FIG.59. R. sataense at Mt. Takakuma. Note the rounded, convex leaves. Photo by Frank Doleshy

FIG.60. R. sataense at Mt. Takakuma, growing in partial shade. Photo by Frank Doleshy

Near the road, one has the feeling that an effort has been made to improve on nature by moving in selected azaleas, spacing them carefully, and removing some of the Vaccinium, Ilex and Hydrangea. This is a nice show, but a more interesting area is found by climbing up and over a bald ridge and starting down the west side. Almost immediately, the trail enters a thicket of large R. sataense. As Mr. Kunishige had told us, these wild plants are only found within a narrow slice of elevations - perhaps 625-710 m. at this place. Yet in any open area within these limits, they seem to be completely indifferent to variations in exposure, steepness, soil mixture, or any other environmental circumstance, and they appear to be exuberantly competitive and invasive. One area looked like a group of old rice fields, separated by earth walls about 2 m. high, and buried under ash. The R. sataense had colonized the sides and tops of the walls and the flat fields between, apparently without preference. Along a power line maintenance trail they practically formed hedges. And, on a fairly flat area which was being reforested, these azaleas had apparently been slashed down to give the small Cryptomeria trees a better start, but they had grown back from the stumps as dense, rounded plants, covered with flowers; these could, I think, be entered in any Rhododendron show as Kurumes without raising an eyebrow, and they might win the awards.

The color range was the now-familiar diversity of salmons, reds, pinks, roses and purples. Flowers sometimes formed a large cluster at the end of each branch, but they were more often scattered thickly on and within the foliage as on many cultivated Kurumes. Mature plants in the open were 1¼ to 2 m. tall, for the most part, but those partially within scrub woodland ranged up to 3 m.

Other Azalea Mountains

Takakuma completes the circuit of well-known azalea areas near Kagoshima, in southern Kyushu. However, in other parts of Kyushu, there are additional stands of wild plants resembling the Kurumes. Creech and Kunishige have mentioned the existence of R. kiusianum and R. kiusianum - R. kaempferi hybrids at Unzen, Mt. Haneyama, and Mt. Daisen (east peak of Mt. Kuju complex). Whether these played a part in the development of the older Kurumes is, I suppose, a matter of historical accident and accessibility. The people who today continue the development of new Kurume azaleas at the Horticultural Station are certainly able to use these additional stocks (unless they feel constrained to use only R. sataense as the source for Kurumes). Their collections of plants from these other locations are large and are very interesting because of the distinctive nature of each one.

The Unzen R. kiusianum, from elevations no greater than the 1360 m. summit, are considerably different from the high-elevation (1000 m. plus) Kirishima plants. Those from Unzen grow larger, to about 1½ m., some flowers are a watermelon red or almost salmon, and some have a distinctive pink throat. Perhaps, one may speculate, the high-elevation R. kiusianum at Unzen was destroyed by the substantial volcanic eruption of 1792 (or an earlier one), and the R. kiusianum which migrated back into the higher areas was somewhat intermixed with R. kaempferi.

Mt. Daisen is part of the highest mountain group in mainland Kyushu, and the R. kiusianum from this mountain, as cultivated at Kurume, is low growing and has flowers of a cool violet color. Dr. Creech also notes that Daisen is a place where white-flowered forms are found (but it should not be assumed that these are frequently seen, since Mr. Kunishige tells us that their field people usually manage to find only a few new white-flowered plants per season).

Mt. Haneyama differs from all the other places discussed because it is not a volcano or ash field. Instead, it is structured from the immense flows of andesite lava which make up much of northeastern Kyushu. The R. kiusianum on this mountain is notably compact in habit; intermediate, pink-flowered plants growing at the very moderate elevation of 500 m. are small and dense, and they retain this habit even when transplanted to the broad, lowland valley where Kurume is located. Mr. Kunishige states that the Haneyama R. kiusianum is the best there is, and we cannot disagree. Perhaps the reason for the consistent low-growing habit is that this colony has never been destroyed by volcanic activity in the higher part of its home mountain and thus retains the original characteristics of the species.

Differences in Geographical Distribution

We have reviewed all the stands of wild azaleas which are likely to have contributed to the development of the Kurumes. Now we can take a look at the geographical limits of the territory where each species is found, and this may turn up some information of interest to the breeder of azaleas.

First, there is R. kaempferi, found on the hills and mountains of all four main islands of Japan. Growing on the lower peninsulas of Kyushu, it is at the latitude of southern Georgia or the boundary area of western Mexico. And, growing from bottom to top of 811 m. Mt. Apoi, Hokkaido, it is at the latitude of Boston or southern Oregon. It is very common and variable; besides the usual orange-red or salmon, white forms have been found, but kaempferi like purples are fairly certain to be hybrids.

Mr. Wada has reminded me that some interesting kaempferi affiliates with variable flower color grow in the wild on the main island of Honshu, between Hamamatsu and Nagoya. These may be the source of certain cultivated azaleas resembling the Kurumes. However, with regret, I have to consider these beyond the scope of the present discussion.

Our second species is R. kiusianum; this, the authorities agree, is limited to the higher mountains of Kyushu only. The compact form with uniform, rose-purple flowers apparently is found only above 900 or 1000 m., although forms or hybrids with compact habit and variable flower color are found lower on Mt. Haneyama. We have talked with several Japanese plantsmen about the reason for this distribution pattern, and the general opinion is that R. kiusianum was at one time distributed throughout Kyushu but now survives only on the high mountains as a resu1t of climatic change. One change which suggests itself is the warming trend at the close of an ice age, but it is hard to see how this would explain the extraordinary retreat of R. kiusianum, which is cultivated with success in the warm climate of Kurume. A better explanation, I think, is that the mountain top or polka dot distribution came into being at a time when heavy, continuous tree cover allowed little light to reach low-level shrubs, and survival was a matter of being able to spread quickly onto areas denuded by volcanic eruption or land slippage. If this explanation is correct, the scattered stands may be the survivors of a cold period, with conifer forest replacing open grassland and shrub land.*

*Another interpretation is possible, i.e., that R. kaempferi and R. kiusianum are simply two extreme phases of an assemblage of plants which share a common pool of genes (or, roughly speaking, are a single species). This would be close to Wilson's views and, also, would make it easy to explain why R. kiusianum survives in active volcanic areas; we could say that it migrates back from the common pool after being destroyed on a mountain top, and that the low-growing plants are the ones sorted out by ecological factors as it climbs back up the mountain. I used to favor this interpretation, but there are two things wrong with it. First, it does not account for the remarkable uniformity of R. kiusianum on the highest mountains of Kyushu; second, it fails to explain why there is no such thing as R. kiusianum on high peaks outside of Kyushu, even where the surrounding hills are full of R. kaempferi.

Next, we need to consider the Kirishima intermediates and the similar intermediates at Unzen, Haneyama, Kuju, and probably elsewhere. These are generally attributed to natural hybridization of R. kaempferi and R. kiusianum, and I must agree, since no other explanation appears to account for these plants.

R. sataense does not differ a great deal in general appearance from the Kirishima intermediates, but it is treated by the Japanese azalea specialists and by Dr. Creech as a somewhat different entity. As mentioned, Mr. Kunishige suggests that it can perhaps be considered an ecological form of R. kaempferi, Present-day opinion seems inclined toward use of the R. sataense name for all of the Osumi Peninsula plants which resemble R. kaempferi but have smooth and generally smaller leaves and varying flower colors (i.e., considerable diversity outside of the orange-red and salmon range).

If R. sataense can, indeed, be considered an ecological form of R. kaempferi, it is difficult to see how the separation could have taken place in today's environment. As Mr. Kunishige points out, R. sataense usually grows in more open, sunny places than R. kaempferi. Yet, at present, the two are found in such close proximity that it would be hard to imagine the emergence of distinct races in response to local conditions. Instead, it would seem that they must have developed their distinctive characteristics at a time when they were more effectively isolated from each other. One possible explanation is that the R. kaempferi on the Osumi Peninsula was cut off by the huge pumice flows from a vent about 10 km. north-northeast of present-day Mt. Sakurajima. According to Kuno, in Catalogue of the Active Volcanoes of the World, these flows covered an area not less than 120 km. by 100 km., to a depth of 100 to 200 m. (no surprise to anyone who has looked at road cuts and stream banks near Kagoshima). The result may have been a long-lasting barrier which allowed the cut-off R. kaempferi to evolve into the fairly distinct and stable new race we call R. sataense. However, more favorable conditions eventually allowed ordinary R. kaempferi to cross the barrier, and this latter plant is now widespread on the peninsula.

Hardiness: A Plan for Action

The Kurume azaleas perform admirably in warmer parts of Britain and in much of the U.S., including sunny portions of Seattle and Portland gardens. However, most of them don't seem to have a great deal of tolerance for cold or dark weather, and British writers have mentioned a failure to set flower buds unless planted in a warm location. This suggests that they are largely the offspring of wild plants which evolved in a warm, sunny climate - and R. sataense may well be such a plant. Yet, among the individual Kurumes, quite substantial differences in hardiness have been reported. This suggests the interesting possibility that a Kurume which is primarily derived from the Kirishima intermediates, instead of R. sataense, may be more tolerant of cool summers and cold winters.

Obviously, testing for temperature tolerances should continue, and the whole group of Kurumes will be more useful when we are more certain of the ones adapted to particular climates. Also, carrying this approach further, why not a new group, specially bred for the cool climates of eastern Scotland, Scandinavia, and portions of the U.S. and Canada? Immediately, I must add that I am aware of the hundreds of existing clones-probably hundreds too many. Yet, if it is possible to develop a Kurume-type race for growers in these cooler climates, there is no reason why they shouldn't have such plants. A developmental plan with every chance of success would be to take far-northern R. kaempferi, perhaps from an elevation of several hundred meters in the cold, foggy Hidaka Mountain Range of central Hokkaido (at the latitude of New England), and cross this with high-elevation forms of R. kiusianum. Neither of these parents should cause any problems of heat demand in summer or winter. Yet this would be nothing more than a new version of the hybridization which is considered to have yielded the Kirishima intermediates and, perhaps, a portion of the Kurumes.

The resources for making such crosses are probably at hand now, in this country. Dr. Creech, in 1961, made no less than seven collections of R. kaempferi plants in Hokkaido, under USDA Plant Introduction Numbers 276093-276097 and 276141-276142. Or, if these plants are widely dispersed and now difficult to locate, there should be no special difficulty in getting replenishment stocks from the same Hokkaido areas. The ordinary, high elevation rose-purple R. kiusianum is in ample supply, at least in the Seattle area, from seed collections by Mrs. Doleshy and myself: No. 8, from pumice flats of Ebino Plateau, Kirishima, and No. 37, from summit of Mt. Tsurumi. White-flowered clones are available commercially, and all the ones we have seen appear to be high-elevation P. kiusianum. Also, beautiful pink-flowered selections are little known in the U.S. but apparently popular in Japan, and these should certainly be included in the program. The seeds from these crosses should be grown from the beginning in the cool-summer or cold-winter localities where the new strains are particularly needed. Testing will thus begin when the small plants are moved out-of-doors, and there will be no mistake as to the attainment of hardiness or bud-set objectives.

In selecting clones, it would be highly desirable for the Kurume Research Station and/or the U.S. Department of Agriculture to act as referee, in order to prevent any more of the confusion which has resulted from too many named clones of the Kurumes.

This proposed new race is something for the future, and you, the reader, may be interested in selecting some Kurumes at present. Advice is plentiful in Rhododendron publications, general gardening books, and even the newspaper garden columns. I can add little. However, because of the differences in climatic adaptability, I'd recommend that you get well-informed local advice, rather than depending on recommendations from another part of the country or another part of the world.

ACKNOWLEDGMENTS

I wish to express appreciation to the following individuals for their comments and suggestions concerning this paper: J. F. Caperci, Seattle; Dr. J. L. Creech, Washington, D.C.; E. Dzurick, Seattle; Dr. A. E. Kehr, Washington, D.C.; M. Kunishige, Kumme; K. T. Lauhon, Seattle; B. M. Swenson, Seattle; K. Wada, Yokohama. Also, T. Takeuchi, of Kawasaki, contributed by sending new information. However, I retain responsibility for accuracy and also for all opinions and interpretations expressed herein.

LIST OF SPECIES AND SYNONYMS, as interpreted by the author:

KAEMPFERI Planch. YAMA-TSUTSUJI

KIUSIANUM Makino. MIYAMA-KIRISHIMA

obtusum (Lindl.) Planch. KIRISHIMA-TSUTSUJI

=hybrids of natural or garden origin, involving KAEMPFERI and other species

obtusum f. amoenum Wilson =selections of KIUSIANUM

obtusum f. japonicum Wilson = KIUSIANUM or natural hybrids between KIUSIANUM and KAEMPFERI

obtusum var. kaempferi Wilson = KAEMPFERI SATAENSE Nakai.

SATA-TSUTSUJI (also perhaps called KIRISHIMA-TSUTSUJI in the past)

VOLCONIC ACTIVITY DURING HISTORIC TIMES, compiled by the author, mainly from information in Kuno's Catalogue of the Active Volcanos of the World, Part XI:

SAKURAJIMA. Records of frequent eruptions extend back to AD 708. Much of the surface of the mountain, from summit to sea level, is a patchwork of the lava and other materials from these recorded eruptions. KIRISHIMA. Records of frequent eruptions extend back to AD 742. Apparently, only scoria and small, solid fragments have been ejected during this interval. Most eruptions have been at Ohachi crater, near Takachiho Peak; also, some have been at Shimmoe Peak, between Takachiho and Karakuni, and the April, 1946 eruption probably involved Karakuni.

TAKAKUMA. Not a volcano; appears to receive substantial ash fall from Sakurajima.

UNZEN. Records of eruptions extend back to AD 860. Lava extruded in 1657 and 1792. Violent 1792 activity caused 14,300 casualties on May 21st, from collapse of peak and resulting wave action in the sea. HANEYAMA. Not a volcano.

KUJU. The only activity known with certainty in historic times is gas emission.

TSURUMI. Latest eruption of blocks and ash was in AD 867; steam rose from old crater in 1949.

SELECTED REFERENCES

CREECH, J. L., Plant Explorations of Ornamentals in Southern Japan, U. S. D. A. Crops Research publication ARS 34-1, 1957. (Pub. by Agric. Research Serv. in cooperation with Longwood Gardens.)

CREECH, J. L., Ornamental Plant Explorations - Japan, 1961, U. S. D. A. Crops Research publication ARS 34-75, 1966. (Pub. by Agric. Research Serv. in cooperation with Longwood Gardens.)

CREECH, J. L., Ornamental Plant Explorations - Japan, 1961, Quart. Bulletin Amer. Rhod. Soc., 21: 2-9, January 1967.

HARA, H., Occurrence and Distribution of Rhododendrons in Japan, Royal Hort. Soc. Rhododendron Year Book, 3: 112-127. 1948.

INGRAM, C., Is Rhododendron kaempferi a Good Species?, Royal Hort, Soc. Rhododendron and Camellia Year Book, 10: 28-30. 1956.

KUMAZAWA, S., Kurume Azalea. 25 pp., illus., 1958. Photo-reproduced; distributed by Kumme Branch, Horticultural Research Station, Miimachi, Kumme, Japan.

KUNG, H., Catalogue of the Active Volcanos of the World including Solfatara Fields, Part XI, Japan, Taiwan and Marianas. 332 pp., maps and diagrams. Rome, International Ass'n. of Volcanology, 1962.

MAKING, T., (Rewritten under Maekawa, Hara, Tuyama), Makino's New Illustrated Flora of Japan. 1149 pp., illus. Tokyo, Hokuryukan Co. Ltd., 1962. (In Japanese.)

OHWL, J. Flora of Japan (in English). 1067 pp., some illus. Washington, D.C., Smithsonian Institution, 1965.

STEVENSON, J. B., ed., The Species of Rhododendron, 2nd ed. 861 pp., some illus. London, The Rhododendron Soc., 1947.

TAKAI, F., MATSUMOTO, T., and TORIYAMA, R., Geology of Japan. 279 pp., maps and diagrams. Tokyo, Univ. of Tokyo Press, 1963. Esp. see part 10, The Quaternary.

WILSON, E. H., and REHDER, A., Publications of the Arnold Arboretum, No. 9, A Monograph of Azalea. 219 pp. Cambridge, Mass., The University Press, 1921. WILSON, E. H., Plant Hunting: Vol. II, The Tropics; The Orient. 276 pp., illus. Boston, The Stratford Co., 1927.

(Also, my own accounts of trips to Japan are in various issues of the Quarterly Bulletin American Rhododendron Society, beginning in volume 20, 1966.)