Rhododendron trichocladum Franch.

A deciduous or partly evergreen shrub up to 5 ft high, with usually stiff erect branchlets, clothed their first and part of their second year with long, pale bristles. Leaves oblong to obovate and oval, often rounded at the apex, 1 to 11⁄2 in. long, 1⁄3 to 3⁄4 in. wide, dark dull green and glabrous except for a few bristles above; scaly beneath with bristles on the midrib when young; margins bristly. Flowers as many as five in a terminal compact cluster, open in May. Calyx five-lobed, the lanceolate lobes 1⁄6 in. long, fringed with long bristles; flower-stalk 1⁄2 to 3⁄4 in. long, scaly and very bristly. Corolla open and flattish, 1 to 11⁄4 in. wide, yellow tinged with green, woolly in the throat, slightly scaly outside, five-lobed, the lobes rounded. Stamens ten, scarcely 1⁄2 in. long, downy towards the base, anthers brown. Ovary scaly; style glabrous, abruptly bent over. Bot. Mag., t. 9073. (s. Trichocladum)

Native of N.W. Yunnan westward through upper Burma to the eastern Himalaya; discovered about 1884, by the Abbé Delavay; introduced by Forrest in 1910. R. trichocladum is no great beauty, its growth being curiously stiff; still its soft yellow, flatly open flowers have some attractiveness. Forrest found it at 11,000 ft altitude and upwards and it is quite hardy. Flowers frequently open in autumn.

R. trichocladum is the oldest of a small group of species all resembling it in general aspect and distinguished from it and from each other by secondary characters such as the amount of bristles and scales on the pedicels and calyx, the size of the latter, the presence or absence of hairs on the upper surface of the leaf. Of possibly greater significance is the length and thickness of the style. Even in a single collecting there is variation in the characters at present considered to be of specific value. The following are perhaps the commonest in cultivation, though few are met with outside the gardens where they were raised from the wild seed:

From the Supplement (Vol. V)

According to Dr Cullen, the key distinction between this species and R. mekongense is that the scales on the underside of the leaf are all more or less equal in size, golden, large, and distant from each other, so differing from those of its ally, for which see below. Of less diagnostic value is that the bristles are more widely spread over the various parts of the plant, usually appearing on the upper sides of the leaves as well as the lower, and are always to be seen on the pedicels and calyx. The former are also shorter (1⁄4 to 1⁄2 in. long) and the latter rather better developed than in R. mekongense (Rev. 1, pp. 152–4). Dr Cullen gives central and south-west Yunnan and northern Burma as the area of R. trichocladum, but he also identifies Kingdon Ward’s 8259 as this species, though it was actually collected in the Delei valley of north-east Assam.

R. trichocladum was collected by Forrest in the Tali range (Cangshan) in 1906, when working for H. K. Bulley (F. 4145), but the first successful introduction is thought to have been in 1910 (F.6755). He sent seed later on many occasions, mostly from the Yunnan-Burma borderland. The plant in the Windsor collection that received an Award of Merit as R. lophogynum was probably raised from F.15658 or F.17501. Both were collected in this area, as too was F.27117, represented at Nymans in Sussex by a form with deep yellow flowers (Rhod. Cam. Year Book, No. 25, p. 40). There is still an original plant at Borde Hill in the same county from Farrer 876, collected on the Burma side of the frontier in 1919 and originally grown as R. oulotrichum. Kingdon Ward’s collections include KW 8259, mentioned above, and KW 6806, from the Seinghku valley of north-west Burma. But according to Dr Cullen, his 21079 from the Triangle of northern Burma is not this species as stated in the R.H.S. Species Handbook, but R. mekongense.

R. mekongense – The only essential difference between this species and R. trichocladum is that the scales on the undersurface of the leaves are unequal in size, the larger twice as wide as the smaller, which are brownish, greyish or purple on mature leaves, and densely arranged. See also R. trichcladum above. Four varieties are distinguished by Dr Cullen (Rev. 1., pp. 154–5):

var. mekongense – R. viridescens Hutch.; R. rubroluteum Davidian; R. trichocladum subsp. nepalensis Hara – The leading characters of the type-variety are that bristles are present on the lower surface of the leaves as well as on their margins and petioles; and that both the pedicels and the calyx are bristly. This variety has a wide range from north-west Yunnan through north-east Burma to south-east Tibet, with an outpost in Nepal where it was collected by Adam Stainton. According to Dr Cullen, the Nepal plant differs in no way from those of China, with the consequence that Dr Hara’s R. trichocladum subsp. nepalensis, which is founded on the Stainton collection, becomes a synonym.

R. mekongense var. mekongense was introduced by Forrest, probably under F.19930, though this collection was originally identified as R. trichocladum. However, most of the garden plants derive from Kingdon Ward 5829, for which see in the main work under R. viridescens. This species is sunk in R. mekongense var. mekongense by Dr Cullen, but whatever the botanical status of the plants raised from KW 5829, they are horticulturally distinct and could be distinguished as R. mekongense Viridescens Group. Mr Davidian retains R. viridescens as a separate species and to it he also refers Ludlow, Sherriff and Elliot 12505, which too was collected in south-east Tibet and originally identified as R. trichocladum. Plants from this collection may be in cultivation. Other reintroductions are Rock 122 in 1949 from Yunnan and KW 21079 in 1953 from the Triangle of northern Burma.

R. rubroluteum was described from a cultivated plant said to have been raised from Kingdon Ward 5489. This collection is mentioned on page 789, misprinted as ‘5849’. According to Dr Cullen, it differs from R. mekongense var. mekongense only in having the flowers tinged with red. See also Peter Cox’s remarks in The Smaller Rhododendrons, p. 134.

var. melinanthum (Balf.f. & Ward) Cullen R. melinanthum Balf.f. & Ward; R. chloranthum Balf.f. & Forr.; R. semilunatum Balf.f. & Forr. – In var. mekongense the pedicels and calyx are clad with bristly hairs, but in this variety the calyx is glabrous, as are the pedicels, though these may sometimes have hairs at the base (Rev. 1. p. 154, key). However, it is doubtful whether this distinction is worth maintaining, in view of the strong probability that one and the same population would contain both variants and intermediates between them.

R. melinanthum was described from a specimen collected by Kingdon Ward in 1913 on the upper part of the Mekong – Salween divide, a short way to the north of the type-locality of R. mekongense, and was introduced by him in the same year (KW 406). The portrait in the Botanical Magazine (t.8903) was made from a plant raised at Edinburgh from this introduction, but this seems to have been intermediate between var. melinanthum and var. mekongense. An excellent representative of the subsection Trichoclada grows at Borde Hill in Sussex. It was originally introduced under the label R. melinanthum KW 406, but has bristles on both pedicels and calyx and has been identified as R. mekongense var. mekongense. The flowers are rich yellow, usually opening in late April, and the young foliage is bronze-tinted.

Plants under the synonymous name R. chloranthum were raised from Forrest collections and are of no horticultural interest.

The other two varieties recognised by Dr Cullen are var. rubrolineatum (Balf.f. & Forr.) Cullen and var. longipilosum (Cowan) Cullen (Rev. 1, pp. 154–5). Peter Cox has suggested that some specimens identified as var. rubrolineatum may be natural hybrids between R. trichocladum and R. racemosum. Indeed, the type-specimen of R. rubrolineatum might itself be such a hybrid (The Smaller Rhododendrons, p. 136).