Abies chensiensis Van Tiegh.


A. ernestii (Rehd. & Wils.) Rehd.; A. beissneriana Rehd. & Wils., not Mott.

Article sources


New Trees

A tall tree, attaining 200 ft in W. Szechwan but apparently smaller in other parts of its range. Young shoots glabrous, rarely slightly hairy, yellow or yellowish grey. Leaves 12 to 134 in. long, spreading slightly upward from the shoot or almost vertically, but with a V-shaped opening between the two ranks; they are pointed or blunt at the apex, or slightly emarginate, and with whitish or glaucous bands of stomata beneath. Cones cylindric to ovoid, 2 to 334 in. long, green at first but reddish brown when ripe, sessile or short-stalked. Bracts shorter than the scales and concealed by them.

Native of W. China; discovered in the Chin Ling Shan, Shensi province, by the Abbé David in 1872. In 1907 Wilson found in the Tapao Shan, W. Szechwan, a finer form of this fir which is sometimes treated as a distinct species (A. ernestii). A. chensiensis has a wide range in W. China from Shensi to the Burmese border and probably the majority of the trees in cultivation in this country are from Forrest's seed, collected in Yunnan. Plants sent out by Messrs Hillier around 1940 are almost certainly from seed distributed by the Sun Yat Sen University; they agreed so well with Wilson's specimens from the Tapao Shan that the late E. Hillier retained the name A. ernestii for them.

A. chensiensis is well-marked by its stout yellowish shoots and by its crowded leaves separated into two ranks by a V-shaped depression. It is very rare, however. The tallest authentic specimen traced is at Stanage Park, Radnor, 42 ft in 1959.

From the Supplement (Vol. V)

The synonyms A. beissneriana and A. ernestii should be deleted; see further under A. recurvata, in this supplement.

specimens: Borde Hill, Sussex, in Gores Wood, pl. 1937, 52 × 2 ft, and in the Tolls 66 × 134 ft (1981); Burnside, Angus, pl. 1936, 40 × 2 ft (1974); Crarae, Argyll, 42 × 414 ft (1976).

† subsp. salouenensis (Bordères-Rey & Gaussen) Rushforth A. salouenensis Bordères-Rey & Gaussen – The Salween fir occupies the south-westernmost part of the species area and differs from typical A. chensiensis in its much longer leaves. main work. But Keith Rushforth has argued convincingly that these two firs differ in more characters than those mentioned by Craib, and his views are based on a far wider range of specimens than were available for study in 1919 when Craib published his note.

The shoots of A. fabri are paler than in A. delavayi, fawn or yellowish brown, glabrous or sometimes with fine pubescence in the shallow grooves. The leaves are more or less pectinate below the shoot, parted above, rounded and notched at the apex, up to 1 in. long, distinctly revolute at the margin but less markedly so than in A. delavayi, silvery glaucous beneath, but the waxy coating covering the stomatal bands only. The cones are of about the same length as in A. delavayi, but barrel shaped and with the exserted bract-cusps reflexed.

A. fabri, in its typical state, is confined to central western Szechwan. (Forrest collections from Yunnan identified as A. fabri by Orr are mostly A. delavayi.) A. fabri was introduced by Wilson in 1903 when collecting for Veitch, and it may be that the Headfort tree is of this provenance. But most of the trees originally planted in Britain were from his 1910 seed, and of these few remain. Some younger trees are grafts, distributed by Messrs Hillier in the 1930s. Trees measured recently are: Speech House, Glos., from W.4078, pl. 1916, 50 × 412 ft (1975); Dawyck, Peebl., pl. 1918, 50 × 514 ft (1982); Headfort, Co. Meath, Eire, 56 × 434 ft (1980); Birr Castle, Co. Offaly, Eire, pl. 1931, grafted, 42 × 214 ft (1980).

A tree at Werrington, Cornwall, from W.4086, was propagated by grafting before it died recently.

subsp. minensis (Bordères-Rey & Gaussen) Rushforth A. minensis Bordères-Rey & Gaussen; A. faxoniana Rehd. & Wils., in part (type excluded) – From subsp. fabri (i.e., A. fabri sens. strict.) this differs in its honey-brown against yellow-brown shoots; longer leaves, to about 138 in. long, not inrolled at the margin and with greenish white stomatal bands beneath. Also, the leaves are more widely parted above than they are in subsp. fabri.

A. minensis was described in 1947 from a specimen collected by Yu west of Sungpan, which is situated on the Min river of northern Szechwan. Although on the whole amply distinct from A. fabri, some specimens are intermediate, and this fir is consequently reduced to the status of a subspecies by Rushforth (Notes Roy. Bot. Gard. Edin., Vol. 43 (2), p. 273 (1986)).

It will be noted that A. faxoniana appears above among the synonyms of subsp. minensis. This species, as originally interpreted by Rehder and Wilson, contained two distinct elements. The type of their species (W.4060) is now included in A. fargesii (see further under that species, in this supplement). But they cited three other Wilson collections from the same expedition (W.4052, 4069 and 4070), and these are A. fabri subsp. minensis. Seed was distributed in Britain under all four numbers and, not surprisingly, it was remarked on several occasions during the Conifer Conference of 1931 that some plants raised from seed of A. faxoniana differed markedly from the original description and type. Most were in fact A. fabri subsp. minensis, and of these the following originals still exist:

Speech House, Glos., from W.4069, 52 × 5 ft (1983); Werrington Park, Cornwall, from W.4069 and W.4052, 42 × 334 ft, 66 × 314 ft and 50 × 5 ft (1977); Dawyck, Peebl., pl. 1918, 70 × 512 ft (1983); Powerscourt, Co. Wicklow, Eire, (probably from W.4069), 72 × 434 ft (1980); Headfort, Co. Meath, Eire, pl. 1915, 92 × 5 ft (1980).

Wilson's 4069 seems to have been the best of the three sendings, and the tree at Werrington girthing 5 ft is probably from this batch, the other two from W.4052.

From New Trees

Abies chensiensis Tiegh.

Shensi Fir

This species was described by Bean (B149, S19) and Krüssmann (K33). Two subspecies have since been recognised, however, and a key to these is provided below.


Longest leaves of sterile branches to 7.5 cm, occasionally longer; China (northeast Yunnan, southeast Xizang), India (Arunachal Pradesh)

subsp. salouenensis


Longest leaves of sterile branches < 4.5 cm



Resin canals in leaves of coning shoots medial; cones 8–10 cm long; China (southeast Gansu, Henan, west Hubei, south Shaanxi, west Sichuan)

subsp. chensiensis


Resin canals in leaves of coning shoots marginal; cones 10–14 cm long; China (Yunnan: Lijiang Shan Massif)

subsp. yulongxueshanensis


Abies chensiensis subsp. salouenensis (Bordères & Gaussen) Rushforth

Salween Fir

Subsp. salouenensis differs from the type subspecies mainly in that the leaves are longer (up to 7.5 cm vs. 2–4 cm in subsp. chensiensis). The resin canals of leaves in coning shoots of subsp. salouenensis are marginal, while those of subsp. chensiensis are medial. Rushforth 1984a, Farjon 1990, Fu et al. 1999c. Distribution CHINA: northwest Yunnan, southeast Xizang; INDIA: northeast Arunachal Pradesh. Habitat Montane forest, between 2000 and 3200 m asl. USDA Hardiness Zone 6. Conservation status Lower Risk. Illustration NT48. Taxonomic note Flora of China treats both these taxa as synonyms of A. ernestii var. salouenensis (Bordères & Gaussen) W.C. Cheng & L.K. Fu. Abies ×chengii Rushforth is now considered to be A. forrestii A. chensiensis subsp. salouenensis (K. Rushforth, pers. comm. 2007).

The Salween Fir was first introduced to cultivation before World War II by George Forrest (for example, F 30668) and Frank Kingdon-Ward (for example, KW 10412), and trees from these early introductions are still to be found in collections in the United Kingdom. These older trees (and propagations from them) are now being joined by trees from the new generation of collectors. Several trees assumed to be grown from F 30668 and planted out at Borde Hill, West Sussex, c.1940, have achieved 14–16 m (the largest with dbh 40 cm), and Alan Mitchell measured one of unknown origin at Burnside, Forfar, Angus in 1990 that was 16 m tall with a dbh of 25 cm. Growth is steady rather than fast, a tree grown by Keith Rushforth (pers. comm. 2007) achieving 12–15 m in 25 years. When doing well, subsp. salouenensis is an attractive tree, with long dark needles slightly drooping from the twigs, and in maturity its bluish cones are another feature. It is said to require a sheltered damp site to do well (Rushforth 1987a), but a young tree at Thenford House, Northamptonshire is growing happily in a comparatively open site, and produces an abundant crop of big blue-black cones that drip copious resin.

Abies chensiensis subsp. yulongxueshanensis Rushforth

Subsp. yulongxueshanensis differs from the type subspecies in the length of the female cones, which range from 10 to 14 cm (female cones of subsp. chensiensis are typically 8–11 cm long, though they can reach 12 cm). Like subsp. salouenensis, the resin canals in the leaves of coning shoots are marginal. Rushforth 1984a, Farjon 1990, Fu et al. 1999c. Distribution CHINA: northern Yunnan, Yulong Xue Shan massif in the Lijiang Shan. Habitat Montane forest between 2600 and 3200 m asl. USDA Hardiness Zone 6. Conservation status Vulnerable. Restricted to a small area of forest.

Abies chensiensis subsp. yulongxueshanensis is extremely rare in cultivation, which is perhaps just as well considering that its name is difficult both to write and to say, but with it also being rare in the wild this is somewhat worrying. It appears to be known only from a collection made by T.T. Yu (15050) growing at Benmore.

A promising young Abies chensiensis subsp. salouenensis at Arboretum Wespelaar, with characteristic long leaves. Image P. de Spoelberch.



Other species in the genus