A dwarf evergreen shrub up to 2 ft high in the wild; young shoots scaly. Leaves obovate, the apex mucronulate, the base wedge-shaped, 1 to 23⁄4 in. long, 1⁄2 to 11⁄8 in. wide, glossy dark green above, pale green and fairly thickly sprinkled over with yellowish shining scales beneath; stalk 1⁄/6 in. long. Flowers opening in May, usually three (sometimes two to four) in a terminal cluster, each on its slender scaly stalk which is 3⁄4 to 1 in. long. Calyx large for the size of the flower, 1⁄3 in. long, cut to the base into five ovate lobes, scaly outside. Corolla bell-shaped, five-lobed, about 1 in. wide, clear pink, speckled with crimson. Stamens ten, hairy on the lower two-thirds; ovary densely scaly; style 1⁄3 in. long, thick and glabrous. Bot. Mag., t. 9358. (s. and ss. Glaucophyllum)
Native of N.E. upper Burma, found in the Shing Hong pass by Farrer in June 1920. It was then in flower and he describes it as a 'particularly charming plant with three- (rarely four-) bloomed inflorescences. Flowers of a clear apple-blossom pink flushed more warmly in the upper lobes, and speckled with crimson; and with a deep rose tube.' Farrer died before he could harvest the seeds of his 1920 discoveries, but four years later Forrest met with this species near the type-locality and introduced it (F.25570 and 25581).
It is a very attractive little shrub, quite hardy, though its expanding flower-buds may be killed by late frost. Often it produces a quite heavy crop of flowers in the autumn, though at the cost of next spring's display.
R. tsangpoense Hutch. & Ward – This species is very near to R. charitopes. The differential characters given by Cowan and Davidian are the narrow-obovate to oblong-elliptic leaves, against broad-obovate in R. charitopes, and the smaller calyx, up to 1⁄4 in. long. It was discovered by Kingdon Ward in 1924 on the Doshong La, S.E. Tibet, in the mountains enclosed by the Tsangpo bend. It is uncommon in cultivation. Award of Merit May 2, 1972, when shown by Major A. E. Hardy, Sandling Park, Kent (clone 'Cowtye').
var. pruniflorum (Hutch.) Cowan & Davidian R. pruniflorum Hutch.; R. sordidum Hutch. – Leaves more densely scaly beneath, slightly overlapping to their own diameter apart (against mostly three to six times their own diameter apart in the typical state). This variety was found by Kingdon Ward in 1926 on both sides of the Irrawaddy-Lohit divide, on the frontier between Burma and Assam and was introduced by him. According to the field notes, the colour of the flowers in KW 6924 (Seinghku valley) was plum-purple and in KW 7188 (Di Chu valley) 'plum-purple or inclining to crimson on the one hand, or to violet on the other'. He later found the same variety on Kaso peak in the Mishmi Hills, Assam (KW 8415), flowers described in field note as claret-coloured; the specimen under this number is the type of R. sordidum, a synonym of var. pruniflorum. As met with in cultivation this variety usually has flowers of a slaty purple.
Cowan and Davidian also describe R. tsangpoense var. curvistylum, with smaller, tubular-campanulate flowers. This was found by Kingdon Ward on the Doshong La (KW 5843) and was considered by him to be a natural hybrid between R. campylogynum var. myrtilloides (KW 5842) and typical R. tsangpoense (KW 5844). In this connection it is interesting to note that Farrer found in Burma some plants which he considered to be hybrids between R. campylogynum var. charopoeum and R. charitopes, to which R. tsangpoense is closely related (see above).
From the Supplement (Vol. V)
(The authors as above, not as printed on page 628.)
subsp. tsangpoense (Hutch. & Ward) Cullen R. tsangpoense Hutch. & Ward; R. tsangpoense var. curvistylum Cowan & Davidian; R. curvistylum Ward, nom. nud. – This has a more northerly distribution than the typical subspecies. With regard to var. curvistylum (page 629), Dr Cullen remarks that while the (flowering) type-specimen belongs to subsp. tsangpoense, plants cultivated under KW 5843 are of the parentage suggested by Kingdon Ward. This is puzzling, as the collector clearly considered that his flowering specimen (KW 5843) was also hybrid (5844 being R. tsangpoense and 5842 R. campylogynum).
R. pruniflorum Hutch. R. tsangpoense var. pruniflorum (Hutch.) Cowan & Davidian; R. sordidum Hutch. – Dr Cullen restores this rhododendron to its original species status. He considers it to be a very distinct species, more closely related to R. brachyanthum than to R. charitopes, despite the difference in flower colour.