A tree 20 to 50 ft high, forming a rounded, wide-spreading head of branches, the lower ones arching or pendulous at the extremities; trunk 1 to 2 ft in diameter. Leaves 11⁄2 to 31⁄2 in. long, about half as much wide, oval or ovate, rounded or tapering at the base, acute or acuminate at the apex, finely and shallowly toothed, glabrous and glossy green above, paler and glabrous beneath; leaf-stalk slender, glabrous, about two-thirds the length of the blade. Flowers white, produced during April in umbels, each flower 11⁄2 in. across and borne on a slender stalk 1 to 11⁄2 in. long. Calyx-tube and calyx-lobes glabrous, the lobes longer than the tube, lanceolate. Fruits about 3⁄8 in. thick, globular, bright red or yellow, slightly hollowed at the base and with a round scar but no calyx-teeth at the top.
Widely spread in nature, this species in its typical state ranges from E. Siberia eastward to the Pacific and southward to Mongolia, N. China, and Korea; but plants from S.W. China and the Himalaya, usually referred to var. himalaica, scarcely differ from the type. It is said to have been introduced to Kew in 1784. In fact, plants were being offered in the London nurseries earlier than that, but whether they were the true species is not known. Indeed the true species has never been common in gardens, most of the trees grown as Malus baccata until recently being forms of M. × robusta (see below). It was reintroduced by Farrer from Kansu under numbers 398 and 778 (the latter grown wrongly as M. theifera).
cv. ‘Gracilis’. – Habit graceful, with pendent branches. Leaves smaller than normal, but otherwise scarcely differing from the typical variety. Raised from seeds collected by Purdom in Shensi. Dr Wyman considers this variant to be superior to the ordinary Siberian or Manchurian crab as an ornamental.
var. himalaica (Maxim.) Schneid. Pyrus b. var. himalaica Maxim. – It is very doubtful whether this variety is worth distinguishing; some Himalayan specimens have the coarsely serrate leaves mentioned in the description, others do not. It is also uncertain whether the species is anywhere truly wild in the Himalaya. Plants raised from Forrest’s 22188 and 30485, collected in Yunnan, China, and grown as var. himalaica, do not differ from the typical variety. A plant cultivated from Ludlow and Sherriff’s 17388, collected in Bhutan, differs in having ellipsoid fruits and the name var. ellipsoidea was proposed for this variant by Yü, but not published.
cv. ‘Jackii’. – A variant of the Siberian crab introduced to the Arnold Arboretum in 1905 by means of scions sent by J. G. Jack from Seoul, Korea. A similar crab was found by Wilson in the Kongo-san, Korea (W.10696). ‘Jackii’ differs from the var. mandshurica in its larger elliptic leaves, up to 41⁄2 in. long, 25⁄8 in. wide, dark green above, glabrous beneath and in its larger flowers and fruits, the latter being about 3⁄8 in. wide. There is a fine specimen at Borde Hill, Sussex, which came from Vicary-Gibbs’s collection at Aldenham. It is very robust, with leaves of a remarkably deep green for a crab.
var. mandshurica (Maxim.) Schneid. Pyrus baccata var. mandshurica Maxim. Manchurian Crab. – Leaves with small, distant teeth, or even entire in the lower half, downy beneath when young; petiole usually tomentose. Calyx-tube downy but sometimes only slightly so; calyx-lobes always downy on the inside. Native of the Russian Far East, central and north-west China, Korea, and Japan. This variety was in cultivation at Kew in 1874, in which year it was figured in Bot. Mag. 6112 (as Pyrus baccata), but had probably been introduced earlier (the figure in Watson’s Dendrologia Britannica (1825) is considered by Rehder to represent this variety). Wilson reintroduced it from N.W. Hupeh in 1901, when collecting for Messrs Veitch.
The Siberian and Manchurian crabs make sturdy trees which in time attain a fair size. Their foliage is healthy and the ivory-white flowers are borne freely in most years, but the fruits are too small to make much display.