Cornus quinquenervis Franchet

One of the most ornamental and elegant of the shrubby dogwoods, Cornus quinquenervis is also the last to bloom, flowering freely in late July and August, 7–10 days after C. macrophylla – ‘a useful time of year’, as Bean (1976) remarked. The species is easily told by its elongated, glossy dark-green, rather leathery leaves with a small number of lateral veins running up the leaf roughly parallel to the midvein. The black fruits (produced only sparingly in Britain) ripen at the beginning of October. The shrub is particularly attractive just before flowering, when the unopened flower clusters are greenish cream in tight bud. Dirr’s assessment (2009) that C. quinquenervis offers ‘probably no potential for everyday landscaping’ seems overly pessimistic, as well as historically inaccurate: in the United States, the shrub was quickly adopted as a landscape plant by both amateur gardeners and landscape architects, valued for its refined character and tendency to hold onto its leaves until late in the year (mid November) (Hicks Nurseries 1920; Hamblin 1923).

Cornus quinquenervis has been compared with C. alba, but with black instead of white berries, and Cappiello and Shadow (2005) report that plants masquerading under this name do indeed often turn out to be C. alba (or C. stolonifera). In fact, the species is commonly, or usually, encountered as C. paucinervis, the name under which it was first described by Hance, who was unaware that this name had already been published as a fossil taxon and was therefore invalid (Eyde 1988). The correct name – Cornus quinquenervis Franch. – has never really caught on, and the species continues to grow happily in numerous botanic gardens (and appear in databases) under Hance’s name.

The species is widely distributed within western and central China, growing in a variety of habitats, but was first noticed as a riparian plant along the banks of rivers up which early western botanists travelled. Handel-Mazzetti described the shrub as capable of thriving despite frequent inundation by rapidly flowing flood waters, often in association with Distylium (quoted in Stapf & [Balfour?] 1927).

The species was introduced into cultivation by Wilson in the early 1900s, and was seen in flower in late July at Veitch’s Coombe Wood Nursery in 1904 (Pettigrew 1905). This was probably material collected in Fangxian, northwestern Hubei in September 1901 (Wilson for Veitch 495) – the location and date are given on the herbarium sheet at the US National Herbarium (USNH 596467); Rehder’s details for this collection number (May 1900, ‘without precise locality’) is probably an error (Rehder in Sargent 1916). Wilson collected the species subsequently from Emei Shan in western Sichuan in 1904 (Wilson for Veitch 4950), and it is not clear which of these two collections was the one that was exhibited in flower (gaining an Award of Merit) at the Royal Horticultural Society in Chelsea in 1911 (Gardeners’ Chronicle 1911).

Wilson collected the species again in 1907 from near Ichang, Hubei for the Arnold Arboretum (Wilson 136), which appears to be the source of much of the North American material subsequently distributed (cf. Hicks 1920, 1921); curiously W 136 is recorded as having been reintroduced to the United States from Vilmorin-Andrieux in France under PI 49665 in 1920 (United States Department of Agriculture 1922). John ‘Lilacseed’ Dunbar, the superindendent of Highland Park Arboretum (Rochester, New York) obtained five seedlings from Sargent at the Arnold Arboretum in 1909 – probably Wilson’s 1904 collection from Emei Shan, since the material is described as Sichuanese, though W 136 is a possibility. Wilson’s description of W 136 is of a tall shrub 1–3 m in height (Rehder, in Sargent 1916), which corresponds to the erect stems of the plants grown at the Arnold Arboretum (Horticulture 1919; Arnold Arboretum 1922), though Dunbar noted that the Rochester seedlings showed ‘great variability in habit’, some low-growing or semi-prostrate (Dunbar 1916, 1917).

Forms in cultivation today show the same variability in habit, from plants with spreading stems such as that grown by the Raulston Arboretum, North Carolina, to the tighter, more upright form described by Dirr (2009). The colour of the fruit varies from the ‘jet-black’ of Dunbar’s specimens, to very dark indigo in plants offered by Pépinière AOBA, France (Pépinière AOBA 2025). Rehder (in Sargent 1916) noted that cultivated specimens collected by Faber (Faber 637) had much broader, elliptic to broadly ovate leaves, which were densely pubescent. Dried specimens collected by Augustine Henry (Henry 218 and 471) from Hubei also have this wider leaf (Forbes & Hemsley 1887), though Xiang and Boufford (2005) describe this condition (leaves ‘oblong-ovate’) as rare in the wild. There have been a few subsequent collections, notably SABE 1629, from the 1980 Sino-American Botanical Expedition to an area of northwest Hubei familiar to Wilson (Bartholomew et al. 1983).

Though it was originally collected at low-altitude from regions with a warm-temperate climate (‘where oranges grow’), assessments of the species’ hardiness have steadily been revised upwards since its first introduction, and material in cultivation can be regarded as fully hardy; Dirr (2009) records its having survived temperatures of –29°C in Cincinnati. For a riparian species subject to frequent flooding, drainage is obviously not a concern, though a moist substrate is not essential, and Dunbar’s material at Highland Park thrived on ‘fairly dry sloping well-drained soil’ (Dunbar 1917). Species adapted to life on the banks of fast-flowing rivers with fluctuating water levels often exhibit markedly plastic morphology and wide environmental tolerances, suggesting the potential landscape value of the shrub in modern planting design.

Seeds are reported easy to germinate if sown immediately and overwintered, germinating the following summer (Dunbar 1917). A spontaneous cross with Cornus amomum at the Arnold Arboretum was described by Rehder as Cornus × dubia (Rehder 1924).