Hydrangea davidii Franch.

The whole concept of Hydrangea davidii is a troublesome one. There are doubts around its relationships, how to recognize it, and whether it even exists as a taxon distinct from H. pottingeri (still better known as H. chinensis) within the H. scandens/pottingeri complex. In Vietnamese material there is further confusion with H. indochinensis. Western enthusiasts sometimes point to a plant as ‘what so-and-so calls davidii’. It is undoubtedly rare in gardens, and everyone seems to agree that if it exists as a distinct entity, the fertile flowers have blue rather than yellowish petals (pink on alkaline soils).

H. davidii was first described by Franchet (1885) from herbarium specimens collected in western Sichuan by the missionary botanist Armand David. Scientific exploration of East Asian hydrangeas was still in its early days, and Franchet’s description contrasts it only to H. hortensis (i.e. H. macrophylla) and H. hirta. His account of the hairy leaves matches the description above, based on Flora of China (Wei & Bartholomew 2001), and the blue fertile flowers with thin, narrow, flat, persistent petals are mentioned. The type had only three sepals per sterile flower, and this is true of some plants in gardens, giving the impression of a ‘Mickey Mouse face’ (Foster 2023); some others in gardens and the wild have a more conventional four (Wei & Bartholomew 2001; Chinese Academy of Sciences 2024). Flora of China (using the name H. chinensis for H. pottingeri) further suggests that H. davidii has more spreading styles than H. chinensis in fruit, not an easy character to compare objectively. (It is worth noting that one of the two Flora of China authors would have preferred to treat the species as part of the variation of H. chinensis.) Flora of China also lists allopatric Chinese distributions, with H. davidii only in the southwest, H. chinensis further east. The Chinese name xi nan xiu qiu (西南绣球 – Southwestern Hydrangea) emphasizes this. Herbarium specimens in China filed under H. davidii seem only to be from the southwest, although some Chinese botanists apparently recognize H. chinensis in Guizhou, Sichuan and Yunnan (Chinese Academy of Sciences 2024); whether this is due only to differences in species concept is unclear. Images on the Plantplus website (Chinese Academy of Sciences 2024) include a small minority without blue fertile flowers, but since there are also a few clearly mislabelled images this might not convey much about current Chinese opinion. One plant named as H. davidii has been included in recent molecular studies; Yang et al. (2024) placed it in Section Hirtae. From the standpoint of Western gardens, Foster (2023) notes ‘I find it difficult to be sure of distinguishing it from forms of H. chinensis [i.e. pottingeri]’.

Ernest Wilson made several early seed collections in Western Sichuan. Plants from W 4343 of 1910 were grown at the Arnold Arboretum, now dead (Arnold Arboretum 2024). Bean (1981) reports that plants from a 1908 collection were in cultivation at one time; this would have been W 1159 or 1249 (Sargent 1913). Both are probably now lost. Several recent collections by the likes of Bleddyn and Sue Wynn-Jones, Peter Wharton, Dan Hinkley and Keith Rushforth are in circulation among enthusiasts of these esoteric plants, tentatively labelled H. davidii. The bulk of these are from Northern Vietnam rather than China. Collections which have been offered commercially by Crûg Farm Plants, UK, include WWJ 11920 and BSWJ 8307, 11692 and 11717, all Vietnamese; at least some were initially labelled H. indochinensis. An evergreen specimen from BSWJ 8307 at Trewithen, Cornwall, had narrowly elliptic leaves with very short, straight to crisped, rather stiff hairs on the veins of both leaf surfaces, and no discernible hair tufts in the vein axils (pers. obs. of old leaves, March 2025). BSWJ 12233b, with narrow, thick textured leaves purple beneath has been distributed under the very tentative and perhaps unpublished name H. davidii f. purpurascens (Crûg Farm Plants 2024); to the author, this plant with its more or less inferior ovary and long styles seems rather closer to H. indochinensis, although the veins on the leaf underside are hairy (pers. obs. 2023); we illustrate this plant under H. indochinensis. Indeed, Crûg Farm currently identifies a second collection from the same wild population (KWJ 12233a) as H. indochinensis! Maurice Foster has found many of these plants hard to establish in Kent, UK, under conditions which suit other, apparently similar hydrangeas: they tend to grow away encouragingly before dying back and dwindling after 2 or 3 years (Foster 2023).

Uncertainty reigns here.