A tree up to 80 ft in the wild, with horizontal or slightly ascending branches, forming a broad crown; bark grey or grey-brown, scaly; buds broadly ovoid, obtuse, not or only slightly resinous; young shoots pale brown, glabrous or sometimes downy in the furrows. Leaves pointing forward on the upper part of the shoot, the lower ones more or less horizontal, a few downward pointing, 3⁄8 to 5⁄8 in. long, rather stiff, apex bevelled, acute with a pale horny tip, four-angled, with lines of stomata on all four surfaces, but these are rather more numerous on the lower surface, which is bluish or whitish green, in contrast to the upper (exposed) surface, which is greyish green – hence no doubt the epithet bicolor. Cones 2 to 4 in. long, cylindrical or slightly tapered from the middle, the scales toothed, variable in shape even on the same cone, some broad and rounded, some tending to a triangular outline.
Native of Japan, where it is confined to the mountains of the central part of the main island (Honshu); introduced by J. G. Veitch in 1861. Veitch procured the seeds, and specimens, when he visited Mt Fuji as unofficial botanist on an excursion to the summit led by the British Minister to Yedo (Tokyo), Rutherford Alcock (see Gard. Chron. (1861), pp. 49-50). From the specimens Lindley described Abies alcocquiana, but it later proved that only the cones belonged to the species described here; the leaves were from P. jezoensis, which grows with P. bicolor on the mountain. The seeds too were mixed and produced both species. The confusion was cleared up by Masters in 1880, but in the meantime species described here had been validly named Abies bicolor by Maximowicz. P. jezoensis, the other party to the confusion, is a quite distinct species belonging to the Omorika or flat-leaved section.
P. bicolor is hardy, but rare and of no ornamental merit. Among the largest examples ‘are: Wakehurst Place, Sussex, 71 × 3 ft (1973); Bicton, Devon, 74 × 71⁄2 ft (1968); Melbury, Dorset, 75 × 51⁄4 ft (1971); Scone Palace, Perth, 52 × 6 ft (1970).
var. acicularis Shirasawa & Koyama – This variety was described from trees growing in the Yatsuga range, and was said to differ from the typical state in having more densely crowded, longer, linear leaves, bluish white on the concealed side; and in its entire cone-scales. A tree at Borde Hill, Sussex, of unknown provenance, is grown under this name but does not agree with the original description. The var. reflexa, described by the same authors from the Shirane-san in the Akaishi range, was said to have reflexed cone-scales. Wilson considered that both varieties were part of the normal variation of the species.