Hydrangea petiolaris Siebold & Zucc.

To most gardeners this is the archetypal climbing hydrangea. It is also perhaps the hardiest. A deciduous species, clinging tightly to surfaces by aerial roots, it is usually seen on walls but can be even more effective grown into large trees or forming bushy mounds out in the open. Sometimes slow to establish, after a few years it grows equally vigorously in sun or shade, given adequate moisture, but flowering is most abundant in more or less full sun. It can be very showy in flower for a few weeks in early summer, although the inflorescences quite quickly age to an insignificant brown. For much of the season the dark green, broadly ovate – almost rounded – leaves are its best feature; good yellow or even white tints in autumn are possible, but are by no means guaranteed.

Key identification features of H. petiolaris are its deciduous, broadly ovate leaves with a sharply toothed margin, and often a cordate base; presence of showy sterile flowers around each inflorescence, each having about four (rather than just one) enlarged white sepal; and each fertile flower having 15–20 stamens (Ohba 2001; Shaw 2012). A member of Section Calyptranthe, the petals of the fertile flowers are fused together to form a cap (the calyptra) which falls away as the bud opens and stamens elongate. The only other member of this Section we recognize is the very similar H. anomala, which has a wider and more westerly distribution from eastern mainland China and Taiwan to the Himalaya; it has fewer stamens (usually less than 15), is far less often grown, and with certain exceptions proves a less showy garden plant.

Throughout its garden career Hydrangea petiolaris has been involved in confusions around naming. Plants labelled Schizophragma hydrangeoides (i.e. Hydrangea hydrangeoides) in the 19th century mostly proved to be H. petiolaris (Bean 1913; Anderson 1933). Once one understands the differences, notably the single enlarged sepal in H. hydrangeoides, the issue is easily resolved, yet a trawl of online images shows that misidentification continues in some well known gardens and nurseries (pers. obs. 2023). A second source of confusion is McClintock’s treatment of H. petiolaris as a subspecies of H. anomala in her influential revision of the genus (McClintock 1957). While the taxonomic pendulum has swung away from her very broad species definitions, many horticultural books cling to this approach, and occasionally one finds statements made about H. anomala which clearly refer to H. petiolaris.

Fertile flowers are dominated by the stamens, many of which are folded in a complex way within the bud. As they unfold they push away the petals, which become fused along their margins during development, so the entire corolla pops off as a unit and falls (Hufford 2001). Each inflorescence is ringed by sterile flowers with about four enlarged white sepals. The size and number of sterile flowers is an important part of the plant’s appeal in flower, and varies quite a bit among seedlings. Foster (2023) recommends holding seed-raised plants back as non-climbing shrubs for four or five years before allowing only the best to climb away, clearly the advice of a man with a large garden. Scientific pollination studies are lacking, but the availability of both pollen and nectar as rewards, combined with the inflorescence structure implies a generalist insect pollination strategy. Diverse flying insects certainly visit in British gardens (pers. obs.), and H. petiolaris is often recommended as a ‘pollinator friendly’ climber (Royal Horticultural Society 2019).

Records of pink-flowered plants (as occasionally seen in H. anomala and H. hydrangeoides) are very few; Hinkley (2003) mentions a supposedly pink or pinkish flowered selection made in France. Siebold & Zuccarini (1835, 1870) illustrate an unambiguously pink flowered Japanese plant as H. bracteata: it has much in common with H. petiolaris but differs in key floral features (10 stamens, petals not forming a calyptra). It is customary to treat H. bracteata as a synonym of H. petiolaris (Royal Botanic Gardens Kew 2024), and to explain the anomalies as illustrator’s errors (McClintock 1957). Whatever the truth, there is too much uncertainty to allow this illustration as evidence of a good pink flowered H. petiolaris.

Even the most ordinary garden forms are handsome in leaf and tend to remain a fresh dark green even through hot summers. Wild plants vary a great deal in leaf form and size, but contemporary botanists are wary of describing points in the continuum as distinct varieties (e.g. Ohba 2001; Ohba & Akiyama 2014; Royal Botanic Gardens Kew 2024). Some such names still occasionally crop up in horticulture at various taxonomic levels, some published, some not. The most important of these are var. cordifolia (Siebold & Zucc.) Franch. & Sav., var. ovalifolia Franch. & Sav. and the shadowy plant originally described as H. tiliifolia H.Lev.. Regardless of the original concept, garden plants labelled var. cordifolia have relatively small, rounded leaves (van Gelderen & van Gelderen 2004). Some island collections (see below) have desirable foliage, especially those from Yakushima with red petioles, and large-leaved plants from Jeju-do (Crûg Farm Plants 2024). Shaw (2012) attempted to equate Jeju-do collection BSWJ 8846 with var. ovalifolia following Nakai (1926), without giving defining features, and described H. anomala subsp. petiolaris var. megaphylla J.M.H.Shaw (no published combination in H. petiolaris) from Ulleung-do collections BSWJ 4400 & 8497, but did not show how these might fit within wider patterns of variation in the species. We are loth to convey a false sense of understanding by giving any of these variants taxonomic status at any level. A thorough molecular and morphological study of Section Calyptranthe including both H. anomala and H. petiolaris from across their full geographical range is badly needed.

Autumn colour is a short-lived but sometimes excellent feature of this species. Mild autumns in southern England with only modest diurnal temperature fluctuations seem to result in no more than a pleasant yellowish tint. Rich yellow is possible however, especially in eastern North America: Dirr (2021) mentions seeing ‘superb fall color’ from Maine to Georgia. In Britain at least, cold nights can sometimes result in the leaves turning a ghostly pale cream before falling, a fine but chancy effect (see image below).

Hydrangea petiolaris is a root climber, attached to trees or rocks by aerial roots. As in other climbing species, seedlings develop branching horizontal stems rooting in the soil or litter layer. From these, erect side branches may colonize vertical surfaces they meet. The fine-scale distribution of wild plants may have as much to do with this juvenile stage as with the more obvious climbing phase. H. petiolaris and H. hydrangeoides frequently grow together in Japanese forests, but almost never colonize the same trees. Kato et al. (2014) concluded that this is because erect shoots in H. petiolaris were much more numerous in brighter parts of the forest floor, while H. hydrangeoides showed no such preference. Once on the tree, extension or ‘searcher’ shoots colonize branches, attaching by aerial roots. This attachment is strong, and in gardens there is rarely a problem with a heavy, established shoot framework pulling away from a wall. However, the grip is even stronger where the shoot system is allowed extend onto the flat top of a wall (Brown & Kirkham 2004). In subsequent years shorter, non-rooting side or ‘ordinary’ shoots grow out from this framework, away from the branch. These are the flowering shoots, which also carry the bulk of the leaves. Like H. hydrangeoides (Ichihashi & Tateno 2011), the physical scale of the extension shoots probably limits it to inner parts of the tree canopy with thicker branches, and renders spread to neighbours impossible, making it an unthreatening climber for large, established trees in cultivation.

Like other climbing hydrangeas, it has a reputation for being slow to establish (Dirr 2021). Bleddyn Wynn-Jones’ general advice on establishing climbing hydrangeas applies here. He recommends pegging down the shoots to encourage buds to break and produce horizontal extension shoots which grow along the soil; they or their branches soon colonize the wall (pers. comm. 2023). Pruning is unnecessary for the plant’s own good, but may be desirable where it extends too far. Side shoots may in time become awkwardly long, so a proportion of the shoots each year may be pruned back quite hard in spring to a suitable bud. Extension shoots which are outreaching their welcome can be pruned back in summer as they appear (Brown & Kirkham 2004). Removing parts of an established framework from a wall risks leaving behind unsightly trails of aerial roots, or even pulling away strips of paint or render ‘like the dark tracks of some giant centipede’ (Foster 2023). In the absence of vertical surfaces, H. petiolaris is sometimes recommended as a leafy ground cover for rock piles and the like (Nevling 1964). As a clump around a tree stump or single boulder (Bean 1981) it can prove very attractive and free flowering under brighter conditions, and can be pruned or clipped to limit its ambition (Foster 2023). H. petiolaris is rarely seen in an intimate mixture with other climbers, but Haworth-Booth (1984) suggests growing the slightly earlier flowering, scarlet, herbaceous climber Tropaeolum speciosum through it. This would probably work best in areas with cool, moist summers.

Hydrangea petiolaris was among the first Japanese plants described scientifically by Philipp von Siebold and Joseph Zuccarini in their long running Flora Japonica project (Siebold & Zuccarini 1835). As so often, it is unclear whether it came into European cultivation from Siebold’s Japanese botanic garden, but it was certainly in circulation among European collectors by the 1870s. RBG Kew had received it twice by 1884 (Hooker 1884), from German horticulturalist Max Leichtlin in 1874, and subsequently from George Joad, an English plant collector and ‘obsessive hoarder’ (Page 2011), whose vast collection passed to Kew on his death. It reached North America around the same time (though mislabelled Schizophragma [Hydrangea] hydrangeoides, a very common error in the 19th century), initially through Thomas Hogg, an American diplomat in Japan who also had an interest in a New York nursery firm (Nevling 1964). There have been numerous subsequent introductions from the wild (e.g. Royal Botanic Garden Edinburgh 2023); none apart from the distinctive island forms have been particularly influential, although it is notable that all the wild origin plants currently growing at the Arnold Arboretum, Massachusetts, are of northern (Hokkaido) provenance (Arnold Arboretum 2023).

Several cultivars have been selected, mostly for leaf colour or variegation; the variegated forms often show reduced vigour (Hinkley 2003). One interesting interspecific hybrid (a climber, parentage H. petiolaris × H. seemannii) is in general cultivation, ‘Inovalaur’ (= SEMIOLA™). French nurseryman Henri Cayeux successfully crossed H. macrophylla with H. petiolaris before 1922; the resulting plant was intermediate between the parents, but it was apparently never distributed, and all material was destroyed in the devastating bombing of Le Havre in 1944 (Haworth-Booth 1984).