31 Davidsonia 14:2

A Plantsmans Observations on the Genus

Hydrangea
The hydrangea family, Hydrangeaceae, is a respectable conglomeration of
shrubs, vines and herbaceous perennials resulting from a recent extraction
from a behemoth institution known as the Saxifragaceae. The herbaceous
components, in the respective genera of Cardiandra and Deinanthe proffer
ornamental species for cultivation although it is the woody taxa of shrubs
and lianas of this family, in the genera Hydrangea, Deutzia, Philadelphus,
Platycrater and Schizophragma, that are by far the most familiar.
Other than the exceptions of Hydrangea paniculata and H. quercifolia, the
prototypical hydrangea inflorescence is a corymb; a rounded disc of numerous
small, fertile flowers that possess insignificant sepals and four or five small
often white, pink or blue petals. This cluster of utility is surrounded by the
advertising agency of sterile florets, or ray flowers, that provide the stuff of
ornament. The ovaries are, as a rule, inferior (i.e., enclosed in the receptacle),
while the dehiscent capsules, sometimes in the shape of a Grecian water jar,
gave rise to the genus name; hydro, water and angeion, a vessel. The foliage is,
without exception, arranged in pairs.
This floral strategy of sacrificing the fertility of a few flowers to provoke
a bit of curiosity by commuting pollinators has co-evolved in other non-
related groups, most notably in the genus Viburnum. Thus, it would not
come as a complete surprise to those who have grown a double-file viburnum,
Viburnum plicatum, that the second Asiatic hydrangea to be noted by Western
botanists (in Japan, in 1777) was named Viburnum serratum. That nascent
nomenclatural error commemorates what would become a long and
complicated excursion into a field plethoric with taxonomic landmines.
The first hydrangea described from Asia, also in 1777 by Thunberg while
in Japan, was actually a mophead or Hortensia cultivar, with the bulk of its
fertile flowers clustered into heads of embellished yet mostly sterile florets
(see under H. macrophylla for the convoluted genesis of Hortensia). Having
Daniel J. Hinkley, Heronswood Nursery, 7530 288th N.E., Kingston, WA, USA, 98346,
heron@silverlink.net
32
never encountered such a lovely creature before, it is understandable that
Thunberg christened this Viburnum macrophyllum. It is important to note
that this type of hydrangea flower mutation is different from a doubling or
trebling of the sepal numbers in each ray flower, which results in rose-like
florets surrounding the central core of usually fertile florets.
I consider that the confusing taxonomy of the hydrangeas that exists to
this day can be ascribed to a quartet of unfortunate circumstances. First,
much of the fi rst named materi al , i ncl udi ng Thunbergs Vibur num
macrophyllum, was based on clonal selections with no botanical standing.
Second, many Hydrangea species have immense geographical ranges and the
natural, often significant variation found within each taxon was always
problematic, even in the relatively few cases when live material was readily
accessible for study. Remote and politically insular hotspots of Hydrangea
speciation (i.e., Japan and China), have not until recently offered much useful
data. Third, although Elizabeth McClintocks highly regarded monograph
(McClintock, 1957) remains the most comprehensive to date, it is fair to
point out that many of her astute observations were made in spite of a
paucity of oven-dried herbarium specimens. The fourth contributors to the
confusion have been nurserymen and lay authors, in the midst of whom I
find myself, who have simply taken matters into their own hands, laying low
in great sweeps the systematists attempts to provide a robust grip on the
genus.
In this paper, I will attempt to place my personal observations of species
in the wild and those of known provenance that I have grown within the
broader academic context provided by McClintock. I have purposely avoided
the minefield of the complex of H. macrophylla and H. serrata cultivars, but
I have discussed forms of numerous species, which I feel are important. I
leave the taxonomy to my more academic brethren hoping that they can
solve the riddles of this remarkable genus of plants, many of which are not
at all well known to horticulturists
Section Hydrangea
This section contains the deciduous members of the genus, including the
vines, and distinguishes them from the much lesser known, evergreen and
primarily southern hemispheran taxa, which are placed in Section Cornidia.
Both sections are divided further into subsections.
33 Davidsonia 14:2
Subsection 1 Americanae
The two North American species of Hydrangea have achieved more than a
modicum of acceptance in Western horticulture. They were recognized by
native Americans for their curative properties and used by the Cherokees
and early American settlers for the treatment of calculus. They contain the
cyanogenic glycoside, hydrangin, although ingestion of raw hydrangea
vegetation does not seem to result in the typical clinical signs of cyanide
poisoning.
Hydrangea arborescens L. occurs along the eastern coast of the USA from
Florida to New York and west/southwest to Iowa and Louisiana, and is often
found growing under exceptionally shaded conditions. It is a deciduous shrub
rising to 3m (10 feet) in height carrying ovate, smart, green leaves of a papery
texture. The mostly fertile, dingy white flowers of the northern populations
are hardly awe-inspiring in their pure state, but at least two selections have
considerable hardiness as well as frilly heads of sterile flowers. Hydrangea
arborescens Grandiflora and the more compact H. arborescens Annabelle each
begin flowering in June with tasteful tones of lime green, ripening to pure
white, and changing later to verdant tones. The former was the first hydrangea
that I grew as a young gardener in Michigans frigid USDA Zone 4 interior,
where its somewhat sloppy growth was overlooked through my innocent
eyes of ignorance. The somewhat stronger stemmed, later blossoming
Annabelle, with flower heads to 30cm (12 inches) across, was selected at the
University of Illinois by J.C. McDaniel. It has justifiably superceded
Grandiflora in commerce, though I still find both exceptional contributors
to my mixed borders in USDA Zone 8.
In the southern Appalachians, the brilliant white indumentum on the foliage
undersurface of H. arborescens subsp. radiata (Walter) McClintock (Figure 1)
makes up for its lackluster floral display. It looks best growing atop
embankments or retaining walls. It remains one of my favorite hydrangeas
and deserves much greater recognition.
Hydrangea arborescens subsp. discolor (Ser.) McClintock is poorly represented
in cultivation. It is best known as its cultivar Sterilis, which has frosty
heads comprised of mostly sterile florets. In June of 2003, I observed
populations of this taxon at 2000-3500 in elevation near Asheville, N.C.
together with numerous individuals of H. arborescens subsp. arborescens, i.e.
34
without the grayish tomentum found on H. arborescens subsp. discolor. Sterile
florets may or may not be present.
The oak-leaf hydrangea, Hydrangea quercifolia Bartram had a prolonged rise
to favor among North American horticulturists in those USDA Zone 5-9
landscapes where it can be grown. Dr. Michael Dirr is probably responsible
for much of this interest, due to his selection work at the University of
Georgia. Though one of only two Hydrangea species that possess flowers
born in panicles rather than corymbs, it is the foliage of H. quercifolia that
sets it aside. Very leathery in texture, the boldly lobed and jagged-edged
leaves to 20cm (8 inches) in length do indeed resemble those of a red oak
and develop intense and prolonged tones of glossy burgundy in autumn. I
have seen this Gulf State native growing in Mississippi in semi-shaded sites
with more than adequate water, though in cultivation it appears to tolerate a
great deal of drought when established. Though Dirr recommends always
providing this species some protection from full sun, this does not seem
necessary in the Pacific Northwest where it becomes annoyingly rangy if
grown in too much shade, with leaves that fail to develop autumn tones or
abscise uniformly.
Subsection 2 Asperae
The subsection Asperae has outstanding ornamental potential, but it
contains many barriers and pitfalls in proper classification.
Hydrangea sikokiana Maxim. (Figure 2) is shockingly scarce in cultivation
considering its clearly handsome foliage and large white lacecaps to 30cm
(12 inches) across. I first observed it in the autumn of 1997 while in the
moist, cool highlands of Central Honshu on the Kii Peninsula with colleagues
Bleddyn and Sue Wynn-Jones and Darrell Probst, the Epimedium specialist.
It was on that particular day that I became aware of the enormous diversity
of the Hydrangeaceae on the Japanese archipelago. I could count no fewer
than eight hydrangea species or their close relatives without altering my
position. Among the group, growing on the shore of a rapidly moving
mountain stream was a tall shrub with large felted, light textured, jagged
edged leaves that appeared superficially similar to those of H. quercifolia. As
we had spent longer a week on Shikoku Island specifically searching for this
species without success, and were nearing the end of a second rainy, cool
autumn day, I was certain I had found what we sought. The moment of
35 Davidsonia 14:2
recognition was accompanied by that rarified charge of electricity that such
experiences are known to bring. From seed collected at that time, we were
able to re-introduce H. sikokiana into cultivation in North America and
Europe, but we have received few reports as to how it has fared. It is proving
to be a distinctive, mid to late summer blossoming addition to our cool,
shaded woodland.
Hydrangea involucrata Sieb. (Figure 3) was not among the species that we
saw on the Kii Peninsula on that particular day, although I had encountered
it on numerous occasions during prior travels in Japan, most notably in 1995
along the coastal forests of the Chiba Peninsula south of Tokyo. As its
specific epithet implies, H. involucrata sets itself aside from all other species
in Section Hydrangea by possessing involucral bracts that enclose its plump,
rounded flower buds before opening. The bracts dehisce as the flattened
corymbs of lavender blue fertile flowers, surrounded by creamy white ray
flowers, open. The ovate foliage to 13cm (5 inches) in length is held along
stems to 3m (10 feet) in height. There are several forms in cultivation that
possess an aberrant doubling of the sepals in the ray flowers, in particular
the lovely cultivar Hortensis, which also surrenders a greater number of its
fertile florets to the cause of ornament.
Hydrangea involucrata hybridizes readily with H. aspera D. Don, a complex
taxon found in the eastern Himalaya and the mountains of western China
south to Taiwan, Java and Sumatra. Following McClintock, I have grouped
numerous closely related species as subspecies under this umbrella. I do this
with a certain degree of discomfort because those I have observed in their
wild state appear unquestionably distinct from one another.
Hydrangea aspera subsp. strigosa (Rehder) McClintock will be the first
hydrangea encountered by anyone traveling to many areas of Asia, where it
occurs at lower elevations in mountain valleys. I have seen it in eastern
Nepal at elevations of 1800m, in Vietnam at 1550m and in Sichuan Province
at 950m. It is easily recognized by its very narrow foliage with strigose hairs
on the undersurface and a very late blossoming habit; in fact, I have often
seen it still in blossom in October. Use of the strigose hairs alone as a
diagnostic tool is problematic. In 1998, I established cuttings from a specimen
collected on the lower slopes of Mt. Emei clinging precariously to a small
plot of ground between an asphalt highway and a steep cliff to a river below
(collection number: DJHC 98464). This plants globose heads were nearly
36
entirely comprised of ray flowers. Michael Haworth-Booth reported that
Ernest Wilson collected a similar type from Emei in the early years of the
20
th
century (Wilsons number 4902), but it has apparently been lost to
cultivation. We have introduced this plant to cultivation under the clonal
name of Elegant Sound Pavilion (Figure 4) to commemorate the Buddhist
temple near the collection site. The plant was still intact and in good health
when I visited Emei in 2000.
Hydrangea aspera, as most know it in cultivation, is from the higher elevation
taxon known as H. villosa Rehd. (or H. aspera subsp. villosa Hort), but which
was reduced to synonomy by McClintock under subsp. aspera. There are few
more beautiful deciduous shrubs than this, with its velvety villose leaves borne
along stems up to 3m (10 feet). In July and August, the plant seems to be
smothered with 25cm (10-inch) lacecaps of lavender-purple fertile florets
surrounded by creamy white ray flowers. By late summer, the woodland
floor beneath the plant is covered with a beguiling purple snow as the petals
dehisce. My collection of this form (number: DJHC 636) from the Wolong
area of Sichuan, at 3000m, has proved to be a sensational plant in our garden
with cymes to 30cm (12 inches) across. In 1998, I collected DJHC 98443
from a similar elevation on Emei Shan and described it as possessing ovate,
villose foliage to 20 cm (8 inches) in length and 18cm (7 inches) wide.
Though McClintock chose to reduce H. sargentiana to a subspecies of H.
aspera, Rehder gave it species rank. It was introduced from western Hubei
Province by Ernest Wilson in 1908, but I have not visited this area of China
and am unable to confirm if subsequent collections have been made. It
differs from H. aspera in what has been well described as a moss-like coating
of hairs on its young stems, as well as large, ovate, substantive leaves of
deep green covered with a seductive indumentum. The cymes, produced in
mid to late summer are large and flattened. The purple fertile flowers are
surrounded by sterile flowers with somewhat concave sepals of cream. I
have raised this species from seed collected under cultivation and have seen
little variation amongst the progeny, except for a nicely golden variegated
seedling, which we have named Binti Jua (daughter of sunshine). It is a
beautiful addition to the garden but requires proper pruning of the young
plant to produce its fullest potential.
On moderate to high elevations in Taiwan, the Wynn-Joneses and I have
37 Davidsonia 14:2
collected seed from H. kawakamii Hayata (like H. villosa, reduced to synonomy
by McClintock under the subsp. aspera), which impressed us in both foliage
and flowers. One particular specimen at 2395m had felted foliage to 25cm
(10inches) long and 20cm (8 inches) wide while the cymes to 40cm (15 inches)
across were held atop sturdy stems to 6m (20 feet) in height. It does not
come as surprise to find that Haworth-Booth described this subspecies as
the most spectacularly fine member of this subsection. However, it has
suffered from untimely cold spells in USDA Zone 8 and overhead protection
should be considered in colder microclimates.
It was not until the autumn of 2000, while north of Boaxing in Sichuan
Province at elevations of 2010m, that I finally encountered a misunderstood
taxon known as H. aspera subsp. robusta (Hook. f. & Thoms.) McClintock (H.
longipes Franch). While hiking through a damp wood in a mostly degraded
agricultural area, I found three specimens of a startling hydrangea with
colossal foliage carried on 30cm (12-inch) petioles. The cordate leaf blade
was 30cm (12 inches) long and 32cm (13 inches) wide. In the autumn of
2002 in eastern Nepal, the Wynn-Joneses, J. Kincaid and I found this same
taxon, though at elevations of 2655m. Bleddyn Wynn-Jones reports this
plant to be common at similar elevations in Sikkim. There is some confusion
caused by Bean (1973), who recognized the name H. longipes but distinguished
it from the closely related H. robusta, which he admits never to have seen in
leaf.
Subsection 3 Calyptranthe
This subsection, containing some of the deciduous climbing hydrangeas,
appears taxonomically straightforward, although I have found several pitfalls.
The members are well known, if perhaps not well named, by the horticultural
community at large and are amenable to deep shade in USDA Zones 5-10,
where they will climb and use aerial roots to adhere to any compatible
substrate. The subsection is segregated from other hydrangeas by the unique
abscission of the corona (flower petals ) in one bonnet-like structure.
I have observed Hydrangea anomala D. Don subsp. anomala and H. anomala
subsp. petiolaris (Siebold & Zuccarini) McClintock (back cover) in numerous
localities, collected seed and grown the resulting seedlings, and I feel quite
satisfied with McClintocks reduction of H. petiolaris to subspecies rank. The
differences that distinguish the two subspecies (fewer stamens, a slightly more
38
domed inflorescence and more coarsely toothed leaves in H. anomala subsp.
anomala) do not seem sufficient to justify separating them as distinct species
especially when one considers the breadth of their geographical range.
H. anomala subsp. petiolaris is the more northerly of the two and is presumed
to be the more hardy. Most climbing hydrangeas grown in North America
and Europe are given this name, whether it has been legitimately applied or
not. This taxon occurs from Sakhalin in N.E Russia through the Japanese
archipelago extending southward to Taiwan. It is commonly encountered
on Ulleong Island (Korea), Cheiju Island (Korea), the central highlands of
Hokkaido, throughout Honshu, Kyushu, Shikoku and Yakushima as well as
the North-Central Alps of Taiwan. In 1993, I found it on Ulleong Island,
off the eastern coast of Korea, co-mingling with its near look-alike and close
relative Schizophragma hydrangeoides, whereas in Taiwan it sprawled over stumps
and up trees amidst the stems of S. integrifolia. On more than one occasion,
I have admired the shine of this vines golden autumn foliage as it emerges
through the fog climbing high into the overstory on virtually every available
tree trunk.
A well grown specimen of this hydrangea in blossom in June is hard to
improve upon, with its shag of lacey white heads held amidst glossy, deep
green foliage. To my knowledge, no selections have been made based on
floral attributes, but I have seen a small non-blossoming specimen in a garden
in the U.K. that is reportedly a cutting from a pink flowered selection from
France. The amount of red pigment in the leaf blade leads me to believe
that eventually this will indeed produce pinkish flowers, and recently it did.
This is not the case in regard to variegated foliage, because after years in
cultivation there has been no departure from the norm in two nearly identical
golden variegated sports, which occurred spontaneously at virtually the same
time on both coasts of North America. Hydrangea anomala subsp. petiolaris
Mirranda and Firefly each possess orbicular foliage nearly identically
emargined with yellow. Though their entrance into commerce caused
excitement, we have thus far been disappointed by their lack of vigor in the
garden.
In 1922, Monsieur Henri Cayeux exhibited in Paris a cross he had created
between H. anomala subsp. petiolaris and H. macrophylla var. rosea Hort, which
showed intermediate traits between the two. All plants of Hydrangea
hortentiolaris were reportedly destroyed during the bombardment of Le Havre
39 Davidsonia 14:2
during the end of WWII.
Cheiju-do (Quelpart Island) is a large volcanic outcrop approximately 50
nautical miles from the southern tip of the Korean Peninsula. It boasts a
large inventory of endemic species in addition to sharing a number of species
with the Japanese Archipelago that lies directly to the south. Geologically,
this location seems a somewhat bizarre mongrel between tropical Fiji and
the Salisbury Plain of England, with tidy rock walls surrounding mossy green
pastures and numerous volcanic cones that rise from the flattened landscape
like druid mounds. Among the endemics that grow here, Hydrangea
quelpartensis, a deciduous climbing species, is frequently found clambering up
the oaks and pines in the dry woodlands of the volcanic slopes. I do not
believe it to have any botanical standing despite the fact that I have collected
seed under this name while on Cheiju-do and purchased plants under this
name while in Europe. The claim that it possesses foliage one quarter the
size of H. anomala subsp. petiolaris can be attributed entirely, it seems, to
juvenility. I find it important to raise this rather esoteric fact because of my
culpability in distributing a deciduous self-clinging vine for several years under
the name of Schizophrama hydrangeoides Brookside Littleleaf . I received it
under this name from J.C. Raulston of North Carolina, who in turn had
received it from Brookside Gardens in Maryland. After numerous years of
climbing a tree in the garden, by which time it proved itself quite capable of
producing entirely normal sized leaves, it blossomed, authenticating itself as
H. anomala.
Hydrangea anomala subsp. anomala appears virtually identical to H. anomala
subsp. petiolaris, at least superficially and to my eyes. It too is quite commonly
encountered in the mountains of western China and the eastern Himalaya. I
have collected seed of this species above 2300m in central and eastern Nepal,
at 2860m in Sichuan and at 2700m on the Cangshan in Yunnan. The latter
collection, made in 2000 under the number DJHC 00-0455, came from an
individual that I noted as having very lustrous foliage and cymes to 25cm (10
inches) across. The resultant seedlings have not yet flowered. Finally now
afforded numerous collections of this continental counterpart, with known
provenance, a comparative evaluation of the differences exhibited in
cultivation is underway, though it will be many years before data is
forthcoming.
40
Subsection 4 Petalanthe
This subsection of wholly Asian species is characterized by the position
of the ovary (half superior) as well as the fact that the seeds are not winged
or nearly so. Ostensibly, it is an assemblage of vastly under-known but rather
remarkable species. I feel that it is taxonomically complex, riddled with
nomenclatural anomalies and represented in the West mostly by dried
herbarium specimens.
The Kei Peninsula is a large landmass that juts into the Pacific Ocean,
with Osaka on its upper northwest corner and Nagoya to the upper northeast.
We centered our activities from the centrally positioned city of Hashimoto
and explored the large mountain range rising to the southeast.
It was here on our first day that we encountered Hydrangea hirta (Thunb.)
Siebold a species virtually unknown in western cultivation. The rounded
leaves of this species, to 8cm (3 inches), are very distinct in possessing an
extremely dentate leaf margin. It grew in great abundance here in moderate
shade beneath a canopy of deciduous and evergreen trees at approximately
1000m elevation. In May, profuse terminal flowers of light blue are formed
in compact corymbs to 8cm (3inches) across atop 1.25m (4-foot) stems,
lacking the expanded sepals on sterile florets that we have come to associate
with the hydrangeas as a whole. Though I already possess plants of this
species, raised from seed provided by the Tsukuba Botanic Garden in Japan,
this was my first encounter with it in the wild and I found it a handsome
shrub that certainly deserves more evaluation for use in Northwest gardens.
It is not held is very high esteem in most literature.
In this area, we found what I have come finally to regard as Hydrangea
scandens Ser., another rare species that certainly deserves more recognition
than it currently enjoys in the West. It has a wide range throughout east Asia
and the nomenclature compressed into this variability is in a constant state
of fusion and fission. We had already collected seed of this species on Shikoku
Island, where we found it common as an understory shrub in a wide range
of elevations. The foliage is quite narrow, to less than one inch, with serrations
present on the leaf margin along the terminal quarter of its total length of
10cm (4 inches). In full sun situations, the foliage had transformed to lovely
tints of burgundy in good complement to its characteristic brownish purple
stems. This species has already flowered in my woodland at Heronswood,
48
where I raised it from seed nearly 15 years ago. It performs quite admirably
in the climate of the Puget Sound area where it often flowers before June
producing quantities of fragrant, 10cm (4-inch) corymbs surrounded by
creamy white, fading to yellow, sepals of the sterile florets. I cultivate both
a variegated as well a double-sepaled forms of this taxon originating from
Japan.
Hydrangea scandens is known from numerous variants in herbaria and
reported in the literature, most of which have originated on the mainland of
China. One of those, referred by McClintock to H. scandens subsp. chinensis
(Maxim.), is H. angustipetala Hayata (H. yayeyamensis Koidzumi). It is this
plant that I have observed in blossom in very early spring both on extreme
southern Kyushu as well as Yakushima further to the south. It is easily
recognized, with lance-like, extremely dentate leaves to 10cm (4 inches) in
length, appearing more like the foliage of a forsythia than hydrangea. The
airy inflorescence is of a respectable size with loose corymbs, to 12cm (5
inches) across, of white flowers surrounded by white, fading to pink, ray
flowers.
While in Sichuan Province in the autumn of 2000, on the border with
Yunnan, I observed a hydrangea in full blossom in mid-October at lower
elevations, possessing narrow, coriaceous foliage and large flattened cymes
of white flowers surrounded by white florets. These were collected under
the numbers 00-0492 and 00-0499 and assigned temporarily to H. scandens
subsp. chinensis (Figure 5). In the autumn of 1999, while on Fan Si Pan, the
tallest mountain in Viet Nam at elevations of over 3000m, we collected seed
from an evergreen species, common on the lower slopes. The lance shaped
leaves to 12cm (5 inches) were glabrous above and a striking purple beneath,
while the cymes of white fertile flowers and lavender sepals were quite
handsome. This was collected under HWJ 99736 and assigned the provisional
name of H. scandens subsp. indochinensis (H. indochinensis Merrill). Superficially,
it resembles H. lobbii (Maxim.) - another entity attributed by McClintock to
H. scandens subsp. chinensis - which we observed at lower elevations in Taiwan
a month later. I have grown this evergreen shrub in our USDA Zone 8
woodland and it thus far seems to be quite content.
Subsection 5 Heteromallae
Whereas Subsection 4 Petalanthe contains mostly unknown hydrangea
49 Davidsonia 14:2
species, Heteromallae, characterized by terminally winged seeds and globose
seed capsules, includes one of the best known and cherished of all flowering
shrubs. Hydrangea paniculata Sieb., known to the lay audience as the peegee
hydrangea, is one of the hardiest and most handsome of hydrangeas grown
in North America and Europe. It has a large geographical range, from
Sakhalin and NE China to Taiwan through the entire Japanese Archipelago.
The flowers, as one would expect, are carried in panicles with sterile florets
spread throughout the inflorescence. These are born terminally on each
branch and are produced on current years wood.
In the autumn of 2001, while on the chilly mountain slopes of Mt. Daisen
in Japan, I collected seed from specimens rising to nearly 7.5m (25 feet) in
height. In 1999, those plants we observed in Taiwan, were more demure
with an intriguing bluish green colored foliage. Sargents collection from
Hokkaido in 1893 led to development of Praecox, which blossoms nearly
six weeks earlier than the much celebrated H. paniculata Grandiflora, the
above mentioned peegee, emitting a fragrance that has been aptly described
as a heady infusion of horse and Chanel N
o
5. The peegee was imported
from Japan in the latter years of the 19
th
century and, with its startling heads
of nearly all sterile florets and tough demeanor, was almost the sole
representative of the species in cultivation for practically 100 years. Selected
forms such as Floribunda, Interhydia (Pink Diamond), Greenspire,
Brussels Lace, and Unique have grabbed market share since. I have not
grown the highly touted H. paniculata Tardiva which is reported to blossom
much later than the others, often not commencing until mid-September.
Hydrangea heteromalla D. Don has become the overall name for a rather
large assemblage of variable ecotypes occurring from the eastern Himalaya
through the interior to eastern China. It grows to immense proportions in
central and eastern Nepal where I collected it in 1995 and 2002 under
numerous numbers at elevations of 2000 to 3500m. The foliage of the
Himalayan forms are ovate linear to nearly 30cm (12 inches) in length and
glabrous above with adpressed hairs beneath. The inflorescence is a terminal
cyme of white fertile flowers and white or pink ray florets. Interestingly, I
have not observed any red pigment in the petioles of the Nepalese plants.
From my observations, this trait enters the picture in the Sikkim populations
and intensifies as one travels further east into western China, where I have
collected numerous forms possessing dark green, nearly orbicular foliage
50
and striking, lipstick-red leaf stems. Hydrangea heteromalla Bretschneiderii
(H. bretschneideri Dipp) represents collections made in the mountains near
Beijing and possesses a framework with handsome exfoliating cinnamon-
colored bark, while H. heteromalla Snowcap, grown under the name H. robusta
for many years in England, is touted for its large cordate foliage and
substantive flattened white cymes. The latter is rare in commerce in North
America.
Subsection 6 Macrophyllae
The confusion surrounding the naming of H. macrophylla and H. serrata is
due mostly to the rules designed by the International Code of Botanical
Nomenclature to prevent such confusion. Through the untangling of a legion
of sticky webs, a rather interesting story unfolds while possessing all of the
parameters necessary for a perfect storm.
As already noted, those hydrangeas first described from Japan, through
Thunberg, were named Viburnum macrophyllum and Viburnum serratum. The
former, a selection later named Otaksa possessed an inflorescence comprised
virtually of all sterile florets, and is thought to have been imported into
Japan from China where it was already a popular garden plant in the courts
of Imperial China. The latter probably came to Thunbergs attention by way
of fodder brought for animals on Deshima Island, an artificial Archipelago
in Nagasaki harbor where all visitors to Japan during the 18
th
century were
forced to live.
1
1
It was on Deshima that Philipp Franz von Siebold fell in love with Otaksa-san, his
Japanese lover whom he was forbade by the Japanese government from marrying.
Von Siebold took Otaksa and the daughter they had produced with him when he
left Japan in 1829, however, a shipwreck led to the finding of a forbidden map of
Japan that von Siebold had acquired. He was arrested and later banished; Otaksa
and their daughter were not allowed to leave with him.
2
There remains a great deal of speculation as to exactly whom the name Hortensia
commemorates. Commersons mistress, Jeanne Baret, who accompanied him dressed
as a man on Bougainvilles round-world voyage, is ruled out. Queen Hortense was
born several years after Commersons death. Mme Lepaute, the wife of a clockmaker
and herself a respected mathematician is often cited, however, her Christian name
was Nicole-Reine. Most likely is the daughter of the Prince of Nassau-Siegen whose
father had accompanied Commerson with Bougainville.
51 Davidsonia 14:2
About the same time, the French botanist Lamarck published the name of
Hortensia opuloides, the name given to a plant that the naturalist Commerson
had sent to Paris from the island of Mauritius, where it grew in the garden
of explorer Pierre Poivre.
2
This invalid genus name is still used to some
degree in French commerce to represent those cultivars possessing mostly
infertile florets.
In 1788, Sir Joseph Banks presented to Kew a sterile-floreted form he had
procured from China. This was properly placed in the genus Hydrangea and
described in 1792 as H. hortensis. This clone is still common along the mild
southern coasts of England where it is now grown as H. macrophylla Sir
Joseph Banks
Though it is now apparent that by trying to sort out and rename these two
hydrangeas as we know them, adhering to the rules of priority, would only
cause more damage than good, it is understandable and meritorious that
many have tried. Ernest Wilson, Michael Haworth-Booth and Elizabeth
McClintock all weighed in heavily to make things right. Though there is still
plenty to be learned, what we currently have is a better understanding of two
species.
Hydrangea macrophylla, with its typical large, glossy, somewhat fleshy leaves,
is common throughout coastal Japan at relatively low elevations, frequently
found hugging windswept beaches. I have observed large populations of
this growing on the Chiba Peninsula, south of Tokyo, in 1995 as well at
1997. Of course, the normal form in the wild possesses both fertile and ray
florets. Hydrangea macrophylla Seafoam, a sport from Sir Joseph Banks,
with rather large cymes of blue fertile flowers and white sterile florets, is
probably the closest thing to true H. macrophylla in cultivation. There have
been subsequent introductions of this from the wild, including Wilsons 1917
collection from Oshima, however, they have made little if any impact in
Western horticulture. Hydrangea macrophylla blossoms on second year wood,
making it thusly inappropriate for use in any climate where winter damage
will occur. For those blessed with warmer gardening climates, there exists a
voluminous listing of cultivars of H. macrophylla to choose from.
Thunbergs Viburnum serratum did indeed represent a different species
hailing from the higher elevations of Japan and Korea. Hydrangea serrata
(Thunb.) Ser. (H. macrophylla subsp. serrata (Thunb.) Mak.) is thusly a much
52
hardier species and possesses the ability to blossom on new wood. I have
encountered and collected seed of this on numerous adventures on Hokkaido,
central and northern Honshu, Cheiju Island and most notably while on Mt.
Chiri on the southern tip of the Korean Peninsula, in 1993 and 1997, at
1500m. Here the hillsides are smothered in this species, at a much greater
degree than I have ever witnessed while in Japan, and must provide a dazzling
sight when in blossom. I have come to expect the typical wild form of this,
from both Japan and Korea, to offer blue fertile flowers in a slightly dome-
shaped corymb, encircled with ray-florets of blue, if provided acidic soils, or
suffused with pink when grown under more basic conditions.
A collection from our time on Chiri-san in 1997, under the number of HC
970416, was a self-layered lower branch from a specimen possessing sterile
florets with double-sepals. Heronswood entered this into the trade under
the cultivar name of Chiri-san Sue, in honor of Sue Wynn-Jones, who
accompanied us on that trip and who had waited a much too inappropriate
length of time for us to return to our pre-ordained meeting spot on that
chilly, rainy day.
It was, however, in Japan that I first observed a vast array of hydrangea
cultivars based upon H. serrata. Unlike the European hybridization of
hydrangeas, which focussed primarily on increasing the flower size of H.
macrophylla, the Japanese breeding programs instead selected for delicate
subtlety, which translates to pure charm in the garden. Unfortunately,
hydrangeas from Japan are forbidden from entry in the United States, though
Europe does not ban their importation. Furthermore, as Hydrangeas can
come into the U.S. through Europe under post entry quarantine, these exciting
new plants will ultimately be made available through the trade, though not
necessarily at the pace I would prefer.
Consider H. serrata Izu No Hana (front cover), one of dozens of classical
cultivars from the Japanese tradition, each with names that are nearly as
enticing as the floral effects themselves. With this cultivar, the flattened disk
of indigo blue flowers are surrounded by sterile florets of similar hue, much
in the same fashion of other lacecap hydrangeas. In this cultivar, however,
each sterile floret possesses a double dose of sepals, creating miniature blue
roses that arc from the flower heads on wiry stems, suggesting that of
intricate fireworks at precisely the moment of ignition.
53 Davidsonia 14:2
This charming flower type is seen in numerous named Japanese cultivars,
including the beguiling pink H. serrata Jogasaki, as well as the large headed,
pure white Miyake-Tokiwa. A single stem of the latter, with miniature white
tassels opening from chartreuse buds, could by itself be carried as a tastefully
exquisite bridal bouquet.
Flower heads possessing sepals richly edged in contrasting hues comprise
another substantial subset of the Japanese mountain hydrangea. The
blossoms of H. serrata Beni Gaku open with sepals colored brilliant white
bordered in rich cerise. As the floral effect matures, the definition between
these two colors dim, while melding to enchanting shades of rose pink.
In foliage as well as flower, H. serrata offers numerous irresistible selections.
The leaves of H. serrata Kiyosumi (Figure 6) emerge in tones of rich
burgundy, retaining these colors throughout spring and early summer. In
near perfect combination, its flowers open from deep crimson buds, later
broadening to heads of startling white picoteed with rose. Hydrangea serrata
Beni Nishiki brandishes intensely colored blossoms of cherry red held amidst
a flurry of sprightly cream-splashed leaves.
Section II Cornidia
The Section Cornidia is comprised entirely of evergreen species, most of
which are scandent shrubs or lianas and naturally occur from the Neo-Tropics
south to southern Chile. The Section is delineated by the presence of cup-
shaped floral bracts that enclose the inflorescence and dehisce upon opening,
superficially not entirely unlike those found in H. involucrata. The inflorescence
is comprised entirely of fertile florets; i.e., the sterile ray flowers are absent.
These are chiefly unknown in cultivation though at least three species have
recently become better known to a realm of USDA Zone 8-12 gardeners
beyond the cognescenti.
Subsection 1 Monosegia
A simple corymb sets this group aside from the only other subsection in
Section Cornidia. I have little experience with or knowledge of three of the
seven species currently enveloped by Monosegia, many of which, to my
knowledge, remain unmolested by the gardeners hand in their native habitats.
Not so, thankfully, with Hydrangea integrifolia Hayata (Figure 7), which I
54
consider one of the finest evergreen vines that I grow. I happened upon a
scrambling terrestrial specimen of this species in the Edmonds, Washington
garden of Marlee Hedges in the late 1980s and the oldest specimen in my
garden originates from the cuttings she offered me at that time. Much later,
in 1999, I collected seed of this species growing in its native range of Taiwan
at elevations of 2395m where it was climbing specimens of Trochodendron
aralioides to 12 m (40 feet) in height. It is the only member of the section
Cornidia to occur in the Old World and naturally occurs in both Taiwan and
the Phillipines.
It is a very handsome species with deep green linear obovate foliage, very
leathery in texture, to 20cm (8 inches) in length, adhering to the stems in
pairs by way of deeply hued red petioles. The leaf margin is slightly crenulate.
As with H. anomala, the stems adher to its host substrate by way of
adventitious roots. In late June, terminal cymes of white flowers open from
buds that are at once both curious in appearance while conjuring forth an
overinflated anticipation. As the attending bracts fall and the flowers expand,
for a short period it appears as if large popcorn balls have been stashed
among the greenery, though fully expanded, the effect is much as with the
better known deciduous species without the outer row of sterile florets.
My original garden plant of H. integrifolia has achieved over 9m (30 feet) in
height upon the lower trunk of a Douglas fir, though has thus far remained
quite stingy in blossom. An additional plant representing seed collections
from Taiwan has proven to be much more generous in this regard despite its
relative youth.
Hydrangea seemanii Riley (Figure 9) has been anything but reticent to blossom
in our woodland, now smothered with flattened cymes of creamy white
flowers without ray florets. It is very closely allied to H. integrifolia despite
the enormous range differential. Hydrangea seemanii is found naturally occuring
in the mountains of SW Mexico in the Sierra Madre Occidental at approximate
elevations of 2000 to 2600m; in fact it is the only member of the genus to
occur within the political boundaries of this country. The leaves on the
clone that we grow are broadly ovate to 15cm (6 inches) in length and 60cm
(2 feet) wide and extremely glossy on the upper surface. It is vigorous in
growth and has climbed over 10.5m (35 feet) in our woodland in less than 10
years.
55 Davidsonia 14:2
Hydrangea asterolasia Diels is the only other American species in this
Subsection to possess flowers of white rather than pink. It is found naturally
occurring from Costa Rica south to the Andes of Ecuador and Columbia.
The ovate foliage of the clone that we currently cultivate, from wild collected
Ecuadoran stock, is ovate and coriaceous to 10cm (4 inches) in length,
decidedly smaller than H. seemanii and H. integrifolia, while both the new stems
and leaf petioles are coated in a characteristic brownish tomentum. Though
I have traveled in the mountains of Costa Rica where this is reported to
occur, my only experience horticulturally is with the young specimen we have
integrated into our woodland garden.
While in Costa Rica, however, I delighted in encountering my first pink-
flowered, evergreen climbing Hydrangea species (Figure 8), both in the high
treetops of Monteverde as well as in high elevations of the Cordillera De
Talamanca south of San Jose. Though I presumed these to be Hydrangea
peruviana Moric., subsequent reading suggested one of two very closely related
species that occur in the same geographical range. Hydrangea oerstedii Briq.and
H. peruviana both possess very showy cymes of pink fertile flowers surrounded
by large sterile florets. The difference between the two is based entirely
upon the length of the stamen and one that may not stand the test of time as
more material becomes available for study. As those I observed were in full
flower, in February, I was unable to gather seed while my attempt at cuttings
also failed.
McClintocks monograph retains three additional species in this subsection;
H. preslii Broiq, H. diplostemona Standley and H. steyermarkii Standley. All
three reportedly possess pink flowers and naturally occur from Guatemala
south to the Andes of Ecuador, Columbia and Peru.
Subsection 2 Polysegia
It is the compound cyme (i.e., one above another), that distinguishes this
subsection from Subsection Monosegia, and, of the four currently adherred
here, my interaction with its constituency is restricted to only one; Hydrangea
serratifolia Engl.
Hydrangea serratifolia has purchased a rather schizophrenic episode in
taxonomy, though as always the case, this malady originates not from the
plant itself but from the progenitor of its name. The young Hooker, having
56
come upon this liana in Chile in 1833, put this to a new genus, Cornidia, while
tagging on an apt specific epithet, integerrima, which translates to smooth or
entire margined. Indeed, the leathery, ovate semi-gloss green leaves of this
species, to 10cm (4 inches) in length, do not possess the degree of jag found
on others in this section. Never mind the fact that the name H. integerrima
might be confused with H. integrifolia; this is a conflict that mature gardeners
can cope with.
Exactly why, however, this species consummated its classification with
exactly the opposite and erroneous reference will never be fully actualized.
Nonetheless, this smooth-margin-leafed hydrangea conjures forth an image
of teethly dentation that does not actually exist.
In the southern Andes in 1998, my friends Kevin Carraibine, Jennifer
Macuiba and I encountered this species growing upwards to 18m (60 feet)
along the main trunk of a Nothofagus, while later observing it to 30m (100
feet) in mammoth specimens of Eucryphia cordifolia. We were able to collect
its seed from the large heads of flowers only in instances when its host tree
had retired to a decidedly less intimidating horizontal stance on the floor of
the Alercean rainforest.
The species has flourished in our woodland at Heronswood, ascending to
12m (40 feet) in less than a decade along the main trunk of a second growth
Douglas fir, though it has yet to grace the garden with blossom. I have no
doubt that it will do this in time.
This subsection currently supports the existence of three additional species,
and although I have not observed the herbarium specimens that meagerly
represent their being in modern science, based on the collection notes and
physical condition of the holotypes and McClintocks reservations of their
preservation, I would not put good money on their timeless qualities.
Hydrangea tarapotensis Briq. is noted from the Andes of Colombia, Bolivia
and Peru from 800 to 1500m. Hydrangea felskii Szyszyl. is found in the Andes
of southern Ecuador and northern Peru, 2100m while H. mathewsii Briq. was
first collected in the Andes of northern Peru.
This paper does not deal with the closely allied, and ornamentally fertile
genera of Schizophragma, Pileostegia, Decumaria and Dichroa, all found within
the realm of Hydrangeaceae and with variable representation in Western
horticulture.
57 Davidsonia 14:2
References and Further Reading
Bean, W.J. 1973. Trees and Shrubs Hardy in the British Isles. Volumne II D-M.
Fully revised eighth edition. In collaboration with The Royal Horticultural
Society. Butler and Tanner Ltd., Frome and London.
Bondurant, C.S. 1887. Botanical Medicine Monographs and Sundry. American
Journal of Pharmacy. 59(3).
Department of Plant Resources, Ministry of Forests and Soil Conservation. 1997.
The Flora of Phulchoki and Godawari. Thapathali, Kathmandu.
Dirr, Michael A. 1983. Manual of Woody Landscape Plants; Their Identification,
Ornamental Characteristics, Culture, Propagation and Uses. Stipes
Publishing Company. Champaign, Illinois.
Gentry, Alwyn H. 1993. A Field Guide to the Families and Genera of Woody
Plants of Northwest South America. University of Chicago Press. Chicago
and London.
Hara, Chater and Williams. 1982. An Enumeration of the Flowering Plants of
Nepal Volume 3. Trustees of British Museum of Natural History. London,
UK.
Haworth-Booth, Michael. 1984. The Hydrangea. 5th revised edition. Constable
and Company Ltd. London, UK.
Hillier Nurseries. 1992. The Hillier Manual of Trees and Shrubs. Redwood Press
Ltd, Melksham, Wiltshire, UK.
L. H Bailey Hortorium, Cornell University. 1976. Hortus Third. Macmillan
Publishing Co., Inc. New York, N.Y.
Lancaster, Roy. 1989. Travels in China. Antique Collectors Club Ltd.Woodbridge,
Suffolk, UK.
Lancaster, Roy. 1995. A Plantsmen in Nepal. Antique Collectors Club Ltd.
Woodbridge, Suffolk, UK.
Lawson-Hall, Toni and Brian Rothera. 1995. Hydrangeas, A Gardeners Guide.
B.T. Batsford Ltd. London, UK.
Li, H. L. 1963. The Woody Flora of Taiwan, The Morris Arboretum and Division
of Biology of the University of Pennsylvania. Livingston Publishing
Company. Narberth, Penn.
McClintock, Elizabeth. 1957. A Monograph of the Genus Hydrangea.
Proceedings of the California Academy of Sciences, Fourth Series.
29(5):147-256.
Mallet, Corrine. 1994. Hydrangeas - Species & Cultivars Volume 2. Centre dArt
Floral. Varengeville s/mer, France.
58
Mallet, Corinne, Robert Mallet and Harry van Trier. 1992. Hydrangeas - Species
& Cultivars. Centre dArt Floral. Varengeville s/mer, France.
Ohwi, Jisaburo. 1984. The Flora of Japan. Edited by Frederick G. Meyer and
Egbert H. Walker. Smithsonian Institution. Washington, DC.
Polunin and Stainton. 1984. The Flowers of the Himalaya. Oxford University
Press. Oxford, UK.
Radford, Ahles and Bell. 1968. Manual of the Vascular Flora of the Carolinas.
The University of North Carolina Press
41 Davidsonia 14:2
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 1. Hydrangea arborescens subsp. radiata.
42
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 2. Hydrangea sikokiana.
43 Davidsonia 14:2
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 3. Hydrangea involucrata.
Figure 4. Hydrangea aspera subsp. strigosa Elegant Sound Pavilion.
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
44
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 5. Hydrangea scandens subsp. chinensis.
45 Davidsonia 14:2
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 6. Hydrangea serrata Kiyosumi.
46
P
h
o
t
o
:

D
a
n
i
e
l

M
o
s
q
u
i
n
Figure 7. Hydrangea integrifolia.
Figure 8. Hydrangea sp. (with an affinity to H. peruviana).
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
47 Davidsonia 14:2
P
h
o
t
o
:

D
a
n

H
i
n
k
l
e
y
Figure 9. Hydrangea seemannii.