W. R. Philipson and M. N. Philipson
Botany Department, University of Canterbury and Botany Division
D.S.T.R., Christchurch, New Zealand
That Rhododendron is centered on the mainland of southern and eastern Asia is well known, and this is true not only of the number of species there, but also of the variety of different divisions of the genus. Of the eight subgenera only the three very small ones - Candidastrum, Therorhodion and Mumeazalea, totaling only four species between them - do not occur there. But it may not be generally realized how comparatively few species of Rhododendron grow in other parts of Asia and on the other continents. In view of this it is of interest to look at the rhododendrons which have escaped from this homeland, which extends from the uplands of the Himalayas and Tibet to the tropical forests of Malaysia and the seaboard of China, and to see how far they have travelled. We shall consider them under the various subgenera.
Azaleastrum. This subgenus, with its two Sections centered round R. ovatum and R. moulmainense respectively, need not detain us, since all the species are confined to southeast Asia (including Taiwan and other coastal islands off China).
Tsutsuji. The evergreen azaleas may be dismissed almost as quickly. Apart from one species which reaches the northernmost island of the Philippines, the only examples to stray outside south-east Asia are confined to Japan and Korea where the group is particularly well represented.
Hymenanthes. This great assemblage, which includes the majority of the horticulturally important species, also presents a surprisingly simple picture. Again excepting the coastal islands off China and also Ceylon, there are two species which have crossed the Strait of Malacca into northern Sumatra. Apart from these two, only one of the twenty-four subsections strays beyond the core area of southern and eastern Asia, namely the subsection Pontica. This subsection maintains a very slight toehold on the core area, there being a single species in Taiwan. Otherwise the species range in a very disjunct fashion over the north temperate part of the globe. The other ten species extend from the Atlantic seaboard of Europe to the Middle East and the Caucasus, through central and eastern Siberia to Korea and Japan, and then across the Pacific to western and also eastern United States and Canada. This remarkable distribution is to be accounted for by the southward extension of cold climates during the ice age. These must have pressed the vegetation into refugia from which centers they have subsequently spread.
Rhododendron. The lepidote rhododendrons present a much more complex picture. By far the biggest outburst took place to the south into the tropical islands of Indonesia, the Philippines and New Guinea. Practically the whole of this eruption consists of the single assemblage of the vireyas, for the only other rhododendrons to have jumped off the mainland of Asia southwards (always excluding Ceylon, Taiwan and other coastal islands off China) are two species (already mentioned under Hymenanthes) in northern Sumatra, and one evergreen azalea in the north of Luzon. The vireyas form a prolific and very distinctive offshoot of the lepidote rhododendrons, and though a small number occur on the Asiatic mainland they are extremely abundant as forest epiphytes and as subalpine scrub throughout the Malay Archipelago, even reaching Queensland and the Solomon Islands.
Comparatively few lepidote rhododendrons have escaped in other directions. To the west, two species occur in Afghanistan, but these can be regarded as outliers of the Himalayas which, of course, is an important sector of the homeland of rhododendrons. Otherwise there is only the European alpine triumvirate R. ferrugineum, R. hirsutum and R. kotschyi. It is understandable, for historical reasons, that the alpine rose is one of the species upon which Linnaeus formulated his concept of the genus, and fortunately it displays no extreme deviation, for, as we shall see later, many of the peripheral species exhibit peculiar characteristics.
Rather more lepidote species occur to the north of the homeland. One species of the Triflora sub-section reaches Japan (R. keiskei) and the very distinctive R. micranthum reaches Korea. More significant distributions occur in subsections Lapponica and Rhodorastra and in Section Pogonanthum.
All but two of the large Lapponica sub-section occur south of Mongolia. Of the two northern species R. burjaticum has a very limited range near the Siberian Lake Baikal, whereas the other, R. lapponicum ranges almost around the arctic, spreading widely in Siberia and in Alaska, Canada, Greenland and Scandinavia, with outliers in Japan and in central and eastern United States. This remarkable range must be the greatest in the whole genus. The Lapponica subsection originated among the alpine habitats of the Sino-Himalayan massif and once the group reached the similar ecological conditions of the arctic, it found conditions which allowed it to spread far and wide. We shall find similar ecological explanations for the geographical ranges of other groups of rhododendrons.
R. dauricum and R. mucronulatum (Subsection Rhodorastra) have rather similar distributions, stretching from eastern Siberia to Japan. This group was regarded by Dr. Sleumer as being sufficiently distinctive to be a separate subgenus, although Dr. Cullen now believes that they may be placed among the main body of the lepidote species. Section Pogonanthum also forms a distinctive group of species and these R. fragrans has a wide distribution in Siberia.
There remains the problem of the only lepidote species in the United States (apart from the arctic R. lapponicum), namely the variable R. minus from the southeastern States. Dr. Cullen considers its nearest relative to be sub-section Heliolepida of China. At first glance this very wide separation appears difficult to accept, but on reflection we recall that several genera (e.g. Magnolia, Liriodendron, Catalpa) occur in the eastern United States and also in south-east Asia, and that parts of these regions have very similar ecological conditions.
Pentanthera. The situation is quite different in the deciduous azaleas. To begin with, they comprise several rather diverse Sections, and secondly none of these have their centre in south-east Asia - indeed only one occurs there. The great majority of the species of Section Pentanthera occurs in North America - mainly to the east but also on the Pacific coast - with one each in China, Japan and eastern Europe. Section Rhodora is exclusively eastern North American, and Sections Sciadorhodion and Viscidula are virtually confined to Japan (R. schlippenbachii extends to adjacent mainland regions of Asia). From this, I think, we must conclude that the deciduous azaleas are made up of remnants of a rather ancient stock which has survived mainly in the zones of deciduous woodland - an ecological niche avoided by other rhododendrons.
Similar considerations apply to the three small remaining Subgenera. Mumeazalea (with the single representative R. semibarbatum) is deciduous and confined to Japan. Subgenus Candidastrum also comprises a single deciduous species, R. albiflorum from northwestern North America. Finally, Subgenus Therorhodion, with two deciduous species (R. camtschaticum and R. redowskianum) spans the cold regions to either side of the Bering Strait. It is probably no coincidence that these geographical outliers are also the most aberrant species within the whole genus. R. camtschaticum and its ally have sometimes been considered distinct enough to form a separate genus. Their inflorescences appear peculiar because their bracts are green and leaf-like, but in reality, as in all rhododendrons, the trusses arise from buds distinct from those that produce the leafy shoots. The flowers of R. semibarbatum have stamens of two sorts, a feature not found in any other rhododendrons, and few people would regard the flowers and foliage of R. albiflorum as at all reminiscent of a rhododendron, although all the essential features of Rhododendron are present.
An overall view of the genus shows it to consist of shrubs or trees mostly suited to growth in evergreen rain forest, in deciduous woodlands and in subalpine regions; The main divisions of the genus show evident correlation with these ecological tendencies, and all of these types occur within what we have called the core homeland of the genus. Groups which have spread beyond this homeland into temperate climates often favor subalpine vegetation or are found in more or less deciduous woodland. The tropical vireyas to the south have exploited both rain forest and subalpine scrub and are so successful that they far outnumber other lepidote species and are as prolific in species as the non-lepidote rhododendrons. Perhaps it is because the species of Hymenanthes and Azaleastrum appear to be more tied to warm moist habitats that they are restricted to the original homeland.