JARS v47n2 - Rhododendron vanhoeffeni Abrom.

Rhododendron vanhoeffeni Abrom.
Bengt Kihlman
Uppsala, Sweden

Ledum - A Subsection of the Genus Rhododendron ?
In 1990 the American botanists Kathleen Kron and Walter Judd (1) suggested that Ledum should be regarded as a subsection of the genus Rhododendron , rather than as a separate genus. Characters that are supposed to separate Ledum from Rhododendron are a corolla with separate petals and a capsule splitting from the base. As pointed out by Kron and Judd, however, the petals of Ledum have been reported in the literature to be actually slightly united at the base.
Although the suggestion by Kron and Judd may not have been received with any greater enthusiasm by European taxonomists, the Finnish taxonomist Harry Harmaja has not only accepted it, but also presented additional reasons why the species Ledum should be placed within the genus Rhododendron (2, 3). One of these reasons is Rhododendron vanhoeffeni . This taxon is mentioned in small type in The Flora of Greenland , a species of Rhododendron that could also be a native hybrid between R. lapponicum and Ledum palustre (4). According to Harmaja (3), the occurrence of such a hybrid points to a very close relationship between Rhododendron and Ledum .
Hybrid or not, I found R. vanhoeffeni a most interesting taxon. I, therefore, wrote to the Botanical Museum of the University of Copenhagen to find out if they had any new information on this plant. The reply came by return from Dr. Bent Fredskild at the Greenland Botanical Survey: a mere curiosity for decades, R. vanhoeffeni is currently in the limelight again.

The Discovery and Rediscovery of R. vanhoeffeni
On the 2nd July 1893 Dr. Emil Vanhffen from Kiel found a plant on the Karajak-Nunatak in West Greenland that appeared to be intermediate between Rhododendron lapponicum and Ledum palustre ssp. decumbens . A pressed specimen, the type, was described in 1899 by Abromeit (5) as a new species of Rhododendron which he called R. vanhoeffeni . He pointed out, however, that the possibility of the plant being a hybrid between R. lapponicum and Ledum palustre could not be excluded.
For nearly a century this was the only known finding of R. vanhoeffeni , but in June 1986 the Dane Lars Ceil found a plant answering to the description of R. vanhoeffeni at a locality a few kilometers east of the airport at Sndre Strmfiord. The following summer Bent Fredskild was able to confirm that the plant was identical with R. vanhoeffeni .
Since then several more specimens of R. vanhoeffeni have been found, both in Sndre Strmfjord and in other places in West Greenland, a fact that suggests that the taxon cannot be extremely rare. When not in flower, it is, however, very difficult to detect. This, and the fact that it flowers early in the summer, before the start of most botanical excursions, explains why it has escaped detection for so long.
All findings of R. vanhoeffeni are from the lowland where it grows together with R. lapponicum and L. palustre ssp. decumbens on hummocky, mossy heaths (Figs. 1 and 2). Dwarf shrubs, such as Betula nana (dwarf birch), Salix glauca (northern willow), and Vaccinium uliginosum ssp. microphyllum (Arctic blueberry) are common at these localities.

My account of the finding and re-finding of R. vanhoeffeni is based on information kindly placed at my disposal by Bent Fredskild. The findings and observations of the last few years have recently been reported by Vilhelm Dalgaard and Bent Fredskild in an article to be published in the Nordic Journal of Botany (6). Dalgaard and Fredskild are convinced that R. vanhoeffeni is a hybrid between R. lapponicum and L. palustre ssp. decumbens . They found the hybrid to be completely seed and pollen sterile, which is not surprising if it is the triploid product of a cross between plants belonging to different genera. Dalgaard and Fredskild, who follow the traditional classification of Ledum as a separate genus, have named the hybrid x Ledodendron vanhoeffeni . If, on the other hand, one agrees with Kron and Judd (1) that Ledum should be regarded as a subsection of the genus Rhododendron , the hybrid should be called R. x vanhoeffeni . Since I am not yet prepared to take a position on this question, I have used throughout this paper the name that Abromeit gave the original specimen, viz., R. vanhoeffeni .

Description of R. vanhoeffeni
As pointed out by Vanhffen and further stressed by Abromeit, R. vanhoeffeni is intermediate between R. lapponicum and L. palustre ssp. decumbens . It is closer to L. palustre in its growth form, being less compact and more spreading than R. lapponicum . The leaves are quite similar to those of R. lapponicum , except that they are narrower, somewhat more pointed and have a more pronounced midrib (Fig. 3). Their upper surface is covered with scales, although not so densely as the upper surface of the leaves of R. lapponicum . In contrast to the leaves of L. palustre , those of R. vanhoeffeni are only slightly, if at all, revolute at their margins and their lower surface lacks the rust-red woolly covering characteristic of the leaves of L. palustre . The inflorescence is a 5-10-flowered umbel (Fig. 4). The corolla has a diameter of about 12 mm which means that it is slightly larger than that of L. palustre , but distinctly smaller than the corolla of R. lapponicum . The five petals are fused at the base and have a colour that varies from pink to nearly white. Like L. palustre , R. vanhoeffeni has 10 stamens which, however, are longer than those of the former taxon. The stamens of R. vanhoeffeni further differ from those of L. palustre by having brownish-red anthers. The stigma is red or brownish-red and the style sometimes of a pinkish colour. According to Dalgaard and Fredskild (6) the morphology of the seed capsules is dependent on their size. Normally developed capsules "intermediate in size, are dehiscing by five basal septicidal slits and are more or less pendulous as in Ledum , whereas minor capsules as in Rhododendron are more or less erect and splitting apically." The capsules I have seen from different specimens at Sndre Strmfjord were all of the rhododendron-type.

Production of Hybrids Between R. lapponicum and L. palustre ssp. decumbens
In his description of the type Abromeit (5) points out that the ultimate proof of R. vanhoeffeni being a hybrid between R. lapponicum and L. palustre ssp. decumbens would be obtained by making a detailed comparison between R. vanhoeffeni and an artificially produced hybrid between the putative parental species. In the autumn of 1991 it so happened that my wife Irma and I had in our garden specimens of both R. lapponicum and L. palustre ssp. decumbens with flower buds. Furthermore, we had plants of different origins for both species, viz., plants derived from seed collected by me in late August 1990 on Tardoki-Jani in Sichote-Alin, Russian Far East, and plants from seed collected in Canada and obtained through the Scottish Rock Garden Club. We decided, therefore, to carry out the crossing suggested by Abromeit.
In the beginning of December we took two plants of each parental species (one from Sichote-Alin and one from Canada) into our greenhouse and a couple of weeks later we had flowers on all four plants and were able to carry out the crossing suggested by Abromeit. Reciprocal crosses were made between pairs of species from the same origin. The results are summarized in Table 1.

The table shows that fertilizations were successful only when L. decumbens was the recipient and R. lapponicum the donor of pollen, independently of the origin of the plants. At the end of April when the seedlings were about two months old, it was possible to establish their hybrid nature by comparing them with seedlings of the same age from the parental species. Like the leaves of R. lapponicum , those of the hybrids were covered with scales, in contrast to the leaves of L. palustre which were hairy.
The mortality of the hybrid seedlings was unusually high during their first months, possibly as a result of a genetic imbalance. The surviving plants, however, 10 of Canadian and 13 of Russian origin, all appear to be healthy and vigorous. This is particularly true for the plants of Russian origin, seven of which already have formed flower buds.
Although the crossing experiments with material from Sichote-Alin and Canada have given us valuable information and experience, the forms of the parental species used in the crosses, particularly those from Sichote-Alin, are too different from the forms occurring in Greenland. (In fact, the form of R. lapponicum from Sichote-Alin is regarded by Russian botanists as a separate species, R. parvifolium var. alpinum .) Ultimate proof of its hybrid origin will come from a comparison between R. vanhoeffeni and the progeny of artificial crosses between Greenland L. palustre and R. lapponicum .
In order to obtain the correct forms of the parental species, Irma and I visited Sndre Strmfjord in Greenland in June 1992. The time was chosen in such a way that we also would have the opportunity to study and photograph flowering specimens of R. vanhoeffeni . The estimation of the correct time was based on the conditions during a normal year. Since the ticket reservations had to be made months in advance, we had no possibilities to consider possible deviations from the normal. Unfortunately summer was at least three weeks late in 1992 and this meant that we did not get any opportunity to see R. vanhoeffeni flower during our visit to Sndre Strmfjord.
As already mentioned, R. vanhoeffeni is very difficult to detect among the abundant R. lapponicum and L. palustre when it is not in flower. Fortunately, Bent Fredskild had provided us with a detailed description that made it possible for us to find one specimen. We took some twigs with flower buds from this plant, and after our return to Uppsala the buds opened, making it finally possible for us to photograph a living flower of R. vanhoeffeni (Fig. 4). Ten days after our visit, Wendy Eisner took some photographs of the plant in Sndre Strmfjord which was then in full flower (Figs. 1 and 2).
The consequences of the summer being so late were, however, not entirely negative. We now had no difficulties in finding specimens of the parental species of a suitable size and with plenty of flower buds. As a result, we were able to carry out the reciprocal crosses under controlled conditions at our home in Uppsala. The crosses and their result appear in Table 2.

As in the previous experiments, the crosses were very successful when L. palustre was the recipient and R. lapponicum the donor of pollen though, in this experiment, it seemed as if some of the crosses in the opposite direction also had succeeded.
The seeds from the reciprocal crosses between Greenland L. palustre ssp. decumbens and R. lapponicum were harvested in September 1992 and sown in October. Controls were obtained by sowing at the same time seed collected in Greenland from the parental species.
After seven to 10 days at 20°C, the majority of the seeds had germinated and two months later, when most of the seedlings had formed at least two pairs of true leaves, it was possible to ascertain that the cross L. palustre (♂) x R. lapponicum (♀) had succeeded (leaves and shoots more scaly than hairy, whereas the opposite was true for the control seedlings of L. palustre ). In contrast, the attempted reciprocal cross produced seedlings indistinguishable from the control batch of R. lapponicum seedlings, suggesting that they are not hybrids but the product of selfing or possibly pseudogamy. If so, it further stresses the importance of using Ledum as the female parent, when a cross between L. palustre ssp. decumbens and R. lapponicum is attempted.

Summary and Conclusions
Although the ultimate proof (the comparison with an artificially produced hybrid between plants from Greenland) is still lacking, most of the available observations suggest that R. vanhoeffeni is a hybrid between R. lapponicum and L. palustre ssp. decumbens . In most respects R. vanhoeffeni is intermediate between these two species which occur abundantly at the localities where the possible hybrid is found. At these localities the putative parental species also flower at about the same time, with L. palustre starting to flower before R. lapponicum has finished flowering. Our crossing experiments have further shown that the two species can easily be crossed, provided that L. palustre is the recipient of pollen. Thus, the primary conditions for hybrid formation are at hand. The fact that R. vanhoeffeni appears to be completely seed and pollen sterile also indicates that it is a hybrid.
So far R. vanhoeffeni has been found only in Greenland, but if it is a hybrid between R. lapponicum and L. palustre ssp. decumbens , it is likely to occur wherever the two parental species grow within the same locality and have overlapping flowering times. Such localities are likely to be found in the Arctic regions of North America, from Baffin Island to Alaska. The hybrid may also occur in the Russian Far East, although on Tardoki-Jani, the place I visited in 1990, the two parental species grew at different altitudes and under different environmental conditions.

Acknowledgements
I should like to thank Dr. Bent Fredskild at the Botanical Museum of the University of Copenhagen for valuable help, advice and support in connection with the preparations for the trip to Greenland, as well for kindly placing at my disposal available information on R. vanhoeffeni . My wife and I are greatly indebted to Ms. Wendy Eisner for all the help she generously gave us during our stay at Sndre Strmfjord. I am also very grateful to her for the photographs of R. vanhoeffeni and for allowing me to use them in this article. My trip to Greenland was financially supported by the Enander Fund, The Royal Swedish Academy of Sciences.

Literature Cited
1.  Kron, K.A. and Judd, W.S. Phylogenetic relationships within the Rhodoreae (Ericaceae) with specific comments on the placement of Ledum . Syst. Bot. 15: 57-68, 1990.
2.  Harmaja, H. New names and nomenclatural combinations in Rhododendron (Ericaceae). Ann. Bot. Fennici 27: 203-204, 1990.
3.  Harmaja, H. Taxonomic notes on Rhododendron subsection Ledum (Ledum, Ericaceae), with a key to its species. Ann. Bot. Fennici 28: 171-173, 1991.
4.  Bcher, T.W., Holmen, K., and Jakobsen, K. The Flora of Greenland. Copenhagen 1968.
5.  Abromeit,J. Botanische Ergebnisse der von der Gesellschaft fr Erdkunde zu Berlin unterLeitung Dr. v. Drygalski's ausgesandten Grnlandsexpedition nach Dr. Vanhffen's Sammlungen bearbeitet. B. Samenpflanzen (Phanerogamen) aus dem Umanaks - and Ritenbenks-Distrikt. Biblioth. Bot 42(2): 1-106, 1899.
6.  Dalgaard, V. and Fredskild, B. x Ledodendron vanhoeffeni (syn.: Rhododendron vanhoeffeni ) refound in Greenland. Nord. J. Bot. 13(3), 1993, in press.

Professor Kihlman is a retired professor of cytogenetics.