bicolor and P. pimgens, are the latest to grow in Germany, not opening their buds until June. It is much injured by squirrels, and will probably be of no economic value, either on the Continent or in England. (A. H.) I372 The Trees of Great Britain and Ireland PICEA BICOLOR Picea bicolor, Mayr, Abiet.Jap. Reich. 49, t. 3, fig. 8 (1890), and Fremdländ. Wald- u. Parkbäume, 323 (1906); Shirasawa, Icon. Ess. Forest. Japon, i. text 19, t. 4, figs. 1-14 (1900). Ficea Alcockiana, Carrière, Cotiif. 343 (1867); Masters, in Gard. Chron. xiii. 212, figs. 41, 43 (1880), and in Journ. Linn. Soc. (Bot.) xviii. 508, figs. 7-9 (1881); Hennings, in Gartenflora, xxxviii. 216, fig. 40 (1889); Kent, Veitch's Man. Conif. 429 (1900); Henry, in Trees of Great Britain, i. 89, 90 (1906). Picea japonica? Regel, Index Sem. Hort. Petrop. 33 (1865). Picea acicularis, Beissner, Nadelholzkunde, 380 (1891). Abies bicolor, Maximowicz, in Mél. Biol. vi. 24 (1866). Abies acicularis, Maximowicz, in Index Sem. Hort. Petrop. 74 (1868). Abies Alcockiana, Gordon, Pinetum, 4 (1875) (not Lindley). Finns Alcoquiana, Pariatore, in De Candolle, Prod. xvi. 2, p. 417 (1868). A tree, attaining in Japan 80 ft. in height. Bark greyish brown, fissuring into small scales. Young branchlets yellowish, glabrous on the lateral branches, but pubescent in the furrows between the pulvini on strong leading shoots ; older branchlets shining reddish brown. Buds, about \ in. long, conic, rounded at the apex, without resin, and with few scales, scarious in margin. Leaves, on lateral branches arranged as in P. excelsa, about f in. long, rigid, curved, ending in a short cartilaginous point, rhombic in section, with two conspicuous white stomatic bands on the upper two sides, each of 5 or 6 lines, and two bands of about 2 lines each on the two lower green sides. Cones, averaging 3^ in. long and i in. in diameter when closed ; bluish red with green margins to the scales when growing, brownish when mature ; ovoid-cylindrical : scales obovate with a cuneate base, about f in. broad, thin and faintly denticulate in the upper rounded margin ; bract \ in. long, spatulate, with a slightly expanded denticulate lamina. Seed, \ to £ in. long, brown ; seed with wing f in. long ; wing widest about the middle, rounded and faintly denticulate at the apex. This species, as its name implies, differs from the other quadrangular-leaved spruces, in the conspicuous white broad stomatic bands on the upper surface of the leaf, contrasting with the green lower surface, and in this respect it simulates the flat- leaved spruces, and has been confused2 with P. hondoensis and P. ajanensis. The leaves of the latter are flat and not rhombic in section, and are devoid of the faint stomatic lines on their lower surface, which are readily seen in P. bicolor. HISTORY This species was discovered in 1860 on Fujiyama by J. G. Veitch, who collected cones of it, unfortunately mixed with twigs of P. hondoensis. Lindley, in 1861, described a mixture of the two species, and his name, Abies Alcoquiana, Veitch,8 1 A name without any description. It is identified in Index Seat. Hurt. Petrop. 3 (1866) with Abies bicolor, Maxim. Seeds were sent from Japan in 1865 by Tschonoski. 2 Cf. Vol. I. p. 90. 3 Ex Lindley, in Gard. Chron. 1861, p. 23. Lindley's description comprises the leaves of P. Iwndoensis and the cones of P. bicolor. The type specimen, in which both these are mixed in one packet, is in the herbarium at Cambridge. Picea cannot stand. Maximowicz gave a correct description of the species under the name Abies bicolor in 1866. As explained in our article* on P. hondoensis, seeds of both species were early distributed as P. Alcockiana ; and in gardens most trees named P. Alcockiana are in reality P. hondoensis. This species was introduced into the St. Petersburg Botanic Garden by seeds sent from Japan in 1868 by Tschonoski under the name Abies acicîilaris? Maxi mowicz. DISTRIBUTION This species occurs only in the main island of Japan, where, like P. hondoensis and P.polita, it is confined to the central ranges between lat. 35^° and lat. 38°. It forms part of the coniferous forest, which covers these mountains at varying altitudes from south to north, usually above the zone of broad-leaved trees ; but occasionally scattered trees are met with in the upper limits of this zone. Mayr never saw any trees over 80 ft. in height, though he thinks that it occasionally attains greater dimensions. This species is rare in collections, the largest we have seen being at Kew, where there are two trees, 25 and 30 ft. high, one of which bore cones in 1900. There are also specimens at Westonbirt, Pencarrow, Murthly, Castle Kennedy, and Glasnevin. Mr. H. Clinton Baker collected cones from the tree at Pencarrow in August 1908; and I saw at Castlewellan in 1907 a tree about 20 ft. high bearing cones. The tree at Blackford Park, Edinburgh, mentioned by Kent, was planted about 1882-1884, and measured 20 ft. by i ft. 7 in. in 1906. The gardener, Mr. Small, states that it is late in starting into growth in the spring, and in consequence escapes late frosts. Probably the finest tree in cultivation is growing in Mr. Hunnewell's pinetum at Wellesley, Mass., U.S.A. It bears cones freely, some of which I gathered in 1906, when the tree measured about 36 ft. high by 3 ft. in girth.3 (A. H.) 1 Vol. I. p. 90. 2 Young plants with slender sharp-pointed needles were distributed under this name. 3 Cf. Sargent, Pinelum at Wellesley, 1905, p. 11. 1374 The Trees of Great Britain and Ireland PICEA MAXIMOWICZII Picea Maximowiczii, Regel, in Index Sem. Hort. Petrop. 33 (1865); Carrière, Conif. 347 (1867); Masters, in Gard. Chron. xiii. 363 (1880), and four». Linn. Soc. (Bot.) xviii. 507 (1881); Mayr, Abiet. Jap. Reiches, 98 (1890). Picea obovata, Ledebour, \sx.jafonica, Beissner, Nadelholzkunde, 370 (1891). Picca Tschonoskii, Mayr,1 Fremdländ. Wald-u. Parkläume, 339 (1906). Abies obovata, Loudon, var. japonica, Maximowicz, in Index Sem. Hort. Petrop. i and 3 (1866); Franchet, Enum. PI. Jap. i. 466 (1875). Abies Maximowiczii, Neumann, Cat. 1865, ex Pariatore, in De Candolle, Prod. xvi. 2, p. 431 (1868) ; Veitch, Man. Conif. 80 (1881). A small tree. Young branchlets reddish brown, glabrous, with the apices of the pulvini all directed outwards and forwards. Buds about 1 in. long, ovoid, acute, with glabrous rounded resinous scales. Leaves on lateral branches radially spreading on all sides at nearly a right angle to the branchlet, but with their tips pointing slightly forwards ; f to ^ in. long, rigid, tapering near the apex which is tipped with a short blunt point ; green, quadrangular in section, with three to five stomatic lines on each surface ; resin-canals two, lateral, close to the epidermis. Cones, if to 2 in. long, i in. in diameter when open, shining brown when ripe, cylindrical, but tapering at both ends : scales numerous, obovate with a cuneate claw, | in. wide ; rounded, entire, and bevelled in the upper margin : glabrous in the exposed part, elsewhere covered with a minute reddish pubescence : bract about \ in. long, oblong, with a rounded faintly denticulate apex. Seeds, not extending to the upper and,lateral margins of the scale, \ in. long, dark brown mottled with lighter streaks ; seed with wing \ in. long ; wing widest near the upper rounded denticulate margin. This species is readily distinguishable by its short leaves radially arranged, and its resinous buds. At Kew it produces new shoots a month earlier than P. bicolor. This spruce is a native of Japan, where it was collected in 1864 on Mt. Fujiyama by Tschonoski,2 a young Japanese collector in the employment of Maximowicz. One of the original specimens from this locality is preserved at Kew, where there is also an imperfect specimen,8 collected in the same year in the province of Senano by Tschonoski, which was recognised by Maximowicz to be the same species.4 It appears to be very rare, and has not since been found by Japanese botanists. Maximowicz considered it to be a variety of P. obovata, from which it is clearly distinct ; but it is rather related to P. bicolor, though differing much in foliage and in cones. 1 Mayr erroneously considered that the tree cultivated as P. Maximmuiczii was different from Tschonoski's Fujiyama specimen. He identified the latter with P. bicolor, and proposed a new name, P. Tschonoskii, for the former. 2 Maximowicz, in Rhamn. As. Or. 17 (1866), gave an account of Tschonoski, who was a Japanese and not a Russian as some authors have supposed. He gathered about 800 species of Japanese plants, and sent seeds of many kinds to St. Petersburg. ' Consisting of a cone and a single leaf. The cones on the Grignon tree, about 2 in. long, are intermediate in size between those of the Fujiyama tree (which are if in. long) and those of the Senano specimen (about 2j in. long). 4 The Senano specimen is labelled Abies obovata, Loudon, var. japonica, Maximowicz ; and the Fujiyama specimen is named Picea Maximowiczii, Regel. Picea Seeds sent to St. Petersburg by Tschonoski in 1865 were distributed by Regel to various botanic gardens on the Continent. The best specimen that I have seen is a tree with ascending branches at the Agricultural School of Grignon near Paris, which is about 30 ft. in height by 19 in. in girth ; but M. Hickel ' tells me that there are still finer trees elsewhere in France. A smaller tree in the Arnold Arboretum, U.S.A., also bears cones, smaller in size than those on the tree at Grignon. Another in Mr. Hunnewell's pinetum at Wellesley, Mass., was n ft. high in 1905. There is also a small specimen2 in the spruce collection at Kew, a bush about 4 ft. high ; and two trees at Handcross Park, Sussex, the taller of which was 32 ft. by 2 ft. 5 in. in 1911. These were planted about thirty years ago, and have not as yet borne cones. (A. H.) PICEA NIGRA, BLACK SPRUCE Picea tiigra, Link, flandb. ii. 478 (1831); Kent, Veitch's Man. Conif. 438 (1900); Clinton-Baker, Illiist. Conif. ii. 41 (1909). Picea Mariana, Britton, Sterns, and Poggenburg, Cat. PL N. York, 71 (1888); Sargent, Si/va N. Amer. xii. 28, t. 596 (1898), and Trees N. Amer. 39 (1905). Picea brevifolia* Peck, Spruces of the Adirondacks, 13 (1897), and in Bull. Torrey Bot. Club, xxvii. 409 (1900). Abies Mariana, Miller, Diet. ed. 8, No. 5 (1768). Abies nigra, Du Roi, Harbk. Baums, ii. 182 (1800); Loudon, Arb. et. Frut. Brit. iv. 2312 (1838). Abies dcnticulata, Michaux, FI. Bor. Amer. ii. 206 (1803). Pinus Mariana, Du Roi, Obs. Bot. 38 (1771). Pinus nigra, Solander, in Aiton, Hort. Kew. iü. 370 (1789). A tree, attaining in America 100 ft. in height and 9 ft. in girth, but usually much smaller. Bark brownish, fissuring into irregular thin appressed scales. Buds small, ovoid, acute ; the terminal buds surrounded at the base by ciliate pubescent scales with conspicuous long subulate points. Young branchlets brownish, with dense short erect glandular pubescence, retained on the dark-coloured branchlets of the second year. Leaves, arranged on lateral branches as in the European spruce, about \ in. long, bluish or glaucous green, slightly incurved, ending in a short cartilaginous point, quadrangular in section, with four lines of stomata on each of the two sides turned towards the branchlets, and with one to two lines on each of the other sides. Cones persistent on the branches for several years, ovoid, acute at the apex, f to ii in. long, dark purple when growing, dull brown when ripe; scales rigid, woody, pubescent, about f in. wide, rounded or rarely pointed at the apex, denticulate T,-J M' "*kd1infonM me that the oldcr tre<* i" France, which were planted about 1868, were originally raised in Thibaut and Iveteleer s nursery at Sceaux, from seed given then, by Carrière, which he received from Kegel Of late years this spruce has been propagated by grafting. ' y • This is perhaps the same shrub as that from which a specimen in the Kew Arboretum herbarium was taken in 1882 labelled «low hush, I to 2 ft. (rounded). Pinetum, Aug. 3, 1882. J. D. Hooker.» The low stature of the shrub at Kew indicates probably an alpine origin for the seed from which it was raised. 3 This is the ordina stunted form of p n^.a> growing on swampsand exposed mounta.n sum and .snot able even as a variety by Sargent, or by Britton and Shafer, N. Amer. Trees, 57 (1908) ; Rehder, in Bailey, Cyc. A^Hort i«- 1334, fig. 1794 (1901), and in Rhodora, ix. 109 (I9o7), has distinguished it as var. bremfolia. VI 1376 The Trees of Great Britain and Ireland in margin. Seeds dark brown, about ^ in. long, with pale brown wings broadest above the middle and very oblique at the apex. Dwarf and fastigiate forms,1 and varieties in which the foliage is variegated with white or golden yellow in colour are mentioned by Beissner. Var. Doumetii, Carrière, Conif. 242 (1855). This variety was first noticed about 1835 in the garden of the Château de Baleine2 near Moulins in France. It is a small tree or large shrub, with short numerous branches, forming a dense conical pyramid of foliage. The leaves are very crowded, thin and sharp-pointed. As seen at Kew this variety is very distinct in appearance. There are remarkable black sprucess in the Wilhelmshöhe and Karslane parks at Cassel in Germany, which are pyramidal in habit and bluish in foliage. Self-layering occurs, and numerous colonies of young plants are produced round the parent trees. DISTRIBUTION The black spruce is widely spread throughout the Dominion of Canada, occur ring as far north as Labrador on the Atlantic coast, and reaching lat. 65° in the valley of the Mackenzie River, whence, crossing the Rocky Mountains, it spreads in the interior of Alaska to the valley of the White River.4 Farther south, it is restricted to the eastern side of the Rocky Mountains, extending throughout Alberta, Assiniboia, northern Saskatchewan, and northern Manitoba (where it attains its largest size) to central Wisconsin and Michigan. It is common in Newfoundland and all the eastern provinces of Canada, except southern Ontario ; and spreads in the north-eastern United States to Pennsylvania, reaching its most southerly point in the Alleghanies in southern Virginia. Towards the northerly part of its range it is abundant, and grows on well- drained alluvial soils and on the stony slopes of barren hills ; while towards the south it is almost entirely restricted to bogs and swamps. Mr. H. E. Ayres in Garden and Forest, vii. 504, fig. 80 (1894), describes and figures it under these conditions in Minnesota, as the " Muskeag" spruce, this being the name by which the sphagnum bogs so common in North America are known. He states that in these swamps the trees grow slowly to a height of 60 ft. with very drooping branches, the trunks never exceeding about 10 in. in diameter. The cones are densely crowded at the summit of the tree, and are sometimes produced on trees only 3 ft. high. (A. H.) CULTIVATION The black spruce was introduced5 into England by Bishop Compton about 1700; but Sir W. Watson, who gave a list of the principal trees which he found in the Fulham Palace gardens in 1751, does not include it. 1 For Picea nigra, var. virgata, Render, see under P. rubra, p. 1378. 2 When I visited this place in 1909, I found that the original specimen, a tree about 30 ft. high, was dead ; but two plants grown from its layers are now about 15 ft. high, with pointed tops ; while others, which were raised from cuttings, form dense dwarf bushes.—(H. J. E.) ~ 3 See Gard. Chron. xi. 8i, Suffi, fllast. (1892). The black spruce appears to layer frequently; and Loudon figures an instance which was noticed in 1828 at Braco Castle, Perthshire. 4 Cf. Sargent, Suva N. Amer. xiv. 106(1902). 6 Aiton, Hort. Kew. iii. 371 (1789). Picea 1377 It is common in cultivation and usually sold in nurseries under the name of blue spruce, but it never attains large dimensions and is of no economic importance. It has been recommended for planting in boggy and marshy situations, but is always much surpassed in growth by Picea sitchensis, and seems to be a short-lived tree in this country. One of the best specimens we know of is the one figured in Plate 346, which grows on the north edge of a plantation of common spruce at Lyde, near Colesborne, on my property. This tree has been favoured by a moist clay soil, a sheltered position and a cold damp climate ; and has attained at about fifty-five years old a height of 56 ft. with a girth of 2 ft. 10 in. As the figure (Plate 346) shows, it has become self-layered under the shade of a hedge, which was cut away to show it ; and one of the lower branches has already attained half the height of the parent stem. Though it has not increased much in the last ten years, this tree is in good health, but several others, planted at the same time on dry land, are not half the size and dead or dying. As usual in England it bears cones in abundance near the top of the tree. I have seen a tree at Woburn about 60 ft. by 4^ ft. ; and there is one at Merton which was about 40 ft. by 5 ft. 10 in. in 1905. In the west of Scotland it grows well, but so far as we have seen never attains a large size; the tallest recorded1 in Scotland in 1891 was 46 ft. by 3 ft. 5 in. at Mount Stuart. Of the numerous trees planted in two groups in 1832 at Keillour, Perthshire, at the lower end of a peat-bog, Henry only found a few surviving in 1904, none exceeding 40 ft. in height. At Dawyck, a tree was 37 ft. by 2^ ft. in 1911. In Ireland the best we have seen was measured by Henry at Fota in 1903, when it was 60 ft. by 4 ft. 10 in. (H. J. E.) PICEA RUBRA, RED SPRUCE ricea rubra, Link, in Linnaa, xv. 521 (1841) (not Dietrich 2) ; Gorrie, in Trans. Pot. Soc. Edin. x. 353 O8?0); Kent, Veitch's Man. Conif. 450 (1900); Sargent, in Bot. Gaz. xliv. 226 (1907); Clinton-Baker, Illust. Conif. ii. 47 (1909). ricea nigra, Link, var. rubra, Engelmann, in Card. Chron. xi. 334 (1879). Picea rubens, Sargent, Sîlva W. Amer. xii. 33, t. 597 (1898), and Trees JV. Amer. 41 (1905). rinus rubra, Lambert, Genus Finns, \. 43 (1803) (not Miller3). I'icea acutissima. Jack, in Garden and Forest, x. 63 (1897). At-ies rubra, 1'oiret, in Lamarck, Diet. vi. 520 (1804); Loudon, Arb. et Frut. Brit. iv. 2316 (1838). A tree, attaining in America 100 ft. in height and 9 ft. in girth. Bark, branchlets, and buds, similar to P. nigra. Leaves yellowish or dark green, not glaucous, about f in. long, incurved, acute or rounded at the apex, quadrangular in section, marked on each of the two upper sides by about four stomatic lines, and on each of the two lower sides by two to three stomatic lines. 1 Joui n. Roy. Hort. Soc. xiv. 506 (1892). * Pnea rubra, Dietrich, fl. Berol. ii. 795 (1824) is the common European spruce, Picea excelsa. 3 finas ruera, Miller, Gcud. Diet. No. 3 (1795) is the common European pine, Pùats sylvcstris 1378 The Trees of Great Britain and Ireland Cones, ovoid-oblong, i£ to 2 in. long, green or purplish green when growingj shining reddish brown when mature, usually falling in the second summer : scales £ in. broad, rounded entire or denticulate at the thin upper margin : bract inconspicuous, oblanceolate, about ^- in. long. Seeds mottled dark brown, about £ in. long, with wings broadest near the rounded apex, the total length of seed and wing being about f in. VARIETIES 1. Picea australis, Small, Flora S.E. United States, 30 (1903), is probably a variety1 of the red spruce, which differs in bearing small cones that are said to fall directly after shedding their seed. Large trees of this kind, attaining 130 ft. in height, are reported to occur on the summits and rocky slopes of mountains in Virginia and North Carolina. 2. A solitary red spruce,2 with snake-like branches, similar in habit to P. excelsa, var. virgata, was discovered in 1892, near Williamstown, in north-western Massa chusetts. From it young plants were raised by grafts in the Arnold Arboretum. DISTRIBUTION The red spruce has a much more southerly distribution than the black spruce,3 and does not extend farther north than Prince Edward Island and Nova Scotia. It is widely spread in New England,4 through Maine, New Hampshire, Vermont, and northern Massachusetts ; but is not known in Rhode Island and Connecticut. In New York, especially in the Adirondacks, it forms extensive forests ; and extends through the Alleghany Mountains southward through Pennsylvania and West Virginia to the high peaks of North Carolina. Pinchot5 has given a complete account of this species, which provides the only merchantable spruce timber in the United States ; and states that it is remarkable for its tolerance of dense shade and its capacity for recovering after years of suppression. In the Adirondacks, it ascends to 4500 ft., and is often seen on steep southern slopes ; but elsewhere is mixed with balsam fir, hemlock, white pine, birch, maples, and beech. (A. H.) CULTIVATION The red spruce was first accurately described and figured by Lambert from a tree growing in England, which was said to have been brought from Newfound land. According to Aiton,6 it was cultivated near London by Miller before 1755; but it is doubtful if this tree was distinguished from the black spruce at that date. P. rubra is rare in collections, the only large trees which we have seen being 1 Cf. Britton and Shafer, North American Trees, 58 (1908). 3 Figured in Garden and Forest, viii. 45, fig. 7 (1895). This is identical with Picea nigra, Link, var. virgata, Rehder, in Bailey, Cycl. Am. Hort. iii. 1334 (1901), corrected to P. rubra forma virgata, Rehder, in Rhodora, ix. no (1907). Cf. Mitt. Deut. Dend. Ges. 1907, p. 116. 3 Keiler, Our Native Trees, 470 (1907) says : "Black spruce is a tree of the far north, existing but precariously south of the northern boundary of the United States ; while red spruce is an Appalachian tree, attaining its greatest dimensions in northern New Hampshire and Pennsylvania." * Dame and Brooks, Trees of New England, 14 (1902). 6 The Adirondack Spruce, pp. 1-157 (1898). o Hort. Kew. v. 319 (1813). Picea one at Stanage Park, Herefordshire, which measured in 1911, 72 ft. by 5 ft. 9 in. ; and another at Merton, Norfolk, 63 ft. high and 4 ft. 7 in. in girth in 1909. The large trees reputed * to be of this species at Dropmore are undoubtedly P. excelsa. The only place I know where this tree has been planted in quantity is on the drive of the Rhinefield enclosure in the New Forest, where there are a number of red and white spruce along the south end of the main avenue. The largest of the former was on the north side of the first cross ride on the west and measured 40 ft. by 4 ft. in 1906 ; most of the trees had old and new cones, low down on the ends of the branches, from which I have raised seedlings. The largest of the white spruce at the corner of the second cross ride on the west was 56 ft. by 5 ft. 4 in., and I found cones on only one of the trees. Some small trees sent from America grow very slowly at Colesborne. In 1870 Gorrie2 found a few trees of this species, about 12 to 18 ft. in height, and bearing cones, which were growing on the railway bank near Tynehead in Midlothian at 800 ft. elevation. They had been raised about fifteen years previously from seed obtained in Newfoundland. Some of the seedlings which had been planted two or three miles off in a dry heavy soil had dwindled and died. Dr. A. W. Borthwick visited this place in 1906, and sent me specimens from these trees, from the cones of which I have raised seedlings. The trees are now about 35 ft. in height, growing mixed with common spruce, but not so large as white spruce at the same place. Gorrie also reported in the same year trees about 15 to 20 ft. high growing in Dunmore Park, near Stirling. We have not been able to discover whether these are living. At Avondale, in Ireland, there is an experimental plot, covering about two acres, which was planted in 1907 with red spruce, mixed with a small proportion of white and black spruce. The red spruce in this plot is extremely thriving, being about 6 ft. high in January 1912, and exceeding in vigour a plot of European spruce beside it. In France the species seems to grow remarkably well at Les Barres, according to Pardé, who figures3 a group of three trees about 45 ft. by 4 ft. There are others even larger planted in 1832 which have produced several natural seedlings. Beissner says that there are fine specimens in the Karls-aue at Cassel, at Herren hausen in Hanover, and at Worlitz. (H. J. E.) In 1908 I visited in the Hertogenwald in Belgium, a plantation of red spruce consisting of five acres in two separate plots at an elevation of 1250 ft. The soil here is a loamy clay, on which the European spruce thrives remarkably well. The plots had been accurately measured in 1907, when the trees were fifty-five years old with the following results :— Number of trees per acre, 950. Total volume in the round per acre, 3265 cubic ft. Annual increment, about 60 cubic ft. per acre. 1 Kent, Veitch's Man. Conif. 451, note (1900). a In Trans. Bot. Soc. Edin. x. 353 (1870). 3 Art. Nat. des Barres, 102, pi. 46 (1906). 1380 The Trees of Great Britain and Ireland The two best trees were, in 1908, 50 ft. high and 10 in. in diameter at five feet from the ground. The plantation had not been properly thinned at an early period. The European spruce in the same district, at a higher elevation, about 1800 ft., averaged at forty-five years old 445 trees per acre, with an annual increment of 136 cubic ft. per acre ; and the best trees were 60 ft. high by \2\ in. in diameter. The red spruce had a redder and more scaly bark than the common spruce ; and was more densely clothed with foliage, the improvement of the soil due to the decay of the fallen leaves being well marked. The trees bear cones about every two years ; and I noticed several seedlings in the open ground adjoining the plantation. One of the trees had a sucker from its roots about 3 ft. high. (A. H.) PICEA ALBA, WHITE SPRUCE Picea affla,1 Link, Handb. ii. 478 (1831), and in Linnœa, xv. 519 (1841); Kent, Veitch's Man. Conif. 427 (1900); Clinton-Baker, Illust. Conif. ii. 34 (1909). Picea canadensis, Britton, Sterns, and Poggenburg, Cat. PI. N. York, 71 (1888); Sargent, Sifoa IV. Amer. xii. 37, t. 598 (1898), and Trees JV. Amer. 42 (1905). Picea laxa, Sargent, in Garden and Forest, ii. 496 (1888); Jack, in Garden and Forest, x. 63 (1897). Abies canadensis, Miller, Diet. 8th ed. No. 4 (1768). Abies alba, Michaux, Fl. Bor. Amer. ii. 207 (1803) (not Miller); Loudon, Arb. et Friit. Brit. iv. 2310 (1838). Abies curvifolia, Salisbury, in Trans. Linn. Soc. viii. 315 (1807). Abies laxa, Koch, Dendrologie, ii. 2. p. 243 (1873). Pinus canadensis, Du Roi, Obst. Bot. 38 (1771) (not Linnœus). Pinus laxa, Ehrhart, Beiträge, iii. 24 (1788). Pinus alba, Solander, in Aiton, Hort. Kew. iii. 371 (1789); Lambert, Genus Pinus, i. 39, t. 26 (1803). A tree, attaining in America 70 to ioo ft. in height and 6 to 8 ft. in girth. Bark £ to \ in. thick, with thin greyish plate-like scales. Young branchlets slender, glabrous, glaucous ; becoming greyish or pale brown in the second year. Buds, \ in. long, ovoid, rounded or obtuse at the apex ; with glabrous scales, non-ciliate, rounded and bifid at the tip, and usually loosely imbricated. Leaves on lateral branches arranged as in the common spruce, usually with a disagreeable odour2 when bruised, bluish, 1 The oldest specific name (canadensis of Miller) for this species is not available, as it was previously used by Linnœus for the eastern hemlock, his Pinus canadensis being Tsuga canadensis. Moreover, Jack, in Garden and Forest, x. 63 (1897), gives some reasons for supposing that Miller indicated the red spruce by his name Abies canadensis ; and on this account Jack proposes the name Picea canadensis for the red spruce, and Picea laxa for the white spruce. The latter name is based on Ehrhart's Pinus laxa, which is earlier than Solander's Pinus alba. Voss, in Mitt. Deut. Dend. Ces. 1907, p. 93, proposes Picea glatica for the white spruce, based on Pinus glanca, Moench, which is earlier than any name except Miller's, but was applied to the glaucous variety. In the midst of the confusion, in which the nomenclature of the American black, white, and red spruces is involved, it is most convenient to adopt for them the names Picea nigra, Picea alba, and Picea rubra, which were first used in combination by Link, and which have been in common use for a great number of years. Moreover, these names are unambiguous, as they have always been applied in each case to the same species. 2 Hence the name cat or skunk spruce often given to the tree in America. Usually the odour is only perceived when the leaves are bruised, but in certain states of the air it can be detected at some distance from the tree. Cf. Garden and Forest, x. 63 (1897). Picea 1381 about \ in. long, incurved, ending in a rounded or acute (not acuminate) cartilaginous tip ; quadrangular in section, with three to four rows of stomata on each side. Cones, deciduous in the autumn or winter of the first year after the escape of the seeds, sessile or shortly stalked ; slender, cylindrical but tapering at both ends, about 2 in. long and \ in. in diameter, green when growing, shining pale brown when ripe : scales few, loosely imbricated, thin and flexible, so that the cone can be easily crushed by the hand, orbicular or oval, \ in. broad, rounded or truncate at the entire anterior margin : bract about £ in. long, oblong with a slightly enlarged ovate denticulate lamina. Seed, £ in. long, brown, partly embraced by the inflexed margins of the base of the narrow pale wing, which is broadest near the rounded denticulate apex ; seed with wing, § in. long. The three American species are often confused, though they have been clearly recognised by botanists in Europe since Lambert's time. In America the younger Michaux and Asa Gray united P. rubra with P. nigra ; but all modern American botanists and foresters keep the three species distinct. The best account of their history is given by Dr. G. Lawson of Halifax, Nova Scotia, in Proc. Canad. Institute, 1887, pp. 169-179. Formerly the white spruce was considered to be a native of the Rocky Mountains, but the tree inhabiting Alberta, British Columbia, and Montana is now considered to be distinct, and has been named P. albertiana. P. alba is readily distinguished by its bluish disagreeably smelling foliage and glabrous branchlets, and cannot be confused with P. nigra and P. rubra, which have pubescent branchlets and peculiar buds with long subulate scales. The cones of the white spruce are easily crushed by the hand on account of their thin flexible scales, and are very different in shape from those of the other two species, which have firm rigid scales.1 VARIETIES 1. Var. arctica, Kurz, in Bot. Jahrb. xix. 425 (1895). Abies arctica, Murray, \njourn. Bot. v. 253, t. 69 (1867). Pinus alba, var. arctica, Pariatore, in De Candolle, Prod. xvi. 2, p. 414 (1868). Towards the northern limit of its a'rea the white spruce has thicker leaves and smaller cones, with more concave scales and bracts slightly different in shape. This form was first collected by Seemann in north-western Alaska. According to Sargent2 the branchlets of the white spruce in the interior of Alaska are sometimes slightly pubescent, and in all probability this variety is a connecting link between P. alba and P. albertiana. A few peculiar forms have arisen in cultivation :— 2. Var. nana, Loudon. A round compact bush, rarely exceeding 6 ft. in height. 3. Beissner mentions fastigiate, pendulous, and variegated forms, which we have not seen in England. 1 Trelease, in Bot. Gaz. xxix. 196 (1900) describes remarkable burrs, almost globose in shape and covered with smooth bark, which are occasionally seen on the trunk and branches of the white spruce in the United States. 2 Silva N. Amer. xii. 38, note (1898). 1382 The Trees of Great Britain and Ireland 4. Var. cœrulea, Carrière, Conif. 320 (1867). Pinus glauca, Moench, Bäume Weiss. 73 (1785). Abies rubra cœruka, Loudon, Arb. et frut. Brit. iv. 2316 (1838). Abies cœrulea, Forbes, Pin. IVolurn. 99 (1839). Picea cœrulea, Link, in Linnœa, xv. 522 (1841). Pinus rubra violacea, Endlicher, Syn. Conif. 114 (1847). Picea canadensis glauca, Sud worth, in U.S. Forestry Bulletin, No. 14, p. 37 (1897). A small tree of dense pyramidal habit, with very glaucous leaves closely pressed against the branchlets. This variety, which according to Carrière frequently arises in the seed-bed, appears to have been known since 1785, and is unquestionably a form of P. alba, though it has been by various authorities ascribed to P. rubra. (A. H.) DISTRIBUTION The white spruce is a native of eastern Canada and the northern part of the United States, extending southward to the Black Hills of Dakota, the northern parts of Minnesota, Wisconsin, and Michigan, New York, Vermont, northern New Hampshire, and the coast of Maine as far south as Casco Bay. It is recorded1 for a few stations in Massachusetts, its most southerly limit. Its westerly distribution in the Dominion of Canada is uncertain ; but according to Dr. Lawson, the white spruce is essentially a maritime species, growing along the Atlantic and northern coasts of Canada, and extending by way of the St. Lawrence to the great lakes, as far as Lake Winnipeg. It is common in Newfoundland, Nova Scotia, and New Brunswick, and on the streams which flow from the north into the St. Lawrence, ranging westward through Ontario to the treeless plains of Manitoba, where it occupies sandhills and the dry slopes of river banks. In Labrador it is widely but not generally distributed, growing in the south in well-watered valleys and ascending rocky hills to elevations of 2000 ft. West of Hudson Bay it grows to a large size on river terraces to the borders of the barren lands ; and its stems choke the mouths of every arctic American river, strewing the shores with driftwood and testifying to its abundance on their shifting banks.2 CULTIVATION The white spruce was first described by Miller in 1731, and is said by Loudon to have been introduced into England by Bishop Compton in 1700. Though the name is often found in nursery catalogues and it has no doubt been planted in many places, yet it is nowhere in England so far as we have seen of any special value, either as a timber or an ornamental tree. In some parts of Denmark, however, it has been largely planted as a shelter tree on poor sandy land, in alternate rows with Pinus montana, as it is found to grow on poorer soil and to bear salt sea 1 Dame and Brooks, Trees of New England, 17 (1902). 2 E. T. Seaton, Arctic Prairies, 329 (1912), measured a tree near Fort McKay 118 ft. high. A log here, 84 ft. long, Was 22 in. in diameter at the butt and 15 in. at the small end. At tree limit on the eastern shore of Artillery Lake, a tree, 8 ft. high and I ft. in diameter at the butt, showed 300 annual rings. Picea 1383 wind better than common spruce. From what I saw, however, during our visit to Denmark in igoS,1 it is not likely to become a timber tree of any value here. In a paper on the " Reclamation of Moors in Belgium," * Mr. A. P. Grenfell says that it forms an excellent mixture with common spruce on poor soils, and in exposed situations in that country, and that it is more windfirm than common spruce. The white spruce is extremely hardy, and will thrive in exposed situations on high hills, where the common spruce succumbs to the continued action of cold winds in spring. Annand8 gives an instance of its success on poor peaty soil at a high elevation in the north of Scotland, and recommends it for planting as a wind break. He tells us that at Carragill in Cumberland, where it has been planted in perhaps the most exposed situation in England, it continues to grow as a low tree between 1600 and 2000 feet, where there is practically no soil, and above the zone in which the common spruce can exist. He considers it specially valuable on wet soils ; and says that it has been planted for shelter to a considerable extent in the Moorfoot hills in Midlothian, and in hilly districts in Peeblesshire, Aberdeen- shire, and Caithness. At Durris,4 however, P. sitchensis far surpasses it in growth in such conditions, and appears to be equally hardy. In the Hertogenwald in Belgium P. alba has been planted with'some success in the wettest parts of the peat mosses at high altitudes. At Weston Hall, Staffordshire, on good alluvial soil, a plantation was made of the common spruce in 1868, amongst which are scattered a few P. alba. The best of the white spruce was 45 ft. by 2 ft. 9 in. in 1909, while the European spruce averaged 60 ft. by 5 ft. 3 in. The tallest white spruce in England is probably one at Woburn, which measured in 1909 72 ft. by 4 ft. 6 in. There are two fine trees at Powis Castle, the best of which measured 56 ft. by 5 ft. 3 in. in 1906. At Eastnor, a white spruce measures 46 ft. by 2 ft. ID in. In the Keillour pinetum, Perthshire, which was planted in 1832, the best P. alba measured 52 ft. by 5^ ft. in 1905. In this poor boggy soil, the West American conifers much surpass both P. alba and P. nigra in growth, the growth of P. sitchensis and Abies grandis being astonishing.6 Kent mentions a tree, 45 ft. high, growing on light loam at Dolphinton, Lanarkshire. At Fota, a white spruce was 45 ft. by 5 ft. in 1907. TIMBER OF THE BLACK, RED, AND WHITE SPRUCES In the United States, only the red and white spruce yield merchantable timber, the black spruce never attaining large enough dimensions. In Canada the red spruce is never mentioned by foresters or lumbermen, and only the white and black spruces are said to produce lumber. According to Dr. Lawson, the black spruce is 1 Quarterly Jeumal cfForestry, iii. 75 (1909). 2 jbid- n 2?3 (1908). 3 In Trans. Key. Scot. Arb. Soc. xvi. 473 (1901). 4 Cf. Crozier's account in Trans. Roy. Scot. Arb. Soc. xxiii. 7-16, plate i (1910), and in our Vol. I. p. 95. 6 Cf. our account of the pinetum at Keillour, in Vol. I. p. 96. Complete details of the original planting operations in 1832 at Keillour are given in Free. Hort. Soc. iii. 297 (1863). VI II 1384 The Trees of Great Britain and Ireland famed amongst the Canadian lumbermen as a tree yielding sound, strong, and lasting timber ; while red spruce produces softer wood, less durable " under exposure to the open air, as is known from experience ; every year the red spruce poles have to be replaced more frequently than the black in fences.1" Langelier2 states that the black spruce is the prevailing coniferous tree in northern Quebec, where the forests are estimated to be capable of supplying 400,000,000 cords of pulp wood. White spruce is less abundant in this region, but attains a larger size, and is utilised for lumber, only the tops being converted into pulp wood. In the southern section of the Abitibi territory white spruce attains splendid dimensions over an area of 15,000,000 acres, and Mr. H. O'Sullivan has seen trees over 100 ft. in height and 20 in. in diameter. Dr. Bell is quoted as saying that " white spruce is perhaps the most valuable tree of the district. It grows to a great size everywhere along the rivers and lakes, where it often girths upwards of 6 ft. The timber is sound ; as a rule the trunks run to a great height, and in every respect the white spruce ranks among the very best timber for the manufacture of first-class saw-logs. " J. M. Macoun, in Forest Wealth of Canada, says that the wood of the three species is not separated commercially, and that they are used for the same purposes. The black spruce is perhaps the best suited for masts or spars. Of the white spruce he says that the wood is tougher, stronger, and more elastic than that of pine, and is very largely used as lumber, and for railway ties, fence-posts, piles, and telegraph poles. The wood of these Canadian spruces now supplies the greater part of the material used for pulp-making, which has recently become one of the great industries of Canada. According to a paper on Pulp Wood in Canada, by George Johnson, which was printed for the Minister of Agriculture at Ottawa in 1904, no less than 15,000,000 to 20,000,000 dollars are now invested in this manufacture; and as it is estimated that no less than 450,000,000 acres of land in Canada are covered more or less densely with spruce which reproduces itself very rapidly when cut, there is no risk of the supply failing. Great Britain and the United States are said to consume about 900,000 tons of pulp wood annually, the product of about 90,000 acres. The black spruce is considered better than the white for this purpose and grows mostly on the hills and rocky ground, whilst white spruce loves valleys, where there is more soil. To show the rapid increase in the value of these timber lands it is stated that in 1892 spruce limits were sold in the province of Quebec as low as eight dollars per square mile, whereas in 1899 similar limits realised 150 dollars per mile, and the price has risen higher lately. English papermakers are said to have found out that Canadian spruce pulp makes a stronger and better newspaper than Scandinavian pulp ; and the immense water-power of the Dominion makes both the transport of the logs and the manu facture cheaper than in most parts of Europe. (H. J. E.) 1 Proc. Canad. Inst. 1887, p. 169. 2 Report 6th Meeting Canada Forestry Association, 1905, p. 65. Picea 1385 PICEA ALBERTIANA, ALBERTA WHITE SPRUCE Picea albertiana, Stewardson Brown, in Torreya, vii. 126 (1907); Render, in Mitt. Deut. Dend. Ges., 1907, p. 69; Britton and Shafer, IV. Amer. Trees, 58 (1908). Picea columbiana, Rydberg, in Mem. Neiv York Bot. Garden, i. n (1900) (not Lemmon T) ; M. E. Jones, in Montana Botany Notes, 10 (1910). Picea alba, Mayr, Freindländ. Wald- u. Parkbäume, 319 (in part), fig. 101 (1906) (not Link). A tree, attaining in western North America 160 feet in height. Bark thin, greyish white, scaling off in small quadrangular plates, furrowed at the base of old trunks. Young branchlets greyish or light yellow ; yellow or orange in the second year ; glabrous or more usually with a minute pubescence on the pegs (from which the leaves arise), which is occasionally scattered over the rest of the surface of the pulvini. Buds about \ in. long, ovoid, slightly resinous, with scarious scales rounded and entire in margin ; terminal buds girt at the base with acuminate ciliate keeled scales. Leaves bluish green, in an imperfect radial arrangement on the lateral branches, but more crowded on the upper side of the branchlets ; |- to i in. long, soft or rigid, curved, ending in a short point, quadrangular in section, with three to five stomatic lines on each side. Cones, i to -2\ in. long, cylindrical, obtuse at the apex, sessile, about i in. wide when open, shining light brown when ripe : scales numerous, thin, and flexible, fan- shaped, wide, and rounded anteriorly, cuneate on the sides, flatter than those of P. alba ; upper margin thin, undulate, or faintly denticulate ; light brown and glabrous on the exposed part, minutely pubescent and reddish brown on the concealed part : bract J in. long, with an oblong claw, slightly expanded into a denticulate lamina, which is either emarginate or rounded at the apex. Seed ^ in. long, mottled dark brown ; seed with wing J in. long ; wing contracted just above the seed, widest in the upper two-thirds, ending in an oblique denticulate apex. This species is very variable in the amount of pubescence on the branchlets, which is occasionally absent both in specimens from Montana2 and from Alberta. The cones are also variable in size, and in the faint clenticulation of the margin of the scale. It is most closely allied to P. alba, of which it may be considered the Rocky Mountain form. In P. alba the branchlets are always perfectly glabrous, with less prominent pulvini ; and the leaves are differently arranged in the two species. The buds of P. alba are non-resinous, with scales emarginate or two-lobed, and not entire as in P. albertiana. The cones of P. alba are less rigid, being easily crushed by the hand, and have very fragile scales, entire in margin, more concave internally from side to side, and more reddish brown in colour than those of P. albertiana. The seeds are similar in the two species, but those of P. alba have shorter wings. 1 P. columbiana, Lemmon, is imperfectly described, and is referred by Sargent and by Britton to P. Engelmanni. Lemmon's description may have partly included P. albertiana ; but the latter name, being quite certain, must stand for the species now treated here. Cf. p. 1388. ! Three trees growing together in a clump at 3300 ft. altitude, near Belton in Montana, which I examined in 1906, were precisely alike in size, bark, and habit. One had perfectly glabrous branchlets, silvery leaves, and large cones. Another had very pubescent branchlets, green leaves, and small cones. The third was intermediate. 1386 The Trees of Great Britain and Ireland This species, the exact distribution of which has not yet been clearly defined, ranges from Wyoming1 and western Montana northward to Alberta and British Columbia. It occurs in the Rocky Mountains at lower elevations than P. Engelmanni, extending from 3000 to 5000 ft. The type specimen was collected at Bankhead, Alberta, by Stewardson Brown ; and I have received from Macoun specimens from the neighbourhood of Banff, in the same province. Rehder states that this spruce near Banff occasionally attains 160 ft. in height, and forms extensive forests, in one of which he took a fine photograph, which shows well the habit of the tree, and is reproduced by Mayr 2 in his article on the white spruce. P. albertiana is the white spruce referred to by Sargents as a native of " the Rocky Mountains of Alberta, British Columbia, and northern Montana, where it lines the banks of streams and lakes up to 5000 ft. elevation, attaining a large size, and sending up tall spire-like heads of dark foliage." In Montana this spruce is not found on the east side of the continental divide, but is common in the Flathead4 region, where it forms a low tree in marshy situations ; but on moist alluvial soil, in mixture with the Douglas fir, western larch, and Thuya plicata, it attains large dimensions. It usually occurs in small groups in these mixed forests, occupying the moister ground, and bearing considerable shade. The largest tree which I measured, growing near Nyack on the Northern Pacific railway, was 150 ft. by 10 ft. A tree 114 ft. by 4 ft. 9 in. showed, when cut down, 114 annual rings; another, 15 in. in diameter, showed 160 rings, the bark being only \ in. thick. It is possible that the trees referred to P. Engelmanni, in Idaho, Washington, and Oregon, may wholly or in part belong to P. albertiana ; and a further study of the spruces in western America is desirable, as the variability in P. albertiana points possibly to hybridisation with Engelmann's spruce. This spruce is the finest species in North America, except P. sitcliensis, and is worth a trial as an ornamental tree. It was introduced into England by Elwes, who received seeds from Mr. J. M. Macoun of Ottawa in 1906, which have produced plants, the largest of which in 1912 were about 18 in. high, and which have been distributed to several places in England and Scotland. According to Rehder, it was sent by Baron von Fürstenberg to Germany in 1907. (A. H.) 1 Britton and Shafer, N. Amer. Trees, 58 (1908), give Wyoming as a habitat ; but I have seen no specimens. 2 Fremdliind. Wald- u. Parkbäume, fig. loi (1906). This photograph is also reproduced in Möller's Deut. Gärtn. Zeit. 1905, p. 117. 3 Silva N. Amer. xii. 39 (1898). 1 The spruce described as P. Engelmanni by Whitford, in Bet. Gas. xxxix. 196 (1905). Picea PICEA ENGELMANNI, ENGELMANN'S SPRUCE Picea Engelmanni, Engclmann, in Trans. St. Louis Acad. \\. 212 (1863), and in Gard. Chron. 1863, p. 1035 ; Sargent, Silva N. Amer. xii. 43, t. 599 (1898), and Trees N.Amer. 43 (1905); Kent, Veitch's Man. Conif. 431 (1900); Britton, N. Amer. Trees, 59 (1908); Clinton-Baker, Illust. Conif. ii. 37 (1909). Picea columbiana, Lemmon, in Garden and Forest, x. 183 (1897). Picea pseudopungens, Dieck,1 Verkaufs- Verzeichnis* ZSschen, 28 (1904). Abies Engelmanni, Parry, in Trans. St. Louis Acad. ii. 122 (1863). Abies commutata, Murray, in Gard. Chron. iii. 106 (1875). Pinus commutata, Pariatore, in De Candolle, Prod. xvi. 2, p. 417 (1868). A tree with disagreeably smelling foliage, attaining in America 150 ft. in height and 15 ft. in girth, though usually considerably smaller. Bark reddish, exuding resin, with thin loose scales. Young branchlets greyish yellow, with a sparse minute erect glandular pubescence. Buds conical, about £ in. long, obtuse at the apex ; the terminal bud closely surrounded at the base by the uppermost leaves ; scales scarious, rounded, without resin. Leaves, arranged on lateral branches as in P. excelsa, £ to i in. long, soft and flexible, tapering towards the apex, which ends in a sharp point ; bluish green in colour, with a cat-like odour when bruised ; quadrangular in section, with four to five stomatic lines on each side. Cones horizontal at first, ultimately pendulous, sessile, green tinged with scarlet when growing, light brown when ripe, cylindrical but narrowed at both ends ; very variable in size, i£ to 3 in. long : scales numerous, thin and flexible, rhombic or ovate, minutely pubescent in the lower half, longer than broad, f in. wide, with their upper margin truncate or rounded and lacerate : bract £ in. long, with an oblong claw, and an oval expanded denticulate lamina. Seed about ^ in. long, dark brown ; seed with wmg i3d i"- 6°6 (I844), and in Godman, Nat. Hist. Azores, 224 (1870); Trelease, in Missouri Bot. Gard. StA Ann. Rep. 169 (1897); Kent, Veitch's Man. Conif. 180 (1900). Juniperus Oxycedrus, Linnaeus, var. brevifolia, Hochstetter, in Seubert, Fl. Azorica, 26 (1844). Juniperus rufescens, Link, var. brevifolia, Endlicher, Syn. Conif. n (1847). A shrub or small tree in the Azores, with a stem often 3 to 4 ft. in girth. Branchlets numerous, short, densely clothed with foliage. Leaves all acicular, in alternate whorls of threes, very short and broad, \ to \ in. long, about ^ in. wide, oval-linear, jointed and swollen at the slightly narrowed base, widest about the middle, whence they taper to a rounded or acute (non-acuminate) apex ; upper surface with a narrow green midrib not extending to the apex, on each side of which is a broad white stomatic furrow, bounded by an external green band ; lower surface green, with a prominent midrib ; margin entire. Flowers dioecious. Fruit sub-globose, \ in. in diameter, on scaly stalks about ^g- in. long, dark reddish brown when mature ; scales three, separated at the apex by three radiating prominent lines, and each marked by a minute mucro. Seeds three, embedded in a scanty pulp, ovoid, triquetrous, broadest at the base, gradually tapering to an acute apex, light brown ; outer surface convex with two or three longitudinal furrows, not separated from the inner surface by a winged thin margin, as inj. Oxycedrus. This species is remarkably distinct in the very short glaucous leaves ; and its seeds differ from those of the allied species. This species is limited to the Azores, where it occurs on the islands of Corvo, Flores, Fayal, San Miguel, and Pico, ascending to 5000 ft., and rarely descending below TOGO ft. It is locally known as cedro, and is usually a compact shrub or small tree, becoming a prostrate bush with interlacing branches on exposed hill summits. Formerly it appears to have been a tree of considerable size, as large logs have been found deeply buried under the secondary volcanic debris in the Grotto do Enferno of the large crater known as Caldeira des Sette Cidades. A slab of this was pre sented to the Kew Museum by Dr. Goeze, which was reported to have been excavated from a depth of 100 metres.1 So far as we know this beautiful species has not been introduced2 into cultivation in England. (A. H.) 1 Cf. Goeze in Card. Chron. 1867, p. 929, and Masters mjourti. J?. Hort. Soc. xvii. 3 (1894), who identified this wood with Cupressus lusitanien. The slab at Kew is undoubtedly the wood of this species of juniper. Cf. our Vol. V. 1179, note I. 2 Gordon, Pinetum, 131 (1880), says it is tender in England ; but he gives no particulars as to its introduction. 1414 The Trees of Great Britain and Ireland JUNIPERUS CEDRUS Juniperus Cedrus, Webb and Berthelot, Hist. Nat. Isles Canar. iii., Phytog. Canar. ii. 277, t. 217 (1840); Antoine, Cupress. Gattung. 14, tt. 16-19 (1857); Pariatore, in De Candolle, Prod. xvi. 2, p. 478 (1868); Kent, Veitch's Man. Conif. 180 (1900). Juniperus Oxycedrus grandifolius, Buch, Phys. Beschr. Can. Inseln, 109, 159 (1825). Juniperus canariensis, Knight, Syn. Conif. 13 (1850). Juniperus Webbit, Carrière, Conif. 13 (1855). A tree, attaining a large size in the Canary Islands, with wide-spreading branches and long pendulous branchlets. This is an insular form of /. Oxycedrus, differing mainly from the Mediterranean tree in habit and not in technical characters. Leaves directed towards the apex of the branchlet, and not widely spreading, resembling in this respect J. formosana, thinner in texture than in J. Oxycedrus, and becoming acute or rounded at the apex in adult trees ; glaucous and not bright green on the lower surface, the glaucous tint being present on the narrow midrib and the borders external to the white stomatic bands of the upper surface. Fruit similar to that of J. Oxycedrus, with two or three seeds, which are often acute and not broad and emarginate at the apex, the resin-pits being usually more developed than in the Mediterranean species. Copious specimens received from Dr. Perez show that this tree is scarcely separable as a distinct species from J. Oxycedrus ; and trees of the latter, with pendulous branchlets, which occur on the Riviera, are very similar to, if not identical withy. Cedrus. J. Cedrus is a native of the Canary Islands, where it has been nearly exter minated by the inhabitants, who value its timber highly. It still exists on Teneriffe, Grand Canary, and Palma, but is extinct on Gomera.1 Dr. Perez writes that it was common on Teneriffe at the end of the eighteenth century,2 as Humboldt mentions it as occurring all the way up from Orotava to the canadas, growing with Pinus canariensis ; but only a few stunted specimens now survive, which grow on inac cessible rocks about the canadas, from 7000 to 9000 ft. altitude. Schenck,3 who gives the latest published account, quotes Fritsch,4 who mentions a noble juniper, which formerly grew on the south-west side of the Peak at 8000 ft. elevation. This tree was 18 ft. in girth and nearly 100 ft. high. It also occurs in the crater of Tirijana on Grand Canary, but seems to be most abundant now on Palma,5 where it grows on the inaccessible inner walls of the crater, and outside it to the north-east on isolated rocks at 7000 ft. elevation. Dr. Burchard8 1 Dr. Christ, in Engler, Bot. Jahrb. vi. 487, 500 (1885). 2 Buch, Phys. Beschr. Canar. Inseln, 109 (1825), mentions a few trees which were then growing at 9000 ft. altitude amidst the lava at the foot of the last cone of the Peak. 3 Wiss. Ergeb. Deut. Exped. 'Valdivia,' ii. pt. I, p. 375, figs. 63, 64 (1907). 4 Reisebilder, 6 (1879). 5 Bornmllller, in Engler, Bot. Jahrb. xxxiü. 398 (1904), states that he found trees also on the south side of the crater, in the Barranco de las Angustias at 1300 and 2700 ft. elevation ; and at Cumbrecita at 4700 ft. 6 In Mitt. Deut. Vend. Ges. 1911, p. 296, fig. Juniperus says that these trees have trunks almost completely bare of bark, and over a metre in diameter ; and reproduces the photograph of a fine old female tree, which he took in June 1910. Schenck1 also figures a very old tree, with a short bole, a few snaggy branches, and very little foliage. (A. H.) It is doubtful if this plant was introduced until recently, as it was not mentioned by Loudon2 in 1838, and was included by Knight and Perry in 1850 amongst the kinds of juniper of which little was known. Kent says that it is not hardy in England ; and the only plant now living at Kew is one in the Temperate House, about 7 ft. high, raised from seed sent by Sir Daniel Morris in 1893. Sir John Ross-oi- Bladensburg, K.C.B., however, informs me that a plant in his garden at Rostrevor survived without protection the winter of 1909-1910, which was exceptionally severe in the north of Ireland. As it grew well during the following summer, he looks upon it as hardy; in December 1911 it was 6 ft. high with a good leader, and slightly pendulous branches. A small plant at Glasnevin bore last winter 12° of frost without injury. J. Cedrus is readily propagated by cuttings at Kew ; but there seems to be a great difficulty in raising it from seed.8 Beissner,4 however, has raised young plants from seed which I saw in his collection at Bonn in 1908. (H. J. E.) JUNIPERUS FORMOSANA Juniperus formosana, Hayata, m Journ. Coll. Sei. Tokyo, xxv. art. 19, p. 209, pi. 38 (1908). Juniperus oblonga pendulaf Knight and Perry, Syn. Conif. ii (1850)5 Carrière, Conif. 20 (1867). Juniperus taxifolia, Masters, m Journ. Roy. Hort. Soc. xiv. 215 (1892), and Journ. Linn. Soc. (Bot.) xxvi. 543 (1902) (in part); and m Journ. Bot. xli. 268 (1903) (not Hooker and Arnott) ; Kent, Veitch's Man. Conif. 191 (1900). A tree, attaining in China 40 ft. in height. Branchlets triquetrous, with three narrow ridges, yellowish green in the first year, reddish brown in the second year. Leaves all acicular, spreading, in whorls of threes, linear-subulate, about ^ in. long and Jjj to ^.j in. broad, jointed and swollen at the base, ending in a sharp spine-like 1 Beil. Kennt. Veget. Canar. Inseln, fig. 63, in Wiss. Ergeb. Deul. Exped. 'Valdivia,' iSçS-iSçç (1907). This figure is a reproduction of a photograph taken by Prof. Simony. A figure of a similar tree, also growing on Palma, is given by Webb and Beithelot, Hist. Nat. Isles Canar., Atlas, t. 8 (1838). 2 Cf. note 5 below. 3 Dr. Geo. V. Perez of Orotava, Teneriffe, wrote in Card. Chron. xl. 14 (1906), and xli. 134 (1907), that none of the seed which he sent to Kew and elsewhere germinated. He finds that seeds, soaked in water at 70° Fahr, for 15 to 30 days, germinate freely in about six weeks. Probably germination occurs, under ordinary conditions, in the second year after sowing, or in the wild state after the seeds have been eaten and voided by birds. Correvon, in Card. Chron. xlii. 209 (1907). reports that seeds, which had been soaked in a weak solution of acetic acid, germinated well at Geneva. Dr. Perez informs me that Mr. Lister has raised, from seed sent from Orotava, eight plants in the Government nursery at Pretoria, which are now 4 ft. high.—(A. II.) * Mill. Deut. Dend. Ges. 1906, p. 91. 6 Knight and Perry's plant is undescribed, but is stated to have come from China and Japan, and is evidently the Chinese species here described, as is confirmed hy Carrière's description. It is apparently not the same plant as /. commuais oblonga pendula, Loudon, Art. et Friit. Brit. iv. 2489, fig. 2345 (1838), applied to a shruh at Kew, then 5 ft. high, with fastigiate branches and pendulous branchlets. Gordon, in Gard. Chron. 1842, p. 652, describes the latter as : "Trained to a single stem, if left to nature, it will not rise more than three feet, but will spread over a large space of ground. It is quite hardy and a native of the Caucasus." Webb, Phyt. Canar. ii. 277 (1840), identified this plant with /. Cedrus, which is, however, not hardy at Kew. It is impossible now to identify Loudon's plant, but in all probability it was a pendulous variety of_/. communis. VI vr 1416 The Trees of Great Britain and Ireland point ; lower surface convex, keeled ; upper surface concave, with two broad white stomatic bands, separated by a very narrow green or glaucous midrib, extending from the base to near the apex, where the white bands coalesce. Fruit globose, ^ in. in diameter, ripening in the second year, shining dark reddish brown when ripe ; smooth on the surface, with three deep radial furrows at the summit, in the centre of each of which is a dark line, showing the separation of the three scales of which the fruit is composed ; outer edge of each furrow with a thin mucro, overhanging a slight depression ; base of the fruit hollowed out at the insertion of the short scaly stalk. Seeds three, elongated-ovate, triquetrous, mucronate at the narrow thin apex, with several circular resin-pits at the base, above which on the outer surface are three or four larger elongated oval pits. In cultivated specimens in Europe the branches are ascending, but the branchlets are very pendulous, giving a weeping appearance to the tree. In China it assumes various habits, but is often very pendulous, and occasionally shrubby. The Chinese species now described, as pointed out by Hayata, is distinct from J. taxifolia,1 Hooker and Arnott, with which it has been confused. The latter, so far as we can learn, has never been introduced into cultivation, and is confined to Bonin Isle, where it was discovered by Capt. Beechey in 1827, and to the Liu Kiu Islands. I cannot find any particulars of the size and habit of this species. J.formosana is widely spread throughout the mountains of China, and is also commonly cultivated in temple grounds, being known as the Tz'e Pok, or "prickly cypress." It is represented in the Kew herbarium by numerous wild specimens from the provinces of Szechwan, Hupeh, Chekiang, and Fokien ; and was collected in Shensi by Père Giraldi. It has lately been found on Mt. Morrison and the adjacent ranges in Formosa, between 8000 and 13,000 ft. altitude.2 This species, which is usually known in cultivation as J. oblonga pendula, is stated by Kent to have been introduced from China by Fortune in 1856; but this is incorrect as regards the date. It was for sale in Knight and Perry's nursery in 1850, and was probably one of the plants sent home by Fortune in 1844. Knight and Perry describe it as a very elegant drooping shrub from China and Japan, which they supposed to be possibly identical with J. rigida ; but the latter was not introduced till 1861. J, formosana is now rare in collections, the best specimen that we have seen being a tree at Bicton, 30 ft. by 2 ft. A vigorous tree at Eastnor Castle is about 28 ft. high. A smaller specimen at Bayfordbury is reputed to have been planted 1 Juniperus taxifolia, Hooker and Arnott, Bot. Cap. Beechey s Voyage, 271 (1841) ; Siebold and Zuccarini, in Abh. Akad. Wiss. München, iv. 3, p. 233 (1846) ; Miquel, Pral. Fl.Japonica, 331 (1867) ; Pariatore, in De Candolle, Prod. xvi. 2, p. 481 (in part); Masters, \njourn. Linn. Soc. (Bot.) xviii. 496 (1882); Hayata, in Journ. Coll. Sei. Tokyo, xxv. art. 19, p. 210, fig. 6 (1908). Apparently a shrub. Leaves thicker than those of J. formosana, rounded or acute at the apex, and not ending in a spine-like point, about § in. long ; upper surface with two white stomatic bands, separated by an elevated green midrib from base to apex. Fruit globose, J in. in diameter, yellowish, rugose on the surface, with three prominent radial ridges at the apex, overhanging three furrows, external to each of which is a mucro. Seeds three, similar in shape and resin-pits to those off. formosana. This species is closely allied to J. formosana, mainly differing in the blunt and not spine-tipped leaves. 2 Specimens kindly sent me from Formosa by Mr. T. Kawakami are identical with specimens which I collected in central China in the mountains of Hupeh. Elwes gathered it in February 1912, at about 8000 ft. on the ridges above Arisan ; but saw it only in a bushy form. Juniperus 1417 in 1845. Another at Glasnevin, which bore fruit in 1911, was obtained some years ago from Messrs. Veitch, who have, however, no longer the plant for sale. It has been much confused with/, rigida ; and all the reputed trees1 of/, oblonga pendula on the continent are referable to this Japanese species. (A. H.) JUNIPERUS DRUPACEA Juniperus dntpacea, Labillardière, Icon. PL Syr. ii. 14, t. 8 (1791); Loudon, Arb. et Frut. Brit. iv. 2494 (1838); Lindley, in Gard. Chron. 1854, pp. 387, 455, fig.; Webster, in Gard. Chron. xix. 519, fig. 80 (1896); Kent, Veitch's Alan. Conif. 173 (1900). Arcenthos drupacea, Antoine and Kotschy, in Oester. Bot. Wochblatt, iv. 249 (1854). A dioecious tree, attaining about 60 feet in height, usually in cultivation columnar in habit, but in the wild state broadly pyramidal.2 Young branchlets triangular, with three prominent linear ridges and three grooves, due to the decurrent bases of the leaves. Older branchlets, from which the leaves have fallen, terete and smooth, with a brown scaly bark. Buds, one terminal and usually two or three in the axils of the leaves on the branchlet of the first year, about \ inch long, surrounded by minute sharp-pointed lanceolate scales, the outermost of which persist on the apex of the second and third year's branchlets. Leaves, all acicular, spreading in whorls of threes, about \ to f in. long, T\y to \ in. broad, jointed at the base, decurrent on the branchlet to the next whorl, linear- lanceolate, widest near the base, gradually tapering to the apex, which ends in a sharp cartilaginous point ; lower surface green, convex, with a linear prominent keel ; upper surface concave, with a broad green midrib deeply furrowed near the base, and two white stomatic bands, not extending to the margin, which is entire. Staminate flowers, five or six in a head, on a short scaly stalk, arising in the axil of a leaf on the second year's branchlet ; stamens nine to twelve in each flower. Fruit ripening in the second year, larger than in any other species, f to i in. in diameter, on a short scaly stalk, ovoid or nearly spherical, brown or bluish with a glaucous bloom, usually composed of nine fleshy scales, in whorls of threes, united together, ovate, thickened and often mucronate at the apex ; enclosing a large globose hard bony stone,8 with three small cells, each containing a minute oblong kernel, one of which is often aborted. This remarkable species is a native of the mountains of Asia Minor, Syria, and Greece. It is found throughout the Taurus range in Asia Minor, several localities being mentioned by Boissier,4 between Karaman and Ermenek, in the Bulgardagh, and in the Akkerdagh, close to Marasch. It grows at elevations of 1600 to 5600 feet, and either forms small pure woods or is scattered amidst the forests of cedar and 1 Specimens sent by Späth from Berlin, and by Tardé from Nancy and Les Barres. 2 Walter Siehe, in Garttnßora, xlvi. 207 (1897), states that this tree in the Cilician Taurus never assumes the narrow columnar form which is so common in cultivation. Young trees in the wild state are pyramidal in habit ; whilst older trees, which are mainly females preserved by the peasants on account of their fruit, have a rounded head of foliage. 3 Antoine and Kotschy, loc. cit. consider the bony stone to arise from the union of the testa of the seeds with the inner part of the three upper scales. « Flora Oriental's, v. 706 (1881). 1418 The Trees of Great Britain and Ireland Abies cilicica. Siehe mentions1 enormous trees between Namrun and Güllek in Cilicia, which are over 3 ft. in diameter. It also occurs in Syria, on Mount Cassio, and on the Lebanon and Anti-Lebanon ranges. In Greece it appears to be con fined to Mount Malevo, in the southern part of the Morea, where, according to Halacsy,2 it forms a small wood at 3700 to 4000 feet altitude. According to Boissier,3 the reported occurrence of this species in Crete is erroneous. It is known to the Turks as Andys or Habhel, and is called Duffran by the natives of Syria, who collect and eat the fruits, which have a pleasant though resinous flavour. This species is said by Loudon4 to have been introduced in 1820, but he acknowledges that he had only seen young plants, and these were probably incorrectly named, as Lindley6 in 1854 speaks of J. drupacea as a new plant. It is generally believed to have been for the first time introduced into western Europe in that year by Kotschy, who collected it in Asia Minor in 1853. So far as we know it has never produced fruit in England, where all the trees in cultivation are supposed to be males. It is normally dioecious, but M. Allard5 states that a tree in his arboretum at Angers, which bore staminate flowers for a long time, ultimately produced fruit, and afterwards remained moncecious. A female plant at Angers has, however, never produced staminate flowers. M. Mottet6 states that a small tree at Verrières, only 6 ft. high, has produced fruit, though no other tree of the same species is near it, and he supposes that it must have been fertilised by the pollen of another species. (A. H.) This species usually forms a narrow column and is one of the most beautiful of the junipers, yet is rarely seen in collections. It is perfectly hardy, and, judging from the way it grows at Colesborne, thrives in a limestone soil. The best specimen which I have seen is at Eastnor Castle, where there is a well-shaped tree, 31 ft. by 3^ ft. in 1908. This produced staminate flowers in May 1899, of which there are specimens in the British Museum. At Scorrier, in Cornwall, a fine specimen was 36 ft. by i ft. 9 in. in 1911. There are two good trees at Kew about 30 ft. high. Henry saw at Holkham in 1911 two very narrow columnar trees, about 40 ft. high by 3 ft. in girth. At Brickendon Grange, Hertford, a fine specimen measured 36 ft. high in 1912. Smaller trees occur at Highnam, Tortworth, Chiltley Place near Liphook, Young's nursery at Milford near Godalming, and other places. In Ireland the best specimen that we have seen, a tree about 30 ft. high, is growing at Woodstock, Kilkenny. In France it attains a larger size, and at Angers and Montpellier has produced fruit. A tree in M. Allard's arboretum at Angers, nearly 40 ft. high, is figured in Bull. Soc. Dend. France, 1908, p. 109. Another7 at Antibes, about 30 ft. high, has a leaning stem and wide-spreading branches, being very different in habit from the usual form of this species in cultivation. (H. J. E.) 1 In Mitt. Deut. Dend. Ges. 1911, p. 305. 2 Comp. Fl. Gnzc. iii. 455 (1904). There is a specimen in the Cambridge Herbarium collected by Orphanides on Mount Malevo. 3 p/ora Orientalis, \. 706 (1881). 4 Encycl. Trees, 1084 (1842). 6 Gard. Chron. 1854, p. 455. 6 Cf. Mottet, in Rev. Hort. 1904, p. 356, figs. 147, 148, where a tree is figured in the park of Baron Mallet at Château des Côtes, near Versailles, which Elwes found to be 36 ft. high in 1909. 7 Erroneously labelled J. Oxyccdrns, var. macrocarpa. Juniperus 1419 JUNIPERUS RECURVA Juniperus recurva, Buchanan-Hamilton, ex Don, Prod. Fl. Nepal. 55 (1825); Loudon, Arb. et Frut. Brit. iv. 2504 (1838); Masters, in Gard. Chron. xix. 468 and 574, fig. 69 (1883), and in Jmirn. Linn. Soc. (Sot.) xxvi. 542 (1902); J. D. Hooker, Fl. Brit. India, v. 647 (1888); Kent, Veitch's Man. Conif. 185 (1900) ; Gamble, Indian Timbers, 698 (1902); Brandis, Indian Trees, 694 (1906). Sabina recurva, Antoine, Cupress. Gatt. 67, tt. 88, 90, 91 (1857). A tree, attaining in the Himalayas 30 ft. or more in height. Bark brown, thin, peeling off in long fibrous strips. Branches curved, more or less pendulous. Young branchlets marked between the whorls by the decurrent bases of the leaves, which are separated by three grooves. Older branchlets from which the leaves have fallen smooth, with a light reddish-brown scaly epidermis. Leaves all acicular, densely imbricated in whorls of threes, their basal part being decurrent on the branchlet, loosely appressed, directed forwards, scarcely spreading ; linear-lanceolate, ^ to ^ in. long, ^ to ^5 in. broad, ending in a sharp cartilaginous point ; outer surface convex, channelled longitudinally from the insertion to about the middle, greyish green ; inner surface concave, whitened throughout. The leaves persist for several years, becoming brown in the third and fourth years, the mixture of green and brown leaves giving the foliage a peculiar appearance. Flowers monoecious.1 Staminate flowers, terminal or solitary in the axils of one or two of the leaves on the ultimate branchlets, oblong, about \ in. long, with twelve to sixteen stamens. Fruit axillary, ripening in the second year, subsessile, bracteate at the base, dark purplish brown, ovoid, about f in. long, composed of three or six united scales, each bearing a triangular spreading mucro, prominent near the umbilicate apex of the fruit. Seed solitary, occupying the greater part of the berry, ovoid, with a thin narrowed apex, marked on the surface with two or three large depressions for resin-glands. This species is a native of the eastern Himalayas, occurring in Sikkim and Bhutan, between 9000 and 12,000 feet. It is represented in the north-western Himalayas, China, and Japan by the closely allied species or variety, J. squamata. (A. H.) In the very moist climate of the interior of Sikkim, where it is common in the Lachen and Lachoong valleys, from about 10,000 to above 13,000 feet, Sir Joseph Hooker, whose sketch 2 of it has been reproduced in Veitch's Conifères, fig. 58, gives 30 ft. as its height ; but if my recollection is correct, I saw much larger trees above Lachoong ; and G. A. Gammie, in his account of a botanical tour in Sikkim,3 says that in the Sebu valley he saw large trees at 11,000 feet; and at 13,000 feet in the same valley it was the only arborescent vegetation. 1 This species appears to be always moncecious. Kent, Veitch's Man. Conif. 187 (1900), states that a tree at Fota is a male. We have specimens of this bearing both staminate and pistillate flowers on different branchlets of the same branch. 2 Himalayan Journals, ii. 45, fig. (1854). 3 Published as a Government paper, No. 4! B.S.I., dated Calcutta, 26th July 1893, reprinted in Kew Bulletin, October- November 1893, p. 311. 1420 The Trees of Great Britain and Ireland According to Gamble its growth is slow, about twenty-two rings per inch of radius for the Sikkim tree ; and the wood is very good, equal to the best pencil cedar, but is not used except to burn as incense in the Buddhist temples. This is the most ornamental of the junipers on account of its graceful drooping habit ; and though introduced1 in 1830, and hardy enough to grow well and ripen its berries in Scotland, it is not common in cultivation, and is seldom found in nurseries.2 Its success in cultivation seems to depend principally on sufficient moisture in summer, all the best specimens that I have seen being in districts where the rainfall is heavy. The largest I know in England is at Bicton, where there are two trees about 40 and 35 ft. high by 3 ft. 4 in. in girth, which bear abundance of berries. At Hafodunos in Denbighshire I saw in 1911 a very fine tree with three stems from the ground, almost equal in height, 40 ft. to 41 ft., and each about 2 ft. in girth. At Bodorgan in Anglesea there is a good-sized tree. Even in the drier climate of Gloucestershire there is a thriving tree at Highgrove, near Tetbury, the seat of Arthur Mitchell, Esq., from the berries of which I have raised plants ; and another occurs at Highnam. There are also good specimens at Pencarrow and Menabilly in Cornwall, Mamhead in Devon, Bayfordbury and High Canons, Herts, Rotherfield Park, Hants, and Holkham, Norfolk. In Scotland there is a small tree at Murthly, from which I have raised seedlings ; and a thriving shrub at Drumtochty Castle. Another at Dalkeith was 14 ft. high in 1907. In Ireland there is a large bushy tree8 with nine main stems, 40 ft. high and thirty-seven paces round at Castlewellan (Plate 349). At Salterbridge, Co. Waterford, the seat of Major Chearnley, I saw in 1910 another of the same type and almost as large; and at Fota I measured a tree 38 ft. high in 1910. (H. J. E.) JUNIPERUS SQUAMATA Juniperus squamata, Buchanan-Harailton, in Lambert, Genus Finns, ii. 17 (1824); Don, Prod. Fl. Nepal. 55 (1825); Loudon, Arb. et Frut. Brit. iv. 2504 (1838); Endlicher, Syn. Conif. 18 (1847); Koch, Dendrologie, ii. pt. ii. 121 (1873). Juntperus religiosa? Royle, Illust. Him. Plants, i. 351 (1839) (name only). Juniperus densaf Gordon, Pinet. Suffi. 32 (1862). Juntperus recurva, Don, var. squamata, Pariatore, in De Candolle, Prod. xvi. 2, p. 482 (1868) ; Brandis, Forest Flora N.W. India, 536 (1874), and Indian Trees, 694 (1906); Hooker, Fl. Brit. India, v. 647 (1888); Masters, v\ Journ. Linn. Soc. (Bot.) xxvi. 543 (1902). Juniperus morrisonicola, Hayata, in Gard. Chron. xliii. 194 (1908), in Journ. Coll. Sei. Tokyo, xxv. art. 19, p. 211, fig. 7 (1908), and xxx. art. i. p. 307 (1911), and in Journ. Linn. Soc. (Bot.) xxxviii. 298 (1908). Sabina squamata, Antoine, Cupress. Gatt. tt. 89, 90 (1860). A shrub, with long decumbent stems, running over and under the surface of the ground, from which arise numerous short erect branches. Young branchlets green, 1 According to Loudon, Trees and Shrubs, 1089 (1842). Seeds were subsequently sent home to Kew by Hooker in 1850. 2 It is known in some nurseries as/, repanda, Hort, ex Carrière, Conif. 27 (1867). 3 Figured by Earl Annesley, Beautiful and Rare Trees, 54 (1903). 1 This is identified with/, squamata by Hooker, Fl. Brit. India, v. 647 (1888). 6 Gordon's account is confused, as he states that the berries are three-seeded ; otherwise his description applies to /. squamata. Juniperus 1421 with three grooves separating the decurrent pulvini of the leaves. Leaves all acicular, densely imbricated in whorls of threes, appressed or slightly spreading, decurrent on the branchlets, broader and shorter than in J. recurva, the free part \ in. long and ^ in. wide, curved, gradually tapering to an acute apex, which is tipped with a sharp cartilaginous point ; ventral surface concave, whitened, usually with a faint or obsolete midrib ; dorsal surface convex, green, with a median furrow extending from the base to near the apex. Older branchlets stout, reddish brown, covered with persistent reddish brown acicular leaves. Fruit ellipsoid, reddish brown at first, turning black when ripe in the second year, smaller and of a different shape from that of J. recurva, about \ to \ in. long, somewhat less in diameter, composed of three or six scales, each with a triangular mucro, umbilicate at the apex. Seed solitary, ovoid, broadest above the base, and tapering to an apiculate apex, nearly filling the cavity of the fruit, with about four ridges running from base to apex, and three or four depressions below the middle for resin-glands. J. squamata, differs mainly from J. recurva in habit and in having stouter broader needles ; but it is readily distinguishable, and has a much wider distribution. It varies considerably in the colour of the leaves, and appears occasionally to become an erect instead of a prostrate shrub. It occurs in Afghanistan, the Himalayas, and the mountains of China and Formosa. It grows at a high elevation in the Himalayas, being most common in the north-west ; but is also found in Sikkim, where Gammie states that it attains 15,000 ft. altitude. Brandis describes it as a gregarious shrub, often covering large areas, either pure or mixed with J. communis, with decumbent stems, at times six inches in diameter, running over the ground and giving off numerous short branches, which make it very difficult to traverse such thickets. It is also found at high elevations in China, in the provinces of Hupeh, Szechwan, and Yunnan. In Hupeh, where I saw it in 1888, it is a shrub about a foot high, usually growing on rocky ground, and spreading over the surface to a radius of six feet or more. It resembles in habit the dwarf form of J. communis, but is readily distinguished by its broader shorter leaves and one-seeded berries. J. squamata grows in Formosa on Mt. Morrison, near the summit at 13,200 ft. altitude. J. squamata was introduced1 into England about 1836, and is occa sionally cultivated in rockeries, being known occasionally as J. pseudosabina? J. densa, etc. In its typical form, it has leaves of a pure green tint, which are occasionally nearly as long as those of J. procumbens, Siebold. There are specimens at Kew, Bicton, Bayfordbury, and Glasnevin. The following is probably a variety of J. squamata ; but in the absence of fruit I hesitate to assign it to that species :— 1 Gordon, in Loudon, Gard. Mag. xvi. 10 (1840), states that it was raised in the Chiswick Garden from Indian seed sent three or four years previous to 1840. 2 J- pseudosabina, Fischer and Meyer, is a Turkestan shrub. Cf. p. 1423, note l. 142.2, The Trees of Great Britain and Ireland Juniperus procumbens, Siebold, in Ann. Soc. Hort. Pays-Bas, 1844, p. 31, and in Siebold and Zuccarini, Fl. Jap. ii. 59, t. 127, fig. iii. (1870) (not Sargent1). Juniperus ehitiensis, vax. procumbens, Endlicher, Syn. Conif. 21 (1847). A prostrate shrub similar to J. squamata in habit, but differing in the branchlets being glaucous-white on the edges of the pulvini : leaves longer, their free part \ in. l°ng. gradually tapering to an acuminate spine-like apex; upper surface concave and covered except along the margins with a white stomatic band, divided except near the apex by an elevated and usually green midrib ; lower surface convex, bluish, spotted with white, and with a median furrow which is variable in length. Fruit not seen. This beautiful shrub was first described by Siebold,2 who stated that it was wild in the mountains of Japan, and was cultivated in gardens and temple woods at Nagasaki. It has been collected since only by Faurie,8 who found it at high elevations in Hondo. Siebold considered it to be perhaps J. nipponica, Maximowicz,4 a species with which it has no affinity ; and subsequent botanists confused it with J. chinensis^ a totally different species. It resembles J. squamata very closely, differing only in the glaucous tint of the leaves and branchlets ; but in the absence of fruit cannot be safely united with that species. J. procumbens is said6 to have been introduced, by living plants, into the Botanic Garden at St. Petersburg in 1864 ; but does not appear to have been known in England until of late years.7 It is now imported largely from Japan, and was a striking feature in the exhibit of Japanese plants at the Anglo-Japanese Exhibition of 1909. It is the most ornamental of the creeping junipers, and is occasionally sold under the erroneous name oïj. litoralis? a totally distinct species. So far as I know it has not yet produced fruit in England. (A. H.) 1 /. procumbens, Sargent, Forest Flora of Japan, 78 (1894), and in Garden and Forest, x. 421 (1897), is a variety of J. douerais, described on p. 1432. 2 One of Siebold's original specimens, a branch without fruit, of /. procumbens is in the Kew herbarium, where it was sent from the Leyden Museum. 3 Masters, in Bull. Herb. Boissier, vi. 2 74 ( 1898), refers two specimens collected by Faurie, ' ' No. 47, summit of Scnnintoge, and No. 3409, summit of Ckokkai," to /. recurva, var. squamata. No. 3409 is in the Kew herbarium, and is identical with J. procumbens, Siebold. 4 y. nipponica, Maximowicz, in Mél. Biol. vi. 374 (1867), is a remarkably distinct species, of which little is known, except the original specimen described by Maximowicz. This species has not been introduced into Europe. 6 Siebold's plant has been much confused with J. chinensis, var. japonica, which is also cultivated in Japan. Gordon's specimen of J. japonica procumbens in the Kew herbarium is J. chinensis. 6 Bretschneider, Hist. Europ. Bot. Disc. China, 610 (1898). 7 The plant at Kew was introduced from Japan in 1893. sy. litoralis, Maximowicz, in Mél. Biol. vi. 375 (1867), is a sea-shore plant, which grows abundantly on the shore of Hakodate Bay in Yezo, and near Honjo on the west side of Hondo, where it was found by J. Veitch in 1892. It also grows in Kiusiu and the Liu Kiu Islands. It has three-seeded berries, and has some affinity with J. rigida. It has never been introduced into England so far as we are aware. Bretschneider, op. cit. 610, referring to it as J. conferta, Parlatore, says that it was introduced into St. Petersburg in 1864, along with J. procumbens, Siebold. Juniperus JUNIPERUS WALLICHIANA Juniperus Wallichlana, J. D. Hooker, ex Pariatore, in De Candolle, Prod. xvi. 2, p. 482 (1868); Brandis, Forest Flora N.-W. India, 537 (1874), and Indian Trees, 695 (1906). Juniperus pseudosabina, J. D. Hooker, Fl. Brit. India, v. 646 (1888) (not Fischer and Meyer1); Kent, Veitch's Man. Conif. 184 (1900); Gamble, Indian Timbers, 698 (1907). A tree, attaining in the Himalayas 60 ft. in height. Leaves dimorphic. Adult foliage with tetragonal ultimate branchlets, about •£$ in. in diameter, densely covered with scale-like leaves, which are arranged in four ranks in decussately opposite pairs, closely appressed, narrowly ovate, about ^ in. long, tapering to an acute apex, bright green with a whitish margin, marked on the back with a linear glandular furrow extending from the base to near the apex. Leaves on the main axes, larger, up to £ in. long, tipped with acuminate points. Juvenile foliage, often preponderant on adult trees ; leaves acicular, in threes, decurrent, densely clothing the branchlet in successive whorls, slightly spreading, about £ in. long, sharply mucronate, whitened on the inner (upper) surface, usually marked on the back with a longi tudinal furrow. Flowers dioecious. Fruit, ripening in the second year, on the ends of short curved branchlets, ovoid, f in. long, ^ in. broad near the base, dark purplish brown, becoming quite blue when ripe ; smooth on the surface except for the minute mucros which indicate the three to five component scales ; depressed at the summit with a minute transverse rhomboidal apiculate umbo. Seed, one in each fruit, large for the genus, 5 in. long, ovoid, compressed, with a narrow thin pointed apex, and two or three depressions for resin-glands about the middle of each surface. This species is a native of the Himalayas from the Indus to Bhutan, occurring between 9000 and 15,000 ft. elevation. In the western part of its range, it is a large gregarious shrub ; but in Sikkim, it becomes a large tree, sometimes 60 ft. in height, with a stout trunk and dark branches and foliage. An illustration of it is given by Hooker,2 who calls it the " Black Juniper." Mr. J. Claude White3 saw a large " weeping cypress," at Chalimaphe in Bhutan, which was 50 ft. round the trunk at five feet from the ground ; and this remarkable tree in all probability was J. Wallichiana. J. Wallichiana was introduced in 1849, when Sir J. D. Hooker sent seeds from India to Kew.4 It is very rare in cultivation, the only specimens which we have seen being one in the Juniper collection at Kew, about 20 ft. high ; and another of the same size at Leonardslee, which bore fruit in 1911. (A. H.) 1 J. pseudosabiaa, Fischer and Meyer, in Index Sem. Hort. Petrop. 65 (1841), and Plant. Schrenk. ii. 13 (1842), diners in appearance from the Himalayan tree, the scale-like leaves being less acute, and the fruits smaller and often globose. It was described from specimens gathered in the Altai and the Tarbagatai mountains in Turkestan. It appears to be a low shrub, like J. Sabina in habit, and has not apparently been introduced. * Him.Joum. ii. 55, fig. (1854). 3 Sikkim and Bhutan, 131 (1909). No specimens of this enormous tree appear to have been collected. 4 See Kew archives, "List of Seeds received from Dr. Hooker during his Travels in India," where "No. 78, 1849 (No. 152), Juniperus, large tree," is evidently J. Wallichiana. VI N 142-4 The Trees of Great Britain and Ireland JUNIPERUS PHŒNICEA Juniperus phœnicea, Linnaeus, Sp. PL 1040 (1753); Loudon, Art. et Frut. Brit. iv. 2501 (1838); Parlatore, in De Candolle, Prod. xvi. 2, p. 486 (1868); Vallot, mjourn. de Bot. ii. 329 (1883); Mathieu, Flore Forestière, 517 (1897); Ascherson and Graebner, Syn. Mitteleurop. FI. i. 250 (1898); De Coincy, in Bull. Soc. Bot. France, xlv. 432 (1898); Kent, Veitch's Man. Conif. 182 (1900); Kirchner and Schröter, Lebenges. Blütenpfl. Mitteleuropas, i. 316 (1906); Albert and Jahandiez, PI. Vase, du Var, 451 (1908). Juniperus Lycia, Linnseus, Sp. PI. 1039 (1753); Loudon, Art. et Frut. Brit. iv. 2502 (1838). Sabina phœnicea and Lycia, Antoine, Cupress. Gattung, tt. 42, 44 (1860). A shrub or tree, attaining usually about 20 ft. in height. Foliage dimorphic. Leaves on young plants, and very rarely on isolated branches of adult trees, acicular, spreading in whorls of threes, not jointed at the base, decurrent on the branchlet, about \ in. long, with two stomatic lines on both the upper and lower surfaces. On adult trees, branchlet systems two- to three-pinnate ; ultimate branchlets terete, about -fa in. in diameter ; leaves scale - like, either in four ranks in opposite decussate pairs, or in six ranks in alternating whorls of threes, closely appressed, ovate-rhombic, about -^ in. long, blunt at the apex, serrulate in margin, rounded on the back, which is often marked with a longitudinal or oval furrow. Flowers usually monoecious, rarely dioecious.1 Fruit very variable in size and shape, ripening in the second year, on short scaly stalks, shining yellow or reddish brown, with remarkably fibrous yellowish flesh ; composed of six to eight scales, with no distinct lines of separation between them, each marked by a minute or obsolete mucro ; in the typical form, globose or sub-globose, £ in. to £ in. in diameter. Seeds variable in number, three to nine, shining brown, separable with great difficulty from the adherent yellow flesh, triquetrous, furrowed longitudinally with two or three depressions for the closely adherent glands. This species is remarkably variable in the size, shape, and colour of the fruits. Five or six varieties can be distinguished in specimens gathered by Jahandiez 2 near Hyères, which were sent to me by Lord Walsingham. The typical form of the species has globose berries ; but these vary in size from \ in. to \ in. in diameter, and in colour from dark reddish brown to yellow or orange brown. i. Var. turbinata, Parlatore, in Fl. Ital. iv. 91 (1867). Juniperus turbinata, Gussone, Fl. Sic. Syn. ii. 634 (1844). Juniperus oophora, Kunze, in Flora, xxix. 637 (1846). Fruit ovoid or shortly conical. Seeds deeply furrowed from base to apex, and compressed at the summit into a sharp transverse edge. This variety is met with in 1 This is De Coincy's opinion. Most authors say that it is usually dioecious and occasionally monoecious. Some of the trees we have seen in cultivation are certainly monoecious. 2 Albert and Jahandiez, PI. Vase, du Var, 451 (1908), state that shrubs with large globose berries grow on the maritime sands, whilst those with small berries occur in rocky situations. De Coincy, in Bull. Soc. Bot. France, xlv. 432 (1898), refers the form with large globose berries to^. Lycia, Linnaeus, which may be named, if considered worth distinguishing, as J. phœnicea, var. Lycia, Loiseleur, Notiv. Duham. vi. 47, t. 17 (1812). Juniperus 14*5 nearly all the localities where the typical form occurs, and, like it, is variable in the size, shape, and colour of the berries. 2. Var. filicaulis, Carrière, Conif. 51 (1855) and Conif. 52 (1867). Juniperus myosuros, Sénéclauze, Catalogue, 1854, p. 35. A shrub with elongated twisted branches and slender pendulous branchlets. Leaves scale-like, as in the type ; but occasional branchlets bear acicular juvenile foliage. The parent plant, which was 3 ft. high in 1867, is said by Carrière to have originated from a seed of /. phœnicea, which was sown by Sénéclauze in his nursery at Bourg-Argental (Loire), sometime before 1854. Sénéclauze, however, in his Catalogue, 1867, p. n, calls this plant J. thurifera hybrida myosuros; and its origin must be considered doubtful. We have seen no specimens. J. phœnicea is widely spread throughout the Mediterranean region, occurring in Spain and Portugal, south-eastern France, Corsica, Sardinia, Italy, Sicily, Dalmatia, Greece, Rhodes, Cyprus, Crete, and the Crimea ; but appears to be unknown in Asia Minor. It is also common in Algeria, Morocco, the Canary and Madeira Islands. It usually grows in arid situations on rocky hills, and often forms extensive and impenetrable thickets, as in La Camargue at the mouth of the Rhone. It ascends in the Riviera to 4500 feet. In Algeria1 it is common on the coast, and on the southern slopes of the mountains of the interior, where it is often the only arborescent vegetation, ascending to 6000 feet. In the Canary Isles, J. phœnicea was formerly one of the characteristic trees of the coast-region between 600 and 2000 ft. elevation ; but has been much destroyed on account of its use for firewood. Dr. Burchard2 states that it is still plentiful on the north side of Gomera and Hierro ; but is nearly extinct on Grand Canary and Teneriffe, where only a few specimens remain in the south. On Gomera, it is usually seen as a globose bush on the cliffs, but becomes a tree when old. On the west point of Hierro, there are specimens supposed to be 1000 years old. Dr. Burchard2 reproduces photographs of two of these remarkable trees, which have short stems, 4 to 5 ft. in diameter, with enormous crowns, spreading for an immense distance on one side of the trunk, as the result of the continuous influence for centuries of the north-east trade-wind. Dr. Grabham tells us that in Madeira, J. phœnicea was formerly widely distri buted from sea-level to the highest summits of the mountains, but is now nearly extinct. The wood of this species is still to be seen in enormous beams and slabs in old buildings, and its fragrant roots of great size are often found underground. According to Aiton, J. phœnicea was first cultivated in Britain in 1683 by James Sutherland, curator of the Edinburgh Botanic Garden. It usually forms a pyramidal shrub or low tree, dense in habit ; but is now rare in cultivation in this country. There are specimens at Highnam, Bicton, and Rostrevor. These bear small globose orange-coloured fruits. It loves a warm climate, and is scarcely hardy in Germany ; but a specimen,8 1 Lefebvre, Les Forêts de l'Algtrie, 431 (1900). 2 In Mitt. Deut. Dend. Ges. 1911, pp. 286, 287. 3 Kirchner and Schröter, op. fit. 316 (1906). 142,6 The Trees of Great Britain and Ireland sheltered by other trees and 6 ft. high, at Tübingen, bore, without any injury but a slight browning of the leaves, a minimum temperature in winter of — 29^° Cent. (A. H.) JUNIPERUS FLACCIDA Juniperusflaccida, Schlechtendal, in Linnœa, xii. 495 (1838); Sargent, Silva N. Amer. x. 83, t. 519 (1896), and Trees N. Amer. 89 (1905)5 Kent, Veitch's Man. Conif. 177 (1900). Juniperus fatida, var. flaccida, Spach, in Ann. Sc. Nat. xvi. 300 (1841). Juniperus gradlis, Koch, Berl. Allg. Gartenzeit. 1858, p. 341 (not Endlicher). Sabina flaccida, Antoine, Cup. Gatt. 37, tt. 49, 50 (1857). A tree, attaining 30 to 40 ft. in height, with brown bark separating into thin scales ; branches widely spreading, with long pendulous branchlets. Leaves dimorphic. Adult foliage : leaves in opposite decussate pairs, slightly spreading, ovate-lanceolate, about ^ in. long, decurrent, ending in a sharp cartilaginous point, rounded on the back, which is marked with a linear sunken gland, often exuding resin. Juvenile foliage, usually on the ends of some branchlets of adult trees, acicular-subulate, spreading, usually in whorls of threes, rarely in pairs, decurrent, about ^ in. long, gradually tapering from the base to the very sharp cartilaginous apex ; upper surface concave, with inflexed margins, and with two narrow stomatic lines ; lower surface marked near the base with a linear gland, often exuding resin. Similar spreading acicular leaves, in pairs or in threes, are borne on the main axes of the branchlet-systems, and like these turn reddish brown in the third and fourth year, and fall in succeeding years, leaving the branchlets smooth with a scaly bark. Flowers monoecious. Fruit, ripening in the second year, on a short (^ in. long) scaly stalk, four- to six-bracteate at the base, sub-globose, about \ in. in diameter, reddish brown with a glaucous bloom, and marked on the surface with a few minute tubercles ; composed of six to eight opposite scales, each indicated by a re flexed triangular mucro. Seeds, six to twelve, several often aborted, embedded in a resinous pulp ; cotyledons two. (A. H.) This species is a native of Texas and Mexico. It is limited in Texas to the Chisos Mountains, where it was discovered in 1888 by Dr. V. Harvard. It is common in north-eastern Mexico, at elevations of 6000 ft. to 8000 ft., on the hills to the east of the tablelands, ranging from Coahuila to Oaxaca, and extending eastward to about a hundred miles from the coast. It was introduced1 in 1838 from Mexico by Hartweg, but is probably too tender for our climate in most parts, as the only specimen which we know of in Britain is a fine tree at Bicton, which I found to be about 40 ft. by 3 ft. 10 in. in 1906. It grows in a sheltered hollow, and bears fruit regularly, which is smaller in size than in native specimens, and contains only imperfect seed. Carrière states2 that it is not hardy at Paris, but he mentions a tree at Angers 1 Loudon, Card. Mag. xv. 241 (1839), and xvi. 10(1840). 2 Conif. 49 (1867). Juniperus 1427 io in. in diameter. It is said 1 to be occasionally cultivated in the south of France and in Algeria; but we have seen no specimens. There is a small tree in the Botanic Garden at Genoa, and a larger one in the Botanic Garden at Naples, which was bearing ripe fruit in March 1910. (H. J. E.) JUNIPERUS THURIFERA Juniperus thurifera, Linnseus, Sf. PI. 1039 (i753); Loudon, Arb. et Frut. Brit. iv. 2503 (1838); Pariatore, in De Candolle, Prod. xvi. 2, p. 487 (1868); Laguna, Fl. Forest. Hispan., i. 103 (1883); De Coincy, in Bull. Soc. Bot. France, xlv. 430 (1898) ; Kent, Veitch's Man. Conif. IQI (1900). Juniperus hispanica, Miller, Gard. Diet. ed. 7, No. 13 (i75?)> and ed. 8, No. 13 (1768). Juniperus sabinoides, Endlicher, Syti. Conif. 23 (1847) (in part). Juniperus cinerea, Carrière, Conif. 35 (1867). A tree, attaining in Spain a height of 30 ft. to 40 ft. Leaves dimorphic. Adult foliage, with flattened branchlet-systems, pinnately divided mostly in one plane. Young branchlets tetragonal, slender, ^ in. in diameter ; leaves in opposite pairs in four ranks, appressed but free at their acuminate apices, ovate, about TV in. long, adnate to the branchlet in their basal half, marked on the back with an oblong glandular depression, minutely denticulate in margin. Juvenile foliage often present on adult trees ; leaves in opposite pairs in four ranks, spreading, acicular, decurrent, T^ to £ in. long, whitened on the upper surface. Flowers dioecious. Fruit on short scaly stalks, ripening in the second year, sub-globose, £ in. or a little more in diameter, dark blue with a slight glaucous bloom when mature ; composed of six scales in opposite decussate pairs, two at the base, the upper four scales meeting at the apex, which is marked with their lines of separation ; each scale with a minute mucro. Seeds, two, three, or four, immersed in a granular sweet fragrant flesh, ovate, triquetrous, £ in. long, shining brown, smooth, narrowed at the apex to a curved point, with two or three oblong resin-pits at the base. i. Var. gallica, De Coincy, in Bull. Soc. Bot. France, xliv. 232 (1897), and xlv. 430 (1898) ; Holmes, Pharmac. Soc. Museum (London) Report, 1907, p. 26. A small tree, apparently differing only from the type, in the one to three seeds, being striate and not smooth on the surface, more prominent at the apex, and less angular. This variety, which is the French form of the species, was first noticed in 1830 by Mutel, who gave it the name of J. Sabina, var. arborea? It appears to be confined to the Dauphiné, where it occurs at a few stations in the valley of the Isère, in the immediate neighbourhood of Grenoble ; and in the valley of the Durance, chiefly near Embrun. Near Grenoble it is found on the 1 Sargent, Silva N. Amer. x. 83 (1896). 2 Mathieu, Flore Forestière, 519 (1897) refers the trees at Saint Crépin to/. Sabina. Cf. also Vidai, in Bull. Soc. Bot. frame, xliv. 51 (1897). 1428 The Trees of Great Britain and Ireland mountains of Comboire, Néron, and Saint Eynard. In the valley of the Durance it is more abundant, and grows at Guillestre, Saint Clément, Saint Crépin, in the valley of Ubaye, near Gap, and at Remollon. M. Ph. Guinier informs us that in all these localities it grows on dry limestone soil, and usually as isolated trees. At Saint Crépin, however, it forms a small wood above the village, at 3500 ft. to 4000 ft. altitude. It is usually a small tree, 20 ft. to 25 ft. high, but in rare cases attains 40 ft. in height. The trunk is short, irregular, and deeply furrowed ; and is frequently 6 ft. to TO ft. in girth—one tree at Saint Crépin being as much as 17 ft. in girth. It attains a great age, a section in the Museum of the Forestry School at Nancy, 0.94 metres in girth, showing 175 annual rings, and another 1.48 metres in circumference showing 169 annual rings. J. thurifera is a native of south-eastern France, Spain, Portugal, Sardinia,1 Morocco, and Algeria. It is common in the mountains of central and southern Spain, occasionally forming pure open woods, one of which in the Sierra de Albarracin is figured by Willkomm ;2 or growing mixed with other conifers, as in the fine forests of Pinus Laricio in the Serrania de Cuença. Here,2 on the Muela de S. Juan, near Tragacete, it attains 25 ft. to 35 ft. high, and 10 ft. to 13 ft. girth.3 Laguna, who gives many localities for this species in Spain, states that it never ascends to the high altitudes occupied by J. Sabina, nor descends to the hot and sandy plains, where J. phœnicea is often seen. It inhabits the zone between 2500 and 3500 ft. altitude, where there are abrupt changes of temperature, which it supports well. It is always met with on soils which are either pure limestone, or contain lime in considerable quantity. In Algeria,4 J. thurifera grows mainly in the cedar forests at high elevations, where it is usually a small tree, not exceeding 20 ft. in height. It was collected in southern Morocco by Sir J. D. Hooker. (A. H.) Although this species was cultivated by Miller in 1752, it has never become common, and according to Kent only thrives in warm and sheltered situations. Loudon records a tree at Boyton 28 ft. high in 1837, and another at Croome, forty years planted, which was 30 ft. high in 1838, but we have not found these specimens now living. The largest tree in England is one at Bicton, about 40 ft. high, 4 ft. 4 in. in girth, conical in shape, and bearing male flowers in April 1911. There are two trees at Kew, about 30 ft. high, which were planted in 1870. Another at Bayfordbury, 30 ft. high, was planted in 1841. Smaller trees exist at Highnam and Leonardslee. I saw a tree bearing ripe fruit at Simon-Louis's nursery, Metz, which was about 40 ft. by 3 ft. in 1908. (H. J. E.) 1 Grisebach, Veg. der Erde, i. 572 (1872), states that it occurs in Sardinia. Cf. Nyman, Consp. Fl. Eur. iii. 676 (1881). 2 Willkomm, Pflanztnvcrb. Iber. Halbinsel, 160, 185, fig. II (1896). 3 Dillwyn, Hort. Collinson. 27 (1843), quotes a letter written to Collinson in 1766 by Bowles, an engineer in Spain, who states that large trees, girthing 14 ft., with wide-spreading branches like a beech, grew in the mountains near the source of the Tagus. Willkomm confirms this. 4 Lefebvre, Forêts de l'Algérie, 431 (1900). Juniperus 1429 JUNIPERUS PACHYPHL^A Juniperus fachyphfaa, Torrey, Pacific R. J?. Rep. iv. pt. v. 142 (1858) ; Sargent,Suva JV. Amer. x. 85, t. 520 (1896), and Trees N. Atner. 90 (1905) ; Kent, Veitch's Man. Conif. 181 (1900) ; Britton and Shafer, North American Trees, 113 (1908). Juniperusplochyderma, Pariatore, in De Candolle, Prod. xvi. 2, p. 492 (1868). Sabina pachyphlœa and plochyderma, Antoine, Cupress. Gatt. 39, 40, t. 52 (1857). A tree, attaining in America 60 ft. in height and 15 ft. in girth. Bark, different from that of all the other junipers, f to 4 inches thick, deeply divided into small square scaly plates. Branchlets slender, angled, becoming light brown, terete, and scaly after the fall of the leaves. Leaves dimorphic : on vigorous branchlets, acicular, spreading, in threes and in opposite pairs, \ to \ in. long, tipped with slender elongated cartilaginous points ; upper surface concave and whitened, lower surface greyish green and keeled. The juvenile foliage gradually passes into the adult foliage ; ultimate branchlets tetragonal, £s in. in diameter, with scale-like leaves in opposite pairs, imbricated, closely appressed, about ^ff in. long, ovate - rhombic, rounded at the narrowed apex, minutely toothed in margin, convex on the back, which is marked with a depressed oval gland, often exuding resin ; leaves on the older branchlets tipped with a sharp point. Flowers dioecious. Fruit ripening in the second year, sub-globose, nearly J in. in diameter, sub-sessile, ebracteate, reddish brown covered with a glaucous bloom, tuberculate on the surface, with six to eight scales each marked by a slightly reflexed mucro. Seeds four, nearly filling up the cavity of the fruit, ovoid, angled, shining brown ; flesh scanty, fibrous, yellow. Reputed juvenile forms of this species, vars. conspicua, elegantissima, and ericoides, differing in habit and with blue or whitish-blue foliage, have lately been obtained by Barbier ' at Orleans ; and are now in cultivation at Kew2 and Glasnevin. This species grows on dry arid mountain slopes, at 4000 to 6000 feet elevation, from the Eagle and Limpio Mountains in south-western Texas, westward along the desert ranges of New Mexico and Arizona, and southwards into Mexico, where it occurs along the Sierra Madre to the state of Jalisco and over the mountains of northern Sonora and Chihuahua. It was discovered in 1851 by Dr. S. W. VVoodhouse in eastern New Mexico, and is considered by Sargent to be the most beautiful of all the west American Junipers, its thick checkered bark being unlike that of any other species. It is uncertain when it was introduced into England. It is extremely rare, the only specimen which we have seen being a tree in Kew Gardens, about 20 ft. high, showing the peculiar bark, and producing on its stem several epicormic branches. This has not as yet produced flowers. (A. H.) 1 Mitt. Deut. Vend. Cts. 1910, pp. 139 and 289. * Kew Bulletin, 1911, p. 101. 1430 The Trees of Great Britain and Ireland JUNIPERUS CHINENSIS funiperus chinensis, Linnœus, Mantissa, \. 127 (1767) ; Loudon, Arb. et Frut. Brit. iv. 2505 (1838) ; Siebold et Zuccarini, Fl. Jap. 58, tt. 126, 127, (1844); Pariatore, in De Candolle, Prod. xvi. 2, p. 487 (1868); Franchet et Savatier, Enum. PI. Jap. i. 472 (1875); Masters, mjourti. Linn. Soc. (Bot.) xviii. 497 (1881), and xxvi. 541 (1902), and in Journ. Bot. xli. 268 (1903); Beissner, in Mitt. Deut. Dend. Ges. 1896, p. 69, and 1898, p. 32, and in Bull. Soc. Bot. Ital. Firenze, 1898, p. 167 ; Kent, Veitch's Man. Conif. 169 (1900) ; Shirasawa, Icon. Ess. Forest. Japon, i. text 29, t. 12, figs. 14-27 (1899); Diels, in Engler, Bot. Jahrb. xxix. 220 (1901). Juniperus cernua and dimorpha, Roxburgh, Fl. Ind. iii. 839 (1832). Sabina chinensis, Antoine, Cupress. Gattung. 54, t. 75 (1857). Sabina Cabiancœ, Antoine, Cupress. Gattung. 41, t. 54 (1857). A tree, attaining in China and Japan a height of 6o ft. Leaves of two kinds : on adult trees scale-like ; ultimate branchlets ^s in- m diameter, clothed with four ranks of leaves in opposite pairs, which are imbricated, closely appressed, narrowly rhombic, ^ in. long, tapering to rather an obtuse apex, adnate to the stem, entire in margin; outer surface convex, green with a pale margin, and marked with a depressed oval or oblong gland ; interiorly concave, with a raised narrow midrib, glaucous. On older branchlets the scale leaves are larger, about ^ in. long, con spicuously glandular on the back, persistent four or five years. On young trees and on occasional branches of old trees, the juvenile foliage is linear-acicular, J in. long, spreading, either in whorls of threes or in opposite pairs, tipped with a rigid spine- like point, adnate to the branchlets, swollen on the upper surface near the base, but not jointed ; concave above, with a green midrib and two glaucous bands ; green and convex beneath. Flowers dioecious. Staminate flowers bright yellow, very numerous.1 Fruit ripening in the second year, borne on the ends of short branchlets, which are covered with ordinary scale-leaves ; brown covered with a thick white mealy bloom ; variable in shape, commonly sub-globose, but widest and usually depressed at the summit, averaging J in. in diameter, composed of four to eight scales. Seeds two or three, rarely four or five, immersed in a resinous pulp, shining deep chestnut brown, smooth, broadly ovoid, with a wide base, gradually tapering to a sharp thin- edged apiculate apex, compressed from before backwards, each surface convex, with a longitudinal groove near the thinner outer edges. This species is readily distinguishable by the pale margins of the scale-like leaves, which mark the ultimate branchlets with a series of white crosses. In nearly all adult trees, acicular foliage with the leaves either ternate or in opposite pairs can be found on some of the branches. VARIETIES I. This species is very variable in habit in the wild state ; and, as Beissner 2 points out, in the mountains of Shensi in China, both male and female trees exist, 1 On certain trees at Kew, and in wild specimens of Shensi (ßde Beissner) staminate flowers are borne on branchlets with acicular as well as with scale-like foliage ; and this seems peculiar to/, chinensis. 2 In Mitt. Dent. Dend. Ces. 1896, p. 69, and 1898, p. 32. Juniperus I431 which bear exclusively acicular foliage ; whilst others occur in both sexes with the leaves mostly scale-like. There are no grounds for supposing that the sexes are distinguished in nature by any peculiar habit ; but in cultivation, owing to long- continued propagation by cuttings from trees of different habit, many female trees differ in appearance from that commonly met with in male trees. This is by no means universal, as there are two trees of the same habit, but of different sexes, in the Cambridge Botanic Garden. A common staminate form, with preponderating acicular foliage, and dense branches, forming a conical pyramid, was formerly dis tinguished as/, struthiacea, Knight, Syn. Conif. 12 (1850). A pistillate form, known at first as/, flagelliformis, Loudon, Trees and Shrubs, 1090 (1842), was introduced from Canton in 1839 by J. Russell Reeves, and was subsequently named J. Reevesiana, Knight, Syn. Conif. 12 (1850). II. The following are either closely allied species or varieties of J. chinensis :— i. Jtmiperus sphœrica, Lindley, in Lindley and Paxton, Flower Garden, i. 58, fig- 35 (1850). Juniperus chinensis, Linnseus, var. Smithii, Gordon, Pinetum, 119 (1858) (not Loudon1). Juniperus Fortunii, Van Houtte, ex Gordon, Pinetum, 119 (1858). A tree, 30 to 40 ft. high, discovered by Fortune 2 in the hills north-west of Ningpo and near Shanghai, where it is frequently planted around graves. The type specimen, preserved in the British Museum, does not differ from J. chinensis in the foliage, which is all scale-like, no acicular leaves being present ; but is monoecious, and bears fruit, quite spherical in shape and larger than that oij. chinensis, -fa in. in diameter, smooth, dark purple, scarcely glaucous, containing five seeds, which are larger than, but similar in shape and colour to those of J. chinensis. This is kept separate from /. chinensis, but with some doubt, by Parlatore,3 Kent,4 and Masters ;5 and is probably only a variety of that species, differing mainly in the larger spherical fruit, not covered with a whitish bloom, and containing numerous seeds. The branch collected by Fortune is monoecious ; but this is perhaps an abnormality. A specimen (No. 6576) which I collected in Fang district in the province of Hupeh, with large spherical glaucous berries, resembles Fortune's plant, but is dioecious and with only three seeds in each fruit. Wilson found in the same province another specimen with smaller four-seeded fruits.6 Fortune sent seeds in 1850 to Standish and Noble, who probably raised/. sphœrica in their nursery ; but I have found no living specimens, either monoecious or with the large spherical berries of Fortune's plant. The trees now known in cultivation either as J. sphœrica or J. sphœrica Sheppardi, Veitch, Man. Conif. 290 (1881), usually prove to be female trees of/, chinensis, with a rather spreading habit. 1/. chinensis Smithii, Loudon, Arb. et Frut. Brit. iv. 2505 (1838), described (long before Fortune's discovery of /. sphœrica) as monœcious with angular fruits, was supposed to be of Nepalese origin, and may have been/, religiosa. 2 Residence amongst the Chinese, 63, 140(1857). 3 In De Candolle, Prod. xvi. 2, p. 488 (1868). 4 Veitch's Man. Conif. 190 (1900). 6 \\ijourn. Linn. Soc. (Bot.) xxvi. 543 (1902). 6 The fruits of cultivated trees of/, chinensis are usually three-seeded ; but occasionally four or five seeds are present, the fruits in this case being small, covered with whitish bloom, and depressed at the apex, and not in the least like the large spherical bluish fruits of /. sphœrica. O VI 1432 The Trees of Great Britain and Ireland 2. Var. Sargenti, Henry (var. nova}. Juniperus procumbens, Sargent, Forest Flora Japan, 78 (1894), and in Garden and Forest, x. 421 (1897) (not Siebold). A sea-shore plant,1 forming dense mats, and sending out for long distances prostrate creeping stems, which bear foliage similar to that of J. chinensis : branchlets tetragonal, covered with minute scale-like appressed leaves, furrowed on the back ; no acicular leaves being present on adult plants. Berries bluish, covered with a slight glaucous bloom ; seeds three, like those of/, chinensis. This is said by Sargent, who has kindly sent a dried specimen, to grow on the coasts of Korea and Japan, on low grassy bluffs freely exposed to the ocean gales. A few plants were raised in the Arnold Arboretum from seeds gathered in 1892 near the Aino village of Horobetsu on the coast of Yezo. I have seen no living specimens. III. The following varieties are of horticultural origin :— 3. Var. albo-variegata, Veitch, Man. Conif. 288 (1881). A compact shrub, differing from the type in many of the branchlets being creamy white at the tips. It usually bears adult scale-like foliage ; but a form with acicular foliage is also in cultivation. It is said to have been first introduced from Japan by Fortune, and subsequently by J. Gould Veitch. It is known by several names, as var. variegata, Fortune ; var. argentea, Gordon ; and var. argenteo-variegata, Rehder. 4. Var. aurea, Young, ex Gard. Chron. 1872, pp. 8, 1193. An upright form, with adult scale-like foliage, having the whole of the young growth suffused with a deep golden yellow, which gradually turns green in the summer. The colour is heightened by exposure to the sun. This originated in Maurice Young's nursery at Milford, Godalming, where the original plant was 12 ft. high in 1872, and when Elwes saw it in 1909 was still a small tree. The best specimens we know of this are at Burnham Park, the residence of Sir Harry J. Veitch, and are about 18 ft. high. 5. Var. japonica aurea, Masters, \njourn. R. Hort. Soc. xiv. 211 (1892). Juntperus japonica aurea, Carrière, Conif. 32 (1867). A straggling shrub, with long decumbent branches ; branchlets tinged with golden yellow. This is said to have been first introduced by Fortune from Japan, and subsequently by J. Gould Veitch. 6. Var. japonica aureo-variegata, Masters, injourn. R. Hort. Soc. xiv. 211 (1892). Juniperus japonica variegata, Carrière, Conif. 31 (1867). A dense dwarf shrub, with many of the branchlets of a deep golden yellow. Also of Japanese origin. 7. Var. Pßtzeriana, Späth, Catalogue, No. 104, p. 142 (1899). A broad pyramidal shrub, with dense horizontal branches, and long and slightly pendulous branchlets, clothed with glaucous foliage. This originated in Späth's nursery at Berlin, where the original plant was 10 ft. high in 1901. 8. Var. japonica, Vilmorin, in Hortus Vilmorin. 58 (1906). Juniperus japonica, Carrière, Conif. 31 (1855). 1 y. Thuntergii, Hooker and Arnott, Bot. Beechey's Voyage, 271 (1841), gathered in the Liu Kiu Islands, is represented at Kew by a specimen with acicular leaves, which bears no fruit. It cannot be identified with certainty ; but may be a form ofj. chinensis. Juniperus A diffuse bushy plant, with mostly juvenile acicular foliage in threes ; some of the terminal branchlets being covered with adult scale-like leaves, and occasionally bearing fruit.1 This is a juvenile form of/, chinensis, of which the Japanese make dwarf plants, that are frequently imported into Europe. It was erroneously identified by Carrière with/.procumbens^ Siebold; and this mistake has been copied by Kent and other writers. Sargent describes,3 as J. japonica, a plant of compact habit, with many erect branches and acicular bluish-green needles, which is often cultivated in Japanese gardens, and is very hardy and distinct in appearance. It retains its peculiar compact juvenile habit for several years, but often becomes thin and ragged before it is 12 ft. high, and loses its value as an ornamental plant. Sargent adds that this is one of the most difficult conifers to transplant. Judging from the description, the variety alluded to by Sargent \s J. japonica pyramidalis, Carrière, Conif. 32 (1867), but I have seen no specimen. (A. H.) DISTRIBUTION J. chinensis is a native of China, Mongolia, and Japan. In China, it is frequently cultivated in temple grounds ; but appears to be truly wild in the mountains of Hupeh, Shensi, and Szechwan, where it is usually found growing solitary on cliffs, but occasionally as underwood 4 in the forests. Père David5 found it abundant on the Moni-ula range of the Ourato territory in south-western Mongolia. This is a favourite tree in the parks and temples of Peking, where it attains a great age. The largest I saw were at the temple of Confucius, and were said to be over 700 years old. In a double avenue here, one tree on the left-hand side was about 40 ft. by 17 ft.; and another whose trunk was covered with burrs was 14 ft. in girth. At the Ming Tombs there are many very old junipers and fine specimens of arbor vitse, together with numerous pine trees (Pinus Bungeana and P. funebris). In Japan, it is also common in cultivation ; but is recognised as a native tree by Japanese botanists6 and foresters. Shirasawa states that it is wild in the mountains of the Shinano province in central Hondo, mixed with Pinus densiflora and Quercus serrata, and forming a tree 30 to 40 ft. in height, with straggling contorted branches and greyish green foliage. It occasionally attains a large size, as Sargent7 mentions two venerable trees at the temple of Zenkogi in Nagano,8 which are 70 to 80 ft. high with hollow trunks about 6 ft. in diameter. I never saw it wild ; but I saw several handsome trees in the ancient temple of Tennoji at Osaka, with fine large round heads, the best with a trunk 10 ft. high and 10 ft. in girth, and with a spread of branches of about 14 yards. One had a very twisted and fluted stem. This species is known to the Japanese as Bya Kiishin. 1 This isy. japonica, Carrière, and what is cultivated under that name in Veitch's nursery at Coombe Wood. 2 J- procumbens, Siebold, is a very distinct species. Cf. p. 1422. :1 In Garden and Forest, x. 421 (1897). * Diels, in Engler, Bot. Jahrb. xxix. 220 (1901). 6 Franchet, PI. David. \. 291 (1884), describes, as var. pendula, a form with elongated pendulous branches, found by Père David in Shensi. 6 Matsumura, Index PI. Jap. 10 (1905), gives as localities, Kunasiri in the Kuriles, Rebunsiri in Yezo, Hakoda in Nippon, and the Liu Kiu Islands. 7 Forest Flora of Japan, 78 (1894). 8 Shirasawa states "the temple of Keiichoji in the province of Sagaini." 1434 The Trees of Great Britain and Ireland This tree was first described in 1767 by Linnaeus, who states that it was then cultivated at Upsala. It appears, however, to have been first introduced into England by William Kerr, who sent plants from Canton to Kew in 1804. Next to the Virginian juniper, it is the species now most commonly cultivated in nurseries and private gardens. It is absolutely hardy everywhere, seems quite indifferent to soil, and in many places is a very ornamental shrub or small tree. It ripens seed, which, so far as my observations go, germinate the year after they are sown. The finest trees we have seen are :—At Arley Castle, several old specimens, the largest of which was 48 ft. by 5 ft. 3 in. in 1907 ; at Eastnor Castle, a well-shaped tree, 48 ft. by 3 ft. which was bearing fruit in 1908; at Hardwicke, near Bury St. Edmunds, a fine tree, 38 ft. by 3 ft. 10 in., with abundant ripe berries in 1905 ; at Redleaf in Kent, a tree about 35 ft. high in 1907; at Westonbirt, a tree 32 ft. high and growing fast, which in 1909 was covered with fruit. A tree growing at Rood Ashton, Wilts, about 25 ft. high, wide-spreading in habit, was figured in Gardeners' Chronicle, xlii. 163, fig. 63 (1907). (H. J. E.) JUNIPERUS BERMUDIANA Juniperus bermudiana, Linnasus, Sp. PL 1039 (1753); Loudon, Arb. et Frut. Brit. iv. 2498 (1838); W. J. Hooker, in London Journ. Bot. ii. 141, t. i (1843); Endlicher, Syn. Conif, 29 (1847); Pariatore, in De Candolle, Prod. xvi. 2, p. 490 (1868) (in part); J. M. Jones, Botany of Bermuda, 272 (1873); Hemsley, in Gard. Chron. xix. 656, figs. 105, 106 (1883), in Journ. Bot. xxi. 259 (1883), and in Voy. Challenger, Bot. i. 81, t. 5 (1885); Sargent, in Garden and Forest, iv. 289, figs. 51, 52 (1891)3 Masters, in Journ. Bot. xxxvii. i-n (1899); Kent, Veitch's Man. Conif. 166 (1900). Junipet-us oppositifolia, Moench, Meth. 698 (1794). Juniperuspyramidalis, Salisbury, Prod. 397 (1796). A tree, attaining 50 ft. in height in the Bermudas, with dark red bark and spreading branches. Foliage of two kinds: on adult trees scale-like; ultimate branchlets tetragonal, about ^0 in. in diameter, densely covered with imbricated leaves, which are usually in four ranks, about r]2- in. long, ovate, obtuse at the narrow incurved apex, greyish green or glaucous on the back, which is usually marked with a longitudinal furrow, entire in margin ; on older branchlets, in four ranks or ternate in six ranks, those on the main axes always ternate, up to \ in. long, and becoming acuminate at the apex. Juvenile foliage, occasionally present on some branches on old trees, in alternate whorls of threes, about \ in. long, acicular, slightly spreading ; upper surface whitened with a raised midrib ; lower surface greyish green, very convex, and marked with a longitudinal furrow. Flowers dioecious.1 Fruit ripening in the first year, sub-globose, about \ in. in diameter, dark brown, covered more or less with a bluish bloom, with six to eight scales, each marked by a depression with a minute mucro. Seeds, two to three, immersed in fleshy pulp, shining chestnut brown, ovoid, oblique at the broad base, tapering to the thin-edged apex, with two furrows on the outer surface. 1 Stewardson Brown, in frac. Acad. Nat. Sc. Philadelphia, Ixi. 488 (1909), says that the tree is dicecious : " At the time of flowering in March and April the staminate trees are a golden colour, presenting a strong contrast with the rich blue-green of the pistillate tree." Most authorities say it is monoecious, but specimens with fruit in the British Museum bear no male flowers. Juniperus 1435 This species is confinedl to the Bermuda Islands, where it is the only indigenous exogenous tree. It was formerly abundant2 on the islands, thriving both on the dry limestone hills and in the brackish swamps. The trees grow to a large size in the salt-water marshes, and have much darker heartwood than those on the hills, but not nearly so durable.3 Large trees are no longer common. Sargent4 gives an illustra tion of one in the churchyard of Devonshire parish, which was about 50 ft. high and 15 ft. in girth, and states that only two larger trees were known to exist. Another illustration shows the habit of this species in the Devonshire marshes. The wood was formerly much used in shipbuilding and in making beautiful furniture. " Cedar" chests and cabinets over two hundred years old are preserved as heirlooms by the descendants of the old Bermuda families, who live in houses finished with this wood, which becomes with age a rich dark colour like mahogany. The Bermuda juniper was cultivated6 in England as early as 1684, but it is not hardy in the climate of London. Knight and Perry6 state that it was hardy in their day in Devonshire, and that plants remained uninjured in the open air during the winter of 1849 in Oxfordshire. We have seen, however, no living specimens in England except a shrub at Bicton, about 2 ft. high, which is not thriving, and small plants which were received at Kew7 in 1910. Reputed specimens of this species at Castlewellan,8 as well as some plants that were formerly cultivated at Kew under the namey. bermudiana, turned out to be Ctipressus fuitebris. This species is cultivated in the south of France, Italy, and the Canary Isles. There is a good specimen, which bears fruit regularly, in Dr. Perez' garden at Orotava. Dr. Perez,9 as the result of numerous experiments, finds that seeds of this species germinate speedily when immersed in boiling water for three, six, or ten seconds, and at the end of the time are plunged into water at 65° to cool. Longer exposure to boiling water destroys the embryos. (A. H.) JUNI PERU S VIRGIN I AN A, PENCIL CEDAR Juniperus virginiana, Linnasus, Sp. PI. 1039 (1753); Loudon, Arb. et Frut. Brit. iv. 2495 (1838); Pariatore, in De Candolle, Prod. xvi. 2, p. 488 (1868); Sargent, Silva IV. Amer. x. 93 (in part), t. 524 (1896), and Trees N. Amer. 94 (1905); Kent, Veitch's Man. Conif. 192 (1900); Mohr, U.K. Forestry Bull. No. 31 (1901); Pinchot, U.S. Forestry Circ. No. 73 (1907); White, U.S. Forestry Circ. No. 102 (1907); Clinton-Baker, Illust. Conif. ii. t. 74, fig. 4 (1909). Juniperus caroliniana, Miller, Diet. ed. 8, No. 4 (1768). Juniperus arborescens, Moench, Meth. 699 (1794). Juniperus fragrans, Salisbury, Prod. 397 (1796). Sabina virginiana, Antoine, Cupress. Gattung. 61, tt. 83, 84 (1857). A tree, attaining in North America 100 ft. in height and 12 ft. in girth, often 1 Kent states that pieces of its wood were found 50 ft. below low-water mark, during dredging operations undertaken for the construction of a dock. 2 J. M. Jones, Botany of Bermuda, 272 (1873), states that the trees are becoming extinct, no longer growing in the salt-marshes ; but Sargent's later account does not confirm this. 3 A. Haycock, in Gard. Chron. xxv. 176 (1899). Capt. L. Clinton-Baker informs us that the best trees in 1911 averaged 40 to 50 ft. in height and 9 ft. in girth. 4 Garden and Forest, iv. 289, figs. 51, 52 (1891). 6 For the early history of this species, see Hemsley's account, cited at the head of this article. 6 Syn. Conif. u (1850). 7 A tree in the Temperate House at Kew, which was about thirty years old, was cut down in 1905. 8 List of Plants Hardy at Castlfivellan, 65 (1897). ° In Card. Chron. 1. 127 (1911). ( LIBRA! •fio 1436 The Trees of Great Britain and Ireland with the trunk fluted at the base. Bark, % in. thick, reddish brown, shredding off in long strips. Leaves of two kinds : on adult trees scale-like ; ultimate branchlets very slender, ^ in. in diameter, clothed with four ranks of leaves in opposite pairs, which are imbricated, appressed but free towards the apex, ovate, acute or acuminate with a short point, ^ in. long, adnate, entire in margin ; green and glabrous on the back, which is often marked with a small oval glandular depression. On older branchlets the scale-leaves are broadly ovate, acute, larger, about T\ in. long, and become brown and withered, ultimately disappearing in the fifth or sixth year. Acicular leaves often present on occasional branches of adult trees, spreading in pairs, £ to £ in. long, ending in a spine-tipped apex, adnate, swollen at the base, but not jointed ; concave and glaucous above ; green and convex beneath. The acicular leaves on seedlings and young plants are arranged in alternate whorls of threes. Flowers usually dioecious, rarely monoecious. Fruit ripening in one year, borne at the ends of short branchlets, which are furnished with ordinary adult scale-leaves ; sub-globose, but usually longer than broad, about ^ in. long, bluish, covered with a glaucous bloom, composed of four or six scales. Seeds, one or two, immersed in a resinous flesh, ovoid, broad at the base, tapering towards the apex, smooth, shining chestnut brown, with two indistinct longitudinal furrows, and with or without resin- pits at the base. Seedling ' with two ligulate cotyledons, which have no resin-canals ; primary needles with one resin-canal immediately below the dorsal ridge. ALLIED SPECIES AND VARIETIES This species was formerly supposed to spread over the greater part of the North American continent ; but the geographical forms of the west and south are now considered by American botanists to be two distinct species. I. Jimiperus scopulorum, Sargent, in Garden and Forest, x. 420, fig. 54 (1897), Suva N. Amer. xiv. 93, t. 739 (1902), and Trees N. Amer. 96 (1905). Juniperus excelsa, Pursh, FI. Amer. Sept. ii. 647 (1814) (not Bieberstein). Juniperus dealbata, Loudon, in Gard. Mag. xvi. 639 (1840), and Trees and Shrubs, 1090 (1842) (not Douglas2). (?) Juniperus fragrans, Knight and Perry, Syn. Conif. 13 (1850); Carrière, Conif. 57 (1855). Juniperus bacciformis, Carrière, Conif. 56 (1855). Juniperus Henryana, Brown (Campst.), in Trans. Bot. Soc. Edin. ix. 377 (1868), and in Gard. Chron. 1873, p. 8. Juniperus virginiana, Linnseus, var. scopulorum, M. E. Jones, in Bull. Univ. Montana, Biol. Ser. No. 13, p. 12 (1910). Sabina scopulorum, Rydberg, Flora of Colorado, 10 (1906). A tree about 40 ft. high and 9 ft. in girth, often divided near the base into several stems. Adult foliage like that of /. virginiana, but with a disagreeable pungent smell, and with stouter branchlets and leaves marked on the back by a conspicuous glandular pit. Fruit ripening in the second year, globose, \ in. in diameter, bright blue covered with a glaucous bloom. Seeds, two, triquetrous, reddish brown, prominently angled, and with one longitudinal groove. This tree grows on dry rocky ridges and, except near the coast, usually at over 1 Hill and De Fraine, in Ann. Bot. xlii. 696 (1908). 2 Gordon, in Card. Chron. 1842, p. 562, states that/, dealbata, Douglas, is the same as/. «a«a,Willdanow. Cf. p. 1401, note I. Juniperus *437 5000 feet elevation, from the eastern foothills of the Rocky Mountains from Alberta to Texas, westward to the coast of British Columbia and Washington, and to eastern Oregon, Nevada, and northern Arizona. This species was introduced into England from north-western America in 1839, under the name J. dealbata, Loudon; and was erroneously identified withy, occidentalis, Hooker, a species that has apparently never been in cultivation in this country. Soon after its introduction, J. scopulomm seems to have been known as J. fragrans, a name which still exists in nursery catalogues. It is a rare tree ; but there is a specimen at Kew, not very thriving and about 15 ft. high, which is labelled /. occidentalis fragrans. We obtained in 1911 a fruiting branch from a small plant, named J. fragrans, in Dicksons' nursery, Chester. It is cultivated in Germany,1 at Darmstadt and at Tübingen, where it has borne a temperature of — 24° Cent. \\.Junipems barbadensis, Linnseus, Sp. PI. 1039 (1753); Loudon, Arb. et Frut. Brit. 2504 (1838); Mohr, U.S. Forestry Bull. No. 31, p. 37, plate ii. (1901); Sargent, Silva N. Amer. xiv. 89, t. 738 (1902), and Trees N. Amer. 95 (1905). Juniperus virginiana, Linnseus, var. australis, Endlicher, Syn. Conif. 28 (1847). Juniperus virginiana barbadensis, Gordon, Pinetum, 114 (1858) (in part). Juniperus bermudiana, Lunan, Hort. Jamaic. i. 84 (1814) (not Linnseus). Sabina barbadensis, Small, Flora S.E. United States, 33 (1903). A tree, attaining 50 ft. in height and 6 ft. in girth ; branches and branchlets pendulous. Adult foliage similar to that of J. virginiana, but branchlets more slender, and leaves smaller, acuminate, and conspicuously marked on the back by an oblong or linear oil-gland. Flowers dioecious. Fruit ripening in the first year, sub-globose, |- to £ in. in diameter, bluish with a glaucous bloom ; seeds, one or two, ovoid, pointed, ridged. This species occurs in inundated coastal river swamps from southern Georgia southward to the Indian River, Florida ; and on the west coast of Florida from Charlotte Harbour to the Appalachicola River, often forming thickets under the shade of larger trees. It is often planted in the cities and towns near the coast from Florida to western Louisiana, and is now said to be naturalised on the Gulf Coast. This species also occurs in the West Indies, in San Domingo, the Bahamas, Antigua, St. Lucia,2 and the mountains of Jamaica. It appears to be now extinct in Barbadoes.2 The " red cedar " which grows on the northern Bahama Islands is usually referred to this species, but it has lately been separated3 as Juniperus lucayana, Britton, on account of its depressed globose and somewhat laterally flattened fruit. Little is left of it on account of its use formerly for construction and in more recent times for making pencils. The juniper of eastern Cuba is closely related to, if not identical with, that of the Bahamas.8 III. Junipems Bedfordiana, Loudon, Trees and Shrubs, 1090(1842). Juniperus gracilis, Endlicher, Syn. Conif. 31 (1847). Juniperus virginiana, Linnseus, var. Bedfordiana, Knight, Syn. Conif. 12 (1850); Pariatore, in De Candolle, Prod. xvi. 2, p. 489 (1868); Veitch, Man. Conif. 284 (1881); Kent, Veitch's Man. Conif. 193 (1900). Juniperus virginiana, Linnseus, var. gracilis, Sargent, in Silva N. Amer. x. 96, note i (1896). 1 Mitt. Dent. Dend. Ces. 1906, p. 37, and 1908, p. 144. 2 Cf. Stapf, in Kew Bull. 1911, p. 377. 3 Britton and Shafer, North American Trees, 121 (1908). 1438 The Trees of Great Britain and Ireland A dense low tree, columnar in habit, with slender elongated pendulous branchlets; leaves bright green, acicular, like the juvenile foliage of J. virginiana. This hand some tree was first mentioned by Loudon, who merely states that it closely resembles J. virginiana. Its origin is unknown,1 but it is usually considered to be identical with/, barbadensis, and the fact that it is rather tender in England supports this view. As it has only juvenile foliage, and apparently never bears fruit, the question cannot be decided. IV. Many varieties of J. virginiana, which are always propagated by cuttings, have arisen in nurseries, no less than twenty-one varieties being enumerated in the Kew Hand-List. According to Loudon, it varies much when raised from seed, as at White Knights, where there were hundreds of trees in 1838, differing much in appearance. Some were low and spreading, others were tall and fastigiate, and some had pendulous branches. The foliage varied much in colour, being light green, dark green, or glaucous. The fruit also differed in size. The most important varieties are :— 1. Var. pendula, Knight, Syn. Conif. 12 (1850). This exists in at least three forms :— (a) Branches spreading, branchlets pendulous, leaves scale-like, bearing staminate flowers. (b) Var. pendula viridis. Branches and branchlets pendulous ; leaves scale-like, bright green. (c) Var. Ckamberlainii, Knight, loc. cit. Branches spreading and reflexed ; branchlets pendulous ; leaves mostly acicular, of a greyish tint. 2. \3x.pyramidalis, Carrière, Conif. 47 (1867). Columnar in habit, with either glaucous or bright green foliage. 3. Var. dumosa, Carrière, Conif. 45 (1855). A dense rounded shrub, with both acicular and scale-like leaves. 4. Var. Schottii, Gordon, Pinetum, 157 (1875). A narrow pyramidal tree, dense in habit, with bright green scale-like leaves. 5. Var. tripartita, R. Smith, ^Gordon, Pinetum, 157 (1875). A low spreading bush with dense branches, and bright green, usually acicular foliage. 6. Var. glauca, Knight, Syn. Conif. 12(1850). Differs from the type in its beautiful glaucous silvery foliage, which is most pronounced in spring. 7. Var. Triomphe d'Angers, ßeissner, Nadelholzkunde, 127 (1891). Young branchlets creamy white, contrasting well with the dark bluish-green of the older foliage. 8. There are several variegated varieties, as alba variegata and atirea variegata, Gordon, Pinetum, 157 (1875), which are usually poor in colour. 9. Var. elegantissima, Beissner, Nadelholzkunde, 128 (1891). A pyramidal tree, with the tips of the young branchlets golden yellow. 1 There appears to be no ground for Loudon's statement, Trees and Shrubs, 1118 (1842), that it is identical with /. gossainthanea, Loddiges, as the latter and /. Bedfordiana are kept distinct by Knight and Perry, Syn. Conif. n, 12, 13 (1850). Endlicher states that_/. Bedfordiana is a native of Mexico. Juniperus H39 io. The dwarf forms are often like/. Sabina, and are hard to distinguish in the absence of fruits, except by the much stronger disagreeable odour of the bruised branchlets of the latter species. (A. H.) DISTRIBUTION The distribution of this species, as now limited by Sargent,1 is as follows :— From southern Nova Scotia and New Brunswick southward, often close to the sea- coast, to Georgia, southern Alabama and Mississippi, westward to the valley of the lower Ottawa river, eastern Dakota, eastern Nebraska, Kansas, Indian Territory, and eastern Texas ; not ascending the mountains of New England and New York, nor the high southern Alleghanies ; in middle Kentucky and Tennessee and northern Alabama and Mississippi, covering great areas of low rolling limestone hills with nearly pure forests of small bushy trees. In New England it is very common in the south, rarer in Maine, New Hamp shire, and Vermont ; but nowhere, so far as I saw, attains the size of old trees in England. Dame and Brooks2 give 25 to 40 ft. with a trunk diameter of 8 to 20 in. as the average size, and I saw none larger. It grows here on principally dry, rocky, and exposed hills, but also sometimes in wet ground ; and on the abandoned culti vated fields which are so numerous in the hilly and poorer parts of Massachusetts is taking possession of the soil in many places. At Boston I noticed that both in Prof. Sargent's own grounds and in the Arboretum, pencil cedar was coming up freely from seed ; and I have no doubt it will be planted largely in suitable localities farther south. The rapidly increasing demand for its useful wood has cleared out the accessible timber already in many districts. In Canada, it is a comparatively rare tree, and is confined3 to the limestone districts in the St. Lawrence valley and along Lake Ontario to the Niagara peninsula, where considerable areas were covered with it in 1888. All the timber of any value has now been cut here, as it has been in New England generally. Mohr says that there is hardly any tree in the Eastern States which is so indifferent to soil and climate as the juniper. It thrives in the valley of the St. Lawrence and in New England, often growing on barren hillsides where few trees succeed ; on the exposed arid regions of Kansas and Nebraska, in air and climate with great extremes of heat and cold ; on the limestone plateaux and hills of the south-western States, and on the deep soil of the coast of Georgia, but not ascending the mountains or descending to the alluvial river bottoms. It attains its maximum development south of lat. 36°, where in Alabama it is sometimes as much as IOO ft. in height, but is much oftener 60 to 70 ft. high, and in the north rarely exceeds 40 to 50 ft. and is often a mere shrub. As a rule it is scattered among other trees and forms a small proportion of the forest ; but in the so-called " Cedar Barrens " of Tennessee, it formerly formed an almost pure forest extending over large areas ; 1 Sargent, in Garden and Forest, viii. 61, fig. 9 (1895), gives an excellent illustration of a mature tree near Wawa Station, Delaware County, Pennsylvania. 2 Trees of New England, 27 (1902). 3 Button and Shafer, North American Trees, 117 (1908), state that it also occurs in poor and rocky soil in Nova Scotia and New Brunswick ; and there is a specimen in the Kew herbarium from Newfoundland. VI 144° The Trees of Great Britain and Ireland and in the rocky hills of the Tennessee valley grows in mixture with ash, maple, and oak, and in the prairies of Alabama, with magnolia, lime, and hickory. It grows best on a light loamy soil containing lime, and does not come to perfection on clay or sand. It reproduces itself freely from seed, which it bears every year, the berries being a favourite food of many birds, which scatter it widely. The seedlings endure shade, and spread over abandoned farms in New England and in the south ; but Mohr says that its habit of reproducing itself from suckers seems to have weakened the vitality of the seed, and that under the best conditions only 15 to 25 per cent of the seed will germinate. Excellent illustrations are given in Mohr's paper showing the botanical details and the structure of the wood, with a map giving the distribution of both this species andy. barbadcnsis. CULTIVATION Though described by Parkinson in 1640, and introduced to England before 1664 by Evelyn,1 it has never become an abundant tree in England; and was much more generally planted a century ago than it is now. In most old places it may be found in a more or less damaged condition, for though a long-lived tree it is often broken by wind and snow ; and it varies so much from seed that it is often mistaken for other species. As, however, it is very hardy, and will grow in almost any kind of dry and well-drained soil, but only to a large size where this is also deep and fertile, it should be planted more generally, and in some parts of the south of England might be tried for the sake of its very valuable timber. It is easily raised from seed,2 and grows faster than the common juniper, but it does not produce seed freely in most places ; and I am unable to say what part of America seed suitable for English planting is most likely to come from. Though in the northern United States and Canada it is usually a small and scrubby tree as compared with the large size it attains in the south, I can find no evidence that under cultivation this difference is reproduced. Prof. N. E. Hansen of the South Dakota Agricultural College, how ever, states, in a paper3 on the " Breeding of Cold-Resistant Fruits," that " The red cedar was formerly brought in large quantities from Tennessee, which is well to the south ; northern nurserymen have learned that they must cultivate only the northern form of the red cedar to avoid total failure." It seems to require fairly close planting, as well as pruning, in order to get a clean trunk ; but I do not know to what extent it is capable of bearing shade in this climate. Under favourable conditions in America the growth at first is very rapid, trees only twelve years old having attained in Alabama 25 ft. high and i ft. in diameter. Up to from seventy to one hundred and twenty years the increase continues good, but after that age the trees often begin to decay, though they may live for several centuries, and attain a diameter at the butt of 2 to 4 ft. 1 Aiton, Hart. A'ew. v. 414 (1813). 2 Sargent, in Garden and Forest, via. 61 (1895) says tnat the seecls should be gathered in autumn and then moistened and mixed with sand, and kept in a pit till the following autumn, or the second spring, when they will produce strong plants, 6 to 8 in. high at the end of their second season and ready for transplanting. 3 Report of Conference on Genetics (Key. Hort. Soc.), 1906, p. 402. Juniperus 1441 Sargent says1 that it bears pruning well, and is suitable for formal gardening ; but is not good as a hedge plant, as its branches die when they come in contact with those of a neighbouring tree. There is, however, a thick hedge of this species in Barbier's nursery at Orleans, which is said to grow at the rate of a foot per year. Trees in the American forest are often attacked by two species of Polyporus, which cause white rot and red rot of the timber, spoiling it for commercial purposes. These diseases have been described by Schrenk.2 As a rule, this species, like other junipers, is seen in a bushy form, but on good soil it is capable of producing a clean trunk of considerable length, and this would no doubt be more often the case if planted closer and pruned when young. Mr. A. D. Webster assures me that a tree which grew on sandy loam at Esher, had a trunk with a clean and well-rounded stem free from branches, for 33 ft. in length, and when measured by him contained fully 51 cubic ft. of timber. REMARKABLE TREES Among the finest specimens I have seen in England the largest is an immense old tree at Pains Hill, close to the cedar figured on Plate 128. It measures 13 ft. 9 in. in girth, with a bole about 5 ft. high, dividing into several stems, more or less broken, but one attains 68 ft. in height. At Woolbeding, in Col. Lascelles' grounds, there is a fine tree 65 ft. by 6 ft. 8 in. in 1906. At Bagshot Park a tree with a clean trunk measured 64 ft. by 7 ft. in 1907. At Sherborne, Dorsetshire, a clean-stemmed tree on the ruins of the old castle, 60 ft. by 7 ft. 3 in., is long past its prime. At Coolhurst, Sussex, there is a symmetrical tree 62 ft. by 7 ft. 3 in., and another, 56 ft. by 40 ft. with a clean trunk of 25 ft. At Arley Castle,3 a tree in 1910 measured 68 ft. by 4 ft. 10 in. It is supported by a wire, as it was blown over and pulled upright again several years ago. At Raglan Castle, Monmouthshire, inside the ruins there is a fine old tree 53 ft. by 4 ft. ^10 in. in 1906. At Wimpole, Cambridgeshire, a very handsome tree with pendulous branchlets was 50 ft. high in 1908. In Scotland, the tree does not seem to attain so large a size, the best I have seen being one at Moncrieffe, Perthshire, which4 was 47 ft. by 6 ft. 10 in. in 1907. Another, at Murthly, in the Tayside walk, which in 1906 was bearing many berries, measured 40 ft. by 3^ ft. We have seen no large trees in Ireland; but there is a good specimen of var. glauca at Hamwood. The most remarkable trees that I have seen in Europe are two in the grounds of the Trianon, at Versailles, one of which, not more than about 30 ft. high, has immense spreading branches, which cover an area 57 paces round. The other, close to the château, I could not measure, but estimated it to be 75 ft. high. At Colom- bez, near Metz, there is an avenue of about fifty old trees, 40 to 50 ft. high, with trunks 4 to 5 ft. in girth, growing in an exposed situation. 1 Garden and Forest, x. 142 (1897). 2 U.S. Deft, dgric. Bull. No. 21 (1900). 3 Woodward, Hortiis Arleyensis, No. 4 (1907). 4 Mentioned by Hunter, Woods, Forests, and Estates of Terllishire, 136 (1883), as Cupressus Ihyoidcs viridis. I442 The Trees of Great Britain and Ireland In some parts of Germany the tree grows very well, but does not seem to have any economic value, and in Dr. Mayr's * opinion is only likely to be useful in the south of Europe. TIMBER The timber of the common and of the southern species present no essential difference and are not distinguished in commerce, though at the present time the greater part ofthat imported to Europe comes from the Southern States, that from Florida produced by J. barbadensis being considered the best. Its great resistance to decay makes it very valuable, and formerly, when com moner, it was largely used in America for fencing, telegraph poles, boat- and house building. Now, however, trees large enough for such purposes can hardly be found in the north, and the principal use for which it serves is to make the casing of lead- pencils. Mohr states that for this purpose alone 500,000 cubic feet are annually used in the United States, and 75,000 more exported, most of this going to Germany. At Greenville, Alabama, the logs are cut into pieces of five standard sizes, varying from f to 2^ in. thick, which are packed for shipment in square wooden cases. The waste and sawdust from the mills is made into fine shavings, used for protecting furs and woollen goods against moths, and into paper for underlaying carpets. Cigar-boxes are also made from this wood in Germany ; but most of the so-called cedar in which Havana cigars are packed is the wood of Ccdrela odorata, a very different tree of the West Indies. Though in former times the wood was commonly used in this country for the finer kinds of joiner's work and interior decoration, under the name of cedar, red cedar, or pencil cedar, yet it has, during the last generation, become so scarce and dear, that its use is almost confined to the making of pencils. The greater part of the logs imported at present are from Jamaica, Alabama, and Georgia; and are usually small and faulty. The few large and clean ones which arrive are worth from 6s. to 8s. per cubic foot, whilst small logs and billets are sold at 2s. to 6s. per cubic foot in London. The heart-wood is of a pinkish brown, becoming darker with age, and fading on exposure to the sun, and the sap-wood is whitish. It is distinguished by its fragrance, which, however, is fainter than that of Lebanon cedar, Lawson cypress, or camphor wood ; and not too strong to use for the panelling or ceilings of living rooms. No wood is better adapted for delicate mouldings or carvings, though it is too soft to be used in any positions exposed to friction or contact with furniture. In some cases the heart-wood of old trees is beautifully variegated and twisted, and such pieces would be of the highest value for cabinetmaking, if procurable ; but I have hardly ever seen the waved or curly grain, which is so ornamental in some other conifers, as pitch pine, redwood, or American cypress. One of the best examples of the use of this wood for ornamental work is in the 1 Fremdliind. Wald- u. Parkbäume, 292 (1906). Cf. Schwappach, in Zeitschr. Forst- u. Jagdwesen, xliii. 602 (1911), and in Alitt. Deut. Dend. Ces. 1911, p. n. Juniperus T443 library of Lord Llangattock's house, The Hendre, near Monmouth, which was panelled and ceiled by Messrs. Norman and Burt, from the design of Sir Aston Webb. Overmantels of this wood have been taken out of old London houses, where they have been for probably two centuries, and have realised very high prices ; and on account of its scent, it was a favourite wood in early Victorian times for lining wardrobes, or for matchboarding bathrooms in country mansions. The roof of the fine old church at Bitton, near Bath, is entirely lined with pencil cedar, which was executed under Canon Ellacombe's direction, with wood purchased from a ship wrecked in the Bristol Channel ; and though the odour is not strong enough to be very noticeable, except in damp weather, the effect is very good. Oil of cedar, for which there is a large demand in the United States, is distilled from sawdust and other refuse of the wood, at Cedar Keys in Florida.1 The wood contains as much as 4 or 5 per cent of this oil, which is used as a taenifuge. The shoots ofy. virginiana are sometimes used medicinally in the United States, as a substitute for the true savin, but contain considerably less essential oil.2 (H.J.E.) JUNIPERUS SABINA, SAVIN Juniperns Sabina, Linnœus, Sf. PI. 1039 (1753); Loudon, Arb. et Frut. Brit. iv. 2499 (1838); Parlatore, in De Candolle, Prod. xvi. 2, p. 483 (1868); Bentley and Trimen, Medicinal Plants, iv. t. 254 (1880); Mathieu, Flore Forestière, 518 (1897); Ascherson and Graebner, Syn. Mittehurop. Flora, i. 251 (1898); Kent, Veitch's Man. Com/. 189 (1900); Kirchner and Schröter, Lebengesch. Blutenpfl. Mitteleuropas, i. 320 (1906). Sabina officinalis, Garcke, Fl. Nord- u. Mitteldeutschl. 387 (1858). A shrub, attaining about 15 ft. in height, with foliage of a strong and disagreeable odour, and bitter to the taste. Leaves of two kinds ; on adult shrubs scale-like ; ultimate branchlets very slender, tetragonal, .^ in. in diameter, clothed with 4 ranks of leaves in opposite pairs, which are imbricated, appressed, ovate, acute or blunt at the apex, about ^ in. long, adnate in their lower half, entire in margin, rounded on the back, which usually bears an elliptic depressed resin-gland. On older branchlets, the leaves are more elongated, about \ in. long, acuminate, becoming brown and withered in the third and fourth years. On young plants, and on isolated branches of adult shrubs, the juvenile foliage is acicular, slightly spreading, in opposite pairs, about \ in. long, acuminate at the apex, adnate and not jointed at the base ; upper surface concave, glaucous, and with a prominent midrib ; lower surface, green, convex, marked with a longitudinal depressed gland. Flowers monoecious or dioecious. Fruit ripening in the autumn of the first year or in the following spring, borne on the ends of short scaly recurved branchlets ; irregularly globose or ovoid, about \ in. in diameter, brownish blue, covered with a glaucous bloom, composed of four to six scales, each marked with an obsolete 1 Garden and Forest, ii. 301 (1889). * Fluckiger and Hanbury, P/mrmaccgrafhia, 628 (1879). 1444 The Trees of Great Britain and Ireland mucro.1 Seeds usually two, rarely one or three, immersed in a resinous flesh, ovoid- triquetrous, compressed, narrowed towards the apex, shining brown, with two or three longitudinal furrows, and verrucose on both surfaces towards the summit. VARIETIES The Savin, like most species of juniper, is variable in habit in the wild state in Europe, either occurring as a low prostrate shrub with the branches widely extended and lying on the ground, or as a tall upright dense pyramidal shrub, with horizontal or ascending branches. A fastigiate form has also been observed, similar in appearance to the Swedish variety of the common juniper. The leaves on the adult plant are usually small, scale-like, blunt at the apex, and closely appressed, acicular leaves being not very common in the typical form. The following varieties, differing in foliage, have been described :— 1. Var. lusitanien, Ascherson and Graebner, Syn. Mitteleurop. Flora, \. 253 (1898). Juniperus lusitanica, Miller, Gard. Diet. ed. 8, No. n (1768). Juniperus sabinoides, Grisebach, Spicil. Fl. Rum. ii. 352 (1848) (not Nées2). An upright shrub, with scale-like leaves, which are sharply acuminate at the apex. This variety is common in southern Europe. 2. Var. tamariscifolia, Solander, in Aiton, Hort. Kew. iii. 414 (1789). A low spreading shrub, with mostly persistent juvenile acicular foliage ; leaves on the ultimate branchlets in opposite pairs, slightly spreading, glandular on the dorsal surface, bright green in tint, about -| in. long ; on the older branchlets, occasionally in whorls of threes. This variety, which has been known in cultivation for at least 200 years, is occasionally met with in the wild state in Europe. 3. Var. varicgata, Carrière, Conif. 36 (1855). A dwarf shrub, with adult scale-like foliage, the tips of some of the young branchlets being creamy white. This is mentioned by Loudon, and was known over a century ago in gardens. It is often planted in rockeries. 4. Var. prostrata, Loudon, Arb. et Frut. Brit. iv. 2499 (1838). Juniperus prostrata, Persoon, Syn. PI, ii. 632 (1807); Kent, Veitch's Man. Conif. 183 (1900)5 Britlon and Shafer, N. Amer. Trees, 120 (1908). Juiiiperus Sabina, var./rocumbens, Pursh, Fl. Amer. Sept. 647 (1814); Jack, in Bot. Gaz. xviii. 372 (i893). Juniperns repens, Nuttall, Gen. Amer. ii. 245 (1818). A depressed, usually procumbent shrub, seldom more than 3 ft. high. Leaves usually of two kinds ; the scale-like leaves appressed in four ranks, mucronate at the apex, with a well-marked depressed resin-gland, and similar to those of the type ; 1 The berries are in rare cases open and not coalesced at the summit, the tips of the seeds protruding. This sport is known as var. gymnosperma, Schröter, op. cit. 333, fig. 176 (1906). 2y. sabinoides, Nées, in Liniuea, xix. 706 (1847), is better known as J. mejiicana, Sprengel, Syst. iii. 909 (1826), and is a native of Texas and Mexico. Juniperus H45 the acicular leaves on a few of the branchlets, in opposite pairs, slightly spreading, with a longitudinal depressed gland on the back, about % in. long. Fruit, on recurved stalks, light blue and scarcely glaucous, \ in. in diameter, with one to four seeds. This, which is the American form1 of the Savin, is considered by some botanists to be a distinct species. It is distributed from southern Maine to the shores of Hudson Bay, and westward from Newfoundland and northern New England through New York along the shores of the Great Lakes and through northern Minnesota and south Dakota to the eastern slopes of the Rocky Mountains in Alberta, Montana, and Wyoming. The American Savin grows mainly on sandy soil, as on the seashore of the Atlantic coast and on inland dunes and barrens, and thus differs remarkably from the lime-loving savin of Europe. The American Savin is said to have been first introduced by Loddiges, who called it /. hudsonica in his catalogue of 1836. The shrub cultivated as var. prostrata is low and prostrate, bearing only acicular very glaucous foliage ; leaves in their free part ,\- in. long, spreading, the glaucous bloom appearing on the dorsal surface near the base ; ventral surface whitened ; branchlets of the second and third year bright reddish brown, with persistent needles of the same colour. I have not seen this cultivated shrub in fruit or bearing scale-like leaves ; but it has the disagreeable odour of the savin, and in all probability is, as reputed, of American origin. Sargent states - that it is the hardiest and most beautiful of all the prostrate junipers that can be grown in New England gardens. DISTRIBUTION The Savin is widely distributed, occurring in central and southern Europe, the Caucasus, and North America. It occurs mainly in Europe in extensive thickets on dry rocky sunny mountain slopes ; but is also met with as undergrowth in many pine forests, as those of Pinus sylvestris in the Sierra Nevada in Spain, and of P. Icucodcrmis in Herzegovina. It grows usually on limestone ; but is occasionally seen on other formations. In Europe it is most common in Spain and Portugal, and in the Balkan peninsula. It is widely spread throughout the whole Alpine mountain system, but is rare towards the north, though it is met with in the Bavarian Alps and in a few stations in Switzerland. Its distribution in Russia is remarkable, as it occurs in isolated spots throughout the great plain, reaching as far north as the Baltic coast ; but is a mountain plant, as usual elsewhere, in the Crimea and in the southern part of the Ural range. It also occurs in the Caucasus and the mountains of northern Persia ; but is not met with in Asia Minor, and is totally absent from northern Africa. Its occurrence in Siberia is attested by Russian botanists ; but we have seen no specimens. (A. H.) The Savin was early introduced into England, as it is mentioned in Turner's Names of Herbes, published in 1548. 1 Rehder, in Bailey, Cyd. Amer. Hort. 850 (1900), says that it is sometimes called in America the Waukegan juniper. * In Garden and Forest, x. 421 (1897). Sargent adds that the European savin, if it has ever been tried in gardens'in Massachusetts, has probably not proved hardy. 1446 The Trees of Great Britain and Ireland The finest specimen in Britain is probably one growing in the garden of Stourton Court, Stourbridge, the residence of R. Matthews, Esq. This is well figured in Journ. R. Hort. Soc. xxxiii. 327 (1908), where it is stated to measure 6 ft. high and 57^ ft. in circumference. The soil is Old Red Sandstone. The young green shoots of the Savin are used in medicine, and yield a volatile oil, which is officinal and possesses extremely active properties.1 (H. J. E.) JUNIPERUS EXCELSA Juniperus excelsa? Bieberstein, Beschr. Land. Casp. 204, App. No. 72 (1800), and Fl. Taur. Cauc. ii. 425 (1808); Pariatore, in De CandoIIe, Prod. xvi. 2, p. 484 (1868); Boissier, Fl. Orient. v. 708 (1881); Siehe, in Gartenflora, xlvi. 208, t. 26 (1897); Kent, Veitch's Man. Conif. 174 (1900). Juniferus Sabina, Linnœus, var. taurica, Pallas, Fl. Ross. ii. 15 (1788). Jimiperus Sabina, Linnœus, var. excelsa, Georgi, Besch. Rtiss. Reichs, iii. 1358 (1802). Juniperus fcetida, var. excelsa, Spach (excl. syn. Amer.), in Ann. Sc. Nat. xvi. 297 (1841). Juniperuspolycarpos and isophyllos, Koch, in Linncea, xxii. 303, 304 (1849). Juniperus Olivierii, Carrière, Conif. 57 (1855). Juniperus cegcea, Grisebach, Veg. der Erde, 378, 572 (1872). Sabina excelsa,polycarpos, and isophyllos, Antoine, Cupress. Gatt. 45, 47, 48 (1857). A tree, occasionally attaining in Asia Minor a height of 70 to 100 ft. Leaves dimorphic in wild specimens ; but juvenile foliage is rarely seen on cultivated adult trees of the typical form. Adult foliage ; ultimate branchlets very slender, ^0 in. or less in diameter ; leaves scale-like, closely appressed, in four ranks in opposite decussate pairs, ovate-rhombic, about ^ in. long, acute or obtuse, marked in the middle on the back with a depressed oval or linear gland ; leaves on older branchlets, in pairs or in threes, spreading, mucronate, glandular on the back. Juvenile foliage, when present3 ; leaves acicular, spreading, in opposite pairs, jl to |- in. long, concave above with two stomatic bands, marked on the lower surface with a linear gland at the base. Flowers monoecious or dioecious. Fruit, ripening in the second year, on short scaly stalks, globose, ^ to -J in. in diameter, smooth, dark purplish brown, covered with a bluish bloom when ripe, composed of four or six scales, each marked by a minute mucro. Seeds, about six in each fruit, oblong, more or less triquetrous, apiculate at the apex. i. Var. stricta, RoIIisson, ex Gordon, Pinetum, 144 (1875). A tree, narrowly pyramidal in habit, with glaucous juvenile foliage ; leaves acicular, slightly spreading, about \ in. long including the basal decurrent part, whitened with a stomatic band above, marked with a minute gland near the base on the lower surface. This originated in Messrs. Rollisson's nursery at Tooting, and appears to be 1 Cf. Fluckiger and Ilanbuty, Pharmacographia, 628 (1879). 2_/. excelsa, London, Arb. et Fnil. Brit. iv. 2503 (1838), includes a mixture of junipers from Siberia, the Himalayas, and North America, and does not appear to refer to the true plant from Asia Minor. 3 Described from a native specimen collected by Hausknecht. Juniperus *447 perfectly hardy, retaining its characters in old age. There are three trees at Kew, 25 to 30 ft. high, which were obtained from RoIIisson in 1868. Var. Perkinsii and var. venusta, Gordon, Pinetum, 144 (1875), are similar to var. stricta, only differing in the more glaucous leaves. A specimen of var. Perkinsii at Kew is almost columnar in habit. /. excelsa is a native of the Balkan States, Island of Thasos,1 Crimea, Asia Minor, Syria, Armenia, and the Caucasus. In Thrace, Macedonia, and the Rhodope mountains of Rumelia it is, according to Adamovic,2 a shrub rather than a tree, ascending occasionally to 4000 ft. It is recorded for one station in Bulgaria, near the village of Beli Iskar. Koppen3 states that in the Crimea it is common on the coast side of the mountains, forming pure woods of considerable extent ; but never attaining a large size, the tallest tree noted being about 30 ft. J. excelsa appears to attain its greatest development in Asia Minor, where it forms extensive woods in the mountains, either pure or mixed with Lebanon Cedar and Abies cilicica. Siehe4 gives an illustration of a tree in the Cilician Taurus, nearly 100 ft. high, and states that it assumes two forms, being either a tall narrow pyramidal tree, or a shorter tree with wide-spreading branches ; occasionally with a trunk 4^ ft. in diameter. There are very fine woods at Namrun, Efrenk, and Güllek in Cilicia, where the trees are tall, slender, and dense upon the ground. Siehe considers that the timber will prove of great value for railway sleepers. The Turkish name of the tree is arytsch. The date of introduction is uncertain, as Loudon's account of J. excelsa does not relate to this species; but it was probably brought into England about 1836. It usually forms a narrow columnar or pyramidal tree. A tree at Arley Castle, said to have been planted in 1877, measured5 32 ft. by 3 ft. i in. in 1904. Another at High Canons, Herts, was 32 ft. by z\ ft. in 1908. There are trees of similar size in the botanic gardens at Kew and Cambridge. We have also seen good specimens at Hardwicke, Tortworth, Westonbirt, Highnam, and Eastnor. In Ireland there are trees at Glasnevin and Powerscourt. J. excelsa has been much confused with the two following species, which are not apparently in cultivation. These are remarkably distinct from /. excelsa both in foliage and fruit. I. Juniperus macropoda, Boissier, Flora Orientait*, v. 709 (1881). A tree, occasionally attaining 70 ft. in height, but often shrub-like, which is a native of Persia, Afghanistan, and Baluchistan, where it forms extensive open forests, east of Quetta." This has much coarser foliage than J. excelsa, resembling that of J. chinensis. Fruit globose, \ in. in diameter, brownish purple, tinged with a glaucous bloom, each of the four to six scales with a prominent mucro. Seeds, two to four, ovoid. 1 It is not recorded for any of the islands in the ^Egean Archipelago except Thasos, where it grows in the pine woods on the coast. Cf. Grisebach, Veg. der Erde, 378, 572 (1872). * Veget. Balkanländer, 152 (1909). 3 ffolzgnuäi lise Europ. Kusslands, ii. 423 (1889). 4 In Gartcnjlora, xlvi. 208, t. 26 (1897), and in Mût. Dent. Dend. Ces. 1911, p. 306. 6 Woodward, Hartiis Arleyensis, 19 (1907). « Cf. Lace, in Journ. Linn. Soc. (Bot.) xxviii. 307 (1891). Gamble, Indian Timbers, 698 (1902), gives also some particulars concerning this tree, which he considers to be identical with the Himalayan/, rcligiosa, Q 1448 The Trees of Great Britain and Ireland II. Juniperus religiosa, Carrière, Conif. 41 (1855) (not Royle1). Juniperus gossainthanca? Loddiges, Catalogue, 48 (1836); Loudon, Trees and Shrubs, 1090 (1842). Juniperus chinensis, Pariatore, in De Candolle, Prod. xvi. 2, p. 488 (1868 (in part) (not Linnaeus). Juniperus excelsa, Brandis, Forest Flora, N. W. India, 538, t. 68 (1874) (not Bieberstein). Juniperus macropoda, Hooker, Fl. Brit. India, v. 647 (1888) (not Boissier). A tree, occurring in the inner dry ranges of the north-western Himalayas from Chitral and Kashmir to Nepal, at 5000 to 10,000 ft. altitude. It often attains 50 ft. in height, with a girth of 6 or 7 ft. ; but occasionally trees of enormous girth are met with, one at Lahoul measuring 33^ ft. in circumference. This species has foliage similar to that of J. macropoda and J. chinensis. Fruit obovoid, widest at the apex, which is depressed, \ in. in diameter, bluish black with a soft juicy pulp. Seeds, one to three, ovoid, sharp-pointed, with large resin-glands. (A. H.) ' J. religiosa, Carrière, the first published name with a description, should be adopted for this species. J. religiosa, Royle, Ilhist. Himal. i. 351 (1839), without any description, is possibly J. squamata. Cf. p. 1420, note 4. 2 This name was published without any description, and, moreover, is somewhat doubtful. Cf. p. 1438, note I. ATHROTAXIS Athtotaxis,1 Don, in Trans. Linn. Soc. xviii. 172, tt. 13, 14 (1839) ; Bentham et Hooker, Gen. PI. iii. 430 (1880); Masters, mjourn. Linn. Soc. (Bot.) xxx. 21 (1893). EVERGREEN trees belonging to the division Taxodinese of the order Coniferse. Leaves persistent for several years, spirally arranged, homomorphic, crowded, imbricate, spreading or closely appressed, adnate at the base, free at the apex ; without scaly buds. Flowers monoecious, solitary at the apices of the branchlets. Staminate (lowers catkin-like, with crowded stamens spirally arranged on an axis ; each stamen with a slender stalk and a sagittate connective, which bears two pollen sacs dehiscing longitudinally. Ovuliferous flowers, of ten to twenty-five spirally arranged scales ; each scale with an adnate fleshy disc, bearing three to six ovules. Cones ripening in one year ; scales, ten to twenty-five, woody, spirally arranged, cuneate and narrow at the base, horizontally spreading, dilated into a clavate or peltate lamina, which bears on the back or at the apex a triangular cuspidate process.2 Seeds, three to six, pendulous from the thickened part of the scale below the apex ; oblong, compressed, with a transverse hilum and two lateral wings. Cotyledons two, longer than the primary leaves.8 Athrotaxis is closely related to Cryptomeria ; and Kent states that rooted cuttings of Cryptomeria elegans are used as stocks for grafting scions taken from the different species of Athrotaxis. Fossil remains found in various deposits in Europe have been identified, but perhaps erroneously, with Athrotaxis. C. Reid 4 has lately shown that the remains in the Bovey Tracey lignites belong to Sequoia Couttsiœ, Heer, and not to Athrotaxis, as had been supposed by Starkie Gardner. This genus is confined in the living state to Tasmania, where there are three species 6 distinguishable as follows : — 1 Derived from döpios, crowded, and rd£«, arrangement. Endlicher, Gen. Sufpl. i. 1372 (1841), and Syn. Conif. 193 (1847), gives the erroneous spelling Arthrotaxis, which has been followed by several writers. * This process is the extremity of the scale in the flowering stage, which has coalesced almost completely with the ovuliferous disc, the latter having increased much in size during the ripening of the ovules into seed. 3 Masters, injourn. Linn. Soc. (Bot.) xxvii. 235, 237 (1890). « In Phil. Tram, series B, vol. 201, p. 171, pi. 15, figs. 40, 41 (1910), where the distinctive characters of the epidermis of the leaves of Sequoia and Athrotaxis are made plain. 6 Athrotaxis (?) tetragona, W. J. Hooker, Icon. Plant, t. 560 (1843) belongs to a distinct genus, and is Microcachrys tetragona, J. D. Hooker, in London Journ. Bot. iv. 149 (1845). This is a low rambling shrub, also a native of Tasmania. It is occasionally cultivated in conservatories. 1449 145° The Trees of Great Britain and Ireland I. Leaves spreading, entire in margin, with two continuous white stomatic bands on the ventral surface, and two lateral stomatic depressions confined to near the base on the dorsal surface. 1. Athrotaxis selaginoides, Don. Leaves very spreading, £ in. long, with a rigid spine-like acuminate apex, and an opaque margin. 2. Athrotaxis laxifolia, W. J. Hooker. Leaves slightly spreading, J in. long, obtuse or acute at the apex, and with a translucent margin. II. Leaves closely appressed, scale-like and apparently in four ranks like Cupressus, but really spirally arranged ; margin translucent and denticulate ; ventral surface concealed ; dorsal surface with indistinct stomatic lines. 3. Athrotaxis cupressoides, Don. Leaves rhombic-ovate, obtuse at the apex, \ in. long. (A. H.) ATHROTAXIS SELAGINOIDES Athrotaxis selaginoides, Don, in Trans. Linn. Soc. xviii. 172, t. 14 (1839); W. J. Hooker, Icon. Plant, t. 574 (1843); J. D. Hooker, in Land. Journ. Bot. iv. 148 (1845), and Fl. Tasman. \. 354 (1860); Masters, in Gard. Chron. ii. 724, figs. 140, 141 (1887), and iv. 544, fig. 79 (1888); Kent, Veitch's Man. Conif. 262 (1900); Rodway, Tasmanian Flora, 277 (1903); Bakerand Smith, Pines of Australia, 303, plates on pp. 304, 305, figs. 217-228 (1910). Athrotaxis Gunneana, Carrière, Conif. 207 (1867). Athrotaxis Gunniana, Gordon, finetum, 47 (1875). Cunninghamia selaginoides, Zuccarini, in Siebold, Fl. Jap. ii. 9, note (1844). A tree, attaining a larger size than the other two species, up to 100 feet in height and 10 feet in girth. Bark described by Baker and Smith, as slightly furrowed and fibrous, but not very rough. Branchlets stout, entirely covered by the decurrent bases of the leaves. Leaves spirally arranged, loosely imbricated, widely spreading but incurved at the apex, rigid, coriaceous, about \ in. long, subulate, adnate but not jointed at the base, tapering to an acuminate spine-like apex ; dorsal surface keeled with two lateral depressions near the base, which are whitened by stomatic lines ; ventral surface concave, with two longitudinal white stomatic bands from the apex to the base separated by a green midrib ; margin entire, opaque. Cones ovoid or globose, about i in. in diameter, composed of 20 to 24 brown woody scales, which are about \ in. long, with a narrow cuneate base, and an oval or ovate expanded inflexed lamina, which terminates in a triangular thin process. The species, which is known in Tasmania as King William Pine, is said by Rodway to occur in the western mountains, extending from Mount Field, Mount Hartz, Adamson Peak, and Mount La Perouse to the west coast. Baker and Smith state that it is common in the immediate neighbourhood of Williamsford, Athrotaxis at about TOGO ft. above sea-level. It is a prominent tree in the dense scrub which covers this locality, being associated with Phyllocladus rhomboidalu, Nothofagus Cnnninghami, &c. These authors figure an old tree, said to be typical ^ habit which shows a twisted stem, free of branches for three quarters of its height, and surmounted by a small irregular dense crown of foliage. A. selaginoides ascends to 3000 ft. or more, as it occurs on the summit of Mt. Reed and other mountains, usually in a much dwarfed and stunted form. Baker and Smith, who give excellent figures of the structure of the leaves and wood, state that the wood is not unlike that of Sequoia semfervirens, both in general characters and in texture, being open and straight in the grain, easy to work, and very light in weight. It is pale reddish when freshly cut, but becomes lighter in colour on exposure. Possessing great durability, and considerable toughness and strength, it is used in Tasmania for cabinet-work, coach-building, and for making oars. Penny1 states that it occurs in limited quantities; and is apparently never exported. This species was introduced about the year 1857 by Mr. W. Archer of Cheshunt; but appears to be less common in cultivation than A. laxifoha. A thriving specimen at Osborne, Isle of Wight, planted in 1879, was 17^ feet high in January 1912, when it bore both young and old cones. A tree at Lamellen, St. Tudy, Cornwall, which was 26 ft. high, died in 1909. From it Mr. Magor raised a few seedlings, which are still small plants. A specimen at Abbotsbury was killed by drought in the summer of 1911. The finest specimen is at Kilmacurragh, in Ireland, and measured 32 ft. high in March 1912, when it was bearing numerous old cones. A smaller tree is thriving at Rostrevor. (A. H.) ATHROTAXIS LAXIFOLIA Athrotaxis laxifolia, W. J. Hooker, Icon. Plant, t. 573 (1843); J. D. Hooker, in Land. Journ. Bot. iv. 149 (1845), and Fl. Tasman. i. 354 (1860); Masters, in Gard. Chron. xxiv. 584, fig. 134 (1885), ii. 724, figs. 142, 143 (1887), and ix. 144, 147, figs. 37, 38 (1891), and in Journ. Linn. Soc. (J3ot.) xxii. 201, fig. 26 (1886); Kent, Veitch's Man. Conif. 261 (1900); Rodway, Tasmanian Flora, 277 (1903); Baker and Smith, Pines of Australia, 313 (1910). Athrotaxis Doniana, Maule, ex Gordon, Pinet. Suppl. 16 (1862). A tree, attaining about 40 ft. in height. Bark reddish, fibrous, peeling off in long vertical ribbons. Branchlets slender, covered by the decurrent bases of the leaves. Leaves spirally arranged, closely imbricated, slightly spreading, about \ in. long ; incurved at the acute or obtuse, rarely mucronate, apex ; dorsal surface keeled, with two lateral depressions near the base, which are whitened by stomatic lines ; ventral surface concave, with two longitudinal white stomatic bands ; margin entire, thin and translucent towards the apex. Cones sub-globose, f in. in diameter, composed of 15 to 20 brown woody scales, 1 Tasmanian Forestry, xi. 42 (1905). Penny quotes A. O. Green's tests of various Tasmanian timbers. This species is one of the lightest, a cubic foot weighing only 22 Ibs. I452 The Trees of Great Britain and Ireland which are about § in. long, with a slender cuneate claw, and a thickened oval expanded lamina, bearing on the back a large ovate acute process. This species was found by Gunn and Archer at the falls of the Meander river, and along rivulets near the summit of the western mountains in Tasmania ; and is said by Rodway to occur on Field Range and near Mount La Perouse. It appears to occur at higher altitudes than the other species, reaching 4000 ft. A. laxifolia was also introduced by Archer in 1857, and appears to be the most successful of the three species in cultivation in this country. The largest specimens are in Cornwall, where Elwes measured in 1911 a fine tree at Scorrier, 38 ft. high and 3 ft. 9 in. in girth ; from it three seedlings were raised about twenty years ago, which are now 15 to 20 ft. high. Another fine specimen at Penjerrick was 32 ft. by 2 ft. 8 in. in the same year. One in Mr. R. Gill's nursery at Tremough near Penryn, measured 27 ft. high, and was bearing fruit in January 1912. Another, 17 ft. high, occurs at Trewidden near Penzance. There is also a good tree at Menabilly, growing at 100 feet above sea-level in an exposed position, which was planted in 1880, and reported to be a perfect pyramid, 12 ft. high in 1891 ; it was 27 ft. high and 3 ft. in girth in January 1912. A branch of this tree with cones was figured x by Dr. Masters. This species appears to be very hardy, and succeeds as far north as Durris in Kincardineshire, where a tree about 10 ft. high produced cones in 1909, from which numerous seedlings were raised ; some of these have been planted out at Bayfordbury and in the Cambridge Botanic Garden. In Ireland, there is a fine specimen at Kilmacurragh, 28 ft. high in 1912 ; and a smaller one at Castlewellan. (A. H.) ATHROTAXIS CUPRESSOIDES Athrotaxis cupressoides, Don, in Trans. Linn. Soc. xviii. 173, t. 13, fig. 2 (1839); W. J. Hooker, Icon. Plant, t. 559 (1843); J. D. Hooker in Lend. Journ. Bot. iv. 148 (1845), ai'd -^- Tasman. i. 354 (1860); Masters, in Gard. Chron. xxiv. 270, fig. 60 (1885), and ii. 725, figs. 144, 145 (1887); Kent, Veitch's Man. Conif. 261 (1900); Rodway, Tasmanian flora, 277 (1903); Bakerand Smith, Pines of Australia, 313 (1910). Cunninghamia cupressoides, Zuccarini, in Siebold, M. Jap. ii. 9, note (1844). A tree, attaining 40 ft. in height and 6 feet in girth, with ascending branches. Branchlets pseudo-opposite or alternate, densely covered with scale-like leaves, which appear to be in opposite decussate pairs, but in reality are in a spiral arrangement. Leaves on the ultimate branchlets, similar to those in Cupressus, homomorphic, densely appressed, closely imbricated, rhombic-ovate, about ^ in. long, obtuse at the apex ; ventral surface entirely concealed ; dorsal surface keeled and marked with white stomatic dots in lines on the two sides ; margin denticulate 1 Card. Chron. xxiv. 584, fig. 134 (1885) and Journ. Linn. Sac. (Bot.) xxii. 201, fig. 26 (1886). In Card. Chron. xxiv. 660 (1885), it is stated that a tree produced cones in 1875 'n Mr. Robert Loder's garden at Whittlebury, Tovvcester ; but this cannot now be found and was probably cut down some years ago. Athrotaxis H53 and translucent. Leaves on the older branchlets increasing in size, becoming dark brown and about \ in. long in the fifth or sixth year. Cones, nodding on curved branchlets, f in. in diameter; scales 10 or 12, much smaller than in the other species, less than I in. long, with a triangular recurved process on the middle of the expanded lamina. This species is said by Rodway to be a small erect tree, about 40 ft. high, found in the western mountains of Tasmania, near St. Clair and to the west and south-west of Field Range. It rarely exceeds 5 or 6 feet in girth ; but Sir J. Hooker mentions one very old and hollow tree, which was 15 ft. in girth at 3^ feet from the ground. Baker and Smith state that the timber resembles in all respects that of A. selaginoides. It was introduced in 1857 by Mr. Archer at the same time as the other species, and resembles them in cultivation; but it is rare and has not attained in this country as great a height as A. laxifolia. In Cornwall, a tree 20 ft. high was bearing cones in Gill's nursery, near Penryn, in January 1912; and a smaller specimen, also bearing fruit, occurs at Trewidden. Kent mentioned in 1900 a fine specimen at Upcott, near Barnstaple ; which, Mr. W. Harris informs us, has lately lost its leader and is now only 13 ft. high. In Hillier's nursery at Shroner, near Winchester, a tree, planted in 1887, was 12 ft. high in 1910, and had commenced to bear cones in small quantity. There is also a small specimen at Brickendon Grange, Hertford, which was 5 ft. high in 1911. In Ireland, a fine specimen bearing fruit at Kilmacurragh was 21 ft. high in March 1912; while another at Powerscourt1 was 22 ft. high. A thriving tree at Castlewellan, 15 ft. high, bore 18° of frost without injury in February 1912. (A. H.) 1 Erroneously called A. Doniana in Card. Chron. xlix. 219 (1911). FITZROYA Fitzroya, J. D. Hooker, ex Lindley, in Journ. Hort. Soc. Land., vi. 264 (1851), and ex W. J. Hooker, Bot. Mag. t. 4616 (1851); Bentham et Hooker, Gen. PI. iii. 425 (excl. Diselmd) (1880); Masters, in Journ. Linn. Soc. (Bot.) xxx. 17 (1893). A GENUS belonging to the division Cupressineee of the order Coniferœ, characterised by the cones, composed of nine scales, in three alternating whorls, the three lower most scales minute and sterile ; those of the intermediate whorl larger, and either empty or each bearing a single two-winged seed ; the three upper scales largest and fertile, each bearing two to six seeds, which are partly three- and partly two- winged. The apex of the axis of the cone terminates in three peculiar processes,1 the precise morphological nature of which is unknown, but possibly they may represent an aborted whorl of scales. Only one species is known, in the descrip tion of which below, the vegetative and other characters of the genus are given in detail. Diselma, a genus founded by Sir J. D. Hooker,2 was united by Bentham and Hooker8 with Fitzroya, but appears to be sufficiently distinct. In Diselma, the cones are composed of two pairs of opposite scales, the outer pair small and empty, the inner two scales larger and fertile, each with two seeds, which are three-winged. Diselma Arckeri, J. D. Hooker, Fl. Tasman. i. 353, t. 98 (1860), the only known species, differs greatly from Fitzroya in habit, having minute scale-like appressed leaves, like those of a Cupressus, in opposite decussate pairs ; and is a shrub " about 6 ft. high, growing between 3000 and 4500 feet elevation in the western mountains of Tasmania. D. Archeri was formerly represented in the Temperate House at Kew by a single specimen, which is no longer living. At present it is apparently not in cultivation in England. (A. H.) FITZROYA PATAGONICA Fitzroya patagonica, J. D. Hooker, ex Lindley m Journ. Hort. Soc. Loud. vi. 264 (1851), and ex W. J. Hooker, Bot. Mag. t. 4616 (1851); Lindley, in Paxton, Flower Garden, ii. 115 (1852); Kent, Veitch's Man. Conif. 198 (1900); Castillo and Dey, Jeog. Vej. Rio. Valdivia, 27, fig. 8 (1908). An evergreen tree, attaining in Chile 100 to 160 ft. in height, and 10 to 16 ft. in girth; but dwarfed to a low shrub at high elevations. Bark reddish, longitudinally fissured, and peeling off in narrow ribbons. Young branchlets green, 1 These gland-like processes secrete resin and exhale a slight fragrant odour. 2 Fl. Tasman. i. 353, t. 98 (1860). 3 Gen% Plantf ii;> 42 * Cf. Baker and Smith, fines of Australia, 300 (1910). Fitzroya H55 glabrous, flexile, slender, covered by the decurrent bases of the leaves, which are separated between the whorls by three linear grooves. Older branchlets until the seventh year, stouter, reddish, marked by withered leaves and their remains. Buds ovoid or globose, composed of green scales, which are slightly modified and shortened ordinary leaves. Leaves persistent for several years, in alternating whorls of threes, decurrent by their bases on the branchlets ; their free part spread ing, spatulate, incurved at the rounded apiculate apex, about i in. long; upper surface concave, with two narrow white stomatic depressions extending from the apex to the middle of the leaf or beyond, and separated by a raised green midrib ; lower surface convex, with a broad green raised midrib, on each side of which is a narrow white stomatic depression often extending from near the apex to the adnate base of the leaf. On the main axes, the leaves are often J in. long, adnate in greater part to the branchlet, and becoming reddish brown in the third and fourth years. Flowers1 usually dioecious ; sometimes monoecious or hermaphrodite. Staminate flowers solitary in the axils of the leaves towards the apex of the branchlet, cylindrical, subtended at the base by a few scales, composed of 15 to 24 stamens in ternate whorls ; anthers 4-celled. Ovuliferous flowers, solitary and sessile on the ends of short leafy branchlets near the apex of a branch, composed of nine scales, in three alternating whorls ; the three lowermost scales minute and sterile ; the three scales of the intermediate whorl either empty or each bearing a single two-winged ovule ; the uppermost three scales always fertile, each bearing a central three-winged ovule and one to five lateral two-winged ovules ; oblong tubercles at the summit of the axis of the cone, three, yellowish, translucent, about ^ in. long. Cones, scarcely larger when mature than in the flowering stage, sub-globose when closed and about 3- in. in diameter, ripening in one year, with three minute scales at the base of the six large woody scales, each of the latter with a dorsal process, spreading widely to let loose the seeds, which are variable (nine to sixteen) in number (equalling the number of ovules in the flower). Seed with an oblong compressed body and two or three broad lateral membranous wings ; the seed with the wings nearly orbicular and about ^ to 5 in. in diameter. Cotyledons two. This species is a native of South America, occurring in Chile and northern Patagonia. It extends from the coast range immediately north of Valdivia south ward to the island of Chiloe and the mainland opposite in about lat. 42° 40', and reaches inland to the central cordillera of the Andes. It is known to the inhabitants as alerce? and covers immense tracts of marshy and peaty ground with extensive woods, which are called alerzales. These woods are widely distributed, the best known being in the coast range of Valdivia, around Lakes Llanquihue and Nahuelhuapi, in the neighbourhood of Puerto Montt, and in the valley of the river Maullin.8 1 The flowers of Fitrroya, which are complicated and variable, are being investigated at Cambridge by Mr. R. C. Maclean. Monoecious flowers occur in Chilean specimens, as well as on the tree at Hewell Grange. The hermaphrodite flowers of the latter have several whorls of scales ; the scales in the three lowermost whorls bear anthers, those in the upper two whorls bear ovules ; the axis ends in the normal three gland-like processes. 2 Alerce is used in Spain as the name of the larch ; and is derived from the Arabic, al-arzah, signifying cedar. ' Cf. Reiche, Verbreit. Chilen. Canif. 5 (1900), and Pflanzenverb. Chile, 63, 238 (1907). R VI 1 45 6 The Trees of Great Britain and Ireland Castillo and Dey1 state that enormous trees still occur, up to 160 ft. in height and 1 6 ft. in girth ; and mention veterans, which are said when felled to have shown 4000 annual rings; but this seems scarcely credible. Captain Fitzroy2 in whose honour the genus was named, and who commanded the " Beagle " between 1828 and 1836, states on good authority, that a tree in the cordillera on the mainland opposite Chiloe had a stem measuring 76 ft. in length to the first branch, and 30 ft. in girth at five feet from the ground. It yielded 1500 planks. W. Lobb saw on the precipices around Valdivia trees 100 ft. in height and 8 ft. in diameter ; and states3 that it ascends the mountains to the limit of perpetual snow, where it is occasionally only a few inches high. Fitzroya patagonica was first introduced4 in 1849 by W. Lobb, who sent home seeds from Valdivia ; and again by R. Pearce, who collected in Chile for Messrs. Veitch in 1859-1862. It is perfectly hardy in this country, and at Kew5 bore without injury the severe frost of 1908-1909, when the temperature fell to 10° Fahr. ; nevertheless, it is slow in growth, and seems to develop oftener into a bushy shrub than a tree, but this may be due to most of the specimens in cultiva tion being derived from cuttings. The finest trees are in Cornwall and Devon. One at Killerton, planted in 1864, was 34^ ft. by 3 ft. 2 in. in 1911. Another at Bicton was 35 ft. high in the same year. An ill-shaped tree at Penjerrick, dividing into several stems at the base, was about 35 ft. high in 1910. At Coldrenick a similar tree with several stems was 25 ft. high in 1911. Masters figured6 a tree at Pencarrow, planted in 1852 by Sir W. Molesworth, which was reported in 1902 to have been 21 ft. in height, with a spread of branches of 46 ft. At Abbotsbury, a tree was killed by drought in the summer of 1911. At Upcott, Barnstaple, a well-grown and healthy specimen was 28^ ft. high in January 1912. There are smaller trees at Highnam near Gloucester, and at Leaton Knolls, Shrewsbury. In the pinetum at Uplyme, Dorset, there is a fine tree, 30 ft. by 3 ft. 3 in. in 1912. At Belsay Castle, Newcastle-on-Tyne, there is a fine specimen, which was planted about 1856, and measured 28^ ft. high in January 1912. Sir Arthur E. Middleton, Bart., informs me that it is growing in a sheltered place in an old sandstone quarry, where the soil contains a considerable admixture of clay. This tree has been kept from an early date to one leader; otherwise it would have spread in an irregular way. There are also several bushy specimens at Belsay Castle, where this species has never been touched in the slightest degree by frost. Nearly all the cultivated trees which we have seen bear only female flowers, and in consequence the seed, which is freely produced, is infertile. A tree, however, at Hewell Grange, Redditch, which is about 28 ft. high, is monoecious, and bore in May 1912 both male and female flowers in profusion, as well as a few which were hermaphrodite. eg. Rio Valdivia, 28 (Santiago, 1908). Reiche gives the maximum as 180 ft. high, 16 ft. in diameter, and 2500 years old. 2 Narrative of Voyages of the Beagle, \. 275, 282, and ii. 391 (1839). Cf. also Cook, in London, Gard. Mag. xv. 694 (1839). 3 Joarn. Hort. Soc. vi. 262 (1851). 4 Hortiis Veitchii, 38, 46, 340 (1906). 5 Kew Bull. 1909, p. 235. e Card. Chron. xxxi. 392, Supply. Illust. (1902). Fitzroya H57 The only specimen that we know of in Scotland is one at Ardgowan, which was about 25 ft. high in 1909. A tree at Murthly, recorded at the Conifer Conference as 16 ft. high and 30 years old in 1891, was killed in the severe winter of 1893-1894, when the thermometer registered at Murthly n" Fahr, below zero. In Ireland a tree at Powerscourt was 30 ft. high in 1910; and another at Fota was 25 ft. high in 1912. There are also good specimens at Kilmacurragh, Rostrevor, and Castlewellan.1 (A. H.) TIMBER The wood is very valuable ; and is remarkable for the extraordinary straight- ness of the grain, which makes it very easy to rend into thin boards. These are used as shingles for roofing, which after exposure to the weather turn blue and resemble slates; and also for floors and partitions. Castillo and Dey state that it is very durable in contact with water, is unaffected by heat or humidity, and has lasted in roofs without alteration for over 100 years. Captain Fitzroy2 says that "it does not shrink or warp; and, though brittle, is of very close grain and well adapted for furniture. Of this wood, the country people make staves for casks ; and the bark of the tree is used for caulking the seams of vessels, being extremely durable when constantly wet, though it soon decays when exposed to the sun and air." Spars of alerce, which proved to be very strong, were obtainable 80 or 90 ft. in length. I saw mule-loads of these shingles 8 ft. long and very thin, on my journey from Nahuelhuapi to Puerto Montt, and very large stumps of trees in the country near that port ; but I was unable to visit the forest where it grows. The wood is apparently still unknown in Europe ; but I bought three boards in London imported from Chile, which are of a rich reddish colour with very fine and close grain. (H. J. E.) 1 Figured by Earl Annesley, Beautiful and Kare Trees, t. 68 (1903), who states that this species should be planted in deep loam and peat, as it is a deep-rooting tree which grows slowly till it is well established. 2 Narrative of Voyages of Hu Beagle, i. 275, 282, and ii. 391 (1839). Cf. also Cook, in Loudon, Card. Mag. xv. 694 («839). SAXEGOTH./EA Saxegothœa, Lindley, in Jour». Hort. Soc. Lond. vi. 258 (1851); Bentham et Hooker, Gen. PI. iii. 434 (1880); Masters, in Journ. Linn. Soc. (Bot.) xxvii. 299 (1889), and xxx. 10 (1893); Pilger, in Engler, Pßanzetireich, iv. 5, Taxaceœ, 42 (1903); Stiles, in New Phytologist, vu. 209- 222, figs. 28-34 (1908), and Ann. Bot. xxvi. 446, 463 (1912); R. B. Thomson, in Bot. Gaz. xlvii. 344-354, pi- 22-24 (i9°9)- A GENUS belonging to the division Podocarpeae of the order Taxacese, mainly characterised by the yew-like foliage with true scaly buds, and by the female cones composed of spirally and loosely imbricated carpellary scales, the uppermost of which are fertile, each bearing internally near the base a cavity from which hangs a single free minute ovule ; scales ultimately becoming fleshy, coalescing to form an irregular globose head, only a few of the ovules ripening into seeds, which when mature are set free by the gaping apart of the fertile scales. The genus, which has been studied by W. Stiles at Cambridge, is a remarkable one, all parts of the plant having a simple structure, suggesting that it is a primitive type. Saxegothœa is allied to Araucaria as well as to Podocarpus, resembling the latter in leaf, but the former in the female flowers and in the wingless pollen grains. Only one species is known, in the following description of which the other characters of the genus are given in detail. (A. H.) SAXEGOTHŒA CONSPICUA Saxegothœa conspkua, Lindley, in Journ. Hort. Soc. Land. vi. 258, figs. A and B (1851), and in Paxton, flower Garden, ii. in, fig. 190 (1852); Masters, in Gard. Chron. ii. 684, figs. 130, 131 (1887), and v. 782, fig. 125 (1889); Kent, Veitch's Man. Conif. 158 (1900); Pilger, in Engler, Pflanzenreich, iv. 5. Taxaceœ, 42 (1903); Castillo and Dey, Jeog. Vej. Rio Valdivia, 31, fig. 12 (1908). An evergreen tree, attaining in South America 30 to 40 ft. in height, becoming at high elevations a low dense shrub. Bark greyish brown, scaling off like that of a plane tree, leaving the reddish brown cortex beneath exposed in patches. Branches widely spreading, pendulous at the ends, giving off the branchlets in opposite pairs or in whorls of three or four ; young branchlets slender, glabrous, marked by the decurrent bases of the leaves, green with inconspicuous white dots on the lower side. Buds minute, globose, surrounded by three to seven ovate greenish scales, which persist brown and withered at the apex of the branchlet of the second year. Leaves, persistent about five years, arising in spiral order, spreading radially on 1458 Saxegothaea 1459 leading shoots, but thrown into a pectinate arrangement on lateral branches ; linear, straight or curved, \ to f in. long, narrowed into a petiolate base, decurrent on the branchlet, tapering at the apex, which ends in a sharp cartilaginous point ; upper surface dark green with a raised narrow midrib ; lower surface with a narrow green midrib, on each side of which is a broad stomatic band, composed of about twelve close lines of dots, and wider than the green margins. Flowers monoecious. Staminate flowers cylindric, solitary or two or three in the axils of the leaves near the end of the branchlet, shortly stalked, subtended by four to six scales ; composed of numerous spirally arranged anthers, each with two cells, dehiscing longitudinally. Ovuliferous flowers solitary on the ends of the branchlets, on short peduncles bearing a few modified leaves; succeeded by numerous, spirally arranged, densely imbricated, triangular-ovate, pointed carpellary scales, the lower sterile, the upper fertile, each with a depression on the upper surface near the base, in which is borne an inverted ovule. Fruiting head, ripening in the first year, irregularly globose, £ to \ in. in diameter, with fleshy scales coalescent at the base and free at the apex, containing six to twelve ripened seeds, which are set free by the separating of the fertile scales. Seed, about £ in. in width and length, smooth, shining brown, ovoid, compressed from back to front, with two sharp lateral edges ; base broad, marked by the rough surface of the hilum. Cotyledons two. This species is a native of Chile and western Patagonia,1 where it grows in the lower regions of the mountains in the dense forests, composed mainly of evergreen bush and conifers, like Fitzroya patagonica, Libocedms tetragona, Podo carpus chilina, and Podocarpus nubigena. Castillo and Dey say that these two species of Podocarpus and Saxegothœa are known in Chile as maniu, and yield a fine homogeneous wood, yellow in tint, and admirably suited for joiner's work.2 Saxegothœa conspicua was discovered3 by W. Lobb in southern Chile in 1846, and introduced in 1847. It does not appear to have succeeded in our climate, and is now very rare. At Kew, it is perfectly hardy, but grows slowly, and has a stunted appearance. There are two specimens at Strete Ralegh, near Exeter, the seat of Mr. Imbert Terry, both probably original introductions, and about 30 ft. high in 1909, when the larger was 4 ft. 3 in. in girth at two feet from the ground.4 There is a fine specimen at Coldrenick, about 35 ft. high, with a short bole, only a foot long, dividing into about nine stems, 4 to 6 in. in diameter, with wide- spreading branches, which I saw in 1911. Both this tree, and those at Strete 1 Dusen, in Scott, Princetown Univ. Exped. Patagonia, viii. 20 (1903), says it grows in the forests of the middle and lower Aysen valley in Patagonia. Reiche, Verbrat. Chilen. Conif. 5 (1900), gives the Aysen valley, lat. 45° 10', as its known southerly limit, while it extends northwards to the Rio Maule in lat. 35° 20'. It occurs on Chiloe, but not in the Guaytecas and Chonos islands. 2 Capt. Fitzroy, Narrative of Voyages of the Beagle, i. 280 (1839), says :—''Manu, a tree of great dimensions, tall and straight, the leaf is like that of a yew : it is a very useful wood for shipbuilding, for planks, and next to alerce, is the best for spars that the island of Chiloe produces ; but the large trees have a great tendency to become rotten at the heart owing possibly to the humidity of the climate, and to tbe very wet soil. Of twenty trees that were cut down, not one was sound at the heart. The wood is heavy, with large knots, which penetrate into the trunk to a great depth. A great deal of this timber grows in the Gulf of Fenas." It is doubtful what species is here referred to ; but Saxegothsea does not now occur so far south as the Gulf of Penas. 3 Hortus Veitchn, 38, 345 (1906). 4 Cf. Dallimore, in Kew Bull. 1909, p. 336. 1460 The Trees of Great Britain and Ireland Ralegh, produce flowers and fruit abundantly ; but 1 am not aware that seedlings have been raised. In the pinetum at Uplyme, Dorset, there is a remarkably fine specimen with a single stem, which was 40 ft. high and 3 ft. 4 in. in girth, when I saw it in July 1912. At Abbotsbury this species is represented by a shrubby plant, 4 ft. high and 4 ft. across. At Bury Hill near Dorking there is a tree about 20 ft. high. Bean1 saw in 1907 a specimen, about 12 ft. high, at Ochtertyre in Perthshire. The only specimens that we know of in Ireland are a spreading bush at Rostrevor, where, Sir John Ross-of-Bladensburg informs us, it was planted in 1891; and two bushy trees, about 20 ft. high, which Elwes saw at Kilmacurragh in 1908. (A. H.) 1 Card. Chron. xli. 168 (1907) TORREYA Torreya? Arnott, in Ann. Nat. Hist. i. 130 (1838); Bentham et Hooker, Gen. PL iii. 431 (1880); Eichler, in Engler and Prantl, Pflanzen/am, ii. i, p. in (1889); Masters, in Journ. Linn. Soc. (Bot.) xxx. 5 (1893); Pilger, in Engler, Pflanzenreich, iv. 5, Taxacece, 105 (1903); Sargent, in Bot. Gaz. xliv. 226 (1907). Tandon, Rafinesque, Amenities of Nature, 63 (1840); Sargent, Suva N. Amer. x. 55 (1896). Caryotaxus, Zuccarini, ex Endlicher, Syti. Conif. 241 (1847). EVERGREEN trees belonging to the order Taxaceae, with fissured bark and opposite or whorled branches. Young branchlets green, with linear pulvini, separated by slight grooves. Buds, one terminal, and occasionally two to four lateral, clustered at the end of the branchlet, composed of a few decussately opposite scales. Base of the branchlet marked with scars, left by the fall of the bud-scales of the previous season ; occasionally two or three of these persisting brown, unenlarged, and inconspicuous. Leaves spirally arranged, but thrown, by twisting and turning of their bases, into a pectinate arrangement on lateral branches, as in the yew ; persistent three or four years ; stalked, linear, tipped with a bristle-like cartilaginous point ; upper surface green, convex ; lower surface with a raised green midrib, and two white stomatic bands, sunk in longitudinal depressions ; fibro - vascular bundle undivided, with a solitary resin-canal beneath it. Flowers dioecious, or monoecious2 with the sexes on different branches. Staminate flowers solitary in the axils of the leaves of the current year's branchlet, surrounded at the base by several pairs of decussate scales, composed of numerous stamens, in whorls of fours, on a stipitate slender axis ; filament expanded into four pollen-sacs ; connective truncate or crest-like and dentate. Pistillate flowers in pairs on rudimentary branchlets,8 which are solitary in the axils of a few leaves towards the base of the current year's shoot ; each flower sub tended by four decussate scales and a bract, and consisting of a solitary terminal ovule, surrounded at the base by a small disc, the aril, which grows upwards and ultimately becomes confluent with the succulent testa of the seed. Seed, as only one flower of each pair develops, solitary ; ripening in the second year, drupe-like, with an outer succulent resinous coat and an inner woody shell, within which is the ruminate albumen and a minute embryo with two cotyledons. The shell bears a slightly projecting point at the apex, around which there is a dark-coloured oval, 1 According to the Vienna rules of nomenclature, the name Torreya is to be retained for this genus. It had previously been applied to a species of Clerodendron by Sprengel, Nette Entd. ii. 121 (1821). - The flowers are usually said to be dioecious, but the trees in cultivation are monoecious. 3 This occasionally terminates in a third ovule, or produces an extra bract. Cf. Miss Robertson, in New Phytologist, iii. 142 (1904). 1461 1462 The Trees of Great Britain and Ireland circular, or saddle-shaped area, representing the outer surface of the integument where it is free from the aril ; and below this are two opposite minute shield-like prominences, each with a minute aperture. The seed is subsessile or short-stalked, and subtended at the base by six decussate scales,1 from which it separates, when it falls after ripening. Seedling : see under T. californica. All species of Torreya sprout freely from the stump and roots when cut ; and bear pruning freely. They are propagated by cuttings2 or by grafting on the common yew or on species of Cephalotaxus. During the tertiary period the genus inhabited the Arctic region, and spreading southward existed in Europe. All the living species produce handsome, close- grained, pale yellow wood,3 useful for cabinet-making, and durable when placed in contact with the soil as posts for fencing. The species of Torreya superficially resemble those of Cephalotaxus ; but the two genera are readily distinguishable. Torreya.—Leaves with long spines at the apex, and narrow stomatic bands beneath. Buds with few decussate scales, all deciduous or two to four persisting minute and inconspicuous at the base of the branchlet. Cephalotaxus.—Leaves with short-pointed apices ; under surface with broad stomatic bands extending from the midrib almost to the outer margin. Buds with numerous imbricated scales, which persist at the base of the branchlet. The four living species of Torreya are distinguishable as follows :— I. Leaves and branchlets foetid. 1. Torreya taxifolia, Arnott. Florida. See p. 1466. Leaves linear, f to i^ in. long, -| in. broad, pale green beneath, with a broad midrib and narrow scarcely depressed stomatic bands ; petiole, ^s in. II. Leaves and branchlets pungent-aromatic. 2. Torreya californica, Torrey. California. See p. 1465. Leaves linear, ij to 3 in. long, ^ in. broad ; glaucous beneath with a broad midrib and slightly depressed narrow stomatic bands ; petiole, ^ in. 3. Torreya nucifera, Siebold and Zuccarini. Japan. See p. 1463. Leaves lanceolate-linear, f to \\ in. long, ^- to \ in. wide ; green beneath with deeply depressed broad stomatic bands, about as wide as the midrib. III. Leaves and branchlets devoid of any peculiar odour or taste. 4. Torreya grandis, Fortune. China. Seep. 1464. Leaves lanceolate-linear, \ to i in. long, \ in. wide ; thinner in texture than those of 71 micifera, but with similar stomatic bands. (A. H.) 1 These are the four decussate scales and bract of the flower which develops—the sixth scale being the bract and remains of the other flower which does not develop. 2 Rehder states, in Bailey, Cycl. Am. Hort. iv. 1822 (1902), that plants raised from cuttings grow slowly and remain bushy. This may account for the few good specimens of Torreya which exist in cultivation. The seeds are difficult to transport, as they soon become rancid. 3 Figured by Mayr, Fremdländ. Wald-u. rarkbaume, 423, pi. x. 23 (1906). Torreya TORREYA NUCIFERA Torreya nucifera, Siebold and Zuccarini, in Abh. Akad. München, iv. 3, p. 234 (1846); Franchet et Savatier, Enum. PI. Jap. i. 475 (1875) ; Masters, veijourn. Linn. Soc. (Bot.} xviii. 500 (1881), and xxvi. 546 (1902); Shirasawa, Icon. Ess. Forest. Japon, \. text 32, t. 15, figs. 19-34 (1900); Kent, Veitch's Man. Conif. 119 (1900); F. W. Oliver, in New Phytologist, i. 151, fig. 6 (1902); Pilger, in Engler, Pflanzenreich, iv. 5, Taxaceœ, 105 (1903). Taxus nucifera, Linnseus, Sp. PI. 1040 (1753); Thunberg, Fl. Jap. 275 (1784). Podocarpus nucifera, Persoon, Syn. ii. 633 (1807); Loudon, Are. et Frnt. Brit. iv. 2100 (1838). Caryotaxus nucifera, Zuccarini, ex Endlicher, Syn. Conif. 241 (1847) ; Henkel and Hochstetter, Syn. Nadelholz. 366 (1865). Tumion nuciferum, Greene, in Pittonia, ii. 194 (1891). A tree, attaining in Japan 80 ft. in height, with bright red bark, and pungent aromatic branchlets and leaves. Young branchlets green, glabrous, becoming reddish in the second year. Buds prismatic, about \ in. long, with six to eight decussate external scales. Leaves, f to \\ in. long, | to \ in. wide, lanceolate-linear, tapering from near the base to the narrow acuminate spine-tipped apex; shining dark or yellowish green above ; lower surface with two deeply depressed white stomatic bands, about as wide as the midrib, but not so wide as the marginal green bands ; petiole about Tle in. long. Staminate flowers, \ in. long ; connective crest-like, denticulate. Fruit narrowly obovoid, f to i in. long, green tinged with purple ; flesh thin, resinous ; shell light brown, with irregular longitudinal depressions; inner coat dark red, folded for a short distance into the albumen. (A. H.) This species is a native of Japan, occurring as a rare tree in the southern islands and in the forests of southern and central Hondo, where, according to Sargent, it reaches a height of 80 ft., and is remarkable for the beauty of its bright red bark and lustrous dark green foliage. The only place where I saw the tree wild was in virgin forest on the slopes of the Kireshima volcano in southern Kiusiu. Here there were trees up to about 10 in. in diameter, scattered in a dense forest of Abies firma and Tsuga, mixed with oaks, chestnut, and other broad-leaved trees. Seedlings were found in dense shade and could be distinguished by the sharp points of their leaves. I was told by the foresters that much larger trees existed, but only in remote and inaccessible places. Where it was planted in Kisogawa and below Koyasan at 1500 ft. elevation, it attained a height of 40 to 50 ft. with a girth of 8 ft. or more ; and had very much the habit of a yew, but made a cleaner trunk with less tendency to branch. It is known in Japan as Kaya ; and an oil is extracted from the seeds, which is used in cooking. The wood, according to Rein, is light yellowish in colour, and is used for making chests and boxes. The wood is also used for making Japanese chess-men. The Japanese Torreya was described and figured by Kaempfer, Amœn. Exot. 815 (1712); and seems to have been introduced1 into England in 1764 under the 1 Aiton, Hort. Kew. v. 416 (1813). VI S 1464 The Trees of Great Britain and Ireland name Taxus nucifera. Loudon gives the date of introduction as 1820, and states that a tree at White Knights was 13 ft. high in 1834. Siebold sent it to Holland about 1840. It is rare in cultivation, and is never seen except as a shrub. (H. J. E.) TORREYA GRANDIS Torrtya grandis, Fortune, in Gard. Chron. 1857, p. 788, and 1860, p. 170, and in Gordon, Pinetum, 326 (1858); Masters, in Journ. Linn. See. (Hot.) xviii. 500 (1881), xxvii. 323, fig. 28 (1890), and xxvi. 546 (1902), and in Gard. Chron. ii. 681, fig. 117 (1884); Franchet, PI. David. \. 292 (1884). Torreya Fargesii, Franchet, in Journ. de Bot. xiii. 264 (1899); Pi'geri in Engler, Pflanzenreich, iv. 5, Taxacece, 108 (1903). Torreya nucifera, Siebold and Zuccarini, var. grandis, Pilger, op. dt. 107 (1903). Caryotaxus grandis, Henkel and Hochstetter, Syn. Nadelholz. 366 (1865). Tumion grande, Greene, in Pittonia, ii. 194 (1891). A tree, attaining 8o ft. in height in China, with leaves and branchlets devoid of a disagreeable or pungent odour. Young branchlets green, glabrous, becoming yellowish brown in the second year. Leaves, \ to i in. long, \ in. broad, linear- lanceolate, similar to those of T. nucifera in shape, but thinner in texture, with similar deeply depressed stomatic bands, nearly as wide as the midrib, but narrower than the marginal green bands. Fruit broadly ellipsoid, f to i in. long ; flesh not disagreeable in odour ; shell reddish brown, with irregular shallow depressions over the surface ; inner coat only slightly folded into the albumen. Torreya grandis was discovered by Fortune in 1855 in the coast province of Chekiang in China, in the mountains south-west of Ningpo, at 4000 ft. elevation. Here numerous fine trees were seen, many of which were 60 to 80 ft. in height. It was subsequently collected in the adjoining province of Fukien by Père David. The same tree1 also exists in the central provinces of Hupeh and Szechwan, in the mountains between 4000 and 6000 ft. elevation ; where it occasionally attains a height of 50 ft., but it is more commonly shrubby, bearing fruit when only 8 ft. high. This tree is known to the Chinese as fei ; and the kernels, called fei-shih? are sold in the drug-shops of most Chinese towns, being considered a valuable remedy in cases of cough, asthma, etc. They are occasionally eaten like hazel nuts, and though reputed laxative, are considered wholesome. Torreya grandis was introduced by Fortune, who sent seeds in 1855 to Glen- dinning's nursery at Chiswick, where they germinated freely. This species is not common in collections, the only specimen which we have seen being a small shrub in the Cambridge Botanic Garden, which was obtained from Veitch in 1894. (A. H.) 1 The Torreya of Central China is considered by Franchet and Pilger to be a distinct species, T. Fargesii ; but I can see no characters by which it can be separated from the Chekiang species. 2 Cf. Hanbury, Se. Papers, 233 (1876). Torreya TORREYA CALIFORNICA Torreya californica, Torrey, in New York Journ. Pharm. iii. 49 (1854); J. D. Hooker, in Gard. Chron. xxiv. 553 (1885); Masters, in Gard. Chron. v. 800, figs. 126, 127 (1889); Kent, Veitch's Man. Conif. 117 (1900) ; Pilger, in Engler, Pflanzenreich, iv. 5, Taxaceœ, 109 (1903) ; Jepson, Silva of California, 167 (1910). Torreya Myristica, J. D. Hooker, in Bot. Mag. t. 4780 (1854); Murray, in Trans. Bot. Soc. Edin. vi. 217, pi. iii. (1860); Masters, in Gard. Chron. xxii. 681, fig. 116 (1884). Caryotaxus Myristica, Henkel and Hochstetter, Syn. Nadelh. 368 (1865). Tumion californicum, Greene, in Pittonia, ii. 195 (1891); Sargent, Silva JV. Amer. x. 59,1. 513 (1896), and Trees N. Amer. 98 (1905). A tree, attaining in California 100 ft. in height, and 12 ft. in girth, usually considerably smaller. Leaves, branches, and wood pungent-aromatic. Young branchlets glabrous, green, becoming brown in the second year. Buds, up to 5 in. long, prismatic, with eight to ten decussate outer scales, those towards the apex elongated. Leaves, ij to 3 in. long, -| in. wide ; linear, tapering in the anterior third to an acuminate spine-tipped apex ; dark shining green above ; lower surface flat, glaucous, with two slightly depressed white stomatic bands (about 0.3 mm. wide), much narrower than the broad midrib (about i mm. wide), and the two outer glaucous bands (each about 0.7 mm. wide) ; petiole stout, ^ in. long. Staminate flowers \ in. long ; connective truncate, not dentate. Fruit1 ellipsoid or obovoid, i to if in. long ; light green streaked with purple ; flesh thin, resinous ; shell fawn-coloured, smooth or with irregular slight longitudinal ridges ; inner coat reddish, deeply folded into the white albumen. Seedlingz similar to that of Ginkgo biloba, with two thick and fleshy cotyledons, remaining underground ; the stem bearing below a few scales, which are succeeded by ordinary leaves, the transition between the scales and the leaves being gradual. (A. H.) This species is a native of California, growing on the borders of mountain streams, nowhere common, but widely distributed from Mendocino County to the Santa Cruz mountains in the coast region, and along the western slopes of the Sierra Nevada from Eldorado to Tulare County, at 3000 to 5000 ft. elevation. It is most abundant and of its largest size in the northern coast ranges. Hough, in American Woods, pt. vi. p. 50, describes a fine tree overthrown by a flood near the coast in Mendocino County, from which the specimens of wood in his book were cut. It was 85 ft. long to the point, 5 in. in diameter, where its dead top was broken off. Its straight columnar trunk was 12 ft. in girth at eighteen inches, and 8 ft. at thirty- five feet from the ground. Assuming that the growth of this tree had been as uniform as in the section, which shows ten rings to the inch, it would have been from 250 to 300 years old, and the contents of the log about 300 cubic feet. 1 Described from specimens grown at Orton. Cf. F. W. Oliver, in Ann. Bol. xvii. 466, pi. xxiv. (1903), for a detailed description of the remarkable structure of the seed of this species. 2 Miss Chick (Mrs. Tansley) describes seedlings grown from seed produced at Orton, in New Phytologist, ii. 83, plates vii. viii. (1903). 1466 The Trees of Great Britain and Ireland Hough states that it is an excellent, light, and durable wood, well suited for boat building and cabinetmaking, but too rare to be generally known or used. Jepson adds that the wood is strongly odorous, with white sap wood and clear light yellow heartwood, susceptible of a fine polish. This tree was discovered in 1851 by William Lobb, who sent specimens and seeds in that year to Messrs. Veitch, which were described as T. Myristica in 1854 by J. D. Hooker, a short time after it had been published as T. californica by Dr. Torrey, who had received specimens from Mr. Shelton. Though hardy in most parts of Great Britain, this species requires a heavy rain fall together with rich deep soil to grow to any size ; and owing to its being usually planted in open situations, instead of in the densely wooded ravines which it likes, it has a tendency to spread and form branches rather than make a trunk. By far the largest specimen that I know is at Tregothnan, which, when I first saw it in 1905, was 35 ft. by 6 ft.; and in 1911 had increased to 45 ft. high. Its shape is spoilt by three large lower branches ; its trunk shows some large nipple- like protuberances resembling those often seen on the deodar. Another tree at Orton Hall, Peterborough, has produced fruit regularly for the last twenty years, from which numerous seedlings1 have been raised. It measured about 25 ft. by 4 ft. in 1905. At Poles, near Ware (Herts), a tree, which has never flowered, measured 40 ft. by 4 ft. in 1910. It was planted in 1858. At Tortworth,2 there is a handsome specimen, 25 ft. by 4 ft. in 1904, with branches spreading to a diameter of 36 ft. ; and at Westonbirt, there is a tree in a shaded position, 29 ft. by 2 ft. 9 in. in 1907. At Chipping Campden, there is a tree about 30 ft. high, which bore staminate flowers in profusion in May 1911. In Scotland it succeeds as far north as Durris in Kincardineshire, where a healthy specimen is about 20 feet high. In Ireland, the best example is at Verner's Bridge, near Lough Neagh, where a tree measured 40 ft. by 4 ft. 7 in. in 1904. There are also specimens at Fota and Castlemartyr ; the latter, forking near the ground, was 30 ft. high when I saw it in 1908. (H. J. E.) TORREYA TAXIFOLIA Torreya taxifolia, Arnott, in Ann. Nat. Hist. i. 130 (1838); W. J. Hooker, Icon. Plant, tt. 232, 233 (1840); Kent, Veitch's Man. Conif. 119 (1900)5 Pilger, in Engler, Pflanzenreich, iv. 5, Taxacea, 108 (1903). Taxus montana, Nuttall, mjourn. Acad. St. Phil. vii. 96 (1834) (not Willdenow). Caryotaxus taxifolia, Henkel and Hochstetter, Syn. Nadelh. 367 (1865). Tumion taxifolium, Greene, in Pittonia, ii. 194 (1891); Sargent, Silva N. Amer. x. 57, t. 512 (1896), and Trees N. Amer. 98 (1905). A tree, attaining in Florida 40 ft. in height and 6 ft. in girth, with foetid leaves, branches, and wood. Young branchlets green, with occasional minute hairs ; older 1 These grow very slowly at Colesborne and are now at eleven years old under two feet high. 2 A section from another tree which died at Tortworth was sent by the Earl of Ducie to the Cambridge Forestry Museum. Torreya 1467 branchlets yellowish red. Buds as in T. californica. Leaves f to i J in. long, \ in. wide ; linear, tapering in the anterior third to an acuminate spine-tipped apex ; shining green above ; lower surface pale green, with two stomatic bands, scarcely depressed, and narrower than the midrib and the external bands ; rounded at the base, with a short petiole, about ^ in. long. Staminate flowers, ^ in. long ; connective minute, rounded, not dentate. Fruit obovoid, i to i^ in. long, dark purple ; flesh foetid, coriaceous ; shell smooth, light brown ; inner coat brownish and not so deeply folded into the albumen as in T. californica. This species, of which I have seen no living specimen, appears to be very similar to T. californica, but has smaller leaves, with very short petioles, and, according to Sargent, is different in odour. T. taxifolia is restricted to north-western Florida, where it grows on limestone cliffs and in swamps1 along the banks of the Appalachicola River, from River Junction to near Bristol in Gadsden County. The wood, according to Sargent, is hard, strong, clear bright yellow, with thin lighter-coloured sapwood, and is used locally for fence- posts. Owing to the peculiar odour of the whole tree, noticeable also in the wood when burnt, it is known as "stinking cedar." Dr. Torrey informed Arnott that a blood-red turpentine, of a pasty consistence, flows sparingly from the bark, and is soluble in alcohol. This species was discovered in 1833 by H. C. Croom, and was introduced2 into England, in 1840, by A. J. Downing, who sent a living plant to London, which was propagated3 by Masters of the Canterbury Nursery. It probably proved unsuitable to our climate ; and I have seen no specimen which could be identified with this species. According to Sargent4 it can be kept alive in eastern Massachusetts in sheltered, well-shaded situations ; and occasional individuals have survived a number of years near New York and Philadelphia. (A. H.) 1 Cf. Britton, N. Amer. Trees, 126 (1908). 2 Loudon, Gard. Mag. xvi. 658 (1840). * Loudon, Trees and Shrubs, 944 (1842), states that it was propagated by grafting on the yew. 4 Sargent, in Garden and Forest, x. 400 (1897). CEPHALOTAXUS Cephalotaxus, Siebold and Zuccarini, ex Endlicher, Gen. Suppl. ii. 27 (1842); Bentham et Hooker, Gen. PL üi. 430 (1880); Masters, in Journ. Linn. Soc. (Bot.') xxx. 4 (1893), and in Gard. Chron. xxxiii. 227 (1903); Worsdell, in Ann. Bot. xv. 637 (1901); Pilger, in Engler, Pßanzenreich, iv. 5, Taxaceœ, 99 (1903). EVERGREEN shrubs or small trees, belonging to the order Taxaceœ, with opposite or whorled branches. Young branchlets green, marked by white stomatic dots, and with linear pulvini, separated by slight grooves. Buds, with numerous imbricated scales, which persist as a conspicuous sheath at the apex of the branchlet of the second year. Leaves spirally arranged, radially spreading on vertical shoots, but on lateral branches thrown by twisting of their bases into a pectinate arrangement ; persistent three or four years, very shortly stalked, linear, acute at the apex ; upper surface green with a prominent midrib in a depression ; lower surface with two whitish broad bands, composed of numerous stomatic lines, separated by a narrow raised green midrib, and bounded on each outer side by a very narrow marginal green band ; fibro-vascular bundle undivided, with a single resin-canal beneath it. Flowers dioecious.1 Staminate flowers in globose heads, which are solitary in the axils of the leaves of the branchlets of the previous year ; each head with six to eleven flowers, each of which is subtended by a bract and has seven to twelve stamens ; pollen-sacs two or three, dehiscing longitudinally.2 Pistillate heads few> each solitary in the axil of a scale-leaf near the base of the branchlet of the current year, and composed of a stipitate axis, towards the end of which are three or four decussate pairs of opposite bracts ; each bract is cup-shaped at the base and bears two erect ovules side by side. Usually only one or two of the ovules in a head develops, forming a drupe-like seed, with a fleshly outer covering, and an inner hard woody shell, which encloses the albumen and embryo. The seedling3 has two long linear cotyledons, immediately above which and decussate with them on the stem is a pair of primary leaves, which are followed at intervals by either whorls or pairs of larger leaves. Six species* of Cephalotaxus are known, one of which is possibly a hybrid, 1 In rare cases, the flowers are monœcious, as in a shrub of C. Fortmii, described by Carrière in Rev. Hort. 1878, p. 116, fig. 24. 2 Kerner, Nat. Hist. Plants, Eng. Trans, ii. 124 (1898), states that the anthers open and shut periodically. 3 Cf. Masters, in Journ. Linn. Soc. (Bot.) xxvii. 241 (1889). 4 C. Mannii, Hooker, native of the Khasi Mountains ; C. Griffithii, Hooker, of Assam and Manipur ; and C. Oliveri, Masters, of Central China, are not now in cultivation. The plants of C. Griffithii, formerly in the temperate house at Kew, mentioned by Hooker, Fl. Brit. India, v. 648 (1890), died many years ago. The young plants at Coombe Wood, referred to C. Ouvert in Card. Chron. xxxiii. 227 (1903), and Hortus Veitchii, 338 (1906), are C. drupacea. Podocarptis argotania, Hance, a peculiar conifer in southern China, is referred to Cephalotaxus by Pilger ; but is distinct. 1468 Cephalotaxus 1469 natives of China, Japan, Khasi Mountains, Assam and Manipur. As none of the three species in cultivation forms a tree in this country, they do not properly come within the scope of our work ; but are now briefly described, owing to their interest as conifers, which are frequently seen in gardens. All the three species are perfectly hardy ; but succeed best in shady situations, sheltered from the wind. When propagated by scions or cuttings, terminal shoots should be selected, as these form regular plants with whorled branches like seedlings ; whereas cuttings from lateral branches grow into irregular low spreading shrubs.1 (A. H.) CEPHALOTAXUS DRUPACEA Cephalotaxus drupacea, Siebold et Zuccarini, Fl. Jap. Fam. Nat. ii. 108 (1846), and Fl. Jap. ii. 66, tt. 130 and 131 (1870); Franchet et Savatier, Enum. Fl. Jap. i. 473 (1875); Masters, in Gard. Chron. xxi. 113 (1884), and xxxiii. 228, fig. 94 (1903), and Journ. Linn. Sot. (Bot.) xxii. 201 (1886) and xxvi. 544 (1902); Kent, Veitch's Man. Conif. 112 (1900); Shirasawa, Icon. Ess. Forest. Japon, \. text 31, t xiv. figs. 1-12 (1900); Pilger, Taxaceœ, ioo, figs. 19, 20 (1903); Hemsley, in Bot. Mag. 8285 (1909). Taxus baccata, Thunberg, Fl. Jap. 275 (1784) (not Linnseus). A large shrub or small tree, occasionally attaining in Japan 40 ft. in height. Leaves on lateral branches pectinate, but spreading outwards usually in a V-shaped arrangement and not remaining in one plane, linear, f to i^ in. long, straight or falcate, tapering towards the triangular acute apex, which is often tipped with a short spine-like point ; stomatic bands beneath, each composed of about thirteen to fifteen lines. Staminate heads \ in. in diameter, on very short scaly stalks. Fruit brown, f to i in. long, pyriform, broadest at the rounded apex, which has a circular depression bearing a minute mucro, and narrowed towards the base ; kernel light brown, smooth, ellipsoid, \ in. long, | in. wide, rounded at the base, the two sharp lateral edges in the upper half uniting into a slight apiculus at the apex. This species is a native of Japan and central China. In Japan,2 it is generally scattered through the mountain forests, extending northward to central Hokkaido, where it grows on low hills as an undershrub, 2 or 3 ft. high, of the deciduous forest ; while in Hondo, where it ranges between 2000 and 3000 ft. altitude, it becomes a bushy tree, averaging 25 ft. high, and occasionally attaining 40 ft.3 It is known to the Japanese as Inu-gaya. In China, it has been found in the mountains of Hupeh and Chekiang and in the Chusan Archipelago. C. drupacea was introduced by Siebold into the Botanic Garden at Leyden in 1829; but does not appear to have been known* in English gardens till 1844. It i Cf. Rehder, in Bailey, Cyd. Amer. Hort. 276 (1900). * A species of Cephalotaxus, not yet determined, occurs in the mountains of Formosa at about 8000 feet altitude. Cf. Hayata, \njourn. Coll. Sei. Tokyo, xxv. 215 (1908). Elwes saw this as a bush without flowers or fruit on Arisan in 1912. » Mayr, Frcmdland. Wold- u. Parkbaume, 269 (1906), says that in the warmer parts of Japan it becomes a tree, rarely attaining 60 ft. in height. 4 Nicholson, Card. Diet. i. 294 (1884), gives 1844 as the date of introduction into England. It was mentioned as a cultivated plant by Knight and Perry, Syn. Conif. 51 (1850), who gave for it the synonyms Taxus coriacea and Cephalotaxus coriacea. 147° The Trees of Great Britain and Ireland always remains a spreading bush in this country, an old specimen at Kew being about ID ft. high and 15 ft. through, and clothed to the ground with luxuriant dark green foliage. (A. H.) CEPHALOTAXUS FORTUNI Cephalotaxus Fortuni, W. J. Hooker, in Bot. Mag. t. 4499 (1850); Masters, in Gard. Chron. xxi. 114, fig. 2i (1884), and xxxiii. 228 (1903), and Journ. Linn. Soc. (Bot.) xxvi. 545 (1902); Franchet, PI. David, i. 292 (1884), and in Journ. de Bot. 1899, p. 265 ; Kent, Veitch's Man. Conif. 113 (1900); Pilger, Taxaceœ, 103 (1903). A small tree or large shrub, attaining the same dimensions as C. drupacea. Leaves on lateral branches pectinate and spreading in one plane, falcate, i^ to 3 in. long, tapering gradually in the anterior third to an acuminate, usually spine-tipped apex ; stomatic bands beneath conspicuously white, each of eighteen to twenty-one lines, covering nearly the whole surface, the midrib and green margins being narrower than in C. drupacea. Staminate heads, less than \ in. in diameter, on scaly stalks, which are \ in. long. Fruit olive-green, about i^ in. long and f in. in diameter, elongated ovoid, contracted towards the base, and broadest at the rounded apex, which is tipped with a short elevated point, arising from a circular depression ; kernel elongated ellipsoid, about i in. long and | in. wide, light mottled brown, minutely tuberculate on the surface, rounded at the base ; upper half with two sharp lateral edges which unite at the apex to form an apiculus. This species is occasionally monoecious.1 In the Cheshunt Nurseries,2 there was a female plant, which bore fruit in 1862 and 1863, though no flowers were borne on a staminate plant close by ; and it was supposed to have been pollinated by a yew ; but the seeds were not sown, and may have been infertile. C. Fortuni is a native of China, occurring in the mountain woods of Szechwan, Hupeh, Yunnan, Kiangsu, and Chekiang, where it usually grows as a large bush in the shade of broad-leaved trees. It was discovered in the mountains south-west of Ningpo,3 in 1848, by Fortune, who sent seeds to the Bagshot Nursery in that year, which germinated freely. It is perfectly hardy in this country, and may be seen in many collections of conifers, forming a spreading shrub, with handsome foliage. The largest that we have seen is about 25 ft. high at Coldrenick. It appears to be little known in America.4 (A. H.) 1 Cf. Carrière, in Rev. Hart. 1878, p. 116, fig. 24, where a branch, bearing both male and female flowers, is depicted. * Cf. Card. Chron. 1863, p. 1062. 3 Bretschneider, Hist. Europ. Bot. Disc. China, 502 (1908), points out that the statement usually made, that Fortune found it in North China, is erroneous. 4 Sargent, in Garden and Forest, x. 391 (1897). Cephalotaxus 1471 CEPHALOTAXUS PEDUNCULATA Cephalotaxus pedunculata, Siebold and Zuccarini, Fl. Jap. Fam. Nat. ii. 108 (1846), and Fl. Jap. \\. 67, t. 132 (1870); Franchet et Savatier, Enum. PI. Jap. i. 473 (1875); Franchet, PI. David. \. 292 (1884); Masters, in Gard. Chron. xxi. 113, fig. 22 (1884), and xxxiii. 228 (1903), and Journ. Linn. Soc. (Bot.) xxii. 201 (1886), and xxvi. 545 (1902); Kent, Veitch's Man. Corny. 114 (1900); Diels, Flora von Central China, 214 (1901). Taxus Jlarringfonia, Knight, ex Forbes, Pin. Wolurn. 217, t. 66 (1839); Loudon, Gard. Mag. xv. 273 (1839), and Trees and Shrubs, 942 (1842). Cephalotaxus Harringtonia, Koch, Dendrologie, ii. 2, p. 102 (1873). Cephalotaxus drupacea, Siebold and Zuccarini, var. pedunculata, Miquel, Pro!. Fl. Jap. 333 (1867). Cephalotaxus drupacea, Siebold and Zuccarini, var. Harringtonia, Pilger, Taxaceœ, 102 (1903). A large shrub or small tree. Leaves on lateral branches, pectinate, spreading either in one plane or in a V-shaped arrangement, straight or falcate, i^ to 2^ in. long, tapering in the anterior third to an acute apex, which is often tipped with a spine ; stomatic bands beneath each of sixteen to twenty-one lines, not so white as in C. Fortuni. Staminate heads in clusters of two to five, or occasionally solitary, on scaly peduncles, which are % to i in. in length. Fruit olive-green, f to i in. long, ellipsoid, not contracted at the basal end, with a circular depression at the apex from which arises a short mucro ; kernel similar in size and shape to that of C. drupacea, but mottled light brown and slightly tuberculate on the surface. The following varieties have been described :— 1. Var. sphœralis, Masters, in Gard. Chron. xxi. 113, fig. 23 (1884), and in Journ. Linn. Soc. (Bot.} xxii. 203, plate vii. (1886). Fruit smaller, globose, not depressed at the apex, which bears a long mucro. This was described by Masters from a specimen growing in the Rev. J. Goring's garden at Steyning ; and a shrub at Kew has also borne similar fruit. 2. Var. fastigiata, Carrière, Prod, et Fix. Var. 44, fig. i (1865), and Conif. 717 (1867); Masters, in Gard. Chron. xxi. 113, fig. 20 (1884). Cephalotaxus Buergeri, Miquel, Prol. Fl. Jap. 333 (1867). Podocarpus koraiana, Siebold, in Ann. Soc. Hort. Pays-Bas, 1844, p. 34 ; Carrière, in Rev. Hort. 1863, p. 349, fig. 36; Maximowicz, Mil. Biol. vii. 563 (1870). A fastigiate form, similar to the Irish yew in habit ; branches and branchlets directed vertically upwards ; leaves spreading radially on the branchlets. This handsome shrub appears to have originated in Japan, whence it was introduced in 1861 into England, where it is perfectly hardy.1 It has never, so far as is known, borne flowers either in Japan or in Europe ; and is always propagated by cuttings. It frequently produces near the base lateral branches * with normal foliage ; and grafts, that are taken from these branches, reproduce the ordinary form of the species. 1 Sargent, in Gardin and Forest, x. 391 (1897), states that it is not hardy in eastern New England, but there are good plants near New York and Philadelphia. z Figured by Masters, mjoura. Linn. Soc. (Bot.) xxvii. 245, fig. 5 (1889), and Card. Chron. xxxiii. 227, fig. 96 (1903). Cf. De Vries, Mutation Theory, no, fig. 16 (1911), who instances these reverted hranches as showing the phenomenon of atavism by bud-variation. VI T i472 The Trees of Great Britain and Ireland C. pednnculata has been long in cultivation in Japan, where it is known as Chosen-gaya or To-gaya, meaning Korean or Chinese Cephalotaxus ; and was introduced there in ancient times from Korea or China by the Buddhist monks. It is unknown in the wild state, and in all probability is a hybrid between C. Fortuni and C. drupacea, which originated in China, where these two species are both native. It usually resembles more the former species in foliage, and the latter species in fruit ; but differs from both in the clustered staminate heads, which is possibly an abnormal condition. There are plants in gardens reputed to be, but not exactly matching C. Forhmi, which may be seedlings of C. pedun- culata. The latter species has leaves of a darker hue than C. Fortuni and C. drupacea ; and is equally hardy, but is scarcely so ornamental as the true C. Fortimi, which has the leaves much whiter beneath. The original C. pedunculata, long cultivated in Japan, was always a male plant, no doubt propagated by grafts and cuttings ; and it was introduced1 into England in 1837. So far as can be ascertained, the history of the female plant is as follows :—The seeds of C. Fortnni, which were sent by Fortune2 from China in 1848 to the Bagshot Nursery, produced two kinds of plants; one kind with long leaves, identical with the true wild plant of C. Fourni ; and the other kind with shorter leaves, identical with C. pediwctilata, and comprising individuals which bore fruit.3 (A. H.) 1 Cf. Loudon, Trees and Shrubs, 943 (1842). It appears to have been introduced by Siebold into Holland in 1829. 2 Cf. Fortune, in Gard, dirait. 1863, p. 1134. 3 \V. Gorrie, in Card. Chron. 1861, p. $\, points out that the shorter-leaved plants bearing fruit were certainly not C. drupacea. Fortune, believing that these plants constituted a new species, sent specimens from Chekiang in 1858, which are now preserved in the Lindley herbarium at Cambridge. These specimens, however, are simply a Ç branch of C. Fortnni, and a y Mayr to Pseudolarix Fortunei. 1477 1478 The Trees of Great Britain and Ireland Short shoots with annual zones of growth, each zone marked by a depression and a ring of subulate scales. Buds of three kinds, as in the larches ; those terminating (a) the long shoots and (ti) the short shoots, conic, surrounded by acuminate scales, ending in long subulate points ; and (c) lateral buds, solitary in the axils of a few leaves of the long shoots, globose, with rounded or short-pointed scales. Leaves deciduous, solitary and spirally arranged on the long shoots, and in clusters of fifteen to thirty at the apices of the short shoots ; jointed at the base with the tip of a pulvinus, linear, straight or falcate, i^ to 2 in. long, TV m- broad, acute or acuminate, green and slightly convex above ; under surface with a raised green midrib, two longitudinal channels covered with white stomatic lines, and a narrow thin outer margin. Fibro-vascular bundle undivided ; resin-canals three, all close to the epidermis, one in the median line near the upper surface, and two lateral, near the outer edges of the lower surface. Flowers monoecious. Staminate flowers, pendulous, twenty-five to thirty in a cluster, at the apex of a leafless short shoot, each subtended by loose scarious scales, and including the slender stalk about f in. long ; anthers twenty, two-celled, opening transversely ; pollen grains winged as in Pinus, and different from the simple pollen of Larix. Pistillate flowers globose, f in. in diameter, terminating a short leafy branch, which arises from the apex of a short shoot ; ovules, two on each scale, reversed. Cones, erect on the branches, ripening in the autumn of the first year, ovoid, i|- to 2 in. long : scales numerous, imbricated, coriaceous, reddish brown when ripe, f to i£ in. long, ovate, tapering to a blunt, acute, or notched apex, sagittate at the base, with a claw bent upwards at a right angle, which arises by a narrow linear attachment from the axis of the cone : bract, ovate-lanceolate, £ to ^ in. long, acuminate, denticulate, adnate to the base of the scale, and deciduous with it. Seeds, two on each scale, which they completely cover with their short body and long wing ; wing oval-lanceolate on the outer edge, straight on the inner edge, pale brown, trans lucent, enclosing the body of the seed on the front and sides in a cavity ; body detachable from the wing, white, obovate, with two large resin-vesicles ; cotyledons five to seven. As the cone ripens the scales gape apart, showing the wings of the seeds projecting beyond them, and giving them the appearance of a whitish margin. Soon afterwards the scales, bracts, and seeds fall together to the ground, the central axis of the cone being the only part of it left on the branch, as is the case in Cedrus and Abies. This remarkable conifer is a native of the provinces of Chekiang and Kiangsu in eastern China, where it is known from two localities, both in about lat. 29° 30'. Fortune discovered it in 1853, in the mountains south-west of Ningpo, where there were some fine trees growing near the Tsan-tsin monastery1 at 1000 to 1500 ft. elevation ; and in 1854 he found a plantation, about twenty miles westward, in the vicinity of the Quan-ting monastery, on a mountain slope at about 4000 ft. altitude, one of the trees, standing alone and clothed with branches to near the ground, being 130 ft. in height and 8 ft. in girth. The Rev. G. E. Moule also found 1 This monasteiy is about a day's journey from Ningpo. Cf. Bretschneider, Hist. Europ. Bot. Disc. China, 416 (1898). Pseudolarix 1479 some trees1 in the hills west of Ningpo in 1874. The only other locality where this species has been seen is the Lüshan mountains, south of Kiukiang in Kiangsu, where it was discovered by Abbé David in 1868, and afterwards by Maries,2 who mentions immense trees ; but Wilson8 only succeeded in finding in these mountains, some wild, and half a dozen planted trees, none of considerable size. Fortune had been acquainted with this tree for some years previously as a dwarf plant4 in pots, contrived, though only i^ to 2 ft. high, to look like an aged Cedar of Lebanon. It appears to be known to the Chinese, as either chin-sung, " golden pine," or chin-lo-sung, "golden deciduous pine," names applied on account of the beautiful yellow colour of the foliage for a short time before it falls in autumn. Fortune sent seeds from Chekiang in 1853, and again in 1855 to Glendinning's nursery, Chiswick ; but he states6 that of all the packages of seed, which he sent for several years in succession, only one batch ever germinated ; and that the only plants living in England in 1860 were natural seedlings which had been dug up in the woods of China and sent6 home in Wardian cases in 1854. The tree is perfectly hardy, as it withstood the severe winter of 1859-1860 at Ambleside7 and at Hafodunos7; and possibly its rarity in collections is due to the small number of plants actually introduced ; but it appears to be extremely slow in growth in England. It will not endure lime in the soil, as the seedlings raised and planted at Colesborne soon die. (A. H.) REMARKABLE TREES The finest tree that we know of is in a sheltered situation at Carclew, which, when I saw it in 1902, was 35 ft. by 5 ft. In 1910 it was 40 ft. by 5 ft. 2 in. I am told by Mr. Simmons, the head-gardener at Carclew, that he has never found fertile seed on it. There is a healthy tree in the grounds at Hutley Towers near Ryde, which in 1906 was 30 ft. by 2 ft., but it produced no cones either in 1905 or 1906. At Joldwynds, near Dorking, the seat of Sir W. Paget Bowman, Bt., there is a tree, planted about 1879, which is now 27^ ft. by 2\ ft. Though perfectly healthy and branching to the ground it has never produced any seed. There are several trees at Kew, the largest of which near the main gate is probably one of the original seedlings.8 It flowered profusely in June 1907, producing fully developed cones with imperfect seed, no embryo being formed.9 It bore cones freely again in 1910; and measured 31 ft. by 2 ft. 4 in. in 1912. At Tortworth, a tree measured in 1910, 37 ft. by 4 ft. i in. Lord Ducie informs me that it was planted on 3rd November 1858, in a bed of sand over lying carboniferous limestone, deep enough, however, to sustain a heavy growth of 1 Referred by Hance to Larix dahtirica, but evidently, from the Chinese name "chin-sung" used by Moule, he was speaking of Pseudolarix. 2 Cf. Bretschneider, op. cit. 741 (1898). 3 In Card. Chron. xlii. 344 (1907). 4 Var. nana, Masters, injoura. Linn. Soc. (Bot.) xviii. 523 (1881). B In Card. Chroti. 1860, p. 170. 6 Fortune, in Card. Chron. 1855, p. 644. Kent says that until the tree coned at Pallanza, plants were obtained by layering. 7 Card. Chron. 1860, pp. 74 and 386. 8 It is apparently the tree mentioned by J. Smith, Records of Kew Gardens, 290 (1888), as 5 ft. high in 1864. * According to Masters, in Card. Chron. ii. 440 (1887) and Journ. Roy. Hon. Soc. xiv. 68 (1892), this species first produced fruit in England in 1887 at Lucombe and Pince's nursery, Exeter, but this tree is no longer living. VI U 1480 The Trees of Great Britain and Ireland Rhododendron. The tree is very brittle, and has lost many branches, and is somewhat ragged in look although it is in a very sheltered position. It produced cones on four or five occasions ; but in no case did they contain fertile seed. The Rev. Hon. W. Ellis informs me that at Bothelhaugh, near Morpeth, Northumberland, a tree planted over thirty years ago is under 6 ft. high. At Coombe Wood, the largest specimen measured 35 ft. by 3 ft. 8 in. in 1910. At Scorrier, near Truro, a healthy tree was 30 ft. by 3 ft. in 1911. We have seen no trees in either Scotland or Ireland. The finest specimen in Europe is growing in Rovelli's nursery at Pallanza ; and measured 64 ft. by 6 ft. 10 in. in 1909. This tree produced staminate flowers in 1884 ; and since then has coned regularly every two or three years. Large numbers of natural seedlings appear in prepared soil under the tree ; and the seed is said to germinate better where it falls, than when collected and sown in pans under glass. As the seeds and scales fall together and close to the parent tree, young seedlings probably succeed best with considerable shade. In Belgium it is said1 to have attained no less than 46 ft. in height by 3 ft. in girth at the nursery of the Horticultural Society of Calmpthout, near Antwerp, where seedlings grew as fast as those of the common larch; and Dr. Masters mentioned,2 in 1883, a fine tree in Linden's nursery at Ghent. At Verrières, near Paris, one of the original trees8 is about 35 ft. high and 3 ft. in girth. It produces fruit and fertile seed, but in no great quantity. There is a good specimen * at Karlsruhe, about 35 ft. high, which bore cones for the first time in 1896; the seed, however, was unfertile. In the United States this tree thrives well, as it delights in hot summers ; and Sargent states that he never saw a plant which appeared to suffer from heat or cold, fungoid diseases, or the attacks of insects. The largest specimen6 is growing in Parson's nursery at Flushing, Long Island, which was imported from London in 1859 when it was 3 ft. high. It measured in 1895 55 ft. high, with a stem 2 ft. in diameter, and branches 50 ft. across, and has borne seed frequently. Another specimen in Mr. HunnewelFs pinetum at Wellesley, Mass., measured6 in 1905 35 ft. in height and 4 ft. in girth, with a spread of branches of 37 ft. This tree has borne seed since 1887, and many seedlings have been raised from it. Sargent reports6 another large specimen on Mr. Probasco's estate at Cincinnati. (H. J. E.) ' Bull. Soc. Dcndr. France, No. 18, p. 162 (1910). 2 In Card. Chron. xix. 88 (1883). 3 Cf. P. L. de Vilmorin, Hortus Vilmorinianus, 66, fig. ix. (1906). 4 Cf. Mitt. Dent. Dcnd. Ces. 1896, pp. 71, fig., and 113. 6 Garden and Forest, 1895, p. 415. 6 Sargent, Pinetum at Wellesley in 1905, p. 10, and in Garden and Forest, 1897, p. 317. CATALPA Catalpa? Scopoli, Introd. Hist. Nat. 170 (1777); Bentham et Hooker, Gen. PI. ii. 1041 (1876); Bureau, in Nouv. Arch. Mus. Hut. Nat. vi. 169 (1894); Dode, in Bull. Soc. Dendr. France, i. 194 (1907). DECIDUOUS trees, belonging to the order Bignoniaceae. Branchlets stout, with thick pith ; leaf-scars elevated, orbicular, marked with a circle of dots, which are the tiny scars left by the fibro-vascular bundles of the fallen petiole. Buds minute, globose, immersed in the bark, with two to four external scales ; all axillary, no true terminal bud being formed, the top of the branchlet dying in summer, and leaving an elevated circular scar close to the upper axillary bud. Leaves simple, opposite or in whorls of threes, entire or lobed, long-stalked, pinnately-nerved, without stipules. Flowers perfect, in terminal panicles or corymbs.2 Calyx gamosepalous, mem branous, splitting when the flower opens into two broad ovate entire lobes. Corolla gamopetalous, inserted on a nearly obsolete disc ; tube broad, campanulate, oblique, enlarged above into a spreading bilabiate limb, the posterior lip two-partite, the anterior three-lobed. Stamens inserted near the base of the corolla, two, anterior, filaments flattened, anthers bilocular and opening longitudinally. Staminodes similarly inserted, three, posterior, filiform, minute or rudimentary. Ovary sessile, two-celled ; style elongated, divided at the apex into two stigmatic lobes8 ; ovules numerous in several series on a central placenta. Fruit, a long nearly cylindrical capsule, tapering from the middle to each end, persistent on the branches during winter, and ultimately splitting into two valves. Seeds numerous, small, oblong, compressed, inserted in two or four ranks near the margin of the woody septum, with broad lateral wings, notched at the base of the seed, and ending in tufts of long coarse hairs. The leaves of Catalpa show on their lower surface in the axils of the nerves clusters of circular glands which secrete nectar, and are visited by numerous insects, especially ants and bees, the latter getting honey from them as well as from the 1 Catalpa is a corruption of Catawba, the name of an Indian tribe that formerly occupied Georgia and the Carolinas. 2 Bureau divides the temperate species into two sections :— (a) Thyrsoidetz, comprising C. liignonioides, C. sfeciosa, and C. Kaempferi; inflorescence a narrow panicle, the seconds ry axes being branched. (K) Corymbosœ, including C. Bungei and C. Fargesi ; inflorescence a corymb, with simple secondary axes. 3 The stigmatic lobes exhibit sensitive movements, opening and shutting like the leaves of a book, with the visits of bees and other insects. Cf. Masters in Gard. Chron. xiii. 651 (1880), and Kerner, Nat. Hist. Plants, Eng. Trans, ii. 281 (1898). This phenomenon in C. bignonioides has beun studied by Meehan, in Froc. Amer. Assoc. Adv. Sc. 1873, pp. 72, 73, and in Bot Ga;. viii. 191 (1883) ; and in the case of C. speciosa by Antisdale, in Bot. Gaz. viii. 171 (1883). 1481 1482, The Trees of Great Britain and Ireland flowers. These glandular areas, which are large and conspicuous near the base of the leaf, bear no pubescence.1 The seedlings2 of Catalpa have stalked deeply bifid oblate cotyledons raised above ground, and followed on the stem by opposite decussate or ternately verticillate ovate leaves. Catalpas may be propagateds by both stem and root cuttings. Eight species of Catalpa are known, of which three are natives of the West Indies and not hardy. The West Indian species constitute a distinct section, charac terised by lanceolate or elliptic leaves. The remaining five species, with ovate leaves, inhabit the United States and China ; and have all been introduced into cultivation in Europe. They may be arranged as follows :— I. Leaves glabrous. 1. Catalpa Bungei, Meyer. China. See p. 1489. Leaves, with a disagreeable peculiar odour, entire with a long slender acuminate apex, or with one or two long-pointed lateral lobes, or coarsely serrate in margin. II. Leaves pubescent with simple hairs. * Branchlets glabrous. 2. Catalpa bignonioides, Walter. United States. See p. 1485. Leaves, with a disagreeable peculiar odour, usually entire with a short apex, glabrous above, pubescent beneath, the pubescence not covering the whole surface of the midrib. 3. Catalpa speciosa, Warder. United States. See p. 1483. Leaves inodorous, usually entire with a long acuminate apex, glabrous above, pubescent beneath, the pubescence covering the midrib entirely. ** Branchlets with stiff glandular hairs. 4. Catalpa Kaempferi, Siebold and Zuccarini. Wild in Central China, long cultivated in Japan. See p. 1487. Leaves inodorous, usually three - lobed ; pubescent on the upper surface throughout, and on the lower surface on the midrib and nerves. III. Leaves tomentose with branched hairs. 5. Catalpa Fargesi, Bureau. Central China. See p. 1490. Leaves entire or with one or two acute lateral lobes ; tomentose on the lower surface throughout, and on the upper surface mainly on the nerves. (A. H.) 1 Cf. Ryder in Proc. Philad. Acad. 1879, p. 161, and Amer. Nat. xiii. 648 (1879). The glandular areas are greenish in the two species from the United States, and purplish in the three Chinçse species. They are almost entirely confined to the base of the leaf in C. Fargesi and C. Bungei ; but are also present in the upper axils of the leaf in C. Kaempferi, C. speciosa, and C. tignonioides. " Described by Lubbock, Seedlings, ii. 335, 339, fig. 571 (1892). 3 Cf. J. Clarke in Card. Chron. xlvii. loo (1910). Catalp a 1483 CATALPA SPECIOSA, WESTERN CATALPA Catalpa speciosa^ Warder, ex Engelmann, in Bot. Gaz. v. i (1880); Sargent, Suva N. Amer. vi. 89, tt. 290, 291 (1894), and Trees N. Amer. 795 (1905); Bureau, in Nouv. Arch. Mus. Hist. Nat. vi. 184(1894); André, in Rev. Hort. Ixvii. 136, fig. (1895); Hall and Schrenk, U.S. Dep. Agric. Bur. Forestry, Bull. No. 37 (1902); Roberts and Dickens, Kansas State Agric. College, Bull. No. 108 (1902); Dode, in Bull. Soc. Dendr. France, i. 195 (1907). Catalpa cordifolia, Jaume, in Duhamel, Traité des Arb. ii. t. 5 (1802) (excl. text) (not Moench). A tree, rarely attaining in America 120 feet in height and 14 feet in girth, usually smaller. Bark thick, deeply furrowed, and roughened with scales. Young branchlets glabrous. Leaves similar to those of C. bignonioides, but without their peculiar odour, often larger, up to 10 in. long and 7 in. wide, with longer acuminate points ; glabrescent above ; lower surface with the pubescence of simple hairs more marked than in C. bignonioides, spreading over the whole of the midrib and extending to the petioles. Flowers appearing two weeks earlier than those of C. bignonioides, few in open panicles, which are about 6 in. long and broad ; calyx purplish, glandular- pubescent ; corolla white, 2 in. long, 2^ in. wide, often spotted externally with purple near the base ; marked internally on the lower side with two bands of yellow blotches following two lateral ridges, and a few purple spots on the lobes of the lower lip of the limb. Fruit, 8 in. to 20 in. long, ^ in. to f- in. in diameter in the middle, with a thick wall, splitting into two concave valves. Seeds i in. long, ^ in. wide, light brown, with wings rounded at the ends and ending in a fringe of short hairs. C. speciosa under favourable conditions differs from C. bignonioides in habit, forming a narrow tree with ascending branches ; but in the arboretum at Segrez, where there are old trees of both species, they are nearly alike in appearance. They are readily distinguished by their flowers, fruits, and seeds ; but when these are absent, the main distinctive character is the odour of the leaves.2 This species in its natural range is confined to a limited region, extending from the valley of the Vermilion river, Illinois, through southern Illinois and Indiana, western Kentucky and Tennessee, south-eastern Missouri, and north-eastern Arkansas. It comes in contact with C. bignonioides in south-eastern Missouri ; and is abundant and of its largest size in southern Illinois and Indiana. It has become naturalised through cultivation in southern Arkansas, western Louisiana, and eastern Texas. It has been planted in the United States as far north as South Dakota, southern Michigan, and Minnesota, and southern Massachusetts ; and westward to eastern 1 Warder, in Western Hort. Review, iii. 533 (1853), was the first to distinguish C. speciosa, but did not then publish this specific name. It appears to have been first used by Sargent, who, in Card. Chron. xii. 784 (1879), points out that the western Catalpa differs from C. bignonioides ; and says that if distinct, it should be known as C. speciosa. 2 W. H. Lamb, in Proc. Soc. Amer. Foresters, vii. 80, figs. I, 2 (1912), points out that the septum of the pod (the long wrinkled partition along which the seeds are arranged) is nearly circular in section in C. speciosa, and lenticular or narrowly elliptic in C. ligiionioides. 1484 The Trees of Great Britain and Ireland Nebraska, Kansas, and Oklahoma. It has succeeded on irrigated lands in New Mexico, Utah, and Colorado, at low altitudes and where the soil is free from alkali. The range for economic planting appears to be on the fertile alluvial lands of the middle west, south of lat. 41°. This species, though only distinguished by Warder as late as 1853, appears to have been introduced early into France, as it was figured in Nouveau Diihamel in 1802 ; but no trees so old as this are now known in Europe. Prof. Sargent sent seeds to Kew in 1880, and probably about the same time to Segrez and Les Barres in France. At Kew C. speciosa, though forming a better tree than C. bignonioides, is very slow in growth, the tallest example, now thirty years old, from seed, being about 25 feet high. It is perfectly hardy, as it has borne at Kew o° Fahr, without injury, and does not suffer in the severe winters of New England. Bureau states that on M. André's property in Touraine it did not suffer from a temperature of - 26° Cent. It came into vogue in America as a tree for planting to produce timber quickly about 1879 to 1883, when large plantations were made by R. Douglas, near Far- lington in Kansas, which are now owned by the railway company. One forty-acre tract of these plantations is, however, C. bignonioides. (A. H.) CULTIVATION An immense quantity of literature on this species has appeared in America, mainly by Mr. John P. Brown of Connersville, Indiana, who devoted a great part of his magazine, Arboricidtiire, to advocating the economic value of this tree. Though this publication contains many illustrations and details on the growth of the Catalpa in many localities and under varied conditions, it has more interest for American than for British readers, and our space will not allow me to refer to them in detail. There seems to be little doubt that in the rich alluvial valleys of the Ohio, Tennessee, Arkansas, and other tributaries of the Missis sippi, its growth is very rapid when young, and it is one of the most valuable trees for fencing, lumber, railway sleepers, and other purposes,1 on account of the durability of its timber ; but it requires a much longer and hotter summer than any part of our islands afford. The latest account of this species is by Oman,2 who studied the results obtained by four plantations in Kansas, which were cut in 1902-1906. He gives valuable hints regarding the proper mode of planting, and states that the financial returns on deep fertile porous soil are remarkable. This tree endures inundation, one plantation having been completely submerged for a week without injury. It coppices freely, and can even be propagated by cuttings ; but suckers from the roots have not been observed. Large quantities of seedlings have been raised and distributed in this country on several occasions,3 but we cannot hear of a single place in which they show any 1 It is also a suitable timber for furniture, as shown by an arm-chair given me by Mr. Brown at Louis in 1904, which has handsome grain, takes a good polish, and has worn well. 2 Proc. Soc. Amer. Foresters, vi. 42-52 (1911). - About 150 trees of this species, which were raised from seed in the Royal Horticultural Society's garden at Chiswick in 1880, were distributed widely to the memhers ; but we have not seen any of these (Card. Chron. xlvii. 245 (1910)). Catalpa H85 signs of becoming a timber tree. I agree entirely with the opinion that Mr. Bean has expressed in Kew Bulletin, 1907, p. 43, that it is improbable that this tree can be grown anywhere in England with any hope of profit, though as an ornamental tree of small size it may have considerable value in favourable situations. A plantation of it was made near Tottenham House, Marlborough, by the Marquess of Ailesbury, where the young trees in 1907 were 3 to 6 ft. high, but so far as we can learn they grow slowly and do not ripen their wood in autumn, which is the case with those I have raised myself. (H. J. E.) CATALPA BIGNONIOIDES, COMMON CATALPA Catalpa lignonioides, AValter, Fl. Car. 64 (1788); Bureau, in Nouv. Archiv. Mus. Hist. Nat. vi. 175 (1894) ; Dode, in Bull. Soc. Dendr. France, i. 194 (1907); Sargent, in Bot. Gaz. xliv. 226 (1907). Catalpa cordifolia, Moench, Meth. 464 (1794); Nuttall, Gen. N. Amer. PL i. 10 (1818). Catalpa commuais, Dumont de Courset, Bot. Cult. ii. 189 (1802). Catalpa sytingifolia, Sims, in Bot. Mag.1 t. 1094 (1808); Loudon, Art. et frut. Brit. iii. 1261 (1838). Catalpa Catalpa, Karsten, Pharm. Med. Bot. 927 (1882); Sargent, Suva N. Amer. vi. 86, tt. 288, 289 (1894), and Trees N. Amer. 793 (1905). Bignonia Catalpa, Linnaeus, Sp. PL 622 (1753) (in part). A tree, rarely attaining in America 60 ft. in height and io ft. in girth, usually smaller. Bark separating on the surface into large thin irregular scales. Young branchlets glabrous. Leaves (Vol. III., Plate 204, Fig. 5) emitting when bruised a disagreeable odour, ovate, about 5 to 6 in. long, and 4 to 5 in. wide ; cordate, truncate, or cuneate at the base, contracted into a slender acuminate point or rounded at the apex, usually entire or occasionally with one or two slight lateral lobes, glabrous above, pubescent with simple hairs on the nerves and veinlets beneath, the pubescence on the midrib being confined to its edges close to the surface of the blade ; glandular areas pale ; petioles glabrous. Flowers numerous in a compact panicle, about 8 to 10 in. long and broad ; calyx glabrous, green or light purple; corolla white, i£ in. long and wide, marked on the inner surface on the lower side by two rows of yellow blotches along two parallel ridges or folds, and on the throat and lower lobes of the limb by numerous conspicuous purple spots. Fruit, 6 to 20 in. long, \ to \ in. thick in the middle, with a thin wall, splitting into two flat valves. Seeds about i in. long, \ in. wide, silvery grey, with pointed wings, ending in long pencil-like tufts of white hairs. The following varieties have arisen in cultivation :— 1. Var. aurea, Lavallée, Arbor. Segrez. 175 (1877). Leaves pale yellow, retaining their colour throughout the season. One of the best golden-leaved small trees2 in cultivation. Its origin is unknown to me. 2. Var. purpurea, Rehder, in Bailey, Cycl. Amer. Hort. i. 258 (1900). Leaves purplish, with dark purple glandular spots. It is said by Nicholson to 1 The plate was drawn from a branch of a tree growing in Mr. Granger's garden at Exeter in 1808. 2 A specimen at Kew is figured in Gard. Mag. 1910, p. 709. 1486 The Trees of Great Britain and Ireland have originated in the United States, and is possibly a hybrid. It is cultivated by Simon-Louis at Metz. 3. Var. variegata, Bureau, op. cit. 183. Variegated with white or yellow. In var. Koehnei, Dode, op. cit. 206, the leaves are pale yellow, with irregular angular green patches. Cultivated by Simon-Louis. 4. Var. erubescens, Nicholson, in Woods and Forests, 1885, p. 52. Catalpa erubescens, Carrière, in Rev. Hort. \. 460 (1869). This form, which I have not seen, is said to have purplish petioles and glandular spots, with a more compact inflorescence, and a more highly coloured corolla with a less deeply divided limb than the type. It is possibly, as Dode suggests, a hybrid. 5. Var. nana, Bureau, op. cit. 183. A low spreading bush, with crowded branches, occasionally grafted high. The leaves are identical in odour and in all other respects with C. bignonioides ; and there are no grounds for supposing it to be a form of C. Bungei, under which name it is commonly known in nurseries and gardens. It has not yet flowered anywhere, and appears to have been first cultivated at Segrez1 in 1877, where it may possibly have arisen as a sport. C. bignonioides is a native of the eastern part of the United States ; but the exact localities where it is truly native cannot be determined with certainty. It is usually supposed to be indigenous on the banks of rivers in south-western Georgia, western Florida, and central Albania and Mississippi, and to be naturalised through out the south Atlantic States. On account of its handsome flowers it was extensively planted for ornament ; and its dissemination has been aided by its winged seeds, which are borne to a considerable distance by the wind and float on water without injury for a long period. As it bears moderately severe winters it may possibly have been a native of the more northern parts of the Alleghany range, where it is not now met with in the existing forests. It thrives as far north as Philadelphia, but is killed during the winter at Rochester on Lake Ontario, and often succumbs at St. Louis.2 (A. H.) The first account of this species was published in The Natural History of Carolina by Catesby, who introduced8 it into England in 1726. The largest tree mentioned by Loudon was one at Syon, 52 ft. high and 3 ft. in diameter, of which only the dead stump remains, but there is a spreading tree grown from one of its layered branches on the north side of the lake which was in flower in July 1912 when I saw it last. A tree at Kew, which died in 1907, when it was about sixty years old, was 30 ft. high and 6 ft. i in. in girth. A tree in the Terrace Gardens, Richmond, was 35 ft. by 8 ft. i in. in 1912. A fine specimen4 in Mr. Denne Dunn's garden at Canterbury was 32 ft. high in 1876. At Caldrees, Ickleton, near Cambridge, there is a fine tree, which flowers freely every year ; it is about 35 ft. high and 7 ft. in girth. 1 Lavallée, Arb. Segrez. 176 (1877), where it is named C. Bungei, var. nana (fumila). Cf. also Lavallée, Icon. Are. Segrez. ii. 35 (1880). 2 R. Douglas, in Woods and Forests, 1884, p. 566. 3 Alton, Hort. Ketu. ii. 346 (1789). 4 Figured in Card. Ckron. v. 13, fig. 2 (1876). In Card. Chron. xxvi. 257 (1897) mention is made of a large tree at Rosslyn, Stamford Hill. Catalpa 1487 There are many trees of considerable age, but of no great height, in parks and places1 in and around London, the best known of which was one at Gray's Inn, which died a few years ago. This Catalpa was reported by tradition to have been brought from America by Sir W. Raleigh, and to have been planted by Bacon ; but there is no good authority for this, and the tree is not long-lived in England. At Ham Manor, near Arundel, I saw a very well-shaped tree (Plate 350) in 1907 which measured 52 ft. by 7 ft. with a clean bole 15 ft. high. There is a tree at Heywood, Wilts, which, in 1906, was about 30 ft. high and 15 ft. in girth below the branches. A photograph sent me by the then gardener, Mr. Robinson, showed it in full flower as a very beautiful tree. Another at Elbridge, as measured by Mr. Furze in 1904, was 41 ft. high, 14 ft. in girth, and had a spread of 61 ft. A fine old tree at Wilton House, Wilts, was, when I saw it in 1906, showing signs of decay, but measured 53 ft. by 6% ft. (H. J. E.) CATALPA KAEMPFERI Catalpa Kaempferi, Siebold and Zuccarini, in Abhand. Akad. München, iv. pt. ii. p. 142 (1846); J. D. Hooker, Bot. Mag. t. 6611 (1882); Lavallée, Icon. Arb. Hort. Segrez. 33, t. 10 (1885); Hemsley, in Journ. Linn. Soc. (Bot?) xxvi. 235 (1890); Bureau, in Norn1. Arch. Mus. Nat. Hist. vi. 190 (1894). Catalpa ovata,z G. Don, Gen. Syst. iv. 230 (1837); Sargent, Sifua N. Amer. vi. 84, note (1894). Catalpa Bungei, Decaisne, in Rev. Hort. v. 406 (1851) (not C. A. Meyer) ; Carrière, in Flore des Serres, viii. 8 (1852) ; Jacques, in Flore des Serres, x. 188 (1855). Catalpa Henryi, Dode, in Bull. Soc. Dend. France, i. 199 (1907). Bignonia Catalpa, Thunberg, Fl. Jap. 251 (1784) (not Linnaeus). A tree, attaining 70 ft. in height ; bark brown, slightly fissured. Young branchlets with numerous sessile glands and scattered stiff glandular hairs.3 Leaves (Vol. III. Plate 204, Fig. 6) without a disagreeable or peculiar odour, ovate, variable in size, averaging 5 to 6 in. in width and length, cordate at the base, shortly acuminate at the apex ; rarely entire, usually with one or two (occasionally three or four) triangular sharp-pointed lateral lobes ; upper surface covered with a minute pubescence, the nerves often purple and with scattered long hairs ; lower surface pubescent on the nerves and veinlets ; petiole with glands and glandular hairs, as on the branchlets. Flowers numerous in much-branched panicles, which are 4 to 9 in. long ; calyx glabrous; corolla pale yellow, about f in. long and broad, marked externally with two orange bands and numerous purple spots. Fruit, 7 to 12 in. long, cylindrical, \ in. in diameter, with a thin wall, splitting into two concave valves. Seeds, \ in. 1 Mr. Hugh Boyd Watt, in an article on Catalpas in London and neighbourhood, which appeared in The Field, l-"eb. 17, 1912, states that there was abundance of fruit in the autumn of 1911 on the trees in Victoria Embankment gardens, Brunswick Square, Hampstead, Richmond, Kew, and Syon House. Six large trees in Palace Yard, Westminster, bore no fruit, though they flowered in the preceding summer. There was no fruit formed in 1909 and 1910; but there was a good crop in 1906. 2 This is the oldest name of the species ; but it has never been in use. * These peculiar hairs, which are characteristic of this species, are deciduous in the course of the season. VI 1488 The Trees of Great Britain and Ireland long, ^ in. wide, greyish brown, with pointed wings ending in long pencil-like tufts of white hairs. This species is much planted in gardens and around temples in Japan ; but is not a native of that country, according to the Japanese botanists, who state that it was introduced at an early period by the Buddhist monks. It was found wild ' in central China by myself and by Wilson in western Hupeh, and by Giraldi2 in Shensi. It was first made known to Europeans by Kaempfer, who visited Japan in 1690, and published8 in 1712 a good description and figure of the tree. It is usually known in Japan as the Ki-sasage, or "bean-tree" on account of its peculiar pods. Dupont, in a letter to Lavallée, mentions that he never saw this tree in the forest in Japan, but always planted, and records one 75 ft. in height and 5 ft. in girth growing near a temple. It was introduced into Europe from Japan by Siebold in 1849, and has probably thriven best at Segrez, where Lavallée mentions a tree, which covered an area over 40 yards in circumference. It appears to be much hardier both in France and at Boston4 (U.S.) than C. bignonioides. Plants raised from seed8 sent from central China by Wilson to Coombe Wood, and by Père Farges to Les Barres, appear to be identical in all respects " with trees of Japanese origin. The finest specimen of C. Kaempferi known to us in England is a tree at Syon, which measured, when Elwes saw it in July 1912, 62 ft. by 5 ft. i in. It was then in full flower, with capsules of the previous year containing ripe seeds. The oldest tree at Kew, about 20 ft. high, was procured from Volxem in 1879. A remarkable hybrid has arisen between C. Kaempferi and C. bignonioides :— Catalpa hybrida? Späth, in Gartenflora, Ivii. 481, t. 1454 (1898). Catalpa x J. C. Teas, Sargent, in Garden and Forest, ii. 303, fig. (1889); Catalpa Teasiana, Dode, in Bull. Soc. Dend. France, \. 205 (1907). A tree, said to be of remarkably vigorous growth, resembling C. Kaempferi in foliage, but intermediate in flowers and fruit between that species and C. bignonioides. It produces extremely large panicles, 18 to 20 in. long and 10 in. wide, of 200 to 300 fragrant flowers, about i in. long ; corolla tinged with yellow in the throat and marked with broad purple stripes. Fruit 12 to 15 in. long, £ in. wide. This hybrid was raised about 1880 in J. C. Teas' nursery at Baysville, Indiana, from seed out of a peculiar single pod which was observed on a tree of C. Kaempferi. The latter grew near trees of both C. bignonioides and C. speciosa, but the pollen 1 It is figured in the Chih Wit Ming, xxxiii. pi. 48 (1848). 2 Cf. Diels, in Engler, Jahrb. xxxvi. heft 4, p. 98 (1905). 3 Amcen. Exot. 842 (1712). 4 Sargent, in Gard. Chron. xii. 784 (1879). 8 Sargent, in a letter to Kew, dated 2nd February 1900, states that seeds received from Shanghai in 1892 of reputed C. Bungei, produced plants of C. Kaempferi. 0 A young tree at Kew, 10 ft. high, of Chinese origin, bears leaves identical in odour, shape, pubescence, and colour, with an older tree of Japanese origin. Specimens sent from Les Barres show no differences. 7 Dode, op. cit. 204, identifies C. hybrida, Späth, with C. erubescent, Carrière, described above, p. 1486. Rehder, how ever, in liailey, Cycl. Am. Hort. i. 258 (1900), considers Späth's description to refer to Teas' hybrid ; and this appears to be correct. Späth does not say, as Dode asserts, that the hybrid originated in his nursery. Catalpa 1489 appears to have come from the former. As Sargent points out, C. bignonioides and C. Kaempferi flower at the same time, whereas C. speciosa is two to three weeks earlier than C. Kaempferi. Moreover, Penhallow,1 by an examination of the wood of the different species of Catalpa and of the hybrid, has shown conclusively that the latter is a cross between C. bignonioides and C. Kaempferi, in which the characters of the latter are dominant ; and that C. speciosa was in no way concerned in its pro duction. Penhallow adds, that some of the seedlings of the hybrid, but not all, revert to C. Kaempferi. " Teas' Japan hybrid," as it is commonly called, was introduced into Kew gardens in 1891, and flowered in 1900 and succeeding years, but has not produced fruit. At Kew it shows no excessive vigour. C. japonica, Dode, in Bull. Soc. Dend. France, i. 200 (1907), said to have been introduced from Japan in 1886 by Simon-Louis, is unknown to me; but from the description appears to be another hybrid between C. bignonioides and C. Kaempferi. (A. H.) CATALPA BUNGEI Catalpa Bungei, C. A. Meyer, in Bull. Acad. Sc. St. Pctersb. ii. 51 (1837); Maximowicz, Prim. Fl. Amur. 475 (1859); Kurz, mjourn. Bot. ii. 193 (1873); Lavallée, Arbor. Segrez. 176 (1877); Hance, in Joum. Bot. xi. 37 (1882); Franchet, PL David, i. 229 (1884); Hemsley, in Journ. Linn. Soc. (Bot.) xxvi. 234 (1890); Bureau, in Nouv. Arch. Mus. Hist. Nat. vi. 197, pi. 4 (1894); Bean, in Kew Bull. 1907, p. 102. Catalpa syringœfolia, Bunge, Enuin. PI. China Bor. 45 (1835) (not Sims). Catalpa /leterqphylla, Dode, in Bull. Soc. Dend. France, i. 203 (1907). Catalpa Dudouxii, Dode, in Bull. Soc. Dend. France, 1907, p. 201, and 1909, p. 154. A tree, attaining in China 40 feet in height. Young branchlets glabrous, covered with minute glands. Leaves with a disagreeable odour, variable in size and shape, about 5 in. long and 4 in. wide, ovate or deltoid, cuneate or truncate at the base, ending in a long slender acuminate apex ; entire, or with two lateral long-pointed lobes, or with several irregular acute teeth on each side ; upper surface dark green, shining, glabrous ; lower surface pale green, glabrous ; petiole glabrous. Flowers, three to nine, in a loose corymb ; axis and pedicels glabrous or with a few simple hairs ; calyx glabrous, green below, pink above ; corolla, i£ in. long, with a wide campanulate tube, which is tinged with yellow along two projecting ridges and is elsewhere spotted purple, and a white five-lobed limb, with numerous purple dots, which give the whole flower a decidedly pinkish tint. Fruit, described as very long and slender, 25 to 40 in. in length and \ in. in diameter. Seeds greyish brown, | in. long, with narrow pointed wings, ending in pencil-like tufts of pale yellow hairs. The foliage of this tree is very variable—entire or two- to three-lobed leaves occurring on old trees ; whilst those with a dentate margin are characteristic of 1 In Amer. Naturalist, xxxix. 113, figs. 1-8 (1905). 149° The Trees of Great Britain and Ireland branches ending in an inflorescence, and are normally developed on the ordinary branches of young trees. As both forms occur on the same individual, van hetero- phylla, Meyer, cannot be maintained as a distinct variety.1 In northern China, the leaves are quite glabrous ; but in the mountains of central China, there are traces of pubescence (simple hairs) on the upper surface of the leaves and on the petioles. Specimens with more numerous flowers in the corymb, which has one or two of the lateral axes branched, have been considered to be a distinct species, C. Dziclouxii, Dode ; but these are probably trees of greater vigour and not even a distinct variety. C. Bungei is readily distinguishable by its glabrous shining leaves, with longer and more slender points to the lobes than is the case in the other species. This tree is widely spread throughout the mountains of China, from Peking in the north to Yunnan in the south-west, and is also recorded from the coast provinces of Shantung and Chekiang. Wilson found it in Szechwan at 8850 feet altitude, as a "tree 40 ft. in height, with white flowers suffused and spotted with pink." In Hupeh, it grows at about 4000 to 5000 feet elevation in the mixed forests of deciduous trees, and, my notes state, "40 feet high, 4 feet in girth, with pinkish flowers, which, together with the leaves, are of a disagreeable odour." It is much planted in temple grounds, at Peking, Shanghai, and elsewhere, and is usually known as the ch'iu tree.2 It flowers at Peking in May. This species is very rare in cultivation, the plants generally sold under this name by nurserymen being either the dwarf form of C. bignonioides or C. Kaempferi. Bureau knew of only one living specimen in France, a small tree at Segrez, which had not produced flowers. It was introduced3 about 1904 into the Arnold Arboretum by seed procured from Peking ; and a tree sent by Sargent to Kevv in 1905 is now about 8 feet high. (A. H.) CATALPA FARCES I Catalpa Fargesi, Bureau, in Nouv. Archiv. Mus. Hist. Nat. vi. 195, pi. 3 (1894); Dode, in Hull. Soc. Dend. France, i. 204 (1907). Catalpa vestita, Diels, in Engler, Jährt, xxix. 577 (1900). A small tree, distinct from the other species, in the presence of stellate tomentum or branched hairs on the young branchlets, leaves, axes of the inflorescence, pedicels, bracts, and calyx. Leaves about 5 in. long and 4 in. broad, entire or with one or two acute lateral lobes, subcordate at the base, acuminate at the apex ; tomentose on the lower surface throughout, on the upper surface mainly on the nerves. 1 Cf. Bretschneider, Bot. Sinicum, \r\Journ. N. China Branch, K. Asiat. See. xvi. 112 (1882), who states that "the leaves on the same tree are very variable, cordate, entire, lobed, laciniate, triangular, sinuate, etc." 2 Bretschneider gives an account of the Chinese literature of this tree in Bot. Sin. ii. 339 (1882) and iii. 478 (1895). The classical name tee, with which he identifies the Catalpa, possibly indicated Sassafras Tzumti. Cf. vol. iii. P- 5I5- 3 Cf. Rehder, in Mitt. Dcut. Dend. Ces. 1907, p. 76. Catalpa 1491 Flowers few, seven to ten in a compact corymb ; calyx covered externally with stellate tomentum ; corolla, similar in size to that of C. Bungei, white, spotted with reddish brown dots. Fruit 18 to 20 in. long, \ in. in diameter. Seeds yellowish grey, about 1 in. long, with sharp-pointed wings, ending in long silky hairs. C. Fargesi is a native of western China, where it has been found in the mountains of Szechwan and Shensi by Farges, Giraldi, and Wilson. The latter introduced1 it in 1901. Young plants show merely slight traces of the branched hairs on the branchlets and leaves, which are so characteristic of wild specimens ; but as they grow older this peculiar pubescence may increase in quantity. Plants of this species, raised from seed sent by Wilson (Nos. 636 and 640) in 1905, are growing freely at Colesborne, where they have endured 30° of frost without injury, though the wood does not ripen well in autumn. These are now 4 to 6 ft. high. There are also young plants at Kew and Aldenham. (A. H.) i Joum. Roy. llort. Soc. xxviii. 50 (1902). PAULOWNIA Paulownia} Siebold and Zuccarini, Fl. Jap. [.25 (1835); Bentham et Hooker, Gen. PI. ii. 939 (1876); Dode, in Bull. Soc. Dend. France, 1908, p. 159; Schneider, Laubholzkimde, ii. 618 (1911). DECIDUOUS trees belonging to the order Scrophulariacese. Branchlets with chambered pith, showing in winter large oval raised opposite leaf-scars. Buds axillary, no true terminal bud being formed, minute, covered with two or four pubescent scales. Leaves simple, opposite in decussate pairs, stalked, ovate, cordate. Flowers in large terminal erect panicles, opening in spring before the leaves ; calyx five-cleft, campanulate, persistent at the base of the fruit ; corolla gamopetal- ous, inserted on the base of the calyx, with a long slightly curved tube, and five spreading lobes, the three lower lobes longer than the two upper lobes ; stamens four, affixed to the tube of the corolla, didynamous, included, with divaricate anther- sacs ; ovary superior, two-celled, with numerous ovules ; style one, slender, slightly thickened towards the summit, stigmatic on the inner side. Fruit, ripening in one year, a two-celled woody or coriaceous capsule, ovoid, loculicidally dehiscent by two valves ; placentae two, ovate, compressed ; seeds numerous, minute, oblong, sur rounded by a broad translucent striated wing. Two species of Paulownia2 have been clearly distinguished, one of which, little known and not in cultivation in England, may be here briefly described. I. Paulownia Fortunei, Hemsley, mjourn. Linn. Soc. (Bot.) xxvi. 180 (1890). Leaves narrowly oval, longer and more acuminate than in P. tomentosa, covered beneath with a dense whitish tomentum. Flowers longer and relatively narrower than in P. tomentosa ; calyx-lobes deltoid, obtuse, usually brown tomentose through out, occasionally glabrescent except on the borders. Fruit, 3 to 3^ in. long, narrowly ovoid ; seeds £ in. long, much larger than those of P. tomentosa. 1 Named after Anna Paulowna, Queen of the Netherlands. 2 Hayata, in Bull. Congrès Internat. Bot. Bruxelles, 41, pi. 24 (1910), and m Journ. Coll. Sei. Tokyo, xxx. 209 (1911), mentions a possible new species in Formosa. Elwes saw this in 1912 at a village near Horisha in Central Formosa ; but it was not in leaf. There is also a supposed new species from Western China, raised at the Arnold Arboretum from seed sent by E. H. Wilson. It is in cultivation at Kew and Aldenham ; but the young plants cannot at present be distinguished from P. tomentosa. Cf. Gard. Chron. xlviii. 275, fig. 116 (1910).—A. H. I raised seedlings from Mr. Wilson's seeds, No. 769, collected in his journey of 1908 in Western China, which appear at three years old to be hardier and more rapid in growth than those which I have raised from the common species. The seed lings of the latter were killed to the ground for three years after planting out ; whilst the West China form is now, at three years old, 14 ft. high, of which 10 ft. is the growth of 1911. This form seems likely to be a most ornamental tree even in cold parts of England ; but must be planted in warm sheltered places where its immense juvenile leaves, measuring 21 in. by 23 in., will not be torn by wind.—H. J. E. 1492 Paulownia H93 This species, which is probably smaller in size than P, tomentosa, occurs both in the north of China, where it has been collected at Chefoo, and in the south, where it has been found in the province of Kwangtung. P. Duclouxii,1 Dode, in Bull. Soc. Dend. France, 1908, p. 162, is apparently a variety with white flowers, tinged with pink, and not spotted as in the type. This is said to have been raised by C. Sprenger * of Corfu from seed, which he received from Dr. Dode. PAULOWNIA TOMENTOSA Paulownia tomentosa, Koch, Dendrologie, ii. pt. i, p. 299 (1872); Shirasawa, Icon. Ess. Forest. Japon, i. text 129, t. 85 (1900). Paulownia imperialis, Siebold and Zuccarini, Fl. Jap. i. 27 (1835); Loudon, Trees and Shrubs, 671 (1842); W. J. Hooker, Bot. Mag? t. 4666 (1852); Hemsley, in Journ. Linn. Soc. (Bot.} xxvi. 180 (1890). Bignonia tomentosa, Thunberg, Fl. Jap. 252 (1784). A tree, attaining 80 ft. in height and 12 ft. in girth. Young branchlets green, glandular, and pubescent. Leaves broadly ovate, about 7 to 8 in. long, and 6 to 7 in. wide, cordate at the base, shortly acuminate at the apex, undivided or with one or two short lateral deltoid lobes, entire in margin ; upper surface dark green, pubescent with short erect hairs ; lower surface greyish green, covered with a thin tomentum ; petiole 3 to 5 in. long, glandular, pubescent. Flowers violet, the lower lip marked with dark coloured spots and two yellow bands ; calyx with five ovate erect lobes, covered with a dense rusty brown tomentum. Capsules ovoicl, about \\ in. long, and i in. broad ; seeds minute, about \ in. long. i. Var. Fargesii, Henry (var. nova). Paulownia Fargesii, Franchet, in Bull. Mus. Hist. Nat. Paris, 1896, p. 280. Flowers paler in colour. Adult leaves more glabrous than in the type. This was described by Franchet from a specimen sent by Père Farges from the mountains of north-eastern Szechwan ; and appears to be identical with a tree found by me growing on cliffs in the mountains of Hupeh (No. 5346 A). The pentagonal ribbed calyx, noticed by Franchet, appears to be due to drying of the specimen, and not to be specially characteristic of this form, as a similar calyx occurs in some specimens of typical P. tomentosa. Var. Fargesii was introduced into cultivation in France by M. M. de Vilmorin, who raised it from seed received from Père Farges ; and it flowered4 in M. Boucher's nursery at Paris in 1905. P. tomentosa is a native of the mountains of central and western China, where it has been found growing wild at altitudes of about 4000 feet by Père Farges in Szechwan, and by myself in Hupeh and Yunnan. It is the t'ung tree of the Chinese 1 P. tneridionaKs, Dode, loc. cit., described from a tree growing in Laos in Indo-China, is allied to or perhaps a form of P. Fortunei. 2 Cf. Mitt. Deut. Dend. Ges. 1910, p. 246. 3 Figured from a tree in the garden at Bishopstowe, near Torquay, the first which flowered in the open air in England. 4 Cf. Journ. Soc. Nat. Hoit. vi. 324 (1905). C. Sprenger of Corfu, states in Mitt. Deut. Demi. Ges. 1910, p. 247, that he has raised P. Fargesii from seeds received from the mountains of north-west Hupeh in central China. 1494 The Trees of Great Britain and Ireland classics,1 the wood of which was used in ancient times for making lutes. It is largely planted as an ornamental tree throughout China, Korea, and Japan ; and was prob ably introduced into the latter country at an early period by the Buddhist monks.2 This species often produces root-suckers at a considerable distance from the parent tree ; and when cut clown, sprouts vigorously from the stool. The seedling, which has an herbaceous stem, usually dies down at the end of the first year ; but in the following spring a permanent and more woody stem arises from a bud close to the ground. The leaves on young plants (which are very large), on root-suckers, and on coppice shoots differ from the foliage of the adult tree, their margin being furnished with numerous short teeth, while their upper surface is velvety to the touch and very viscid, owing to the presence of numerous glandular hairs and sessile glands. The Paulownia may be raised from seed sown in spring, or from root-cuttings ; and may also be propagated by stem-cuttings under glass, or even from leaf-cuttings, At Kew this species is very effective as a foliage plant in beds out of doors. The plants, when about three years old, are cut down in early spring to within 6 inches of the ground ; and when they start to grow, all the buds except one on each stem are removed. Watered in dry weather and mulched with manure, stems are produced about 6 ft. high, which bear enormous leaves, 12 to 18 in. in diameter. This tree was introduced3 into Europe in 1834, by seeds sent from Japan to Paris by M. de Cussy ; and one of the original trees in the Jardin des Plantes measured in 1904 about 60 ft. in height, and 12 ft. in girth. It succeeds well in France and Italy, where it attains a considerable size, and regularly produces flowers and fruit. It was introduced * into England by seeds from Japan in 1838, and was cultivated in the Horticultural Society's garden at Chiswick ; but it has never attained large dimensions in this country, and probably many of the older trees were killed in the severe winters of 1860 and 1866. It flowers frequently in the south of England, but rarely produces fruit, and very seldom fertile seed. (A. H.) The largest tree I know of in England is one standing near the entrance lodge at Westonbirt, Gloucestershire, which as measured in August 1911, by Mr. A. Chapman, was 56 ft. by 7 ft. at 3 ft. from the ground. The next is at Wilton House, which in 1906 measured 53 ft. by 6 ft. 8 in., but being in a damp situation it has only flowered twice in fifty years, and Mr. Challis informed me that it suffered much in the hard winters of 1860-61, and 1879-80. A fine tree at Linton Park, Maidstone, in 1911 measured 45 ft. high and 9 ft. in girth at 6 ft. from the ground. At Caldrees, Ickleton, Cambridgeshire, there is a tree about 25 ft. high and 3^ ft. in girth, which produces flowers and fruit nearly every year. From its seeds, plants were raised in 1902, one of which was planted at Ickleton Grange, and is now 12 ft. high. 1 Cf. Bretschneider, Bot. Sinic. ii. 348 (1892). It is colloquially known in China as the pao-t'nng, in order to distin guish it from the wu-fung (Sterculia platanifolia), and the f-ung-yu (Alcurites cordata}. 2 It is figured by Kaempfer, Amcen. Exot. 860 (1712). 3 Actes Premier Congres Internat. Bot., Paris, 536 (1900). Cf. also Gard. Chron. 1841, pp. 349, 701, where it is stated that a single plant was raised in the Royal Garden in Paris, from Japanese seed, in 1834. 4 Loudon, Card. Mag. xvi. 635 (1840). It flowered in a greenhouse at Oakfield near Cheltenham in 1843, according to Loudon, Gard. Mag. 1843, p. 649. Paulownia 1495 At Swanmore Cottage, Hants, Mr. Molyneux showed me a well-shaped tree, 35 ft. by 7^ ft., in 1906, which flowers occasionally but never ripens seed. At Ashstead Park, Surrey, the seat of Mr. P. Ralli, there was in 1892 a fine tree, 45 ft. by 7^ ft., which flowers freely in warm seasons.1 At Whitbourn Court, near Worcester, Sir R. Harrington has a tree which he raised from seed gathered in the Vatican gardens at Rome in 1888, which in 1905 was 23 ft. by 4 ft., and with a head 25 ft. in diameter. In Cornwall, where the climate does not suit it so well, the largest I have seen, at Scorrier, was about 25 ft. high in 1911. There are specimens 20 to 30 ft. high in the Kew and Cambridge Botanic Gardens ; at Grayswood, Haslemere ; in several gardens in Kent ; at East Cowes Castle in the Isle of Wight ; at Hursley Park, Winchester ; at Abbotsbury2 ; at Bicton3; at Trevarno,4 Cornwall ; and at Singleton Abbey, Swansea. In Scotland, the only places where I have seen it are at Castle Kennedy, where in 1906 I saw a poor-looking tree about 25 ft. high, which evidently does not like the climate ; and at Tyninghame, East Lothian, where Mr. Brotherston measured a tree 3 ft. 9 in. in girth at 3 ft. from the ground, which has been in bad health for years, so we may conclude that the summers of Scotland are too short and cold for it. In Ireland, there are two old but not very thriving trees at Glasnevin ; but at Mount Usher, there is a fine tree nearly 40 ft. high, and over 6 ft. in girth. In America,6 it does not flower regularly north of New York ; but is fairly hardy in sheltered positions as far north as Massachusetts, where the flower- buds are killed every winter ; and it can be cultivated as a foliage plant even in Montreal. TIMBER In Japan, it is known as kiri; and produces a very light, dull white, shining wood, which is used for making boxes, musical instruments, linings of safes, clogs, doors, and in cabinet work. As large planks are not usually obtainable, Paulownia boards are made by joining small pieces together with paste and bamboo pegs, as shown at the Anglo-Japanese Exhibition at Shepherd's Bush, London. On my last visit to Japan I saw well-made wardrobes of this wood, which is preferred to all others for this purpose on account of its resistance to damp. Clothes kept in drawers of this wood are said to remain free from mould during the rainy season. These wardrobes are sometimes framed in the wood of Diospyros Kaki, the heart- wood of which is black mottled with grey, and very handsome. The large braziers, called Hibashi in Japan, are often made of sections of the trunk of Paulownia turned in such a way as to show its beautiful grain. (H. J. E.) 1 Gard. Chron. xii. 440 (1892). 2 Fruit was sent to Kew from Abbotsbury in January 1902. s Quart. Journ. Forestry, i. 54 (1907). 4 Cf. Gard. Chron. xxvi. 211 (1899). In The Field, 1908, p. 233, the tree at Trevarno, said to be twenty years old and 30 ft. high, is reported to have produced a few years previously a good crop of fruit. 6 Rehder, in Bailey, Cycl. Amer. Hort. 1223 (1901). Button and Brown, Illustrated Flora Northern U.S. iii. 157 (1898), state that it has escaped from cultivation in southern New York, in New Jersey, and in the Southern States. VI ROBINIA Robinia, Linnseus, Gen. PI. 220 (1737) and Sp. PL 722 (1753); Bentham et Hooker, Gen. PI. \. 499 (1865); Schneider, Laubholzkunde, ii. 79 (1907). Pseudacacia, Moench, Meth. 145 (1794). DECIDUOUS trees and shrubs belonging to the division Papilionacese of the order Leguminosse. Leaves unequally pinnate, alternate, stalked : leaflets opposite, rarely sub-opposite or alternate, stipellate, stalked, entire, penninerved. Stipules in pairs, at first setaceous, ultimately either deciduous or developing into persistent spines. No true terminal bud is formed, the tip of the branchlet falling off early in summer and leaving a scar at the apex of the twig. Buds minute, multiple, three to five superposed vertically, not apparent1 in summer and autumn, being concealed by the enlarged base of the petiole ; in winter, embedded in a projection on the branchlet between the stipules, and covered by three scales, which are united together and form the leaf-scar. Usually only the uppermost bud develops—the scales, which are very tomentose within, bursting open, and afterwards persisting at the base of the new branchlet during the following season. Flowers in pendulous racemes, arising from the axils of the leaves, with long pedicels and caducous bracts and bracteoles. Calyx campanulate, unequally five- toothed. Corolla papilionaceous ; petals with short claws, inserted on a tubular glandular disc, connate with the base of the calyx-tube; standard large, reflexed, obcordate ; wings oblong, curved, free ; keel-petals incurved, obtuse, united below. Stamens ten, inserted with the petals ; nine inferior united into a tube ; upper stamen free at the base. Ovary stalked ; style subulate, inflexed, pubescent, ending in a small stigma ; ovules numerous, hanging in two rows from the ventral suture. Pods in pendulous racemes, linear-oblong, compressed ; valves two, thin, membranous ; seed-bearing suture with a narrow wing. Seeds numerous, reniform, oblique, with a persistent incurved stalk. About eight species of Robinia are known, confined to the United States and Mexico. Four, all natives of the United States, occur in cultivation :— I. Branchlets without glands. i. Robinia Pseudacacia, Linnseus. See p. 1497. Branchlets at first slightly pubescent, soon becoming glabrous. Leaflets 1 Occasionally a supra-axillary bud is formed, visible abo