There are no active references in this article.
The text below is that of Bean (Bean 1981) who discussed this taxon under the name Sorbus umbellata. We have created this hybrid article – Bean’s text under the correct modern name, with appropriate synonymy – whilst we await sponsorship to enable a full revision of this genus to be written. We are re-organising the Sorbus sensu lato articles in this way to enable a new revision of Sorbus sensu stricto to commence in 2023, and to bring the nomenclature of this complex group of plants up to date in line with modern treatments.
TC, September 2023.
A small tree in the wild, with a rounded crown, or a shrub; twigs reddish brown, lenticellate, glabrous; winter-buds roundish, about 1⁄4 in. long, glabrous or downy. Leaves simple, 1 to 23⁄4 in. long, varying on the same plant from obovate with a cuneate base – almost fan-shaped – to a more rounded shape, glabrous and slightly glossy above, densely and closely covered beneath with a vividly white tomentum, deeply and jaggedly toothed or double-toothed on the rounded or almost truncate upper part of the leaf, lateral veins in five to eight pairs; petiole 1⁄4 to 3⁄4 in. long. Flowers in late May or early June, pure white, about 3⁄8 in. wide, in loose clusters about 3 in. wide. Anthers pink. Styles two. Fruits ovoid to globose, about 1⁄2 in. long, said to be red on wild plants.
Native of Asia Minor, the Caucasus, the Crimea and parts of the Balkans; cultivated since early in the 19th century under various names. Although uncommon in gardens, it is one of the most ornamental of the Aria group in its jagged leaves intensely white beneath. An example in the Winkworth Arboretum, planted by Dr Fox in 1938, is an intricately branched shrubby tree about 20 ft high (1979). In cultivation S. umbellata ripens its fruits very late – in early November – and perhaps not at all in some seasons. Dr Fox describes the fruits of the Winkworth tree as greenish yellow except on the side exposed to the sun which is red. But plants in their native habitat may become fully red.
S. graeca (Spach) Kotschy Crataegus graeca Spach; Pyrus aria var. cretica Lindl.; S. cretica (Lindl.) Fritsch; S. umbellata var. cretica (Lindl.) Schneid. – This species, often considered to be no more than a variety of S. umbellata, differs in the more numerously toothed leaves with up to nine pairs of veins, more thinly and grey-hairy beneath. The leaves in the cultivated form are obovate, rounded at the apex, cuneate at the base, of rather thick texture, about 21⁄2 in. long. Fruits dark red, with very few scattered lenticels, globose or broadly oblong-ellipsoid, about 1⁄2 in. long, ripening in September.
Despite its name, S. graeca is of wide distribution, from Iraq, the Lebanon and the Caucasus through Asia Minor and S.E. Europe to eastern Central Europe, west to Sicily and N. Africa. It was in cultivation in Britain as early as 1830 but the present stock is believed to derive from an introduction by E. K. Balls from Anatolia in the 1930s. The description of the foliage given above is taken from the cultivated form; in wild plants there is variation in the shape and relative width of the leaves.
In the Winkworth Arboretum this species and S. umbellata were planted by Dr Fox side by side and make an interesting and instructive pair. At least in these two trees there is a marked difference in habit as well as in the other characters mentioned, S. graeca being of rather fastigiate habit in contrast to the bushy habit of S. umbellata.
Considered as a variety of S. umbellata this sorbus would take the name S. umbellata var. cretica, but S. graeca is its correct name at the specific level.
S. rupicola (Syme) Hedl. Pyrus rupicola Syme; S. aria var. salicifolia Myrin ex Hartman; S. salicifolia (Myrin) Hedl. – This species is not cultivated for ornament to any extent but deserves mention as a native of the British Isles (it also occurs in Scandinavia and Estonia). It is similar to S. graeca, but the fruits are more lenticellate and the teeth of the leaves are curved on their outer margin, pointing forward towards the apex of the leaf, whereas in S. graeca they point more or less in the same line as the lateral ribs (Warburg). Also S. rupicola is apparently always tetraploid and apomictic, while S. graeca is said to be mainly diploid and to reproduce sexually. In the British Isles S. rupicola occurs locally in rocky and often inaccessible places, usually on limestone, in Devon, Wales, the Pennines, Scotland and parts of Ireland.
Some minor and very local species of this complex were described by E. F. Warburg in 1957, and descriptions will be found in Clapham, Tutin and Warburg, Flora of the British Isles: S. eminens (Wye Valley and Avon Gorge, type-locality Offa’s Dyke, Tidenham); S. hibernica (local in central Ireland); S. lancastriensis (Lancashire and Westmorland, type-locality Humphrey Head); S. porrigentiformis (N. Devon, the Mendips, Wye Valley, S.Wales, type-locality Offa’s Dyke, Tidenham); S. vexans (Lynmouth to Culbone, type-locality a wood between Lynmouth and Watersmeet). Apart from the last named, all are confined to limestone, and all are known or presumed to be tetraploid or triploid apomicts. Described by Warburg later is: S. wilmottiana (type-locality above the cliffs of the Avon Gorge near Clifton).