JARS v48n2 - Breeding of Winter Hardy Deciduous Azaleas in Finland

Breeding of Winter Hardy Deciduous Azaleas in Finland
Anu Vinl
Helsinki, Finland

Finland is a country situated between the latitudes of 60 and 70 degrees in Northern Europe where the climate restricts cultivation of many plant species. The lowest recorded winter temperatures vary from -35C to -45C (-30F to -48F) from south to north, though winters are normally somewhat milder. Since many landscape plants are imported from Central Europe, they do not always survive the cold winters. Rhododendrons are very appreciated and valued plants in home gardens and annually about 50,000 rhododendrons and azaleas are imported to Finland, mainly from Germany and Holland. Hybrids or cultivars of Rhododendron catawbiense Michx. and R. japonicum (A. Gray) Suring. account for more than 90 percent of these, and even they can only be grown in the southern part of the country. There was a need to do some breeding work among the hardy species and hybrid specimens growing in city parks and arboreta. Hybridization of broadleaf evergreen rhododendrons was started in 1973 by Dr. P.M.A. Tigerstedt and Marjatta Uosukainen at the University of Helsinki and the first six cultivars are now available. 'Elbiira'* and 'Hellikki'* are dwarf plants with red flowers; the others are taller, 'Haaga'* and 'Helsinki University'* with pink flowers, 'Mikkeli'* and 'Peter Tigerstedt'* with white flowers. Less than 2,000 azaleas are planted every year because they are believed to lack winter hardiness. Having several very hardy specimens and knowing the good results of the Finnish breeding work on evergreen rhododendrons, Dr. Tigerstedt and I started to breed deciduous azaleas at the University of Helsinki in 1988. The main goal of this project is to create azaleas suitable for Finland and other areas with similar climatic conditions. Besides winter hardiness, other important characteristics for superior azaleas are: flower color, shape and size, floriferousness, plant form and size, leaf color and shape, autumn foliage color, mildew resistance, and ease of commercial tissue culture propagation.
The first 278 crosses were made in 1988 with Finnish material, and in 1989 220 more crosses were made using the pollen from Finnish, Swedish and American (Minnesota) sources, all together 262 different combinations. The seed parents represented three species ( R. canadense (L.) Torrey, R. japonicum , R. luteum Sweet) and several hybrids, all of which grow in southern Finland. Pollen was collected from 15 different species and several hybrids. From the crosses we raised some 20,000 seedlings that were planted for trial in various Helsinki city parks and the Arboretum Mustila.
One aim of this work was to study which azaleas produce viable progeny when crossed. The success of the crosses was measured by the ability to produce seed, by seed germination, and by seedling survival during the first summer in greenhouse.

Table 1. Crosses where the pollen parent produced seed only with R. canadense , number of flowers pollinated, number of seed capsules harvested, germination (good = more than 40%, poor = less than 20%).
Pollen parent and its number Flower Seed pods Germination
13 Pallas 20 3 good
143 schlippenbachii 15 8 no
141 nudipes 5 4 no
96 yedoense var. poukhanense 8 2 no

Rhodora Is Different
Rhodora R. canadense is one of the hardiest azaleas. It has small flowers and leaves and is a tetraploid, i.e., it has twice the number of chromosomes of most other azaleas ( R calendulaceum (Michx.) Torrey is tetraploid, too). When crossed with a "normal" diploid plant, its progeny are triploid and sterile and can not be successfully used in further breeding experiments. In any case, we wanted to use it because of its hardiness.
In our experiments, R. canadense was unable to pollinate other azaleas. There are several possible explanations. It could be due to its chromosome number or its taxonomic relationship to other azaleas (i.e., it belongs to another section). However, the most probable reason is because the flowers are small and the pollen tube can not grow much longer than its own style length, which is very short.
When R. canadense was used as a seed parent, we got some interesting results. Six of the species used as pollen parents produced seed only with R. canadense . However, four of these did not germinate (Table 1). Of these species, R. schlippenbachii Maxim. and R. vaseyi A. Gray do not cross readily with other species; and R. yedoense var. poukhanense (Lev.) Nakai belongs to a different taxonomic group. Rhododendron albrechtii and 'Pallas' produced seeds that germinated.

Figure 1

All in all, R. canadense produced seed (viable or not) in most cases. Figure 1 shows the number of flowers pollinated and seed capsules harvested. Different hybrids in hybrid groups are counted together. The germination of seeds was divided into four categories: Good, more than 40%; Fair, 20-40%; Poor, less than 20%; and a class "does not germinate," which included those that died before the first transplanting. One third of R. canadense seeds did not germinate; the rest were either good or poor germinators. Even some lots of similar crosses, but different pollen origin, germinated differently, one well and the other poorly.
Figures 2 and 3 give the survival rates of the seedlings from R. canadense crosses showing the number of seedlings transplanted in each cross. Eight of the progeny lots of R. canadense died between the first and the second transplanting (Fig. 2). The number above the pollen parent's name is the number of the specimen, as occasionally pollen of a particular species was used from different origins.

Figure 2
Figure 3

Is Japanese Azalea Self-Sterile?
We self-pollinated 73 flowers of R. japonicum but did not get any seed from these pollinations. We also did not get any seed from the crosses with R. prinophyllum (Small) Millais, which was surprising, for it belongs to the same section and it has been successfully used in the Minnesota breeding program. Seed set was good with R. luteum . We also got some seed with R. calendulaceum and R. occidentale (Torr & A. Gray) A. Gray, as well as with several hybrids. One tenth of the seed lots, mostly with hybrid pollen, did not germinate. In general, we obtained more seedlings in the crosses where the pollen parent was a species instead of a hybrid.

Crosses with the European Species R. luteum
Rhododendron luteum proved to be self-compatible. It did not produce any seed when pollinated by R. prinophyllum , R occidentale , or Mollis hybrids. We got seedlings with some other hybrids and with R. calendulaceum and R. japonicum . From the cross R. luteum x R. japonicum only 6 percent of the flowers pollinated developed seed pods while in the reciprocal cross, R. japonicum x R. luteum we got seeds from 64 percent of the flowers pollinated.

R. luteum x R. japonicum
A 4-year-old seedling from a cross R. luteum x R. japonicum .
Photo by Anu Vinl

Crosses on Hybrids We did not work only on species, but used hybrids as seed parents as well. Most of these plants belonged either to R. x kosteranum or R. x gandavense hybrid groups.. Besides these, our pollen parents also represented the Knap Hill hybrids from southern Sweden, Northern Lights hybrids from Minnesota and other miscellaneous sources. The genetic base in these many different hybrids is wide and may cause incompatibility when crossing these plants with each other. The results from the crosses between different hybrids were variable and no general conclusions can be made.

The Breeding Goes On
These first crosses are the first steps of a long-term plant breeding project. The first new azalea cultivars might be on the market by the beginning of the next century. Before that, the seedlings must grow above the snow level so that their winter hardiness can be screened; they must bloom so that their flower characteristics can be seen; and they must be big enough to determine plant form. In 1993-94, we are starting to select the best plants of our 20,000 seedlings as candidates for new cultivars. Tissue culture methods will be developed for these plants and clone trials established in about five locations in different parts of Finland and possibly abroad.
Although hybrids were included in the breeding program, many crosses were between different species. The variation among the seedlings from these crosses is limited, thus F2-crosses will be made to produce segregating populations with new combinations of characteristics.

1.  Hogenboom, N. G. 1975. Incompatibility and incongruity: two different mechanisms for the non-functioning of intimate partner relationships. Proceedings of the Royal Society of London. B. 188:361-375.
2.  Johansson, T. 1984. Konsten att vlja frldrar. Rhododendronbladet 1984(1):5-8.
3.  Kho, Y. O. & Baer, J. 1973. Improving the cross Rhododendron impeditum x Rhododendron 'Elizabeth' by temperature treatment. Euphytica 22(2): 234-238.
4.  Sakai, A., Fuchigami, L. & Weiser, C. J. 1986. Cold Hardiness in the genus Rhododendron. Journal of the American Society for Horticultural Science lll(2): 273-280.
5.  Uosukainen, M. & Tigerstedt, P.M.A. 1988 Breeding of frost hardy rhododendrons. Journal of Agricultural Science in Finland 60: 235-254.
6.  Whiting, G. C. & Ascher, P. D. 1974. Interspecific hybridization between closely and distantly related Rhododendron species: pollen tube growth and seed set. HortScience 9(3): 276.
7.  Williams, E. G. & Rouse, J. L. 1988. Disparate style lengths contribute to isolation of species in Rhododendron. Australian Journal of Botany 36(2): 183-192.
8.  Williams, E. G., Kaul, V., Rouse, J. L. & Palser, B. F. 1986. Overgrowth of pollen tubes in embryo sacs of Rhododendron following interspecific pollinations. Australian Journal of Botany 34(4):413-423.
9.  Williams, E. G., Rouse, J. L. & Knox, R. B. 1985. Barriers to sexual compatibility in Rhododendron. Notes from the Royal Botanical Garden, Edinburgh 43(1):81-98.

Anu Vinl (M.Sc.) has worked as research assistant in the rhododendron breeding project at the University of Helsinki and has been involved in the azalea breeding project from its very beginning since 1988. In 1992-93, she was in the United States for a year working at Briggs Nursery, Olympia, Wash., as well as doing plant breeding and cold hardiness research at the Morton Arboretum, Lisle, III., and the Minnesota Landscape Arboretum in cooperation with the Center for Development of Hardy Landscape Plants and the University of Minnesota.

Editors Note: * Name is unregistered but is not in conflict with a registered name.