A modern reference to temperate woody plants, including updated content from this site and much new material, can be found at Trees and Shrubs Online.

Sorbus

Family

Rosaceae

A genus of about 120 species (nearer 200 if European microspecies are included in the total). They are deciduous trees or shrubs, natives of the mountains and moister regions of the northern hemisphere, mostly in the Old World and with a centre in the Sino-Himalayan region.

The leaves are alternate, simple or pinnate (semi-pinnate in some species of hybrid origin); if simple, then toothed or sometimes lobed. Stipules present, usually deciduous, but persistent and forming a conspicuous feature of the flowering shoots in a few species of the section Sorbus. Inflorescence corymbose, more rarely paniculate, terminating a short leafy shoot, the (usually) two uppermost leaves subtending the outer branches of the inflorescence. Flowers up to 1 in. wide in section Aria but more commonly less than 12 in. wide. The body of the flower consists of a concave cone-shaped or urn-shaped receptacle (hypanthium, calyx-tube). Calyx-lobes (sepals) five, triangular or sometimes lanceolate. Petals five. Stamens most frequently twenty, in three whorls. Carpels two to five, partly or wholly enclosed within the tissues of the receptacle and more or less connate, each with two ovules. Styles as many as the carpels, free or partly connate. Fruit composed of the thin-walled carpels, each with usually only one seed, embedded in a dryish or juicy flesh developed from the wall of the receptacle. In the section Sorbus (Aucuparia) both the free upper part of the receptacle and the calyx-lobes persist in the fruiting stage and thicken, becoming in effect an integral part of the fruit; at the other extreme, in section Micromeles, the rim of the receptacle with its lobes is shed. In the other sections the calyx-lobes usually persist but do not become fleshy.

The genus is a polymorphic one, held together mainly by technical characters of the flower and fruit, but extraordinarily diverse in outward appearance. Quite an extensive area could be devoted exclusively to Sorbus without the planting becoming monotonous – except at flowering time, the genus having no great claim to floral beauty. But by summer its ally Malus, which vastly excels it in its flowers, has nothing to offer, while Sorbus comes into its own, with its endless variations of habit and foliage. Not many of the simple-leaved species are outstanding either in their fruits, which are no more ornamental than those of the hawthorns, or in autumn colour, and from August onwards the section Sorbus becomes the standard-bearer of the genus. Only a few in this section do not give autumn colour, and in some, such as S. sargentiana, S. commixta and ‘Joseph Rock’ it is outstanding; all have ornamental fruits, ranging in colour from scarlet as in the common rowan to yellow, orange, pink, crimson or pure white.

The cultivars and hybrids are propagated by grafting. Seeds are used by commercial growers only to raise the necessary stocks, but many species are apomictic and come true when raised in this way, while the others should do so unless the seeds are gathered in large and mixed collections. The seed is best sown in pans as soon as ripe, and overwintered in a frame or cold greenhouse.

Most members of the genus are undemanding, needing only a reasonably good soil and an open position, but the rowan group as a whole are sensitive to drought and excessive heat and may suffer in dry summers if planted in full sun and in a shallow, light soil. Only S. insignis is tender. The worst enemy of Sorbus, as of all members of the sub-family Pomoideae, is Fire Blight, for which see Vol. I, p. 730.

CLASSIFICATION

sect. Sorbus (Aucuparia). – This section, which comprises all the fully pinnate-leaved species except S. domestica, is the only one to occur in North America (see S. americana, S. decora and S. sitchensis) and in western Eurasia is represented only by the variable S. aucuparia. Although the type-species of the section, it is not really representative of the group as a whole. Its nearest allies in eastern Asia would seem to be S. esserteauiana, S. pohuashanensis and S. amurensis, which resemble S. aucuparia in their red fruits and in having white hairs on the bud-scales, on the under-surface of the leaflets and in the inflorescence. Despite the white hairs of its leaflets and its red fruits in broad trusses, the Chinese S. scalaris is not closely related to S. aucuparia and belongs to the same group as the Himalayan S. foliolosa (wallichii). S. americana and S. commixta are both distinct from S. aucuparia in being almost glabrous, and the hairs on the latter, when present, are brown. In their very different ways S. sargentiana of Szechwan and S. gracilis of Japan are unique in the genus.

In the Sino-Himalayan region, where the section Sorbus is at its most varied, red-fruited species are uncommon, and very rare or even absent from Yunnan, their place being taken by what might be called the ‘pernettya-fruited rowans’, whose fruits ripen from green to porcelain white or from dark brownish red to crimson, light pink or rose-tinged white. The flowers of these are mostly borne in laxer and less branched inflorescences than in the red-fruited species, and are often pink or even crimson. Forrest made extensive collections of these interesting rowans and sent seeds on many occasions, but few of his introductions have survived. He himself did not care for them: ‘In all my years in Yunnan, I haven’t seen a member of the group to equal our own mountain ash, when well grown’ (field-note under F.26547). Yet S. hupehensis, whose presence in gardens we largely owe to Forrest, is considered by some judges to be one of the most ornamental species. It is the tallest of the white- or pink-fruited rowans in cultivation. Mostly they are small trees or shrubs, or even dwarf shrublets (S. reducta and S. poteriifolia). Although introduced early this century, S. vilmorinii is rarely seen outside collections, though it is a delightful species, with its fern-like leaves and handsome fruits, and the same is true of S. prattii. A latecomer to gardens is S. cashmiriana, less ornamental in foliage than the others mentioned, but with large rosy flowers and white fruits the size of marbles.

sect. Cormus. – The one species in this very distinct section, S. domestica, is confined to southwestern Eurasia. Its leading characters are the pinnate foliage, the five completely united carpels, and the large fruits with numerous stone-cells in the flesh.

sect. Aria. – The members of this section range in size from large trees to shrubs, with simple, toothed or sometimes lobed leaves, which in most species have a thin to dense felt or a woolly tomentum beneath. The flowers range from 12 to 1 in. in width, white. Styles two to five, free, more rarely partly connate. Fruits mostly 12 to 1 in. long, sparsely to densely dotted with lenticels, red, brown or yellowish. Calyx-lobes usually persistent (if deciduous, then falling late), not becoming fleshy in fruit.

This section is mainly represented in cultivation by S. aria, the European whitebeam, and its cultivars. This has numerous relatives in Europe and southwest Asia, some of them microspecies of often very limited distribution, others, mainly in southwest Asia, of wider range. But these have scarcely been tested in gardens.

The Himalayan and Chinese members of the section Aria differ from most of the European species in having fruits that never become an honest red when ripe, but mostly they have fine foliage. See S. cuspidata, S. megalocarpa, S. pallescens and S. thibetica.

sect. Chamaemespilus. – This section has one species, S. chamaemespilus, endemic to the mountains of Europe. The flowers are pink, with erect sepals and petals.

sect. Torminaria. – Like the preceding, and the section Cormus, this is monotypic, with one species, S. torminalis, in Europe and southwest Asia. Leaves sharply lobed or lobulate. Styles, two, united for more than half their length. Fruits brownish, densely lenticellate, with two chambers, which are surrounded by a dense layer of stone-cells.

sect. Sorbus/sect. Aria. – Intermediate between these two sections there is a group of species, which are believed to derive from hybridisation between S. aucuparia and tetraploid members of the section Aria such as S. rupicola. Some are known to be tetraploid or triploid, and to reproduce themselves apomictically. S. hybrida, the first of this group to be described, occupies a central position between S. aucuparia and S. rupicola and shows the influence of the former in the frequent occurrence of free leaflets at the base of the leaves, while in S. meinichii, as yet scarcely known in British gardens, the influence of S. aucuparia predominates. In the most ornamental of the group, the British endemics S. anglica, S. arranensis and S. minima, the influence of S. aucuparia is less obvious, and the leaves are merely deeply lobed. The status of S. intermedia is uncertain, some botanists holding that the deep lobing of the leaves derives from S. torminalis and not, as others believe, from S. aucuparia.

These species of hybrid origin should not be confused with the casual hybrids between S. aucuparia and S. aria, for which see S. × thuringiaca.

sect. Aria/sect. Torminaria. – This group is a counterpart to the preceding, but the genes of the section Aria are here united with those of S. torminalis. The senior species of this group is S. latifolia, which makes a good specimen tree and has interesting though not showy brown fruits. But in other species, notably the British endemic S. bristoliensis, the influence of S. torminalis has brightened the colour of the fruits to orange-red.

Casual hybrids between S. aria and S. torminalis occur; see S. × vagensis under S. latifolia.

sect. Micromeles. – In this polymorphic Asiatic section the leaves are simple, toothed, the lateral veins either straight and running out to double teeth or lobules, or curving. Flowers in most species very small, to about 38 in. wide, in mostly rather small often paniculate inflorescences. Carpels two to five, connate and enclosed in the receptacle. Styles two to five, free or connate in the lower part. Fruits rarely much more than 12 in. long, lenticellate or smooth. Calyx-lobes and top of receptacle usually deciduous, leaving a conspicuous scar on the top of the fruit.

This section has its headquarters in the rainier parts of the Sino-Himalayan region, with eight species in the eastern Himalaya and northeast India, one of which extends as far southeast as Sumatra. The species best known in gardens are northern outliers of the group and are unusual in having red fruits. These are S. alnifolia, S. folgneri (the two most ornamental species) and S. japonica. Resembling these in the straight lateral leaf-veins, but with fruits of a nondescript colour, are S. caloneura and S. meliosmifolia. The species with curved lateral veins are represented in cultivation, though very rarely, by S. keissleri and S. epidendron.

SELECT BIBLIOGRAPHY

Fox, W. – Unpublished notes on Sorbus.

At the time of his death in 1962 Dr Wilfrid Fox had brought near to completion a horticultural monograph on Sorbus, a genus of which he had planted a comprehensive collection in his arboretum at Winkworth near Godalming in Surrey, now the property of the National Trust. Dr Fox had been assisted in his work by the late Mrs Madeline Spitta, to whom these notes passed on his death. During the closing months of her life Mrs Spitta put the material in order, and entrusted it to the chief editor for use in preparing the present revised account of the genus, and it has been frequently consulted.

Gabrielian, E. – The Genus Sorbus in Eastern Asia and the Himalayas. Erevan, 1978 (in Russian).

– – ‘The Genus Sorbus in Turkey’, Notes Roy. Bot. Gard. Edin., Vol. 23 (1961), pp. 483-96.

– – Sorbus, in Davis, P. H., ed., Flora of Turkey, Vol. 4 (1972), pp. 147-46.

Gilham, C. M., and McAllister, H. A. – ‘Tree Genera – 6. Sorbus sect. Aucuparia’, Arboricultural Journal, Vol. 3 (1977), pp. 85-95.

Handel-Mazzetti, H. – Symbolae Sinicae, Part VII 3 (1933), pp. 465-75.

Hitchcock, C. L., Cronquist, A., et al., Vascular Plants of the Pacific Northwest, Part 3 (1961), pp. 188-90.

Hedlund, T. – ‘Monographie der Gattung Sorbus’, Kongl. Svensk. Vet.-Akad. Handl., Vol. 35, no. 1 (1901), pp. 1-147.

Hensen, K. J. W. – ‘The Sorbus-collection in the Botanical Gardens and Belmonte Arboretum of the Agricultural University at Wageningen’. Parts I-III in Jaarb. Nederl. Dendr. Ver., 20, pp. 121-39; 21, pp. 180-7; 22, pp. 48-56 (1956-62). Part IV in Dendroflora No. 3 (1966), pp. 60-72 (in Dutch).

– – – – ‘Intermediate taxa between Sorbus aria (L.) Crantz and Sorbus aucuparia L. cultivated in the Netherlands’, Jaarb. Nederl. Dendr. Ver. 21, pp. 189-204 (1958) (in Dutch).

– – – – ‘Het Sorbus latifolia-Complex’, Belmontia, fasc. 13 (1970), pp. 181-94.

Hutchinson, J. – Unpublished Notes on Sorbus.

Early in 1943, Dr. John Hutchinson, of the Royal Botanic Gardens, Kew, undertook at the request of the Royal Horticultural Society to write a botanical account of the genus Sorbus. After drafting descriptions of most species of the section Sorbus (Aucuparia) he abandoned the task, which he is said to have found uncongenial, and handed over the completed material to Dr Wilfrid Fox (q.v.), of whose papers on Sorbus it forms a part.

Jones, G. N. – ‘A Synopsis of the North American Species of Sorbus’, Journ. Arn. Arb., Vol. 20 (1939), pp. 1-43.

Karpati, Z. E. – ‘Die Sorbus-Arten Ungarns und der angrenzenden Gebiete’, Fedd. Repert., Vol. 62 (1963), pp. 71-334.

Koehne, E. – Sorbus, in Plantae Wilsonianae, Vol. I (1913), pp. 457-83 (treatment of sect. Sorbus (Aucuparia)).

Kovanda, M. – ‘Flower and Fruit Morphology of Sorbus . . .’, Preslia, Vol. 33 (1961), pp. 1-16.

– – – – ‘Taxonomic Studies in Sorbus subgen. Aria’, Act. Dendr. Čech., Vol. 3 (1961), pp. 41-83.

– – – – ‘Spontaneous hybrids of Sorbus in Czechslovakia’, Act. Univ. Carol., 1961, pp. 41-83 (in English).

Liljefors, A. – ‘Cytological Studies in Sorbus’, Act. Hort. Berg., Vol. 17 (1955), pp. 47-113.

Mcallister, H. A., and Williams, C. M. ‘Sorbus aucuparia in Town and Country’ (notes on the section Sorbus, privately circulated, 1975).

Rehder, A. – Sorbus, in Plantae Wilsonianae, Vol. II (1915), pp. 266-79 (sections Aria and Micromeles).

Regelingscommissie Sierbomen N.A.K.-B. – Sorbus, in Dendroflora No. 2 (1965), pp. 28-44.

Richards, A. J., Sorbus, in Stace, C. A., ed. Hybridisation and the Flora of the British Isles (1975), pp. 233-8.

Schneider, C. K. – Illustrierte Handbuch der Laubgehölze, Vol. I (1906), pp. 667-98 and pp. 700-4.

Yü, T. T. – Sorbus, in Flora Reipublicae Popularis Sinicae, Vol. 36 (1974), pp. 283-344.

– – and Kuan, K. C. – ‘Taxa Nova Rosacearum Sinicarum’, with a classification of the Chinese species of Sorbus by Dr T. T. Yü, Act. Phytotax. Sin., Vol. 8 (1963,), pp. 207-210, 221-225.

Warburg, E., F. – Sorbus, in Clapham, Tutin and Warburg, Flora of the British Isles, ed. 2 (1962), pp. 423-37.

– – and Karpati, Z., E. – Sorbus, in Flora Europaea, Vol. 2 (1968), pp. 67-71.


From the Supplement (Vol. V)

No mention was made in the main work of any species belonging to the group which the Russian botanist Eleanora Gabrielian separates from sect. Aria as sect. Lobatae. Most have simple leaves as in S. aria but these are always more or less pinnately lobed and a few may have one to three free leaflets at the base. They differ from sect. Aria in many other characters and are almost certainly of hybrid origin. Their main distribution is in Anatolia, Armenia, the Caucasus and northern Iran, though the first to be described – S. turkestanica (Franch.) Hedl. – extends into Central Asia. Two species of this group have been introduced recently – S. takhtajanii and S. tamamschjanae – both described by Gabrielian. Some older species in this group are S. persica Hedl., S. armeniaca Hedl. and S. caucasica Zinserling. The last-named was at one time confused with S. intermedia. All are shrubs or small trees.

Species articles