Paulownia Siebold & Zucc.

Introduction

The shape of each Paulownia bloom, the tall towers in which they are borne, the big, woolly leaves and the fruit-capsules fully of tiny seeds are all features which can recall the common European foxglove (Digitalis purpurea); both genera were traditionally placed within a large amorphous family, Scrophulariaceae, and the ‘Foxglove Trees’ have indeed been held up as an example of how a largely herbaceous family may contain one group which has evolved an arboreal habit. In 1949, however, the Japanese botanist Takenoshin Nakai was the first to suggest that Paulownia was better placed within its own family Paulowniaceae, an opinion which has since been thoroughly endorsed by phlyogenetic analyses (Zhao, Li & Shang 2019; Xia, Wen & Gao 2019) and which accords well with the casual observer’s sense that here is a highly distinctive and easily recognised little group of trees. Taxonomic advances have largely demolished the Scrophulariaceae of old, and many genera once placed here have been transferred to other genera within the order Lamiales (Olmstead et al. 2001). Within Lamiales, Paulownia is considered closest to the Orobanchaceae – a family which includes the broomrapes, parasitic herbs whose overall appearance scarcely hints at this evolutionary relationship.

Paulownia is still often confused with the Bean Trees Catalpa (family Bignoniaceae), also placed in the Lamiales though the families are not closely related: both genera share conspicuously large and soft-textured leaves which are usually heart-shaped and which flush late in spring and fall without any autumn colour; they both have stout twigs with tiny opposite buds, a wide-branching habit, and showy flowers in tall ‘candles’. Catalpa flowers are never blue or violet, and they open among the leaves in summer; their brownish, ruggedly-fissured bark differs from the grey rather finely textured bark of Paulownia, and the pith in their twigs is not closely divided into horizontal chambers, as it is in Paulownia.

Although Paulownia, as a genus, is so easily recognised, its species are remarkably alike and are only separable with confidence by observing quite subtle details of the structure of the flower-head, the depth of toothing on the calyx, and the size and shape of the seed-capsule. This account follows Plants of the World Online (Royal Botanic Gardens, Kew 2022) in recognising seven wild species and the nothospecies P. × taiwaniana T.W. Hu & H.J. Chang, which seems to have originated as a natural hybrid. All of these are established in cultivation in the west except for P. laotica Aver. (“from Laos”), which was first described in 2019 as a subtropical tree endemic to that country (Averyanov, Nguyen & Maisak 2019). P. fortunei (Seemann) Hemsley has also been recorded from Laos, as well as in Vietnam and southern China; the remaining species all have different distributions within China, Vietnam and Taiwan, with the range of P. tomentosa (Thunb.) Steud. questionably extending north and east as far as the Korean peninsula. Paulownia fossils have been found as far afield as western North America, southern Europe and the Caucasus (Wikipedia 2022), indicating that this is one of many kinds of tree whose range has drastically contracted during the last few million years of glaciation-related climate change.

Like many another genus, Paulownia has come to be represented in horticulture by one dominant species to such a degree that many European and American gardeners will be quite unaware of the existence of the other six or seven. (In the sphere of forestry, particularly in warm-temperate regions, this is not so much the case.) The one species which tends to stand in synecdoche for its genus within our gardens is P. tomentosa, which was the first taxon to be introduced to the west and is probably the most adaptable and the most tolerant of northern Europe’s relatively short, cool summers. Individual trees of P. tomentosa may indeed represent almost the entire diversity of their genus in gross, visual terms: the flowers can be predominantly white, or soft mauve, or violet-blue; the mature tree can be broadly-domed, or tall and upstanding; its leaves can be deep green or grass-green, and may be broad and almost regularly pentagonal, or gracefully long-pointed.

In botanical detail, the inflorescence of Paulownia tomentosa (a thyrse) consists of an erect stem or rachis with a number of short branches. Along these branches, and also along the rachis, are carried an indeterminate number of ‘cymes’ – clusters of three to eight flowers opening individually one after another; just a few flowers are carried singly along the thyrse. These cymes have a brown-woolly common stalk (peduncle) about as long as each flower’s individual stalk (pedicel). This detailed description is offered here partly because two of the major sources which Trees and Shrubs Online aims to supersede – Trees and Shrubs Hardy in the British Isles and New Trees – fail to observe, or at least to clarify, the distinction between the entire thyrse and its component cymes. The architecture of the thyrse and of the cyme is also key to differentiating the various species of Paulownia: P. fargesii Franch., from southwestern China and northern Vietnam, and P. kawakamii T. Itô, questionably endemic to Taiwan, differ from P. tomentosa in bearing thyrses with just a few, long branches, and their cymes have scarcely any peduncles. In cultivation, P. fargesii has been persistently confused with a variant of P. tomentosa, described below as the Lilacina Group; these two taxa can be distinguished by checking this floral detail.

The other cultivated species of Paulownia have inflorescences which resemble those of P. tomentosa in overall structure. Among these, P. fortunei has significantly the largest seed capsules (more than 6 cm long, with a woody wall). P. elongata S.Y. Hu and P. catalpifolia T. Gong ex D.Y. Hong, two closely allied taxa from southern China which were only separated at species level in 1976, differ from P. tomentosa in their flowers’ less hairy calyces, which are lobed to less than half their length; of this pair, P. catalpifolia has flowers with narrower corollas (less than 35 mm broad). P. × taiwaniana, the hybrid between P. fortunei and P. kawakamii, shows intermediate features in the structure of its inflorescences.

It is not possible to identify the various Paulownia species from their vegetative features, though P. fortunei can occasionally be recognised by its slightly more upright habit and P. kawakamii by its slightly glossier leaves; P. catalpifolia has leaves which are unusually slender (about twice as long as broad). However, as Paulownia flowers are mostly carried towards the tops of quite big trees, the critical floral features can be hard to spot (and are seldom visible in photos of flowering trees to confirm a claimed identification). Nearly all the species of Paulownia have long been widely planted in China, making their original and distinctive wild distributions difficult to determine (Hong et al. 1998) and Valder (1999) suggests that many cultivated Chinese tress are of hybrid origin.

The first phylogenetic analysis of all seven cultivated species of Paulownia was made by Zhi Xia et al. in 2019 (Xia, Wen & Gao 2019). This confirmed the expectation that here is a close-knit genus, but suggested that P. fargesii is something of an outlier within the group and that the similarities in its flower-structure with those of P. kawakamii are an example of convergent evolution and should not form the basis of a division of Paulownia into Sections, as had previously been proposed by the Chinese botanist Shiu-Ying Hu (Bean 1976).

Paulownia is one of few arborescent genera – and one of very few genera whose distribution extends to cool-temperate regions – which can perform photosynthesis based on C4-cycle enzymes (Sage & Sultmanis 2016). ‘C4 photosynthesis’ has evolved many times among tropical plants, especially grasses, and, given high light levels, is significantly more efficient than the C3-cycle on which other plants depend. However, the majority of photosynthesis in Paulownia, particularly in temperate areas with lower light levels, is still performed using C3 activity.

History of cultivation

Paulownia tomentosa was first described by Carl Peter Thunberg (as Bignonia tomentosa) in his 1784 Flora Japonica. No Paulownia is native to Japan, but P. tomentosa had long been grown here, both as an ornament and for its valuable timber; it was traditional to plant this fast-growing tree upon the birth of a daughter, so that it could be felled to provide the wood for her marriage-chest or tansu (Roman 2016; Wikipedia 2022). This same species was introduced (from Japan to Holland) by Philipp von Siebold in 1829–30 (Jacobson 1996), but may not have survived the long sea voyage; its seed, again from Japan, reached France in 1834, where a single plant was germinated by Joseph Neumann within the hothouses of the Jardin des Plantes in Paris (Bean 1976; Philadelphia Historic Plants Consortium 2012). Siebold, writing with Joseph Gerhard Zuccarini and apparently unaware of Thunberg’s description of the tree, created a new genus for the ‘Empress Tree’ the following year, naming it Paulownia imperialis to honour Anna Paulowna (née Anna Pavlovna Feodorovna), who was then the Crown Princess of the Netherlands and became Queen five years later. (She never was an Empress, though she was the eighth child of the Russian Tsar Paul I and the Empress Maria Feodorovna. The story that in old China a Paulownia could only be planted on an Empress’ grave (Johnson 2015) appears to be a modern invention.)

After a couple of years, Neumann found that his plant did better outside, even in the cold winters of 19th century Paris; it flowered for the first time in 1841, and being self-fertile it provided an abundance of seed for the European market, although in 1838 there had been another importation of seed directly from Japan to England (Bean 1976). By 1843, P. tomentosa had reached Parsons’ Nursery in New York (Philadelphia Historic Plants Consortium 2012).

All Paulownia like a long, warm growing season. Despite the southern, low-altitude origins of most species, they are surprisingly hardy, but will struggle to grow into trees if chilly and dark autumn weather sets in when they are still making their summer growths. Their unusual habit of carrying next spring’s blooms half-developed within large woolly buds over the winter also means that severe or late frosts can compromise their role as flowering ornamentals. Except in the warmest parts of the UK, it used to be recommended to treat P. tomentosa as a herbaceous plant, cutting it back each winter to exploit the strong coppicing abilities of the genus and to enjoy the extravagantly big sunflower-like leaves of the regrowth the following summer; these leaves have an interesting soft texture, and are often velvety with sticky, aphid-trapping hairs. In recent years, climate change has made P. tomentosa, at least, a reliable flowering tree well into the English Midlands, and W.J. Bean’s assessment of the species at Kew a century ago speaks to us poignantly from an age of chillier weather, and of limpid prose: “Few more beautiful flowering trees than this exist, but although the tree is fairly hardy and sets its flowers, they often do not develop in this country, owing to its curious habit of exposing them in bud through the winter. Perhaps they do not derive sufficient stamina from our dull summers, but more likely the unrest of our winters, with their alternate frosts and mild spells, prevents their proper development.”

In climates a little milder again, Jacaranda species will begin to compete with Paulownia as spectacular, blue-flowering trees. (Jacaranda flowers are less scented, and even become foul-smelling as they drop.) But Paulownia are planted and thrive even within the tropics, where they become evergreen: P. REVOTROPIX^® is a hybrid of P. fortunei with P. tomentosa selected for its performance as a forestry crop in Malaysian Borneo (Pim 2021).

All Paulownia dislike waterlogging (Dirr 2009), and can struggle in heavy clay; they do need a fairly fertile soil to thrive, and they quickly grow deep, fleshy and rather brittle roots, meaning they are not among the easiest of trees to transplant. P. tomentosa, however, is sometimes offered at large standard sizes (Barcham Trees 2022). Paulownia are generally pollution tolerant, but are sensitive to the salinity that can build up in irrigated soils (Jakubowski 2022). The thick, soft leaves are palatable to deer and other animals, and the grass which grows in the rather light and transient shade of Paulownia plantations in climates such as New Zealand’s can only be grazed if each tree is protected by electric fencing (Dean 2022); but in the right conditions the tree is so vigorous that the top leaves will be out of most grazing animals’ reach by the end of its first summer.

As classic pioneer trees Paulownia grow fast and die young; in ideal sites, forestry plantations can be mature and ready for cutting in as few as five years. Nevertheless, the United States’ champion P. tomentosa, at the Reitz Memorial High School in Indiana, was planted in 1926 and was a huge but by no means moribund tree with a forking trunk 2.43 m broad at the grand old age of 96 (monumentaltrees.com 2022). In a more challenging climate, 50 years seems a likelier average lifespan.

With their huge leaves and rather soft stems, young Paulownia are vulnerable to gale damage and to windblow. Their ornamental value is also maximised if they are planted in a sheltered valley, where the blossom can be admired from the side or even from above; the pale violet to which the flowers usually approximate can almost seem to disappear against a blue sky, and most of the blooms are carried high in the tree’s upper canopy. In northern regions Paulownia should be planted to enjoy full sun, but in forestry trials in the high light levels of Idaho, USA, sunscald to the young branches has been identified as a problem (Barkley 2007).

Paulownia species are resistant to honey fungus (Armillaria spp.) (Huxley, Griffiths & Levy 1992), although David Ewins suspects this was the main cause of the demise in 2022 of a 20-year-old specimen of P. tomentosa Lilacina Group within his UK National Collection (D. Ewins pers. comm). In East Asia the most serious threat is witches’ broom disease, which is caused by phytoplasmas belonging to the Aster Yellows group (Candidatus Phytoplasma asteris) and which results in a proliferation of branches with tiny yellowish leaves, followed by catastrophic retrenchment; almost 80% of plantations of P. × taiwaniana in Taiwan were infected and lost in the 1970s (Liu et al. 2013; Yang et al. 2010). The pathology of this disease remains little understood, but it is not (yet) an issue of concern in northern Europe.

For a century, Paulownia tomentosa was the only member of its genus to be grown in western gardens. Since the 1930s, all the other species bar one have been introduced, though their remarkable visual similarlity means that none of them have yet begun to oust P. tomentosa in popularity; of the newer introductions, P. kawakamii has so far attracted the most interest, thanks in part to the performance of an excellent and conspicuous representative outside the Water Lily House at the Royal Botanic Gardens, Kew. Its assessment by the International Union for Conservation of Nature as Critically Endangered (IUCN 2022) – alone within this genus – has also encouraged its planting for the sake of ex situ conservation.

Since the members of Paulownia can only be separated with confidence through details of their flowers and fruit, it is inevitable that some wild collections in western gardens have not yet been identified to species level. In England these include Barry Clarke 38 at the Sir Harold Hillier Gardens, TH 2026 and TH 2773 at Tregrehan in Cornwall, and at Lamellan – also in Cornwall – KR 3866, plus a second Keith Rushforth collection of great vigour which was 17 m tall after just seven years (Tree Register 2022).

The UK’s National Collection of Paulownia is held by David Ewins in his garden outside Bath, Somerset, and has been built up since 2001 using mostly commercially sourced material; the rather alkaline and stony free-draining soil here, with early springs and warm summers, seems to suit the genus very well (D. Ewins pers. comm). David has very kindly sponsored this revision of the genus for Trees and Shrubs Online.

Paulownia as an invasive plant

Gardeners in parts of our area may be surprised to find that in large parts of the eastern United States Paulownia tomentosa is now a notorious invasive weed (Ecosystem Gardener 2022); it has also been identified as potentially invasive in parts of central Europe (Essl 2007). It is sometimes claimed that only P. tomentosa behaves in this way, but, given the very similar ecology and physiology of the different Paulownia species, it is probably more accurate to suggest that P. tomentosa is the only one to have so far been given the opportunity, thanks to very widespread planting. Invasive behaviour is also reported in P. fortunei which has recently gained popularity in the west as a forestry crop (Jakubowski 2022). It is now illegal to sell P. tomentosa in Connecticut (Ecosystem Gardener 2022).

Paulownia trees tend to be self-fertile (Plants for a Future 2022), and one mature tree can produce more than a million seeds each year, which with their tiny wings can travel more than a mile from their parent. (A figure of 20,000,000 is often cited, but meticulous counts of the seeds per pod and the pods per tree made by David Ewins in his UK National Collection of Paulownia suggest that this is an over-estimation, by a factor of at least ten (D. Ewins pers. comm.).) However, because the seeds are so light, the conditions in which they can germinate are quite narrowly delimited. The fruit capsule typically splits open in spring – a year after flowering and at least twenty months after the flower-buds were set; the fruiting process thus spans two winters. The shed seeds will will quickly die if they dry out, but will not germinate if they get buried; germination is triggered by bright light and by an average (day/night) temperature of around 11° C (Jensen 2016). In the wild, seedlings often spring up in the wake of wildfires, and are likeliest to succeed in near-sterile soil conditions; this is a pioneering species, which cannot invade shady forests and which tends to lose out in competition with longer-lived trees that ultimately grow taller. However, the deep roots and strong coppicing abilities of established Paulownia trees allow them to survive some forest fires (Wikipedia 2022).

In the current climate of the UK, Paulownia seeds nearly always die (or get eaten) before spring temperatures rise far enough to trigger germination; any seedlings are also likely to be eaten by slugs or rodents. In his National Collection in Somerset, David Ewins noted spontaneous seedlings for the first time in the very warm early summer of 2022 (D. Ewins pers. comm.); in 2021, the author observed a three-metre tall sprout of P. tomentosa growing from the ballast of the elevated railway tracks east of Waterloo East Station in the central London heat island. In 2000 and also in the warmest corner of the UK, Chris Gibson studied a seedling in the tarmac of the carpark of the Colchester office of English Nature (as was), though his identification of the then very rare P. kawakamii on the basis of foliage characteristics was naive (Gibson 2003).

Some Paulownia hybrids are infertile, and can consequently be planted with confidence even in areas where P. tomentosa has proved aggressively invasive. (Others of course are fertile, allowing the ongoing breeding of clones with a complex ancestry.) Most of the clones which are now commercially available for forestry use are at least claimed to be sterile, and a few comparable hybrids (ARCTIC^®, ORIENTA^®) are now aimed at the North American horticultural market. In future, it seems likely that most Paulownia sold around the western world will be sterile trees of hybrid origin. In the Far East, although most of the species have been widely planted and have naturalised extensively (Hong et al. 1998), they remain in better balance with their environment.

Uses of Paulownia

Even though in northern Europe and North America Paulownia have been grown for nearly two centuries very largely as ornamentals, they are remarkably useful trees. Leaves, flowers, fruit and bark are utilised extensively in Chinese traditional medicine. The soft, fleshy foliage contains unusually high quantities of proteins and is valuable as a fodder crop, and for green manure. The roots can penetrate to great depths and Paulownia is sometimes planted to reclaim landslips and opencast mining sites; the rapid growth – powered in part by the tree’s use of ‘C4 photosynthesis’ – quickly sequesters heavy metal pollutants (Jakubowski 2022).

Despite this fast growth, Paulownia produces a highly valued timber: there is only a slight trade-off, with slow-grown older wood being preferred for some purposes, such as the construction of musical instruments for which its resonant properties seem particularly well suited. Paulownia sapwood is only one or two rings thick; the attractive heartwood is grey-blonde, with an even grain and almost microscopic vessels (Jakubowski 2022), and it can be cut into very thin veneers. Being so light, it can be air-dried in just three months, seldom warps or cracks or splits, and has good insulating properties. It is about twice as dense as Balsa wood but has a much higher strength-to-weight ratio; this characteristic has led to its being called legno alluminio in Italy – ‘the aluminium of woods’ (Paulownia Italy 2022). Paulownia can also be planted for biomass production and for carbon capture, though the low timber density means that they are not as effective in such roles as growth-rates along would imply (Jakubowski 2022). Its fire-retardant nature has made it a valuable wood in China and Japan for centuries (Li & Oda 2007)

Paulownia is one of very few forestry trees to be insect pollinated, and a highly valued honey is produced by bees in Paulownia plantations in the Far East (El Cortijuelo 2022). (It is sometimes claimed online that the hybrids bred for modern Paulownia forestry do not flower, but this is a misinterpretation of the statement that they never set fertile seed. In a modern plantation with a fast rotation, they will however be cut down almost as soon as they are old enough to begin to bloom.)

Paulownia in contemporary forestry

Most of China’s indigenous Paulownia species have long been exploited for their timber in the Far East; their relatively open canopies allow them to be interplanted in particular with the evergreen tea plant, Camellia sinensis (Li 2001). Interest in Paulownia as plantation trees in Europe, North America and Oceania has been much more recent, and the following synopsis is really just a snapshot of a fast-developing and potentially important industry.

Paulownia silviculture rather resembles that of the traditional English bat willow (Salix alba var. caerulea). It is recommended to coppice the young plant so that strong straight stems are grown the following summer; one of these is singled, and rubbed clean of buds to create a length of knot-free timber, which may be six metres long in ideal conditions; the tree often branches widely above this point. In such sites, trees can be ready for felling in as few as five years; cut to the base, they will re-coppice with equally strong growths and the cycle can theoretically be repeated ad infinitum, though it is sometimes suggested that the stool will become exhausted after a period equivalent to the lifespan of a standard specimen (less than a century).

It needs to be re-emphasised here that all Paulownia appreciate a long, warm growing season. The first plot outside the Far East may have been a 1.3 ha plantation of P. elongata established in Polk County, North Carolina by 1979 (Hall 2008); since then, small-scale timber industries have developed across the south-eastern United States and in California and north-west Mexico, and in Spain, Italy, the Balkans, Germany, Morocco, South Africa, Malaysia, Australia and New Zealand. A trial of three Moroccan and six Spanish clones in the very different climate of Hillsborough in Northern Ireland (Olave et al. 2015) was a resounding failure, and climate change in northern Europe still may not have progressed far enough to turn this particular foresters’ dream into a reality. England’s first experimental Paulownia plantations were made in 2022 (Haugh 2022).

Contemporary Paulownia timber crops usually involve P. fortunei, which is potentially the largest- and straightest-growing wild species. Selected clones are available, but fashion now favours hybrids of P. fortunei with other species. This allows the companies supplying the micro-propagated saplings to patent their own unique varieties; more importantly, the hybrids can be infertile and non-invasive, as well as carrying traits of hybrid vigour.

In the species entries in Trees and Shrubs Online, these forestry hybrids are grouped according to their species parentage (with the parent species arranged alphabetically even when the first is known to have been the seed parent). In a few cases, the clone’s parentage is not known – some will have arisen through open pollination – and then it receives its own species entry. The patented trade names often have to be used, since few of these forms have registered cultivar names. Only clones which currently have some prominence in forestry are listed; in coming years, new and even more vigorous selections seem bound to come into favour. All of these clones are very similar in general appearance and for this reason their descriptions are abbreviated or omitted; their longer, straighter trunks are a consequence of the conditions in which they are cultivated as much as they are evidence of genetic differences between them and the kinds of Paulownia familiar in gardens.

Probably the most important group of forestry hybrids is that of Paulownia fortunei with P. elongata, another southern Chinese species renowned for its vigour and which is also often used in East Asian plantations. Clones with this parentage include BIG^®, ‘Oxi’, PHOENIX ONE^®, ‘PWCOT 2’, and SUN TZU 11^®. The hybrid has been back-crossed with P. elongata to create BELLISSIA^®, crossed with P. kawakamii to create PAULEMIA^®, and crossed with P. tomentosa to create clones such as ‘PW 105’ and ‘T121’. Meanwhile, the cross of P. fortunei with P. tomentosa has produced GANTIER^®, NORDMAX 21^®, REVOTROPIX^®, ROYAL TREEME^® and ‘Shan Tong’; this hybrid has been crossed with P. elongata to create CAROLONIA^® (which therefore shares the three parents of ‘PW 105’ and ‘T121’), and crossed with P. kawakamii to create ‘Pao Tong Z07’. Hybrids of P. fortunei with P. catalpifolia – a slightly less vigorous species but with more valuable timber – include ‘H2F3’ and ‘H2F4’. P. ‘9501’ is a more complex hybrid (fortunei × elongata) × (fortunei × tomentosa); it seems likely that many of tomorrow’s selections will feature increasingly elaborate lineages.

The following key refers only to wild species.