The Trees of Great Britain & Ireland/Volume 1/Pyrus

by Henry John Elwes and Augustine Henry

 

PYRUS

 

Pyrus, Linnæus, Gen. Pl. 145 (1737); Bentham et Hooker, Gen. Pl. i. 626 (1865).

Malus, Ruppius, Fl. Jen. ed. 3, 141 (1745); and Medicus, Phil. Bot. i. 138 (1789).

Sorbus, Linnæus, Gen. Pl. 144 (1737).

 

Trees and shrubs belonging to the sub-order Pomacese of the order Rosaceae. Branchlets of two kinds, long and short shoots, the flowers in certain species being borne on the latter only. Leaves deciduous, alternate, stalked, simple or pinnate; stipules deciduous. Flowers in cymes or corymbs, regular, perigynous or epigynous, calyx-lobes 5, petals 5. The receptacle (the end of the axis) is hollowed out, the ovary being attached to its interior. A disc is present, either annular or coating the receptacle. Ovary with 2 to 5 cells, each cell containing 2 ovules. Fruit, a pome, the external fleshy part being formed of the receptacle, while the interior or core is the developed ovary; cells 2 to 5, with a membranous or cartilaginous endocarp, each containing i or 2 seeds, though occasionally some are empty.

The genus Pyrus has been divided variously into sections, which some botanists treat as distinct genera. The following arrangement is perhaps the simplest:—

 

A. Leaf in the bud rolled inwards towards the midrib.

1. Pyrophorum. Flowers in corymbs on spur-like branchlets, ovary with 5 cells, styles free. Fruit pyriform or hollowed out at the base, flesh granular. Leaves simple. Pears: confined to Asia and Europe.

2. Malus. Flowers fascicled or umbellate on spurs, ovary with 3 to 5 cells, styles united at the base. Fruit with a cavity at the base, flesh homogeneous. Leaves simple. Apples: species in North America as well as in Europe and Asia.

3. Aronia. Flowers in terminal corymbs, ovary with 4 to 5 cells, styles free or united at the base. Fruit small, not hollowed at the base, endocarp very thin, flesh almost homogeneous. Leaves simple, crenate, with the midrib glandular on its upper side. Two North American shrubs.

 

B. Leaf folded in the bud. Flowers in terminal corymbs.

4. Hahnia. Ovary with 2 to 3 cells, styles united below. Fruit crowned by the calyx, and having a hard, almost bony endocarp, flesh granular. Leaves simple, lobed. Pyrus torminalis, the only species.

5. Sorbus. Ovary with 3 or 5 cells, styles free. Fruit crowned by the calyx, endocarp membranous or coriaceous. Leaves pinnate. Includes two sub-sections:—

Aucuparia, with 3-celled ovary and small globular fruit, and

Cormus, with 5-celled ovary and large pear- or apple-shaped fruit.

6. Aria. Ovary with 2 to 5 cells, styles free. Fruit crowned by the calyx, endocarp membranous, flesh granular. Leaves simple. Includes the whitebeam and its allies.

7. Micromeles. Ovary with 2 to 3 cells. Fruit small, globose, umbilicate, endocarp membranous or coriaceous, calyx-lobes deciduous. Leaves simple. Includes several Asiatic species.

 

Synopsis of the Principal Species in Cultivation exclusive of
Pears and Apples.

(Cf, Plates 43-45. where leaves of most of the species are shown.)

I. Leaves regularly pinnate, the leaflets being separate and never decurrent by their bases on the rhachis.

A. Aucuparia. Mountain ashes. Leaflets unequal-sided at the base. Fruit small, not exceeding ⅓ inch.

(1-3) Winter buds white-tomentose.

1. Pyrus Aucuparia, Gaertner. Europe, Northern Asia, Japan. Young branchlets and leaves pubescent, adult leaves glabrous or only slightly pubescent beneath. The common mountain ash.

2. Pyras lanuginosa, DC. South-eastern Europe. Only differs from the preceding in the adult leaves being densely woolly beneath.

3. Pyrus thianschanica, Regel. Chinese Turkestan. Young branchlets and leaves glabrous; adult leaves quite glabrous beneath and conspicuously veined on the upper surface.

(4-5) Winter buds shining, glutinous, glabrous or sparingly pubescent, the pubescence oppressed and of a rusty colour.

4. Pyrus americana, Torrey and Gray. North America. Leaflets long, narrow, acuminate, glabrous beneath.

5. Pyrus sambucifolia, Chamisso and Schlechtendal. Manchuria, North-East Asia, Japan, North America. Leaflets broader than in No. 4, obtuse or acute (not acuminate), more or less pubescent beneath.

B. Cormus. True Service. Leaflets nearly equal sided at the base. Fruit large, inch diameter or more.

6. Pyrus Sorbus, Gaertner. Central and Southern Europe. Winter buds greenish, viscid, pubescent only at the tip; under surface of the leaves slightly pubescent in spring, soon becoming glabrous.

 

II. Leaves pinnate, but upper 3 or 5 leaflets coalesced or decurrent by their bases on the rhachis.^[1]

7. Pyrus Aucuparia, Gaertner, var, satureifolia, Koch. A hybrid. Differs from the common form in the coalescence into one large segment of the 3 upper leaflets; leaflets glabrous beneath.

8. Pyrus Aucuparia, Gaertner, var. decurrens, Koehne (Pyrus lanuginosa, Hort. non DC.) A hybrid. Upper leaflets decurrent on the rhachis by broad bases, often the upper 3 or 5 coalescing into one segment; leaflets tomentose beneath.

 

III. Leaflets pinnate or deeply cut at the base, with 1-4 pairs of segments, the upper part of the leaf lobed or serrate; leaves very variable in shape.^[1]

9. Pyrus hybrida, Moench. A shrub of hybrid origin. Main axis of the leaf glandular above, under surface of the leaf sparingly pubescent, the parents being Pyrus aucuparia and Pyrus arbutifolia.

10. Pyrus pinnatifida, Ehrhart. A hybrid. Axis of the leaf without glands, under surface densely grey tomentose.

 

IV. Leaves simple, lobed.

A. Under surface of the leaf glabrous, or nearly so, light green in colour.

11. Pyrus torminalis, Ehrhart. Europe, Algeria, Asia Minor, and the Caucasus.

B. Under surface of the leaf grey tomentose.

12. Pyrus cratægifolia, Savi. Italy. Leaves small, resembling those of a hawthorn, on each side 4-6 triangular-ovate toothed lobes.

13. Pyrus latifolia, Boswell Syme. Britain, France, Spain, Central Europe. Leaves broad-oval with a wide base; lobes decreasing from below upwards, small triangular, separated by sinuses, which form a right angle and are not narrowed. In some forms the lobes are mucronate, in others cuspidate; and, in var. decipiens, the outline of the leaf is like intermedia, but the lobing is different.

14. Pyrus intermedia, Ehrhart. Europe. Leaves elliptic, with a usually narrow base, lobes decreasing from below upwards, rounded, mucronate, separated by narrow sinuses, which are very acute or almost closed at their bases. This includes several forms:—

Mougeoti. Leaves with 9-12 pairs of nerves, lobes shallow.

Scandica. Leaves with 7-9 pairs of nerves, lobes deep, with sharp teeth.

Minima. Leaves with 5-7 pairs of nerves, smaller and narrower than in the preceding varieties.

Certain forms of Pyrus pinnatifida closely resemble scandica, but the lobing in these will be found always irregular and often very deep.

15. Pyrus lanata, D. Don. Himalayas. Leaves large, broad oblong, woolly underneath, nerves 12-15 pairs, lobes regularly serrate.

C. Under surface of the leaves, which are orbicular in outline, snowy-white tomentose.

16. Pyrus Apia, Ehrhart, van flabellifolia. Greece. Leaves with 3-5 pairs of nerves.

17. Pyrus Aria, Ehrhart, var. græca. Greece, Asia Minor. Leaves with 6-10 pairs of nerves.

 

V. Leaves simple, not lobed, and only occasionally obscurely tabulate.

A. Aronia. Leaves finely serrate in margin, with glands on the upper surface of the midrib. This section comprises 2 North American species and a hybrid of garden origin, small shrubs, only referred to here to prevent their being confused with other species of Pyrus.

18. Pyrus arbutifolia, Linnæus fil. North America. Leaves beneath whitish grey tomentose, with about 6 pairs of nerves directed forwards at a very acute angle.

19. Pyrus alpina, Willdenow. A hybrid between Pyrus Aria and the preceding species. Leaves densely grey tomentose beneath, with 9-10 pairs of nerves directed outwards at an angle of 45°.

20. Pyrus nigra, Sargent. North America. Leaves glabrous beneath or very slightly pubescent.

B. Leaves without glands on the midrib.

(21-22) Leaves glabrous beneath. These 2 species, of which the first is a shrub and the other a small tree, are only referred to here to distinguish them from other species.

21. Pyrus Chamæmespilus, Linnæus. Vosges, Jura, Alps, Pyrenees. Leaves sessile or nearly so, elliptic, with 6-8 pairs of nerves.

22. Pyrus alnifolia, Franchet and Savatier. Japan and China. Leaves stalked, broadly ovate, with 9-12 pairs of nerves.

(23-26) Leaves white pubescent beneath.

23. Pyrus Aria, Ehrhart. Europe, Caucasus, Siberia, Central China. Leaves oval or elliptic with very slight lobules or only doubly-toothed, the teeth or lobules diminishing in size from above downwards; nerves 7-12 pairs, very prominent on the lower surface, pubescence snowy white.

24. Pyrus Aria, Ehrhart, var. rupicola. British Isles (Europe?) Leaves obovate-oblong (broadest above the middle), lobulate above, the lobules and teeth acute, nerves 5-9 pairs, pubescence at first as white as in the type, but ultimately becoming greyer.

25. Pyrus Aria, Ehrhart, var. Decaisneana. Origin unknown. Leaves large, elliptic, or oblong, with margin serrated almost uniformly, nerves 12-15 pairs.

26. Pyrus Hostii, Hort. A hybrid. Leaves like 23, but with very sharp, irregular teeth and tomentum thin, white to greyish white.

(27) Leaves grey, densely-woolly pubescent beneath.

27. Pyrus vestita, Wallich. Himalayas. Leaves very large, elliptic, serrate, and occasionally obscurely lobulate in margin; nerves 15-18 pairs.

 

As many of the species mentioned above are merely shrubs or very small trees, they do not fall within the scope of our work. For this reason, Pyrus hybrida, cratægifolia, Chamæmespilus, alnifolia, and the three species of the section Aronia, will not be further referred to. Pyrus Hostii, a hybrid of inconstant origin, will be briefly mentioned in connection with Pyrus Aria.

Pyrus Aucuparia and its allies will be dealt with in a subsequent part.

The two following species are not known to us to attain timber size in cultivation in the British Isles; but Mr. H.C. Baker tells us that at Chilternhouse, near Thame, there is a specimen of P. vestita 50 feet by 6 feet 5 inches.

 

Pyrus lanata, Don.^[2]

Known in gardens as Sorbus majestica. A tree of the eastern temperate Himalayas; leaves large, oval, oboval, or broadly oblong, with serrate lobes, glabrous above when adult except for some pubescence along the midrib, greyish woolly beneath. Flowers white in densely woolly corymbs; petals glabrous within; styles 2 to 5, free, densely tomentose. Fruit large, about an inch in diameter (½ to 1½ in.), narrowed to the base, red, edible. Judging from wild specimens the foliage is very variable; and the cultivated specimen at Kew bears leaves (figured in Plate 43), which differ from those of wild trees in being less deeply lobed.

 

Pyrus vestita, Wall.^[3]

Often known in gardens as Sorbus nepalensis or Sorbus magnifica. A tree of the temperate Himalayas from Garwhal to Sikkim. Leaves (cf. Plate 43) very large, ovate-acute or elliptic, lobulate-serrate, densely covered with white wool when they first appear, but later in the season becoming shining green and glabrous above, remaining densely woolly beneath. Flowers in very woolly corymbs; petals woolly within; styles 3-5, tomentose only at the base. Fruit large, about ¾ inch in diameter, globose.

 

 

PYRUS SORBUS, True Service^[4]

 

Pyrus Sorbus, Gaertner,^[5] De Fruct. ii. 43, t. 87 (1791), Loudon, Arb. et Frut. Brit. ii. 921 (1838).

Pyrus domestica, Ehrhart,^[5] "Plantag," 20, ex Beiträge zur Naturkunde, vi. 95 (1791); Smith, Eng. Bot. t. 550 (1796).

Sorbus domestica, Linnæus, Sp. Pl. 477 (1753).

Cormus domestica, Spach, Hist. Vég. Phan. ii. 97 (1834).

 

A tree, attaining a height of 60 to 80 feet. Bark, like that of the common pear, dark brown, Assuring longitudinally, and scaling off in narrow, rectangular plates. Leaves pinnate: 6 to 9 pairs of sessile leaflets and a terminal stalked leaflet. Leaflets linear oblong, almost equal-sided at the base, and acute at the apex, serrate with acuminate teeth, except towards the base where they are entire; dull green above, paler below, glabrous on both surfaces when mature, some pubescence often, however, remaining underneath. Flowers white, in short pubescent corymbs; styles 5, united at the base and woolly in their whole length. Fruit either pear- or appleshaped, generally green, tinted with red on one side, 5-celled, about an inch in diameter.

The fruit apparently varies much in flavour, but in good varieties is agreeable though astringent. The French proverb, Ils ne mangent que les cormes, applied to destitute persons, would indicate that the fruit was poor; and this is doubtless often the case. In parts of France a perry is made from them, and they are also preserved dry like prunes. At Vevay^[6] in Switzerland there are avenues planted, consisting of service trees of various kinds; and the brilliancy of the fruit and of the hues of the foliage in October give a very fine effect.

 

Varieties

Two well-marked forms occur, one maliformis,^[7] with apple-shaped fruit, the other pyriformis,^[7] with pear-shaped fruit. There would seem, however, to be in France, though little known to planters in general, varieties which produce fruit of a superior kind. Two of these are strongly recommended by a writer in the Journal of the French National Horticultural Society:^[8] one discovered on the estate of M. Dufresne, near Bordeaux, which has large pyriform fruits of a carmine yellow, produced in large bunches and excellent in flavour, as soon as they commence to mellow; the other was also found growing wild in woods belonging to M. Lafitte at Agen, which has fruit of a bright pink colour.

 

Identification

In summer the tree is only liable to be confused with the mountain ash and its allies. The bark is, however, different, being rough, scaly, and dark-coloured in the true service tree, smooth and grey in Pyrus Aucuparia, etc. In Pyrus Sorbus the leaflets at the base are practically symmetrical, and their serration is very acute. Buds if present are the best distinction, as explained below.^[9] In winter Pyrus Sorbus is distinguished by the following characters, shown in Plate 45:—

Twigs: long shoots glabrous, round; leaf- scars, crescentic with 5 bundle dots, set parallel to the twig on projecting cushions. Terminal buds larger, side buds coming off at an acute angle; all ovoid, densely viscid, shining, generally pubescent at the tip. Bud scales few in number, greenish, sometimes reddened, viscid, quite glabrous, the margin without cilia. Short shoots ringed, glabrous, ending in a terminal bud. The viscid greenish buds, 5-dotted leaf-scars, and rough scaly bark, distinguish this species from other kinds of Pyrus.

 

Distribution

The Service tree is largely cultivated in central and southern Europe; and in many places, where it is recorded as wild, is really only an escape from cultivation. It is met with in the forests of France which rest upon limestone; but in the north and east it does not produce fruit every year, and is doubtfully wild except in the south and west. Willkomm considers it to be wild in the southern parts of the Austrian empire (Dalmatia, Croatia, Banat, Carniola, and South Tyrol), in the valley of the Moselle, in the Jura and Switzerland; also in southern Europe and Algeria. In France it is occasionally met with as a standard in coppiced woods.

Mouillefert says that the tree may live to be 500 or 600 years old, and that it was uninjured by the severe frost of 1879–80, when the thermometer fell to –25° Reaumur. He says, also, that it prefers a rich calcareous soil, but will grow on sand if not too dry.(A.H.)

 

Remarkable Trees

Pyrus Sorbus is not a native of Britain, though a single specimen which grew in a remote part of Wyre Forest in Worcestershire was long considered to give it a claim to be introduced into the British flora. This tree was mentioned in the Philosophical Transactions^[10] as long ago as 1678 by Mr. Pitt, who says that he found it in the preceding year as a rarity growing wild in a forest of Worcester, and identifies it with the Sorbus pyriformis of L'Obelius, a tree not noticed by any preceding writer as a native of England. Pitt says nothing about the size of the tree, merely observing: "It resembles the Ornus or quicken tree, only the Ornus bears the flower and fruit at the end, this on the sides of the branch. Next the sun, the fruit has a dark red flush, and is about the size of a small jeneting pear. In September, so rough as to be ready to strangle one. But being then gathered and kept till October, they eat as well as any medlar."

Ray's account^[11] in 1724 is as follows: "The true Service or Sorb. It hath been observed to grow wild in many places in the mountainous part of Cornwall by that ingenious young gentleman, Walter Moyle, Esq., in company with Mr. Stevens of that county. I suspect this to be the tree called Sorbus pyriformis, found by Mr. Pitt, alderman of Worcester, in a forest of that county, and said to grow wild in many places of the morelands in Staffordshire by Dr. Plot, Hist. Nat. Stafford, 208." In modern times the tree has, however, never been found wild in any part of Cornwall or Stafford, and probably it was confused with Pyrus latifolia.^[12]

Nash,^[13] in 1781, refers to the tree in the Wyre forest as occurring "in the eastern part of Aka or Rock parish, about a mile from Mopson's Cross, between that and Dowles Brook, in the middle of a thick wood belonging to Mr. Baldwyn, which I suppose to be the Sorbus sativa pyriformis, mentioned by Mr. Pitt in the Philosophical Transactions for 1678, called by the common people the Quicken pear tree."

This tree was figured by Loudon,^[14] t. 644, from a drawing sent him by the Earl of Mountmorris. The Rev. Josiah Lee, rector of the Far Forest, told Mrs. Woodward that the old inhabitants of the district, where it was called the " Whitty Pear tree," used to hang pieces of the bark round their necks as a charm to cure a sore throat. Lee's Botany of Worcestershire, 4, gives a good figure of this tree "from a sketch taken many years ago," and another as it appeared in 1856; and says that in 1853 it was in a very decrepit state, producing a little fruit at its very summit. It was burnt down in 1862, by a fire kindled at its base by a vagrant. In a note Lees says that he thinks the tree must have been brought from Aquitaine and planted beside a hermitage in the forest, of which no trace is left but a mound of stones overgrown by brambles. He found the privet and Prunus domestica occurring near it, and nowhere else in the forest.

A seedling (Plate 46) from the Wyre forest tree is growing on the lawn at Arley Castle, near Bewdley, formerly the property of Lord Mountmorris, now the residence of Mr. R. Woodward. I measured it in 1903, when it was 55 feet high by 7 feet 4 inches in girth, and quite healthy, though a large hole in the trunk has been filled with cement. A few seedlings have been raised from it at Arley, but grow very slowly.

There is a large healthy tree in the park at Ribston Hall, Wetherby, the seat of Major Dent, of the pyriform variety, which in 1906 I found to be about 65 feet high by 9 feet in girth, and bearing fruit. This tree was probably brought from France by the same Sir. H. Goodricke who sowed the original Ribston pippin in 1709.

At Croome Court, the seat of the Earl of Coventry, there are two good-sized trees in the shrubbery, one of which is 59 feet high and 6 feet 2 inches in girth. The other, with a clean stem, about 50 feet by 7, is beginning to decay.^[15] Lady Coventry told me that the fruit, which is only produced in good seasons, makes excellent jam when fully ripe, but some seeds which she was good enough to send me did not germinate.

Loudon mentions a tree at Melbury Court, Dorsetshire, estimated to be 200 years old, and 82 feet high, with a diameter of 3 feet 4 inches, growing in dry loam on sand. If this was really a true sorb, it must have been the largest on record, but I learn from the gardener at Melbury that it has long been dead.

There are two good-sized trees at Painshill, and another at Syon which Henry found to be 44 feet high and 6 feet 9 inches in girth, but on this heavy soil the tree does not seem to be so long lived, and is dying at the top.

In the Botanic Gardens at Oxford are two well-shaped trees of this species, which were laden with fruit in 1905, and supposed to have been planted by Dr. John Sibthorp, who was Professor of Botany in 1784-95. The largest measures about 50 feet by 5 feet, and is of the maliform variety. Its fruit, which ripens and falls about the middle of October, is very sweet and pleasant to eat, much better than medlars, whilst the fruit of the other, which is the pyriform variety, does not turn red, is smaller, and ripens later. I have raised seedlings from both of these trees.

In the Cambridge Botanic Garden there is a tree with very upright branches, which measured, in 1906, 42 feet by 3 feet 4 inches.

At Tortworth there is a healthy, well-shaped tree, not more than 40 years planted, which is about 40 feet by 5 feet 1 1 inches. This is in a rather exposed situation, and it had no fruit in 1905.

At Woodstock, Co. Kilkenny, Ireland, there is a tree which seems to be the largest now living in this country. Henry measured it in 1904 and found it 77 feet high by 10 feet 8 inches in girth, with a bole dividing into three stems at 10 feet from the ground and bearing fruit.

 

Timber

A large tree was blown down at Claremont Park, Surrey, the seat of H.R.H, the Duchess of Connaught, in 1902, which I am assured by Mr. Burrell, the gardener there, was a sorb.^[16] Its trunk was sent to Mr. Snell of Esher, to whom I am indebted for two fine planks of its wood. These show a very hard, heavy, compact surface of a pinkish brown colour with a fine wavy grain, which takes an excellent polish, and this wood has been used with beautiful effect in the framing of a brown oak chest, made for me by Messrs. Marsh, Cribb, and Co., of Leeds. Mouillefert says it is one of the hardest and most valuable woods grown in France, and is especially sought for by engravers, carvers, turners, and gun-makers. It seems to be difficult both to propagate and to grow, at least in its youth, and Loudon says that though it may be grafted on the pear or the mountain ash, it is one of the most difficult trees to graft, and that it will not layer successfully, and that it grows very slowly from seed, not attaining more than i foot high in four years. Seeds^[17] sown in autumn germinate in the following spring. The young seedling has two oval entire-margined cotyledons, and attains about 4 inches in height in the first year's growth. Plants may be had from the French nurserymen. Mr. Weale, of Liverpool, reports as follows on a sample of this wood which I sent him:—"The wood is close and homogeneous in texture, tough, but inclined to be brittle. Rays on transverse section invisible, and rings only to be distinguished by the difference in colour of the spring and autumn wood. Harder than whitebeam, seasons well, without warping or splitting, and with little shrinkage. A reliable wood when thoroughly dry."(H.J.E.)

 

 

PYRUS TORMINALIS, Wild Service

 

Pyrus torminalis, Ehrhart, "Plantag." xxii. ex Beit. zur Naturkunde, vi. 92 (1791); Loudon, Arb. et Frut. Brit. ii. 913 (1838); Conwentz, Beob. über Seltene Waldbäume in West Preussen, 3 (1895).

Cratægus torminalis, Linnæus, Sp. Pl. 476 (1753).

Sorbus torminalis, Crantz, Stirp. Austr., ed. 2, fasc. ii. 85 (1767).

Torminaria Clusii, Roemer, Synopsis, iii. 130 (1847).

 

A tree, attaining exceptionally a height of 80 feet, but more generally only reaching 40 or 50 feet. Bark smooth and grey at first, but after fifteen or twenty years of age scaling off in thin plates, and ultimately becoming fissured. Leaves long-stalked, broadly oval, nearly as broad as long, with a cordate or truncate base and an acute apex; with 6-10 triangular acuminate serrate lobes; shining and glabrous above, obscurely pubescent beneath; nerves pinnate, 5-8 pairs. Flowers white in corymbs. Styles 2, glabrous, united for the greater part of their length. Fruit ovoid, brownish when ripe, with warty lenticels, vinous in taste when in a state of incipient decay; cut across transversely it shows a ring of white hardened tissue, forming a mesocarp around the core.

The leaves are generally described as glabrous on the under surface, but in all specimens traces of pubescence may be observed, which is much more marked on coppice shoots and epicormic branches.

 

Varieties

None have been obtained in cultivation so far as we know, and wild trees vary very little in any of their characters. A variety, pinnatifida, with the lobing of the leaves very deep, is described by Boissier,^[18] from specimens occurring in Asia Minor and Roumelia.

 

Identification

The leaves in summer are unmistakable (see Plate 44), and can only be confounded with certain forms of Pyrus latifolia; but in the latter species the under surface of the leaf is always plainly grey tomentose, and the lobes are much shorter than in P. torminalis. In winter the following characters, shown in Plate 45, are available.

Twigs: long shoots, glabrous, shining, somewhat angled, with numerous lenticels; leaf-scar semicircular with 3 bundle traces, set parallel to the twig on a greenish cushion. Buds almost globular, terminal larger, side-buds nearly appressed to the twig; scales green with a narrow brown margin, glabrous, with the apex double-notched. Short shoots slightly ringed, glabrous, ending in a terminal bud. (A.H.)

 

 

Distribution

The most complete and recent account of the tree is a monograph cited above with maps by Prof. H. Conwentz, who describes at great length the various places where the tree is found and the conditions under which it grows. It is widely distributed throughout most of the woods and forests of Europe, but does not occur in Scandinavia, Holland, or the greater part of Russia, where it is only met with in the southern provinces. It also occurs in the Caucasus, Asia Minor, Syria, and in Algeria, It is found on most geological formations, including granite in the Vosges, gneiss in Siberia, and basalt in Austria; and it prefers a soil rich in humus, Willkomm says that on mountains it is commoner on limestone than on other soils, but the French foresters say that it is practically met with on all soils that are not very dry or very wet. It is a tree of the lowlands and hills, attaining 700 metres altitude near Zurich, 1200 in Herzegovina, and 1900 in the Caucasus, It occurs more or less rarely over all parts of Germany, especially in the north-east, and it attains its maximum size in the royal forest of Osche in West Prussia. The largest tree known to Conwentz was "25 metres high, with a clean stem of 12 metres, and a girth at I metre from the ground of 2 metres." The age of this tree was estimated from the rings in the broken trunk of another tree at 235 years.

The scarcity of the tree, as a rule, cannot be accounted for by any deficiency in reproductive powers, for the fruit is produced in some abundance in good years; and being eaten by many birds and animals, among which the waxwing chatterer, the nutcracker, and the fieldfare are mentioned, the seeds must be widely dispersed, while the freedom with which the roots produced suckers is remarked upon.

The timber seems to be much more highly valued in Germany than here, from 18 to 52 marks per cubic metre being given for it, according to quality, in places where hornbeam is only worth 11 marks. It is very hard and durable, and takes a fine polish like that of maple.

In the Hartz mountains and Thuringia it is known as "Atlasholz," and is much used and valued for furniture making.

The fruit is not so much valued as formerly, when it was sold in Prague and Vienna in the winter at the market, and also in Wurtemburg, under the names of Haspele, Arlesbeere, or Adlsbeere.

Conwentz says that the Latin name torminalis was derived from the Latin word tormina, and given on account of the properties of the fruit, to which one of its names in England, "griping service tree," also has reference.

In Upper Alsace a spirit is distilled from the fruit, which tastes something like Kirschwasser.

 

Distribution and Remarkable Trees in England

Pyrus torminalis does not occur as a wild tree, and is rarely planted in Ireland, Scotland, or the North of England. Its range is from Anglesea and Nottingham southwards. It is known as the service or griping service tree; and in Kent and Sussex the fruit is called chequers. It is found in woods, copses, and hedgerows, usually on loam or clay, but does not seem to grow on sandy soils. It attains its greatest size in the Midland counties, where it reaches a height of from 50 to 70 feet. It never seems to be gregarious, and though it reproduces itself by seed or suckers, yet being usually looked on as underwood and not allowed to grow up to its full size, does not attract notice, and is unknown except to the most observant woodmen, even in districts where it occurs. In the vale of Gloucester, on the Earl of Ducie's property, there are probably thirty or forty trees of it scattered over a considerable area. The tallest of these, though not the thickest, is in Daniel's Wood, and is figured (Plate 47). This tree was 62 feet by 5 feet 1 inch in 1904, and is still growing vigorously among other trees and underwood. Not far off is another which may grow to as fine a tree.

In the Cotswold Hills the tree is very rare. I only know a single decaying specimen of moderate size in Chedworth Wood, close to the road leading from Withington to Chedworth Downs.

On Ashampstead Common, in Berks, I found one about 65 feet high and 8 feet in girth, crowded among other trees, which had produced a few suckers. At Rickmansworth Park, Herts, are two fine trees growing together by a pond, which are probably planted (Plate 48), and which Henry measured in 1904, when they were 65 feet by 8 feet 3 inches, and 63 feet by 9 feet i inch respectively. The largest specimen known to me is at Walcot, Shropshire, growing on a bank in good soil with a wych elm crowding it on one side, but probably planted. It measures as nearly as I could estimate about 80 feet high, and is no less than 8 feet 9 inches in girth.

At Cobham Hall, Kent, a tree planted beside a pond measured 55 feet high by 4½ feet in 1906.

In the woods in Worcestershire, as I am informed by Mr. Woodward, it is not uncommon, but is not looked upon as of any value. There is a tree at Arley 5 feet I inch in girth. In Wych wood Forest, Oxon, now nearly all destroyed, I am informed by Mr. R. Claridge Druce, of Oxford, the tree was formerly common enough for its fruit to be collected and sold in Witney market under the name of service berries.

In Cornbury Park, and in the remains of the forest outside it, there are at least six good-sized trees surviving. Of these the largest, just outside the park wall on the south side, is 65 to 70 feet high by 6 feet 6 inches in girth, a well-shaped, vigorous tree, which on 16th October 1905 was covered with unripe fruit. Another, also outside the park, is about 50 feet by 6 feet 10 inches, with a fine clean bole 12 to 14 feet long. I saw no suckers or seedlings near these trees.

In the woods and coppices north and north-east of London, and in Herts, the tree is not unfrequent on clay soil, and Pryor^[19] gives several localities for it.

In Epping Forest, Mr. E.N. Buxton tells me that he does not know of more than thirty trees on an area of 3000 acres, growing on heavy gravelly clay. The largest in his grounds is 40 to 50 feet high.

In Sussex, Sir E. Loder knows it as an uncommon hedgerow tree of no great size, and Mr. Stephenson Clarke, of Borde Hill, also tells me that it occurs there, and more commonly in the Isle of Wight, where, in Bridlesford Copse, Woolton, a wood of about 200 acres, are perhaps two dozen old trees, which differ in appearance from the Sussex ones in assuming a somewhat pendulous habit of branching when well grown. I find no mention of the occurrence of the tree in the New Forest, and the Hon. G. W. Lascelles does not know it there.

We have no record of Pyrus torminalis as a planted tree in Scotland, except that the Rev. Dr. Landsborough^[20] notes a tree in vigorous health in Bellfield Avenue, Kilmarnock, which was 2 feet 9 inches in girth in 1893. He calls it the English service tree or table rowan, and adds that, in spite of its Latin name, the fruit is pleasant. In Ireland the tree is very rare. Henry saw, however, a fine specimen in 1903, at Adare, Limerick, which measured 53 feet by 5 feet 10 inches.

The fruit is ripe late in October, when it falls, if not previously eaten by birds, and the seeds, which only seem to mature in warm summers, should be sown at once, or kept in sand exposed to the weather and sown in spring, when they will germinate the next year. Seedlings raised by me from seed gathered at Les Barres, France, which were sown 7th July 1902, germinated 9th March 1903, and were on 14th October 1904 1 to 2 feet high. The leaves turn a reddish yellow in autumn, when the tree is decidedly ornamental, though, on account of its slow growth, it does not seem to have any value as a forest tree, and is rarely procurable from nurserymen in this country.

 

Timber

Pyrus torminalis is unknown as a timber tree in the trade owing to its scarcity, and is mentioned by Boulger^[21] only as "a small tree, sometimes 30 feet high, with wood practically identical in character and uses with that of the rowan." Stone does not mention it at all, and Marshall Ward, in his edition of Laslett, says nothing worth quoting.

I am indebted to Mr. Stephenson Clarke for a log of the timber, which resembles that of the whitebeam tree, being hard, heavy, and, according to Loudon, weighing, when dry, 48 lbs. per cubic foot.

Mr. Weale, of Liverpool, reports as follows on a sample of this wood which I sent him:—"Of a hardness between true service tree and whitebeam. Rays on transverse section just visible, a little narrower than sycamore, but wood generally exhibits similar characters. Takes a good finish, but this is not lasting, the ring boundaries rising after exposure. Seasons fairly well, shrinks a little, and rather inclined to twist."

Evelyn says that "the timber of the sorb is useful to the joiner, of which I have seen a room curiously wainscotted; also to the engraver of woodcuts, and for most that the wild pear tree serves."(H.J.E.)

 

PYRUS LATIFOLIA, Service Tree of Fontainebleau

 

Pyrus latifolia, Boswell Syme, Bot. Exchange Club Report, 1872–1874, p. 19 (1875).

Pyrus rotundifolia, Bechstein, N.E. Brown in Eng. Bot. iii. ed. Suppl. 164 (1892).

Cratægus latifolia, Lamarck, Flore Française, ed. i. 486 (1778).

Sorbus latifolia, Persoon, Syn. Pl. ii. 38 (1807).

 

A tree, attaining a height of 60 feet in France, with smooth, grey bark, which becomes fissured at the base in old trees. Leaves broadly oval, with a broad, rounded, or truncate base and an acute apex; margin with small triangular lobes, decreasing in size from the base of the leaf upwards, dentate and mucronate, the sinuses opening between the lobes almost at a right angle. The leaves are firm in texture, shining and glabrous above, tomentose and greyish green beneath, with 6 to 10 pairs of lateral nerves prominent underneath. Flowers in moderate-sized corymbs, never long peduncled. Fruit globular, ½ inch diameter, smooth, reddish, marked with brown dots, flesh edible; containing two cells, one seed in each cell, or more often one cell with one seed, the other cell containing two aborted ovules.

The description just given is drawn up from Fontainebleau specimens; and trees absolutely identical are said to occur in various forests in Seine-et-Oise, Seine-etMarne, Marne, Aube, and Yonne.

A series of forms,^[22] however, occur in the forests of the east of France, in AlsaceLorraine, Spain, Switzerland, Austria-Hungary, and Bosnia, which differ slightly in the general outline of the leaf and in the colour and marking of the fruit; and these are supposed to be hybrids between Pyrus Aria and Pyrus torminalis, between which species they oscillate in the characters of the foliage and fruit; whereas, according to French botanists, the tree of Fontainebleau is a true species, as it reproduces itself naturally by seed; and, moreover, one of the supposed parents, Pyrus Aria, is not, according to Fliche, wild in the forest of Fontainebleau.^[23] However, the differences are trifling; and it is convenient, in the present state of our knowledge, to treat these supposed hybrids as varieties of Pyrus latifolia.

 

Varieties

Var. rotundifolia (Bechstein).^[24] Leaves broadly oval or suborbicular, sometimes even broader than long, truncate or rounded at the base, sub-obtuse at the apex; lobes obtusely cuspidate.

Var. decipiens (Bechstein).^[25] Leaves elongated with acute bases, much resembling those of Pyrus intermedia, except that the lobes are triangular pointed, and not rounded as in that species, the sinuses never being acute at their bases.

Var. semilobata (Bechstein).^[26] Leaves oval or elliptic oval, acute at the apex, narrowed at the base, lobes sharply cuspidate.

 

Identification

In summer the leaves are distinguishable from those of Pyrus intermedia by the characters of the lobes and sinuses; while broad-leaved forms differ from Pyrus torminalis in being tomentose beneath, the lobes never being so long as in that species. The tomentum wears off the under surface of the leaf towards the end of the season, and is never so dense or so persistent as in intermedia. On Plate 44 figures are given of leaves from wild trees occurring at Symond's Yat (Fig. 9) and Minehead (Fig. 11), and from a cultivated tree at Kew (Fig. 12). In winter a tree cultivated at Kew showed the following characters, represented in Plate 45.

Twigs: long shoots, shining, round, glabrous, except for a little pubescence near the tip; lenticels numerous as oval prominent warts; leaf-scars set somewhat obliquely on prominent, often greenish cushions; crescentic with three bundle dots, of which the central one is the largest. Terminal bud oval, much larger than the side buds, which come off the twigs at a very acute angle, with their apices bent inwards. All the buds are viscid, pubescent at the tip, and composed of oval scales, which are keeled on the back, ciliate in margin, and short-pointed at the tip. Short shoots ringed, slightly pubescent, ending in a terminal bud. In the specimens examined the leaf-scar at the base of the terminal bud had acute lateral lobes not observed in other species of Pyrus; but these are probably not always present.

 

Distribution

The tree was first discovered in the forest of Fontainebleau,^[27] and was described by Valliant^[28] as "Cratægus folio subrotundo, serrato, et laciniato."

Duhamel du Monceau gave a figure of the leaf in his classic work.^[29] The distribution on the Continent of the type, and of the forms allied to it, has been given above.

In England a small tree, of somewhat rare occurrence, grows wild in woods in Cornwall, South Devon, and Gloucestershire,^[30] which is very near to, if not absolutely identical with, the Fontainebleau tree, as some of the specimens have leaves which resemble rather those of the varieties rotundifolia and semilobata. The South Devon tree produces fertile seed,^[31] which has been planted, and the offspring differs in no respect from the wild trees. In English trees the flowers are reported to have a disagreeable odour,^[32] and the fruit ripens in the end of October or November. When fully grown, but still hard, it is olivaceous brown in colour, with numerous scattered small brown or grey dots; but when quite mature it becomes reddish. At Minehead in Somerset, the Nightingale Valley and Leigh Woods near Bristol, and at Castle Dinas Bran, Denbigh, the variety decipiens occurs.^[33] Mr. E.S. Marshall observed a remarkably fine specimen with good fruit on the Conan river in East Ross-shire; but as no other specimen was seen this tree is probably not wild in that locality. The tree in Earl Bathurst's woods near Cirencester has given rise to some difference of opinion. It was identified at Kew as Pyrus intermedia; but in the specimens which I have seen' the leaves have the triangular lobing and tomentum of Pyrus latifolia, and I have no doubt that it is this species.^[34] Its foliage is very variable, some leaves being broad, with rounded bases like the type, whilst others have narrowed bases, and approximate in outline to the decipiens variety.

 

Remarkable Trees

Pyrus latifolia is seldom planted except in botanical gardens, as at Kew, Edinburgh, and Glasnevin. There are several fine trees at Edinburgh, one of which was figured in the Gardeners' Chronicle^[35] for 1882, when it was 45 feet high by 5 feet 3 inches in girth. Professor Balfour had the tree measured again in January 1904, when it was 45 feet high by 6 feet 6 inches in girth. A year or two before it was considerably pruned on the top branches, and this probably accounts for it not being higher in 1904 than it was in 1882. Professor Balfour kindly sent me specimens of the Edinburgh trees, which, though they differ slightly, are all referable to Pyrus latifolia. He informed me that while the birds eat the fruit off one tree as soon as it is ripe, in another the fruit remains on the tree untouched. The variability of the fruit in this species is remarkable, and points undoubtedly to hybrid origin.

A tree exists at Oakleigh House,^[36] near Keynsham in Somerset, which was planted many years ago.(A.H.) My attention was called by Mr. R. Anderson of Cirencester to a very remarkable tree growing in a part of Earl Bathurst's woods about two miles from Cirencester, known as the Dear Bit. The tree, though it has lost some of its principal branches, is still, as our illustration shows (Plate 49), a very handsome one, and in size exceeds any other of the kind of which we have a record, either in this country or on the Continent. It is, as nearly as I can measure it, about 75 feet high by 11 feet in girth. It grows on dry shallow soil of the Oolite formation, and is close to a ride, which leads me to suppose that it was planted perhaps at the time when the park was laid out. It is near the north-east edge of the wood, and open to the southwest. I have never seen the flowers of this tree, which bears fruit only in favourable seasons near the ends of its uppermost branches, and as the birds are fond of it, and even in good years many of the seeds are immature, I have not until 1904 been able to procure any. A few of these have now produced small plants,

I have been unable to find any self-sown seedlings near this tree, and though there are one or two good-sized P. torminalis in another part of the park, probably planted, none of them approach it in size. As to the possible age of this tree, I can only say that the drive on the edge of which it grows has, as I am told by Mr. Anderson, certainly been in existence over 100 years, and the bank was covered with old beech, which were cut in 1892. The tree has become one-sided from the pressure of a beech which until then closed it in on the south-west side, where it is now open. As these beeches were 150 years old or more, the tree may be now from 150 to 200 years old, and it seems very probable that the person who designed this park had seen the tree at Fontainebleau, and introduced it when Oakley Park was planted by the ancestor of the present Earl Bathurst in Queen Anne's reign.(H.J.E.)

 

 

PYRUS INTERMEDIA, Swedish Whitebeam

 

Pyrus intermedia, Ehrhart, Beiträge zur Naturkunde, iv. 20 (1789); Loudon, Arb. et Frut. Brit. ii. 915(1838).

Pyrus scandica, Ascherson, Fl. des Prov. Brandenburg, i. 207 (1864).

Pyrus suecica, Garcke, Fl. Deutschland, ed. ix. 140 (1869); Conwentz, Beob. über Seltene Waldbäume in West Preussen, 81 (1895).

Sorbus scandica, Fries, Flora Hollandica, 83 (1818).

Sorbus intermedia, Persoon, Syn. Pl. ii. 38 (1807).

Sorbus Mougeoti, Soyer-Willemet et Godron, Bull. Soc. Bot. de France, v. 447 (1858).

Cratægus Aria scandica, Linnæus, Amœn. Acad. 190 (1751).

Cratægus Aria suecica, Linnæus, Sp. Pl. 476 (1753).

 

A shrub or small tree attaining a height of 20 to 50 feet. Leaves stalked, oval or elliptic, rounded or cuneate at the base, pointed at the apex; mai-gin lobed, lobes diminishing in size from the base upwards, rounded, toothed, shortly acuminate, separated by sinuses which are very acute or almost closed at their bases; upper surface green, shining, glabrous when adult, lower surface greyish tomentose. Flowers in branching corymbs, with pleasant odour; petals spreading, tomentose; styles 2, free, tomentose at the base. Fruit oval, red, sweet-flavoured, smooth or slightly dotted.

 

Varieties

1. Scandica.^[37] Leaves less narrowed and almost rounded at the base, deeply lobed, with numerous sharp teeth; 6-8 pairs of nerves. Fruits large, surmounted by the curved and outwardly-reflected calyx teeth.

2. Mougeoti.^[38] Leaves narrowed at the base, slightly lobed, with few short teeth; nerves 9–12 pairs. Fruit very small, surmounted by erect and inwardly-curved calyx teeth.

3. Minima.^[39] Leaves linear-oblong, with 3–4 pairs of lobes, variable in size, but generally deepest at the middle part of the leaf; nerves 6–8 pairs. Flowers—early in June—in loose corymbs, not flat-topped, small, and resembling those of Pyrus Aucuparia. Fruit small, globose, bright red, surmounted by erect calyx lobes.

 

In Plate 44 figures are given of the leaves of var. scandica from Bergen (Fig. 19), of the variety from Great Doward in Hereford (Fig. 10), and of var. minima from Breconshire (Fig. 17).

 

Identification

In summer the greyish tomentum of the leaves underneath, and the rounded lobes, with sharp sinuses which are almost closed at their bases, will distinguish the tree from Pyrus Aria and Pyrus latifolia, the species nearly allied. In winter the following characters are available, as shown in Plate 45:—

Twigs: long shoots, round, glabrous, often with waxy patches; lenticels long, numerous. Leaf-scars: crescentic, with 3 equal-sized bundle dots, obliquely set on a brownish projecting cushion. Buds glistening, pubescent at the tip; terminal much the largest; side-buds arising at an acute angle with their apices directed inwards. The bud-scales have a dark-coloured rim to the ciliate margin, and their apex is scalloped with a central projection ending in a tuft of long hairs. The short shoots are ringed, pubescent, with a terminal bud.

 

Distribution

The variety minima occurs only in Breconshire, on the limestone mountain cliff Craig Cille, near Crickhowell, and at Blaen Onnen, two miles to the west of Craig Cille, and is a small shrub clothing the cliffs up to 2000 feet altitude. The flowers and fruit are very similar to those of the mountain ash; and Koehne supposes it to be a hybrid between Pyrus intermedia and Pyrus Aucuparia, which occur in the same locality.

The form Mougeoti, which is considered by many botanists to be a distinct species, occurs in Lorraine, the Vosges, Jura, Suabian and Western Alps, and in the Carpathians. It never attains a great size, being either a low bush or a small tree 15 to 30 feet in height. In Piercefield Park, Monmouth, Great Doward in Hereford, and a few other localities in the west of England, a shrub or small tree has been found which is near this form.^[40]

Var. scandica has been found in Britain, in a few localities in Denbighshire and Breconshire,^[41] and also at Chepstow^[42] in Monmouthshire, always growing on limestone rocks. It was supposed to grow also in Arran, but Koehne,^[43] as will be seen in our account of the peculiar forms of that island under Pyrus pinnatifida, denies its occurrence there. (A.H.)

This variety is widely spread in Northern Europe. The best account we know of this tree is by Conwentz, who calls it "Pirus Suecica." He says that most authors speak of it as a small tree or shrub—Koehne only gives it as 7 metres high. It grows on granite, gneiss, chalk, and alluvium, and extends from the island of Aland, South-east Sweden, South Norway, and Denmark, to North-east Germany, where, however, it seems to be quite a rare tree and only recently discovered.

It is represented in France, Switzerland, Austria, and Bosnia by P. Mougeoti, which many botanical authors have mistaken for it, and which, according to Conwentz, can only be distinguished in some varieties by the fruit.

In the island of Oesel, in the Baltic, it is much planted, and often attains 2 metres in girth. Conwentz, however, found wild specimens at Soeginina near Karral, at Pajumois near Keilkond, at Wita Jahn, and in other places—mostly small trees, but in some places attaining 10 metres in height. In the Finnish islands of Aland it is found truly wild, in a few places only, sometimes in company with an allied species, P. fennica. Conwentz identified it at Bergö, Skarpnåtö, Labnäs, and elsewhere. The finest specimen he saw at Östergeta, being 12 metres high and 2 metres in girth.

In south-eastern Sweden it is more abundant, but does not occur in any of the provinces north of Wermland, about lat. 60° N. In the neighbourhood of Stockholm it grows at Stockby to 12 metres in height. In Södermanland and the island of Gothland it is more common.

In Denmark the tree has been found in many places, and is undoubtedly wild near Aarhus in Jutland, in the forests of Adslev, Kolden, and Jexen. I believe that I also saw it in the forest of Roldskov near Aalborg, though I did not at that time distinguish it from Pyrus Aria. In the island of Bornholm it is known under the name of "Axelbar."

In Germany it is confined to a limited area on the coasts of West Prussia and Pomerania, where Conwentz has found it living in six places only—Koliebken, Hoch Redlau, Oxhöft, Karthaus, Gr. Podel, and Markuhle near Kolberg. He gives maps showing the position of the trees in these places, and says that whilst P. torminalis grows in the interior, where the hornbeam is predominant, P. intermedia grows in the country along the coast, where the beech is the prevailing tree. It occurs most commonly in a shrubby condition, the tallest wild one being only 13 metres high by I metre in girth, but one tree at Gross Podel in Pomerania is 1.90 metre in girth, and at Wernigerode, in the Harz, a cultivated tree has attained 17 by 3.17 metres, which is the largest known to Conwentz. He thinks that the scarcity of the tree in Germany arises from its not being indigenous, as no geological evidence exists of its having been formerly commoner, and suggests that it has been introduced from Sweden by birds of passage, such as the waxwing or thrushes, which are fond of the fruit, and may have voided the seeds after migration from the north.

The Swedish name is Oxel, and this name being found in many place and family names in Sweden, shows that the tree was probably more common formerly than at present.

In Norway, Schübeler^[44] says that it is wild only in the most southern parts, as at Porsgrund, Grimstad, and Dalen in Eidsborg, in lat. 59° 42' N. There are large trees at Lunde in Stavanger district growing near the church. In the Botanic Gardens at Christiania I have seen a tree which is about 1 2 metres high and over 2 in girth. It has been planted and grows well at Stenkjser, at the north end of the Trondhjem Fjord. The Norsk name is Maave.

Dr. Brunchorst, Director of the Bergen Museum, informed the Earl of Ducie that Pyrus intermedia, as well as P. pinnatifida (P. fennica), were found on the south-west coast of Norway, and that a hybrid which he calls Pyrus Meinickii, P. fennica × Aucuparia, has also been recently discovered in the "Mosterö Bommel Fjord," Dr. Brunchorst, who has paid much attention to this genus, says that three species which he cultivates at Bergen vary much, and perhaps pass into one another. Lord Ducie has brought plants of these to Tortworth, where he grows them under the name of Pyrus hybrida.

 

Remarkable Trees

This species appears to be now rarely planted, except in botanical gardens. The best specimen which we have seen occurs at Syon (var. scandica). In 1904 it measured 48 feet in height by 7 feet 10 inches in girth, with a bole of 7 feet, dividing into 8 large branches, and forming a wide-spreading crown of foliage, about 50 yards in circumference (Plate 50). Another fine tree is growing at Livermere Park, Bury St. Edmunds, Suffolk, specimens and particulars of which have been kindly sent to us by Mr. Stiling. It is now (1905) 45 feet high by 8 feet 5 inches in girth, with a bole of 8 feet dividing into 1 2 main branches, the diameter of the spread of foliage being 45 feet. This tree was reported^[45] in 1889 to have been 42 feet high by 8 feet 3 inches in girth. In August 1905 it was covered with fruit.(H.J.E.)

There is a fine specimen at Stowe, near Buckingham, growing near the bridge over the lake in sandy soil, which measures about 45 feet high by 7 feet 9 in. in girth, with a 7 feet bole. It was loaded with fruit in August 1905.

At Wykeham Abbey, the Yorkshire seat of Viscount Downe, there is a fine tree on the lawn, about 40 feet high, spreading from the ground, where it measures 10 feet 8 inches in girth, into a large and well-shaped head. '

This tree is planted in some of the parks and gardens in London, and grows well at the Botanic Gardens in Regent's Park. I am informed by Mr. A. Stratford, Superintendent of the Corporation Park of Blackburn, that it makes a good shade tree in that smoky town.

 

 

PYRUS PINNATIFIDA, Bastard Mountain Ash

 

Pyrus pinnatifida, Ehrhart, "Plantag." 22, ex Beiträge zur Naturkunde, vi. 93 (1791); Loudon, Arb. et Frut. Brit. ii. 915 (1838); N.E. Brown, in Eng. Bot. iii. ed. Suppl. 168 (1892); Gard. Chron. xx. 493, fig. 78 (1883).

Pyrus semipinnata, Roth, En. Pl. Phœn. in Germ. i. sect. post. 438 (1827).

Pyrus fennica, Babington, Man. Eng. Bot. ed. 3, p. 111 (1851).

Sorbus hybrida, Linnæus, Sp. Pl. 684 (1762); Schübeler, Viridarium norvegicum, ii. p. 476.

Sorbus fennica, Fries, Summa Veg. Scand. 42 (1846).

 

A species of hybrid origin, occurring as a small tree, which may attain 50 feet in height, with smooth, grey bark. Leaves variable in shape, mostly pinnate or deeply cut at the base, with 1-4 pairs of segments more or less separate; the upper part cut into deep sharp-toothed lobes; green and glabrous above, grey tomentose below. Flowers white in loose corymbs; styles 3, woolly at the base; fruit small, globular, coral red, and resembling that of Pyrus Aucuparia.

 

Varieties

This form, the parents of which are P. Aucuparia and P. intermedia, must be carefully distinguished (see p. 143) from Pyrus hybrida, Moench, a shrub of different origin.

Pyrus Thuringiaca, Ilse,^[46] a cross between P. Aucuparia and P. Aria, is generally included under P. pinnatifida, from which it differs only in the leaf, whiter beneath, having its upper part lobulate or dentate and not deeply lobed.

Sorbus arranensis, Hedlund,^[47] is the name given to a form occurring in the Isle of Arran, which is intermediate between P. pinnatifida and P. intermedia, and closely resembles the latter, differing only in the deeper and more irregular lobing of the leaf.

The hybrid forms, which are intermediate between P. pinnatifida and P. Aucuparia, are generally regarded as varieties (var. satureifolia^[48] and var. decurrens^[49]) of the latter species, and will be mentioned in our account of the mountain ash.

 

Identification

Pyrus pinnatifida and the intermediate hybrids are variable and inconstant in the shape of the leaf There is no difficulty, however, in their identification, if it be noted that hybridity may be suspected in all cases where the leaves vary on the one hand from the regularly pinnate separate leaflets of Pyrus Aucuparia, and on the other from the regular uniform lobing or serration of Pyrus intermedia or Pyrus Aria. In winter, specimens of cultivated Sorbus fennica show the following characters represented in Plate 45:—

Twigs: long shoots glabrous, shining, dark brown, with a few scattered lenticels. Leaf-scar crescentic, very narrow, set obliquely on a reddish brown, slightly projecting cushion; it shows a varying number of bundle traces,^[50] 3, 4, or 5, and may thus be distinguished from other species of Sorbus, as Pyrus Aucuparia has 5 dots on the scar, while Pyrus Aria, intermedia, and latifolia have only 3. Terminal bud large, conic, tomentose, especially at the apex. Lateral buds small, either appressed to the stem or diverging from it at an acute angle. Bud-scales few, densely pubescent on the outer surface, and ciliate in margin. Short shoots ringed, pubescent, bearing a terminal bud.^[51]

 

Distribution

The form fennica occurs plentifully in Scandinavia, where it grows wild, reproducing itself naturally by seed, and behaving as a true species. It extends in Norway, according to Schübeler, up to lat. 66° 14' on the west coast as a wild plant, and in Sweden up to 60° wild and 62" planted; it also occurs in Finland, but is not recorded from other parts of Russian territory. In Central Europe it only occurs sporadically, and apparently always in company with the parent species; it is recorded from various mountain stations in France, Switzerland, Germany, and Austria.

The hybrids which occur in the Isle of Arran have attracted much attention and discussion. Formerly it was believed that Pyrus Aucuparia, Pyrus intermedia (var. scandica), and Pyrus fennica, all occurred in a wild state. Koehne,^[52] however, considers that (excepting Aucuparia) all the plants in question on the island are hybrids, there being two sets, one typical fennica, while the other set comprises forms between that and scandica. This view, which excludes one of the parents (viz. scandica), implies that these hybrids, once established, may under favourable conditions reproduce themselves naturally and behave generally as true species.

N.E. Brown says of this species that it is "rare and perhaps not indigenous except in Scotland"; but he has seen specimens from Kent, Sussex, Hants, Somerset, Gloucester, Leicester, Stafford, Cumberland, Roxburgh, Arran, and Dumbarton. He thinks that Arran seems to be the only truly native locality for this tree in the British Isles, and believes that the Arran plant placed under intermedia is a form of it. Watson, however, states in his Compendium, p. 510, that Borrer held it to be wild in North Hants between Farnham and Farnborough, where it was observed sparingly along with Aria and Aucuparia, both more plentifully. A specimen picked by James M'Nab in Darenth Wood, Kent, is, according to Watson, identical with Arran specimens.

There is a fine tree on the edge of a shrubbery close to Wilton House, Wilts, the seat of the Earl of Pembroke, about 50 feet by 5.

There is a tree at Williamstrip Park, Fairford, the seat of Lord St. Aldwyn, which is in a decaying condition. It consists of a large stool measuring 8 feet at the ground, with four ste«ns about a foot in diameter by about 60 feet high. There is also a tree at Arley Castle 50 feet high by 3 feet 5 inches in girth in 1905. One at Bayfordbury is 40 feet by 5 feet 2 inches, branching at 4 feet into four stems, with numerous ascending branches. At Aldenham Cottage, Letchmore Heath, Herts, is a fine tree 44 feet by 6 feet 2 inches, with a bole of 6 feet. At Danson Park, Welling, Kent, the residence of Mr. Bean, there is said to be a tree about 30 feet high, with a girth of 12 feet 4 inches at 1½ feet above the ground, described to be like a large bush with seven main branches.(A.H.)

 

 

PYRUS ARIA, Whitebeam

 

Pyrus Aria, Ehrhart, Beiträge zur Naturkunde, iv. 20 (1789); Loudon, Arb. et Frut. Brit. ii. 910 (1838).

Cratægus Aria, Linnæus, Sp. Pl. 475 (1753).

Sorbus Aria, Crantz, Stirp. Aust. ii. t. 2, f. 2 (1762).

Aria nivea, Host, Fl. Aust. ii. 8 (1813).

 

A tree in woods and on good soil attaining a height of 40 to 50 feet, and rarely 70 feet in height; but in rocky and mountainous situations usually remaining shrubby. Bark smooth and grey, becoming slightly fissured in old trees. Leaves stalked, oval or obovate, rounded, cordate, or cuneate at the base, sharp or obtuse at the apex, biserrate or slightly lobulate with teeth, the lobules largest towards the apex of the leaf; green and glabrous when adult above, but always snowywhite tomentose beneath; nerves, 8-12 pairs, very prominent on both surfaces. Flowers with an unpleasant odour, white, in loose corymbs; the peduncle, receptacle, calyx, and corolla, white tomentose; styles 2, free, pubescent at the base. Fruit globose or ovoid, inch in diameter, shining red with a few brown dots, tomentose at the base and apex; flesh scanty, sweetish acid in flavour.

 

Identification

In summer the leaves, snowy white underneath and with prominent nerves, are a sure guide. The leaves of Aria from a wild specimen growing at Gosford, Kent (Fig. 18); of var. rupicola from a wild specimen from north-west Lancashire (Fig. 13); of var. salicifolia (Fig. 16) and var. Decaisneana (Fig. 8), both from specimens cultivated at Kew, are shown on Plates 43 and 44. In winter the following characters are available, as shown in Plate 45:—

Twigs: long shoots round, shining brown, glabrous except for a little pubescence near the tip, marked with scattered wart -like lenticels. Leaf-scars set obliquely on prominent leaf-cushions, crescentic, with three bundle traces. Buds ovoid, conicalpointed, shining, and somewhat viscid; terminal larger, side-buds coming off at an acute angle. Bud-scales glued together, strongly keeled, glabrous on the surface, densely long ciliate in margin. Short shoots ringed, generally glabrous, and ending in a terminal bud. Viscid buds occur also in Pyrus Sorbus, which is, however, very distinct in its five-dotted scars and glabrous scales.

 

Varieties

Some authors take Pyrus Aria in a wide sense, and under it group Aria proper, rupicola, latifolia, scandica, etc. as sub-species. Most of these, as being readily distinguishable by many characters, have been considered by us as distinct species. Taking Pyrus Aria in a narrow sense, as comprising forms with leaves snowy white beneath, it exhibits a great variety of forms in the wild state, explained by its wide geographical distribution and its occurrence on different soils and in different situations. Moreover, various horticultural varieties have been produced. The type has been described above; the following is a list of the most important varieties:—

1. Var. rupicola.^[53] Differs from the type in the leaves having fewer nerves, generally 7 (5–9) pairs, less prominent; obovate-oblong in shape, widest above the middle, lobulate above, with the tomentum ultimately becoming slightly grey. Fruit smaller, f inch diameter, carmine-scarlet. This variety occurs always on limestone rocks, and is recorded from many stations in the British Isles. It is probably a form due to poor soil and exposure to wind, and other uncongenial conditions.

2. Var. græca, Boissier.^[54] A shrub occurring in Spain, Albania, Greece, Syria, and Asia Minor. Leaves round, thick, almost leathery in consistence, nerves 6-10 pairs, broad, cuneate at the base, lobulate, with large teeth in the upper two-thirds. This form is also known as Sorbus cretica, Fritsch, and Aria grceca, Decaisne.

3. Var. flabellifolia.^[55] Leaves orbicular, cuneate, or rounder at the base, margin with large incisions, sharply toothed, nerves 3-5 pairs. South-eastern Europe and Asia Minor.

4. Var, Decaisneana, Rehder.^[56] Leaves large, 4-6 inches long by 2–3½ broad, elliptic or oblong, narrow or acuminate at the apex, rounded or subcordate at the base, serrate in almost the whole margin with sharp teeth; nerves 12–14 pairs; petiole channelled above, nearly an inch long. Flowers first white, then becoming pinkish; styles glabrous. Fruit purplish, ellipsoid, crowned by the persistent hairy sepals. This tree is of unknown origin; it has been said to be Himalayan, but I am not aware on what authority. It closely resembles Pyrus lanata from that region.

5. Var. sinensis. Leaves narrow, lanceolate or ovate, with acuminate apex and cuneate base; crenately serrate. A series of forms occur in the mountains of Hupeh in China, where the trees are common at high elevations, and vary in size from 10 to 40 feet in height. Seeds were sent home by Wilson to Messrs. Veitch in 1901, and seedlings, very beautiful in foliage and vigorous in growth, are now growing at Coombe Wood.

6. Var. salicifolia. Leaves narrow, ovate-lanceolate, doubly serrate in margin, with long petioles. Origin unknown.

7. Certain horticultural varieties occur in which the leaves are variously coloured, as lutescens, chrysophylla, sulphurea.

8. Var. quercoides. Leaves regularly lobed with their edges bent upwards.

 

Pyrus Hostii,^[57] Hort., may here be mentioned, as it occurs in cultivation and resembles Pyrus Aria. It is of hybrid origin, one parent being either that species or Pyrus intermedia, while the other is Pyrus Chamæmespilus. It is distinguished from Pyrus Aria by the larger and more irregular teeth of the leaves (cf. Plate 44), and its flowers are pinkish white, borne in loose corymbs. Various intermediate forms have been distinguished, as—

Sorbus ambigua, Michalet. Exactly intermediate between Pyrus Aria and Pyrus Chamæmespilus, with the leaves larger than in the second, and smaller than in the first, and the margins having a tendency to lobing. Tomentum whitish.

Sorbus arioides, Michalet. A form intermediate between ambigua and Aria.

Chamæmespilus x Mougeoti. Leaf large, with lobes well marked and rounded; tomentum greyish. These hybrids are common in the Jura and the Alps.

 

Distribution

The whitebeam is a wide-spread species. It occurs throughout Europe generally, reaching in Norway as far north as lat. 63° 52', and in Sweden to lat. 59°. It is met with also in Algeria, Asia Minor, the Caucasus, Armenia, Siberia, and Central China, assuming in some of these regions remarkable varietal forms. It is replaced in the Himalayas and Japan by Pyrus lanata, Don, an allied species.

While it occurs on all soils except those which are wet, it has a decided preference for limestone. In woods and hedges it grows to be a small tree; but in exposed situations on rocky mountains, etc. it dwindles to a mere bush. On the Alps it ascends to 4800 feet

In the British Isles its distribution has not been accurately made out, as many supposed records refer rather to intermedia or latifolia. Apparently, however, as a wild tree, the typical form is almost entirely confined to the southern and midland counties of England and to south Wales. Variety rupicola is recorded from nearly every county from Devon to Sutherland, and is widely spread on the littoral range between Lancaster and Humphrey Head, ascending in Banffshire to 1200 or 1400 feet, where it has been found by Dr. Shoolbred of Chepstow on limestone cliffs near Inchrory in upper Banffshire. In Ireland the whitebeam is rare and local, and both the type and rupicola occur.(A.H.)

 

Remarkable Trees

By far the finest specimen that we know of in England or elsewhere grows on the edge of Camp Wood, near Henley on Thames, on Sir Walter Phillimore's property, where Henry saw it in 1905. It measures 75 feet high by 4 feet 9 inches in girth, with a bole about 35 feet long, and has very smooth beech-like bark (Plate 51).

There is a large and very well shaped tree at Walcot, Shropshire, which in 1906 Elwes found to be 56 feet high and 6½ feet in girth, with a clean bole about 20 feet long.

There is a handsome tree on the lawn at Belton Park, which measures 41 feet by 6 feet 7 inches.

A very spreading, ill-shaped tree in a thicket at Mount Meadow, near Cobham, Kent, is 9 feet 3 inches in girth.

At Stowe, near Buckingham, there are several fine trees near the Queen's Temple, which are about 50 feet high, but the tree when growing wild on the Cotswold Hills, where it is common, rarely exceeds 30 feet with a stem 2 to 3 feet in girth, and is more usually seen as a bush with many stems.

The whitebeam, like the mountain ash, is occasionally found as an epiphyte growing on other trees, where its seeds have been dropped by birds. Though this is more common in the damp climate of the west of England, yet we know of two cases which are remarkable on account of their situation. One is in the Yew Tree Vale in Surrey, where a whitebeam is growing near the top of a yew tree;^[58] the other is near Colesborne in the Cotswold Hills. In this case a large limb has been torn by the wind from a Scots pine, and in the crevice on the east side of the tree, where but very little vegetable matter has yet had time to form, a healthy young whitebeam, now about 3 feet high, grew for seven or eight years, when it began to lose vigour.

Though it is well known that the decaying mossy trunk of a fallen tree is one of the most favourable situations for the seeds of many conifers to germinate and grow, yet in this case the roots of the whitebeam must derive their nourishment almost entirely from the air, the case being very different from those so often seen in the Himalayas and other countries, where a large quantity of moss, ferns, and decaying vegetable matter accumulate in the forks of large old trees.

The whitebeam is easily propagated by seed, which, if sown in autumn, will germinate partly in the following spring and partly in the second year after sowing. The seedlings grow slowly at first, and require five or six years in the nursery before they are large enough to plant out. When planted on good soil the whitebeam is a very ornamental tree, both on account of its leaves and fruit, which is larger and more abundant than when wild. It is, however, so much liked by birds that it is soon eaten up.

 

Timber

The wood is hard, heavy, and even in the grain, and is white in colour, with some dark spots, and in old trees becomes occasionally tinged with red. It is used on the Continent in turnery and in making tools.

Loudon says that it was used for the axletrees, naves, and felloes of wheels, carpenters' tools, and walking-sticks, but that the greatest use of its wood, until iron superseded it, was for the cogs of small wheels. I have felled a tree 18 inches in diameter, which when cut through was perfectly sound at heart, and was considered to be well suited for chair-making.

In Hampshire^[59] the wood is used for making whip-handles, and the tree is known there on that account as "whip-crop."

Mr. Weale, of Liverpool, reports as follows on a sample of this wood which I sent him:—"The wood is of a medium hardness, good length of fibre, and takes a clean finish. Not tough. Rays invisible on transverse section. Grain moderately close and even. Warps badly in drying, and is liable to split."

In a little book on English timber by "Acorn" (Rider and Son, London, 1904), I find the following note on the whitebeam, though the author does not notice either of the service trees:—"In a green state whitebeam has a strong smell, and even after seasoning this is retained to a certain extent. A great many handles for cutlery are made from it, and its hardness is admirably adapted for these, as it is capable of taking a very high polish from the extreme closeness of its grain. It is also used in the manufacture of musical instruments, and the tops and small pieces are always appreciated by the turner." As it is usually available only in small pieces, these would probably, when thoroughly seasoned, be very useful to introduce as blocks in parquet flooring.(H.J.E.)

 

1. These two sections comprise hybrids, the leaves of which vary in shape, not only on different individual trees, but often also on a branch. Hybrid origin may always be suspected when such variation is observed, or when the lobing or cutting is irregular and not symmetrical.
2. Don, Prodromus, 237 (1825). Hook., Fl. Brit. Ind. ii. 375 (1879).
3. Wallich, Catalogue, 679 (1828). Hook., Fl. Brit. Ind. ii. 375 (1879).
4. Service is commonly derived from the Latin cerevisia, a drink said to have been formerly made of berries of the different species of Sorbus, or to have been flavoured with their leaves. C. Woolley Dod controverts this view in Gard. Chron. 1890, vii. 87, and holds that service is simply a corruption of sorbus, and that cerevisia, a drink, according to Pliny, made of cereal grain in Gaul, was ordinary malt ale.
5. Gaertner's and Ehrhart's names were both published in the same year. Gaertner's preface antedates that of Ehrhart by a few days. Nothing is known for certain of the pamphlet "Plantag" cited by Ehrhart. Which name has priority of publication is uncertain.
6. Woods and Forests, July 16, 1884.
7. Loddiges, Catalogue, ex Loudon, loc. cit.
8. Quoted in Garden, 1886, xxx. 89.
9. The stipules of the various species of the section Sorbus differ considerably in shape, as shown in Plate 43; but they are usually quickly deciduous, and can only assist identification in spring.
10. Phil. Trans., abridged edition, ii. 434 (1809).
11. Ray, Synopsis Methodica, ed. 3, p. 542.
12. Miller, Gard. Dict. iii. ed. (1737), under Sorbus sativa, says, "The manured service was formerly said to be growing wild in England; but this I believe to be a mistake, for several curious persons have strictly searched those places where it was mentioned to grow, and could not find it; nor could they learn from the inhabitants of those countries that any such tree had ever grown there."
13. History of Worcestershire, i. {[sc|ii}}.
14. Loudon gives the measurements in 1838 as 45 feet high, with a diameter of trunk at a foot from the ground of 1 foot 9 inches, and states that it was in a state of decay at that time.
15. Loudon gives the measurements in 1838 as 45 feet high, with a diameter of trunk at a foot from the ground of 1 foot 9 inches, and states that it was in a state of decay at that time.
16. "Among interesting trees to be found at Claremont is a good specimen of the pear-shaped service, carrying a heavy crop of fruit. It is rather over 60 feet high and 7 feet 6 inches in girth at 2 feet from the ground." Note by E.B. in Garden, 1883, xxiv. 422. Mr. E. Burrell gives a fuller account of the Claremont trees in the same journal, 1888, xxxiii. 154, in which he states that he thinks the variety maliformis does not increase in height after it gets to be about 30 feet high, whereas pyriformis at Claremont is close on 70 feet high.
17. Mathieu, Flore Forestière, 184 (1897).
18. Flora Orientalis, ii. 659 (1872.)
19. Flora of Hertfordshire, 154 (1887).
20. Annals of Kilmarnock Glenfield Ramblers' Society, 1894, p. 11.
21. Woods of Commerce, 312.
22. These may be called, if their hybridity is considered to be established, Pyrus Ario-torminalis, Garcke, Flora von Deutschland, ed. 17, 207 (1895). Fliche, in Mathieu, Flore Forestière, 177 (1897), sums up the question thus:—Fontainebleau tree not a hybrid, near to Pyrus Aria, a true species, seed germinating readily and producing natural seedlings; Lorraine tree nearer to Pyrus torminalis than to Pyrus Aria, a true hybrid, seeds rarely perfect. Rouy et Fourcaud, Flore de France, vii. 22 (1901), suggest that the Fontainebleau tree is a hybrid fixed and behaving as a true species. See also Irmisch in Bot. Zeitung, 1859, p. 277.
23. Cf., however, p. 156, note 2.
24. Pyrus rotundifolia, Bechstein, Forstbotanik, 152 and 316, t. 5, 1843.
25. Pyrus decipiens, Bechstein, loc. cit. 152 and 321, t. 7.
26. Pyrus semilobata, Bechstein, loc. cit. 152 and 317, t. 6.
27. I visited Fontainebleau in 1905 on purpose to see this tree at home, and found only small trees of it in full flower on 14th May. I was informed by M. Reuss, Inspector of Forests at Fontainebleau, that the tree grows scattered only in the part which is called Montenflammé and Mont Merle, where the sand is covered by the calcareous strata of Beaune, so that the tree is evidently peculiar to calcareous formations. Formerly the trees were cut with the underwood, but are now reserved on account of their rarity, as well as the whitebeam and P. torminalis, which M. Reuss considers to be indigenous at Fontainebleau, and therefore admits the possibility of their hybridising. The largest tree known to him is on Mont Merle, at the corner of the roads d'Anvers et de l'Echo in the 16th série, and is 40 centimetres in diameter, or about 4 feet in girth at 5 feet from the ground. It is known to the peasants at Fontainebleau as baguenaudier or elorsier, but is generally termed by French botanists alisier de Fontainebleau.—(H.J.E.)
28. Botanicon Parisiense, ed. 3, p. 63 (1727).
29. Traité des Arbres, i. 194, t. 80, fig. 2 (1755).
30. "Occurs at Bicknor, Coldwell, and Symond's Yat, which form a single range of wooded limestone rock in West Gloucestershire, about ¾ mile in length."—Rev. A. Ley, Bot. Exchg. Club Report, 1893, p. 415. "French Hales" is the name given to this species in Devon, according to Britten and Holland, Dict. Eng. Plant Names, p. 194 (1886). They state that the fruits are sold in Barnstaple market. These authors call the tree Pyrus scandica, as, at the time they were writing, its identity with the Fontainebleau tree was not established.
31. Briggs, Jour. Bot. 1887, p. 209, and 1888, p. 236.
32. Briggs, Flora of Plymouth, 144 (1880); and Boswell, Bot. Exchg. Club Report, 1872–74, p. 20.
33. Cf. N.E. Brown, loc. cit. 165. Mr. J. White reports a tree 30 feet high in Leigh Wood (Bot. Exchg. Club Report, 1902, p. 45).
34. Mr. Hickel, Inspecteur des eaux et forêts, who knows the Fontainebleau tree well, and to whom I sent specimens, is of my opinion.
35. Vol. xviii. 749.
36. Jour. Bot. 1899, p. 488.
37. Sorbus scandica. Fries, loc. cit.
38. Sorbus Mougeoti, Soy.-Will. et God. loc. cit.
39. Pyrus minima, Ley, Jour. of Bot. 1895, p. 84, and 1897, p. 289, t. 372; Sorbus minima, Hedlund, Kon. Sv. Veten. Akad. Handl. (1901–2) 60.
40. It is called Pyrus intermedia, Ehrh., by the Rev. Augustin Ley. Briggs and Boswell think it is perhaps a form of Aria or rupicola. See Jour. Bot. 1884, p. 216. It is certainly quite distinct, in my opinion, from scandica or latifolia.
41. Jour. Bot. 1903, p. 215.
42. Specimen at Kew.
43. Jour. Bot. 1897, p. 99.
44. Schübeler, Viridarium norvegicum, vol. ii. 477 (1888).
45. Garden, 1889, xxxvi. 342. Note by J.C. Tallack, who named the tree Pyrus pinnatifida.
46. In Jahresb. Bot. Gart. u. Mus. Berlin, i. 232 (1881).
47. In Kon. Sv. Veten. Akad. Handl. 1901–2, p. 60.
48. Koch, Dendrologie, i. 189 (1869).
49. Koehne, Deutsche Dendrologie, 248 (1893). This variety is commonly known as Pyrus lanuginosa, Hort.
50. If the dots are not plainly visible externally, they can be seen clearly on paring off the epidermis of the scar.
51. The twigs in winter described above clearly show the hybrid origin of this species; the varying number of dots on the scar, the pubescence and shape of the scales, etc. show the influence of Pyrus Aucuparia.
52. Koehne, Jour. of Bot. 1897, p. 99. See also Rev. Dr. Landsborough's account of the Arran hybrids in Trans. Bot. Soc. Edin. xxi. 56 (1897).
53. Pyrus rupicola, Boswell Syme, Eng. Bot. ed. 3, t. 483.
54. Flora Orientalis, ii. 658. There is a form in south-east Europe called meriodionalis, which differs only slightly from this variety.
55. Cratægus flabellifolia, Spach, Hist. Vég. Phan. ii. 103.
56. Rehder in Cyclop. Am. Hort. iv. 1689 (1902). Aria Decaisneana, Lavallée, Arbor. Segrez. p. 51. t. 18. Pyrus Decaisneana, Nicholson, Kew Hand-list of Trees and Shrubs, 187 (1894). Sorbus Decaisneana, Zabel, Handbuch Laubholz-Benennung, 199 (1903).
57. Figured in Garden, 1881, xx. 376.
58. Garden, 1882, xxii. 164.
59. Townsend, Flora of Hampshire, 125 (1883).