Potential of Rhododendron yakushimanum in Breeding Commercial Hybrids
Dr. Gustav A.L. Mehlquist, Storrs, CT
Reprinted from ARS Breeder Roundtable 1980
I've been interested in yakushimanum for many years now, and I don't know who to really blame for it. I probably would have to blame this on Dave Leach, and many other people who started to talk about yakushimanum in England in 1960-61, I was convinced that this was something to work with. At that time I had already become imbued with the fact that a good rhododendron, to be successful, first of all should be a good foliage plant because, as many ladies pointed out to me, it only blooms about two or three weeks in a year. The rest of the year you have only the plant to look at, and it's got to be ornamental whether there are flowers on it or not. Furthermore, when I saw that plant in the Wisley trial garden with the new shoots coming out, they were just as beautiful as the flowers. If you can enjoy first the flowers for two or three weeks, and then the new shoots for two or three weeks before they begin to take on normal color, you prolong the season in which you have flowers, or flower-like things, to look at - and then, if the plant is good the rest of the year, you have something much more than just hybrids with very good flowers.
So I like yakushimanum because it has good foliage. There is no question about that. It holds the leaves two years or more. I think I have seen yakushimanum plants that have held the leaves for three years. I think that's very important - that they hold their leaves a long time. In contrast, nearly all the thomsonii hybrids, that is the hybrids from the species in the Thomsonii series, which are the ones from which we have to get most of our yellows, rarely hold their leaves more than a year - in some cases fifteen months - an unfortunate dominant characteristic in breeding with them.
And then we have a felted indumentum, and this is very important. Indumentum, I suppose, in the natural habitat of yakushimanum helps to prevent undue drying from the winds. In the average garden I think it also helps to deter insects. Don Wyman of the Arnold Arboretum I think was the first one to point out that it probably deters the lace-wing bug because they had never noticed much lace-wing damage on any indumented plant in the Arnold Arboretum, and he suggests that it might also deter other insects. For many years in my garden the indumented plants had no red spider mites. One year, however, the red spider did invade the yakushimanums, too, and we had a heck of job finding a spray that would penetrate the indumentum and get rid of the red spider mite. But, by and large, that felted indumentum will help to prevent insects from doing much damage.
In some indumented species, particularly in the Arboreum series, that get tall and lanky, the indumentum has a nice reddish color so that when the wind moves those leaves, you get a very attractive appearance. I can't remember all the species now, but in England I've seen at least 20 species that are indumented with large leaves that twist in the wind so you can see that reddish color behind - and, if you don't think that this color is sufficient, then, of course, you can go from yakushimanum to other forms of the same species so-called - and now that they have merged yakushimanum with metternichii, you have a variety of indumentum from almost white to almost red.
Some breeders claim Rhododendron yakushimanum is dominant in growth habit. I doubt that. Whenever I have crossed two species of sufficiently different size - such that the geometric mean is statistically different from the arithmetic mean - the population has always come closer to the geometric mean than the arithmetic mean. So, therefore, I maintain that it probably isn't dominant, but rather that because it is more compact than most of the other species, the hybrids will tend to be more compact.
The first batch of hybrid seedlings with yakushimanum that I raised was with Rhododendron smirnowii, another indumented species, and they're all beautiful. I made my selections and had to sell most of the others before they bloomed because of lack of space. Most of those went to the Midwest where they were planted in containers outside a bank, and within three months the plants were dead. The Connecticut nurseryman who sold them the plants said they wanted their money back because the plants had all died, but I told this Connecticut nurseryman that he shouldn't have sold them to the Midwest in the first place because we didn't have sufficient evidence that yakushimanum hybrids would withstand the Chicago climate - and certainly they wouldn't stand it in containers in front of a big building - and I didn't think very many other rhododendrons would do it either. I have found that yakushimanum and its hybrids are more susceptible to the ills that you would find in a big city, particularly associated with container-growing in general, so I would not recommend them for that purpose.
Now, as to the results that I've gotten from these crosses. First of all, I have found indumentum always to be recessive. At most you get a bit of puberulence in the first generation, and, curiously enough, that seems to come mostly when you're crossing yakushimanum to other members of its own series - the Ponticum series. You get a very strong puberulence when you cross 'Catalgla' - or any other member of the catawbiense group - to Rhododendron yakushimanum, and you get the puberulence, or something that looks like slight indumentum on the petioles.
On the other hand, if you cross yakushimanum with members of the Thomsonii series or the Fortunei series, then you get hybrids that are normally completely glabrous in the first generation so that, if you wish to have indumented first-generation hybrids, you must cross yakushimanum to another indumented species. Since there are not too many of the other indumented species that are hardy in my area, most of my experience has been with Rhododendron smirnowii as of to date, but now I'm starting with other species.
When I crossed yakushimanum with smirnowii I got hybrids from which I would say approximately 10% of the seedlings are worth looking at from the point of view of future further breeding for commercial propagation. 80% of them were good enough to sell to landscapers for their planting, and about 10% I had to throw away.
When I crossed Rhododendron fortunei a few more had to be thrown away, but by and large the greatest number could be sold. Not so when I crossed it to 'Blue Peter'. I had over 500 seedlings and after 12 years now, there isn't one that I have put a ribbon on to save yet. They're just plain junk - and that was a cross from which I expected a great deal. 'Blue Peter', in my opinion, is a beautiful rhododendron, but it has long, large thin leaves that the caterpillars would walk a mile for, and I had hoped that we could do something about that. Well, I haven't seen any caterpillar damage on the hybrids, but on about 25% of the plants the flowers don't open. They stay closed. The styles stand straight up in the air like something suggesting various things. The bees apparently find something there of interest anyway and pollinate them successfully. Seeds are produced in quantity, but they're not worth anything. And this year when we had what I call a desiccating winter, much of the foliage is so badly disfigured that even if the plants recover they're not going to do me any good because I don't want to introduce any hybrid which in a similar winter would disfigure that way.
After I heard Dave Leach's talk, perhaps I should seriously consider not introducing anything because, if it takes ten to fifteen years to get something to satisfy myself, and then it falls by the wayside, a lot of time and effort would be wasted.
I was in the business of breeding carnations for many years, and after 30 years of carnation breeding and selecting and propagating a number for further testing, I don't think more than five ever reached the market. One of them, however, actually did become a worthwhile variety, but the others didn't get very far. So I know the problems involved there. One shouldn't expect too much out of any one cross.
But the one thing that I would like to emphasize is that you cannot really find out what your cross is going to do unless you have sufficient numbers. In all of these that I mentioned I have 300 plants or more, and some of them as many as 500 or 600 of them. The reason for that being, as Dave Leach indicated, in plants we assume that the bad features will come from both species - both parents - at least most of the time, and I find that now I can improve the cross between yakushimanum and smirnowii by using different smirnowiis. The smirnowii I used was one that I got from Weston Nurseries, which they had gotten at Arnold Arboretum and which we had come to think of as the Arnold Arboretum form. When I spoke to Don Wyman about it, he said that he wished I wouldn't use that term. They had several plants of smirnowii, and when they introduced smirnowii to various people, they didn't distinguish one form from another. So while they would like to get the credit for having introduced that species to the American trade, it probably wasn't good to identify any one plant from the Arnold Arboretum as the AA form because no one form could be identified. I always self both parents if it is at all possible in order to see what the degree of heterozygosity within the species is, and every form of smirnowii that I've selfed so far has given me a wide variation of different types.
I picked up a fairly compact form of smirnowii from Cesarini on Long Island. I think he got it from Mr. Phipps, but I'm not sure about that. I selfed that one, and out of that I'm not only getting compact ones, I'm getting tall ones as well, and some very dwarf ones. And, if you bear in mind that I said, if you're genetically interested, I would naturally make my next cross between yakushimanum and smirnowii involving the yakushimanums that I know something about and a very dwarf form of smirnowii in the hope of getting hybrids that would be even more compact than the ones I've already obtained. In other words, this stresses the importance of using the right forms of the species to start with.
In orchids, where it costs considerable money to grow 1000 seedlings for five or ten years in the greenhouse at 60° minimum temperature, the cost is such that they have kept track of their seed parents and their pollen parents very accurately for a number of years, and they have found that sometimes two forms of a species that look the same for all practical purposes, or commercial purposes, breed quite differently. And it's very important to bear in mind which form you've used because you might find that you will have to go back to that one again.
By all means purify the species if you can. Now, if you wait and purify the species before you make your initial crosses, you'll never make the initial crosses because you won't have the species pure enough in your lifetime. So do it all at once. Make the crosses that you wish to make, but at the same time self the parents in order to find out what forms might be the best ones to use in future crosses.
I'm looking forward to the day when you don't have to propagate certain rhododendrons from cuttings. I maintain that the nurserymen should not be propagating any of the dauricum forms from cuttings that can be grown from seed. All you need to do is to purify the species, select the right parents, isolate them and cross-pollinate them and you get all the 'Cornell Pink' types - or whatever type you want - that you can use. You cannot sell them as 'Cornell Pink' because that is a clonal name, but if they look like 'Cornell Pink', they're going to be just as salable as 'Cornell Pink', and they'll have one advantage over 'Cornell Pink'. They make five or ten stems all at once so that the borers and various other insects can have their share and you still end up with a plant, whereas with 'Cornell Pink' you have only one or two stems, and if the borer takes one, you have a very sad-looking plant until you can regrow it again.
The same thing will probably also happen for many others of the lepidote group. I don't think it's possible for some elepidotes, although I think it's going to be possible to improve smirnowii - or at least use selected forms of smirnowii with selected forms of yakushimanum and other species and get hybrid populations where 99% of them are directly usable to the trade.
I maintain that any seedling is better than a normally-propagated cutting from many points of view. There is something in the seed which tends to produce large numbers of branches from the start and gives you commercial plants in much shorter time. Data from tissue propagation of rhododendrons indicate that plants propagated by this method may approach seedlings in growth habit.
I know some states are now considering outlawing the sale of seedlings, but that's because some people have misused that practice. They are selling seedlings of 'America', but they fail to tell the homeowners that the seedling from 'America' isn't likely to be like 'America' at all. But when you interbreed two selected parents which have been tested individually, you can produce an F1, population that is so similar that no one is going to have any objection to it whatever. We have to thank the flower and vegetable breeders for having developed techniques whereby now they are producing F1, hybrids of vegetables and many flower garden annuals that give you much better results than the old inbred lines used to. To be sure, they still have to inbreed them in order to get suitable parent material for the best selling hybrid form, and I think we're going to reach the same point of view in many rhododendrons. Not all of them, naturally, but there's going to be a market for well-grown seed populations. Of course, what I have said up to now about yakushimanum hybrids is all based on yakushimanum as represented by the FCC form and the Exbury form.
I have obtained seeds from various sources, so I have a total of about 5000 yakushimanums coming along from various sources. I have not found a single population that flowers as early, or is as good in most respects, as the seedlings you grow from crossing the above parents - and it bothered me no end to hear the other day that somebody in this crowd is now saying that there never was more than one clone in the first place.
I selfed yakushimanum FCC and the Exbury form, and it didn't get anywhere near the seed as I did when I cross-pollinated them and the cross-pollinated population grew much better. So I still maintain that there are two of them, and I think that perhaps the people who now say that there is only one do that on the grounds that it is very difficult to tell the two apart unless you have them growing side by side. And even then, if you don't have them correctly named from the start, you may be fooled. I have investigated this problem in England several times, and the people who are my age and were involved in rhododendron breeding at the time this species was introduced still maintain there are two different forms. Some of the younger people say they can't see the difference, so why regard them as two different forms. Well, I can't argue this point further, but I would like to suggest that you obtain both forms for breeding until you have some better forms. Don't neglect growing yakushimanum from every possible source.
I have some in my garden now that in fifteen years have not reached greater height than 6". They have beautiful small leaves and many of the rock plant growers would like to get hold of them for rock garden purposes. But I'm not selling them until I can flower them. I don't know if they'll ever bloom, but I'm still looking for one of these dwarf ones that will flower well.
I don't know why it is that the small ones don't flower as well as the big ones, but Wada told me at the time we had the annual meeting of this society on Long Island that the lanky ones were nearly always the first ones to flower, and the very tight plants that were most ideal from a rock garden point of view tended to be very slow bloomers. Well, he proved the point by selling Cesarini and me some seed from which we've grown now some three to four thousand plants, I guess, together, and most of them have not bloomed even though the seed was planted in 1966. On the other hand, I planted seed in 1965 from the other cross that I referred to, and they're already blooming. But some day somebody's going to come up with a yakushimanum that is dwarf, very tight, beautifully small leaves, and beautiful flowers that will be even better than what we already have.
Now Dave Leach said another thing to me this morning that I think I better share with you. He said there is a vast difference between the propagation from the original plant and the future propagation, and, unfortunately, that is true, but I hadn't thought about it with rhododendrons, frankly speaking, even though I have proved it in carnations.
When I came to this country we were growing a yellow carnation, 'Maine Sunshine', that was very difficult to propagate - therefore very expensive - and everybody said, "Why don't you breed a carnation that's easier to propagate?" Well, I had no difficulty getting new yellows that were easy to propagate the first propagation generation or two, and, if I kept propagating from cuttings all the time, I could propagate for about fifteen to twenty generations without any decrease in their rootability. But if I let the plants bloom and then took the cuttings, it wasn't more than two or three generations before my seedlings were just as hard to propagate from cuttings as 'Maine Sunshine' itself was.
And so I looked at the literature and found out that the avocado growers had paid some attention to this, and they said in effect that any darn fool can propagate avocados from a seedling, but if you let the plant fruit for a few years, it just doesn't propagate very readily. This is apparently a matter of juvenility that has not been studied in rhododendrons, and I'm glad you brought it up because it apparently is going to influence rhododendron production. It simply means that any potential rhododendron breeder would have to test the propagation feature also before he turns them over for commercial propagators because otherwise the loss is going to be much greater than it ought to be.