Small shrub, 40–130 cm, evergreen in the wild, sometimes more or less deciduous in cultivation. Branches hard, pith narrower than the surrounding xylem; nodes without a ring-like scar. Branchlets with short, dense, adpressed, rusty-brown hairs. Bud scales present. Leaf blade spathulate, obovate, or broadly diamond shaped, 2–5 x 1–2.2 cm, both surfaces with short, dense, adpressed hairs, base abruptly constricted, apex described as acute (Peng & Kamelin 1996) or obtuse (Cullen et al. 2011), at any rate not mucronate. Inflorescences terminal or axillary, 5–20 flowered; bracts oblong-lanceolate to oblong-ovate, 6–7.5 x 2–3 mm; bracteoles lanceolate, 5–7 x 1.5–2 mm. Calyx 8–10 x ~1.5 mm, silky-hairy between ribs. Corolla 1.5–1.9 cm, tube purple; lobes blue, obdeltate, 6–7 x 4–5 mm, apex notched and with a triangular mucro. Autumn flowering, August-December in the wild. (Peng & Kamelin 1996; Edwards & Marshall 2019; Cullen et al. 2011)
Distribution Bhutan China SW Xizang, perhaps also Yunnan (see below)
Habitat Warm valleys, 2200–2800 m.
USDA Hardiness Zone 6-9
RHS Hardiness Rating H4
Conservation status Not evaluated (NE)
This is the most shrubby of the species widely grown in our area, with the usual bright blue flowers from late summer. It is a much-branched and strikingly hairy plant. The young stems are thick with red-brown hairs, which are also present on both leaf surfaces. Subjected to only a few degrees of frost on winter, it forms a rather low, spreading shrub suitable for a narrow border at the base of a sunny wall (Bean 1976). In colder areas such as Central Europe, it can survive temperatures as low as –24°C on well drained soils, cut to the base and regrowing in spring (Havlis 2019). Drought-tolerant like other Ceratostigma, annual growth is most luxuriant with a reasonable amount of summer moisture and warmth. The species thrives even with tropical heat in East African gardens alongside the native C. abyssinicum (Cameron 2017).
C. griffithii belongs to a group of rather similar western Chinese species, with C. willmottianum and C. minus. Despite the often-repeated common name, it has not been recorded from Myanmar. It is distinguished from C. willmottianum by its more shrubby habit, much denser hairs, smaller leaves more obovate than diamond-shaped, and by its more westerly range. At least in the usual garden form, C. willmottianum is also less branched, with most flowers in terminal inflorescences. C. minus is less clearly differentiated; while the older literature distinguishes it from the other two by the combination of shrubby habit and a more or less glabrous upper leaf surface, Flora of China (Peng & Kamelin 1996) allows hairier forms in its circumscription, although in these the hairs on the shoots are said to be whitish or pale yellow rather than brown.
There is uncertainty around more easterly plants attributed to C. griffithii, which include our garden stocks. They are said to derive from collections made by George Forrest in Yunnan (Bean 1976), almost certainly Forrest 601 of 1904 and/or 3107 of 1906 (Diels 1912). Forrest’s expedition was sponsored by Liverpool cotton merchant Arthur Bulley, a serious plantsman and owner of Bees Nursery, Chester, which rapidly propagated and disseminated Forrest’s introductions. However, the species was first described from Bhutan, from a poorly preserved specimen collected in 1837 by the British surgeon-botanist William Griffith (Clarke in Hooker 1882). This is certainly a conspicuous member of the Bhutanese flora (Edwards & Marshall 2019), and most taxonomists have given its range as only Bhutan and Xizang (Tibet) (Prain 1906, Peng & Kamelin 1996), though Ludwig Diels was willing to name as C. griffithii a herbarium specimen from Sichuan (Bock & von Rosthorn 3001) and the two Forrest collections from Yunnan (Diels 1900). Ernest Wilson (in Sargent 1916), working after C. minus had been described (in Prain 1906) assigned Bock & von Rosthorn 3001 to C. minus, which he regarded as ‘very close to C. griffithii […] perhaps only a glabrescent variety’, but felt that a much hairier specimen collected in Yunnan (Henry 9586) matched C. griffithii. More work is needed, but is entirely possible that these species intergrade. To summarize the status of most garden plants, they probably originate from Yunnan, and match descriptions of C. griffithii from Bhutan.