Compiled and edited by Ian Donovan, Pembroke, MA
The ARS New York Chapter sponsored another Hybridizers Meeting at Planting Fields Arboretum, Oyster Bay, New York, on September 15, 1979. Thirteen East Coast plant breeders made presentations at the session. We have taken this fourth reprint in our series of Dr. Mehlquist's talks from the published Proceedings of that meeting. Dick Murcott moderated the meeting and Howard Kuhn was the organizing committee chairman. Frank Arsen was on the committee and Ms. Gay Arsen was the typist/transcriber.
Hybridizers' meetings had been taking place since they began in 1973, usually once a year at the annual ARS convention or a regional meeting. Often, a verbatim report of the meeting was published. The challenge was and is that much volunteer effort must be expended to record clearly, transcribe accurately, then edit for both technical clarity and grammatical readability, and finally publish the proceedings. Because of this production burden, limited readership interest, and modest financial return, sponsoring chapters have published reports of only seven sessions since 1973.
Dr. Mehlquist in this fourth talk focuses on his experience of almost twenty years in breeding hardy elepidote rhododendrons for the commercial market place. He gives particular attention to the results he obtained using carefully selected parents as he strives for a hardy, clear red elepidote. He also puts on his teacher's hat when going into the basic calculations for determining how many plants the breeder should grow to obtain a valid sample size. Finally, he discusses the management challenges of growing large numbers of seedlings, formalizing contracts with growers, and introducing new selections. In response to questions from the audience, Dr. Mehlquist ventures his opinions about breeding hardy yellow elepidotes.
Noticeable in the original published paper of this talk is that Dr. Mehlquist no longer referred to species in their then obsolescent Balfourian Series classification arrangement. By 1979 the Series concept was being overtaken by a more botanically sound approach. The first International Rhododendron Conference to report progress in this effort, which Dr. Mehlquist attended, had taken place at the New York Botanical Garden during May 15-17, 1978. We have used the revised Edinburgh classification arrangement in this reprint.
Today, after almost thirty years of work, the taxonomists have completed their revision of the Genus Rhododendron into a hierarchically related set of subgenera, sections, and subsections. The revision of the lepidote section Vireya of subgenus Rhododendron will soon be published. The taxonomists did not sort out until 1987 the naming of the indumented Japanese species in subsection Pontica (the Degronianum Group), which Dr. Mehlquist speaks about in this talk.
Dr. Mehlquist identifies in this talk several elepidote species and hybrid clones that he used in his work. A breeder today would be well advised to seek out these hardy proven performers that deliver, at least in Dr. Mehlquist's experience, predictable breeding results.
Only minor adjustments in punctuation, grammar, and paragraphing; removal of some vocal pauses; identification of people, plants, and references the first time appearing; and some updating of species names with the Edinburgh revision have been made to improve readability and understanding of this verbatim transcript.
Ian Donovan
Some Thoughts On Rhododendron Breeding
Hardy Elepidotes and Indumentum
When I started this rhododendron project I wasn't unfamiliar with rhododendrons. I simply was not prepared to undertake work in rhododendrons as I already had several projects that required all the time I had. However, the head of the department, Professor Howard Rollins, pointed out that he had had numerous calls from nurserymen in Southern New England. They suggested that it was about time that the Connecticut Experiment Station, as represented by the Storrs branch located at the University of Connecticut, did something about getting better rhododendrons for the New England climate.
Particularly wanted were real hardy ones in clear colors and especially some reds. There are no real reds either in species or hybrids that are reliably hardy in New England. I should say that I don't consider 'America' or 'Nova Zembla' as being red. They are deep rose at best or, as they are often listed in Europe, carmine rose. Red in my opinion is something like R. griersonianum, 'Vulcan', 'Mars', and 'Captain Jack'.
At the time I started, I did not consider 'Captain Jack' hardy because of its parentage. It is 'Mars' x R. eriogynum. This species is considered no hardier than H-3, yet this year [1979] when I lost many flower buds, 'Captain Jack' flowered profusely, so obviously it is hardier than I thought and I should have used it. Certainly, I shall do so in the future. [H-3 is an obsolete ARS designation for hardy to -5°F/-21°C.]
I will show you two species that I picked for hardiness: R. catawbiense var. album and R. yakushimanum. I need to emphasize what form of these species I used. [Slides.]
Rhododendron catawbiense var. album 'Glass' is the botanical designation for the white form of the native species R. catawbiense from our southern mountains, as distinct from Rhododendron 'Catawbiense Album', a hybrid of English origin. This hybrid that is so common in the trade is probably, for the most part, of R. catawbiense parentage but may contain some R. ponticum as well to account for its relative ease in rooting from cuttings. The species, R. catawbiense, whether in its normal colored form or the white form that I used, is notoriously hard to root. This is the reason why I urge all Rhododendron enthusiasts to be careful to designate whether they are talking about the hybrid or the species.
As far as I know all forms of the species R. catawbiense are hardy to H-1. Rhododendron yakushimanum, on the other hand, probably is hardy only to H-2. [H-1 is an obsolete ARS designation for hardy to -25°F/-32°C, while H-2 means hardy to -15°F/-26°C.]
The New England nurserymen wanted nothing but H-1 hybrids, and possibly H-2. Anything else they were not interested in, and they were not slow to point out that since this was going to be done with taxpayers money, it better be something that they could use.
Rhododendron yakushimanum is an interesting species in that it holds its leaves longer than most other species. At the other extreme, we have the Fortunea subsection, which holds its leaves rarely more than 15 to18 months. Rhododendron yakushimanum and a few other species of the Pontica subsection, incidentally, hold their leaves for two years or more, and that makes a much better plant for landscaping purposes. Furthermore, Rhododendron yakushimanum is indumented.
You have to distinguish between indumentum and puberulence. There are several different types of indumentum, and if you are interested, read a book on rhododendron leaves by J. M. Cowan, The Rhododendron Leaf [published by Oliver and Boyd of London, 1950]. It has some beautiful photographs illustrating different types of indumentum.
He points out that some of the outstanding indumented species have so-called felted indumentum, which is that of R. yakushimanum. Other forms of the same group like R. metternichii, R. metternianum, and also R. degronianum, have so-called plastered indumentum, quite different, but it's still indumentum. [R. metternichii is now R. degronianum ssp. R. heptamerum var. heptamerum. Rhododendron metternianum, is closely related to R. yakushimanum, and may be a natural hybrid with R. metternichii.]
If you cross any one member of the R. metternichii group, including R. yakushimanum, to other non-indumented species of subsection Pontica, the hybrids will not be indumented in the first generation. They will be puberulent at best. This puberulence is something like granular material that washes off quite readily from the upper surface of the leaves within a few months or so, just like it does in pure R. yakushimanum, but not necessarily from the lower leaves.
I have yet to see a single hybrid coming out of a cross between R. yakushimanum and a non-indumented species that is indumented. At best they are puberulent. If you cross R. yakushimanum to a member of subsection Fortunea or to subsection Thomsonia, there is virtually no puberulence. Things one should bear in mind. On the other hand, crosses between indumented species have yielded nothing but indumented hybrids.
I have used several forms of white catawbiense, for the most part derived from [Joseph B.] Gable's 'Catalgla'. Many people have asked me which one is the best. I used five different forms, and I would say that no one form has been the best in all crosses, but some forms have been distinctly better than others in some crosses.
I went to see Guy Nearing [G. Guy Nearing of Ramsey, New Jersey]. He was a famous breeder, I was told, and I asked him what parents I should use in starting a breeding project. "There are not more than a handful of hardy hybrids that you can start with," he said. 'Kettledrum' and 'Charles Dickens' for reds was his recommendation. For pinks, he recommended 'Scintillation' and 'Rochelle'. In lilac, he recommended 'Everestianum', but he said you have to be careful. "There are two 'Everestianum's in the trade. One of them is much hardier than the other one. Don't get started with the one that isn't hardy."
Then he said, "If you want something that the nurserymen can propagate and grow rapidly, you have to go to something like 'Roseum Elegans', although I wouldn't do it myself. I hate the thing, but bear in mind there are many different clones of 'Roseum Elegans', some of which are hardier than others. If you really want to start out with a very hardy one, I will sell you a cutting of the one I have been growing for years. It's the hardiest and best of them all." I got a small three-inch cutting of it. It took me a while to get it started, and in the meantime I had to breed with the other forms. I have never gotten so much junk in all my life from any crosses as I got from these 'Roseum Elegans' forms.
In the last three years we had three severe winters in a row. Storrs [Connecticut] is still in Hardiness Zone 5A as far as rhododendrons are concerned. I had two 'Roseum Elegans', along with 'America' and 'Nova Zembla', along my driveway. During one winter two years ago [1976-77], 'Nova Zembla' and 'America' froze to the ground. One of the 'Roseum Elegans' froze down to the snow line, and gradually came back but with difficulty. The other 'Roseum Elegans' didn't get hurt at all. It was the plant that I got from Mr. Nearing.
He was right. He had a very hardy form of 'Roseum Elegans', but as he said, "The nurserymen are forever interested in having plants from mutations or sports or what have you of 'Roseum Elegans' that grow more rapidly and have a little better color." 'English Roseum' and 'Roseum Pink' are the ones that are mostly grown today. They are not nearly as hardy as the one he sold me.
He also recommended 'Catanea' (Mr. Nearing's selection of the species R. catawbiense var. album), but he said, "I am so booked up on it that it will be five years before I can let you have a cutting of it." Consequently, I haven't done any breeding with 'Catanea'. I had to start with 'Catalgla', which is simply Mr. Gable's designation for his selection of R. catawbiense var. album 'Glass'. Then I used 'Catalgla' F3 from Mr. Paul Vossberg, 'Catalgla' F4 from Hjalmar Larson in Washington, 'Catalgla' F5 of my own, and 'La Bar's White'. This last is a nearly pure white, but is an ungainly plant to deal with. It is, however, still a good plant. Any one of these has been the best of the lot in some crosses, but no one has been the best in all crosses.
How To Be a Success at Hybridizing
Many people have been influenced by Mr. [David G.] Leach's book [Rhododendrons of the World. New York: Charles Scribner's Sons, 1961.], but I found that some of his ideas didn't necessarily hold.
I had to fall back on the old genetic theories that I had learned in the first course in genetics. For Sid Burns' [Sidney V. Burns of the ARS New York Chapter who was present] benefit, I should say there are a few scientific breeders of rhododendrons. The reason their work hasn't been felt very much is, and you gave the answer yourself, they haven't had time to gain enough experience. You must bear in mind that we haven't had enough experienced breeders in rhododendrons for more than fifty years, whereas rhododendrons have been bred for 150 years.
Likewise, in the grains we have had experienced breeders for only about 100 years, but the grains have been bred for 1,000 years or more. Up to the time that they had the necessary theoretical and practical training, they had to depend entirely on experience with the material itself. I'd say, if you have to choose between technical training and knowledge of the material, knowledge of the material is far superior to technical training alone.
Three-quarters of all the good things we have today have been raised by people who never went to high school, never heard of genetics. They simply knew a good thing when they saw it. They seemed to have the knack for selecting the right parents. They learned at that early age that the value of parents doesn't lie in their beauty, in the ribbons they win at shows, but in the progeny they produce, and that is very important to bear in mind.
I keep telling my students: "Don't start a breeding program until you have read what there is in the literature. The cheapest plant breeding you can do is what you do in the library." Some people hate to take time out to go to the library. They would rather raise seedlings for a few years and find out for themselves. I cannot stop my friends from doing this, but I think it's a waste of time.
The question is often asked whether one parent is more important than the other in a given cross. There is no evidence from plants that one parent is more important than the other, except where we are dealing with cytoplasmic inheritance. Then one plant is definitely more important because very little cytoplasm is carried with the pollen grain.
In animals they always thought the bull to be more important than the cow. I think the main reason for that is that nearly all breeding has been done by men, and men have always had a tendency to present this assumption as fact. There is no evidence, as yet, that either side is more important than the other.
Seedling Population Size
Now to take up the kind of work that I have done. I crossed R. catawbiense var. album in its various forms to R. yakushimanum, and I got quite a population. Although I didn't have a flowering plant of R. yakushimanum at that time, Cecil Smith, from time to time, sent me an envelope full of pollen and I thought that if he was going to all that trouble, the least I could do was to use it. As a result, I ended up with many more seedlings than I needed.
There is always a question of how many seedlings you should raise in order to exhaust the possibilities of the cross. I don't know that. All I can say is, if you are crossing two species, you probably don't have to raise more than 100 seedlings in the first generation cross to get at least a pretty good idea of what is going on. But if you first cross two species, and then go into the F2 generation, that's something else again.
The size of the population you need then depends on the number of genes by which those species are differentiated. Nobody knows exactly. Most taxonomists, however, are in agreement with the idea that distinct species differ by twenty genes or more. If you want to figure out what sort of population you need in order to have a reasonable chance of extracting both parental types in your F2 generation on all possible recombinations, you have this situation:
The basic equation for calculating the number of plants required is:
1- pn = desired probability;
p = probability that the desired
genotype will not occur; then
1- p = the probability that it will occur.
Of course, the desired probability should be at least 50% or 1/2.
If a plant is segregating for three genes, we would have:
1-(63/64)n = 1/2.
Solving for n would give us the number of plants required to have a 50% chance of obtaining at least one of the multiple recessives; in this case, about 44.
In order to calculate the number of F2 plants required to obtain all possible recombinations of characters in a particular cross, one would have to know the exact number of genes involved. This is not known but since calculations based on a larger number of genes than the number of chromosome pairs (in this case 13) are likely to be meaningless, let us base the calculations on 13 genes.
We then have:
1 - (67108863/67108864)n = 1/2
Solving for n we would obtain 1,342,000 +/-.
In other words, one would have to grow this many plants to have a 50% chance of obtaining at least one of the rarer genotypes. Most large-leafed rhododendrons would require about five square feet for each plant in order to develop properly, or more than 150 acres for the whole population. Obviously, this is difficult if not impossible, but the fact remains that the larger number [of plants] one can grow, the better one's chances of obtaining the desired segregate.
You have to have a tremendous population, and then it only gives you very few of the rarer combinations. What is a reasonable chance anyway? A gambler would like to have a little better than 50%. If you insist on at least that, the population required is so large that probably no one has ever done that.
All I can say is that the bigger population you have in the second or third generation, the more likely are you to get what might be of value in that cross. Some people prefer to raise a number of crosses of relatively few plants from each one to find out if the crosses are worth anything or not. Others will raise large numbers all at once. This is where experience and judgement enters in.
Tribulations of Plant Trials,
Selection, and Introduction
I had space, so I raised large numbers because I was faced with an additional responsibility. Not only did the nurserymen want rhododendrons of the type that I have said, but they also wanted them to grow very rapidly--grow like hell in the nursery, so to speak--and then they could slow down as much as they wanted afterwards. They wanted them to flower in a two gallon can with three to five buds in two or three years, or else they couldn't be bothered with them. Two years later they revised that, and now they want something that could be grown in a one gallon can in one or two years, with two or three buds, or more.
The latest advice came from Jim Wells [James S. Wells, Red Bank, New Jersey] and a couple of others. They said that if I really wanted to make a name for myself and see a rhododendron introduced in my lifetime, I would have to produce one that would set one or two buds in less than one year in one quart cans! The way the nurserymen work, if you want to get a new hybrid introduced fast, you have to fit their method of production.
Many of the good rhododendrons we have today [1979] took 25 years to get recognized. Some of them took fifty years because we do not find out how hardy they are until we have a series of bad winters. Until these last two winters [1977-78 and 1978-79], I had a lot more rhododendrons to work with than I have this year. I lost about 25% or 30% of the population, but since I have about eight acres planted with survivors from about 140,000 seedlings that have gone through my hands, I still have enough to look at. 6,000 to 7,000 seedlings are on my property of two acres. All the others are on state property. I don't worry so much if some die as long as there are some left that look promising.
As I talk to the nurserymen, there is yet another important requirement that one should consider. They want the plants to look good 52 weeks out of the year. That means good foliage plants. Then if they have good flowers in addition, that is fine. How many plants with good flowers have that kind of foliage? Not many, and those that have that kind of foliage don't have good flowers, so that introduces more restrictions and calls for bigger numbers to choose from.
When I started this project, the head of the department was very much in favor of it, even though I pointed out that we might not see a solution to these problems in my lifetime. He said: "So what! We have orchards that have been going here for close to fifty years. We don't necessarily discard a project just because the personnel changes." Unfortunately, he retired in ten years, and the next department head who followed him was not nearly as interested in the project as was he.
I started with state funds entirely. I now find that I have to support it partially with personal funds, and that means that I am more concerned about what happens to the plants afterwards.
Three years ago we felt that we had some hybrids that were worth testing, and we contacted some Connecticut nurserymen about that. Yes, they would receive cuttings from us and report back each year for three years, and then we would determine whether they were worth continuing or not.
Everything was fine until they saw the contract. The contract, of course, had to be approved by the State of Connecticut administration and the State of Connecticut almost invariably had a non-discrimination clause included. You know you mustn't discriminate on the basis of race, creed, color, or whatever. The nurserymen said: "The hell with that! We aren't going to sign anything like that." They were adamant. They wouldn't sign it, so I had to go to out-of-state people to get cooperation.
We eventually got four out-of-state nurserymen, including one in Oregon. I found out afterwards, the only reason they signed the contract was that they figured the State of Connecticut couldn't do anything to them anyway.
The trouble with this arrangement is I have to travel to those states if I wish to look at the material, and I have to pay for the travel out of my own pocket. In the meantime I retired but I was permitted to continue with the rhododendron research. But there is no regulation in the world that permits the state to pay me anything. I am retired and it is just their good will that I am allowed to continue with the rhododendron project.
I have been talking to a lawyer friend on mine regarding introduction of my own varieties and he said: "You have to be careful, as I understand you have been thinking in terms of patenting. Bear in mind that the patents are good for only 17 years. From what I know about rhododendrons, it takes about 25 to 30 years to introduce one."
"The only way you can get anything out of a patent with rhododendrons is to build up your stock before you seek a patent so that you literally have enough to saturate the market within five years of the date you get the patent. Then you have to be prepared to sue anybody who infringes on your rights. It doesn't do a bit of good if you take out a patent unless you are prepared to protect your rights."
[This 1979 talk was on the verge of the horticultural industry entering the era of tissue culture, or micropropagation, of woody plants. Today many new woody plant varieties are clonally mass produced in the laboratory. The result is that new variety, container grown woody plants can reach the local garden center in three to five years. A newly introduced plant's product life cycle is typically five years, beginning with lab production of a single variety the first year. This will supply the market with enough plants to meet the demand rapidly when the new plant is introduced. As the horticultural industry becomes more centralized, the large firms are patenting more rhododendrons and azaleas to protect their corporate investments in production and marketing networks. Editor.]
I have used R. catawbiense var. album with R. yakushimanum, and we have a great number of these growing still, about 350 plants. I picked out five as being better than the others, and perhaps in another five years or so we will settle on one or two that are good enough to be introduced. By that time I hope I shall have some of the details of introduction ironed out.
Breeding for Better Reds
I made crosses between R. yakushimanum and 'Vulcan' and 'Mars' in an attempt to get reds, and of course, we got no reds in the first generation. We got no reds when we crossed 'Vulcan' to R. catawbiense. I got pinks and degrees of pinkness, depending on what form of catawbiense was used.
When I used R. catawbiense var. album, they were, for the most part, light to medium pink. When we used R. catawbiense var. rubrum, so called, they were deep pink but not red. When I crossed R. yakushimanum to 'Mars', there was an occasional reddish one, but most of them were light colored. The reason for that is despite the fact that 'Mars' is a very good red, it has a lot of genes behind it that do not promote red. There will be segregation in those so that not so many of the gametes carry genes for red.
When one uses 'Vulcan', one is a little bit better off. 'Vulcan' is one-half R. griersonianum and that species is distinctly red. We get deeper colors, but they are still not red, though there are some deep rose colored ones.
What I perhaps should have done was to use R. griersonianum in crosses to R. yakushimanum. Rhododendron griersonianum, however, is very tender and has many other bad features, so if I were to use R. griersonianum, all I would want to get out of it is the red flower color and the ease of propagation.
If you study some of the propagation books and articles that have been written here in the East, you will find people saying that red rhododendrons are nearly always difficult to propagate. This is not true. It is only the red R. catawbiense hybrids that are hard to propagate. If the red comes from any other species, they are likely to root more readily. With me, any seedling that contains any appreciable percentage of R. griersonianum roots very easily.
If you cross R. griersonianum with R. yakushimanum, there are two characteristics one would hope to get: the red color without much blue in it, and propagatability. From R. yakushimanum you would want a good flower and truss, and hopefully freedom from blueness, plus indumentum.
I'd like to offer the Furmans a suggestion that they shouldn't blame 'Mars' and 'Vulcan' too much. [Robert and Audrey Furman of NY and Cape Cod had made their presentation earlier entitled: "Experiences in Hybridizing" earlier in this 1979 session.] Blame it on R. yakushimanum. That species contains the blue genes that persist through several generations. It's essentially a white species, but it carries all of the bluing genes that exist in the macro-species as it is now constituted, including R. metternichii and R. metternianum. Most taxonomists agree that R. yakushimanum is only a form of R. metternichii. [But see note in Editor's "Introduction".] The top dominant color for that species is rather a non-descript mauve color, and that's why it is so difficult to get clear colors with R. yakushimanum. It takes generations of selection before you get rid of those blue genes.
Toward Yellow Rhododendrons
I worked with carnations for about forty years, and I started that because my father was interested in breeding a yellow carnation. He got hold of one of the best yellow carnations, crossed it with anything he could, and never got a yellow. The various books say you should not inbreed carnations because that deteriorates vigor. That's true, but what those books failed to say was that you cannot recover recessive genes except by selfing. Fifteen years after I started working with them, I knew there were six genes concerned with the color in carnations. The top dominant color, crimson purple, requires six genes and recessive to all of those is yellow. You cannot recover yellow except be selfing or intercrossing combinations.
That's the problem we have with R. wardii and those yellow colors. They are recessive to the anthocyanin colors, and it is going to take a lot of crossing to recover good yellows. Dennis MacMullan [of Long Island, who presented "Breeding for Hardy Yellows" earlier in this 1979 session] asked before why R. wardii didn't produce better colors in the first generation. You must bear in mind that most species breed according to the typical form, and the best forms of R. wardii that we like to breed with are not typical of the species. Rhododendron wardii varies all the way from two or three highly colored, selected forms down to the so-called R. astrocalyx, which is only pale yellow at best.
Rhododendron astrocalyx is now considered to be only a form of R. wardii, and that pale color of astrocalyx seems to set the breeding pattern for the whole species. [Rhododendron astrocalyx is now R. wardii var. wardii.] Unless you cross those highly selected forms to other selected hybrids that don't have the blocking genes, you have to be satisfied in the first generation to get some that are rather light colored. Then one must intercross those hybrids, or inbreed a particular hybrid, in order to recover deep yellow.
It's going to take a good while to get these because we don't have any hardy yellows, as was pointed out. Of all the yellows that I have grown in Connecticut, the only one that I call yellow and which survives with me is 'Odie Wright'. All it does, however, is survive. I have had it for five years and it hasn't gotten any bigger in five years. It hasn't flowered in five years. I grow another as a tub plant in the greenhouse in order to use it for pollination, for it has good color and habit of growth and is almost hardy enough. In my opinion, if I could get a yellow one that was hardy in Connecticut with that color, it would be good enough.
Question Period
Dennis MacMullan: Isn't 'Butter' hardier? [See QB ARS, Fall 1978, v. 32:4, p. 222-223; and JARS, Summer 1994, v. 48:3,
p. 175 for differing origins of 'Butter'. Editor.]
Dr. Mehlquist: No. In my opinion it's less hardy. I haven't tried it, but I saw it at [Ed] Robbins' place and he had 'Odie Wright' and 'Butter' side by side. 'Butter' was not nearly as hardy as 'Odie Wright'. 'Odie Wright' was out in the open, whereas 'Butter' was hidden underneath another big plant. I wish I had taken the cuttings he offered me. He wanted to give me a plant but I didn't want to take it at the time. I knew that he prized it so much, and that if I took it and it got out among other people, he might feel that I had not done him a service. Actually, he let people take the flower trusses home from shows, and I am sure that there are more plants around than most people realize. I know who is propagating it for introduction, and he said he doesn't expect to make any money introducing it. It doesn't grow well and apparently doesn't propagate readily.
Dennis MacMullan: I have a good size plant of R. wardii astrocalyx [now R. wardii var. wardii]. It's hardier than the Kingdon Ward R. wardii variety. Would you think it might be intelligent to put pollen from Kingdon Ward R. wardii on R. astrocalyx and breed a generation that will probably not be as yellow as the Kingdon Ward or might be as yellow, but hardier?
Dr. Mehlquist: It might be yellow, but chances are they will be less yellow. But self that one and I think you can get fairly good yellows that are hardy. The only trouble is you will be at least ten years older by the time you get them.
Dennis MacMullan: It's better than being twenty years older and being nowhere.
Dr. Mehlquist: That is something that the geneticist has to do to produce material for further breeding with sufficiently good requisites so that other people can take it from there. Commercial people cannot afford to do this. For example, Mr. [Edmund] Mezitt of Weston Nurseries told me that he is not raising any crosses on any scale that are not good enough to sell 75% of the population. In other words, 75% of the population has to be saleable as landscape plants, otherwise he can't afford to do it, and he has been very successful so far.
When I look at some of my crosses, the cross between R. yakushimanum and R. catawbiense var. album, 90% of these are saleable as landscape plants. When I look at R. yakushimanum with my selected forms of R. smirnowii, actually 100% of them have been worthwhile as landscape plants. On the other hand, when I look at the 10,000 F2s I have between 'Vulcan' and R. catawbiense var. album, there may be 10% saleable. If I get a good red one out of it that is worth introducing, I don't care so much how many I have to throw away. If I were doing it entirely on my own, without the support of the State, I would have been bankrupt a long time ago. I would have to confine myself to crosses where most seedlings could be sold as landscape plants.
You can't get the government to sponsor a rhododendron breeding program, such as they do for wheat and other food plants. After all, you can't eat the rhododendrons. That's why you have so few geneticists or scientific personnel fooling around with ornamentals. They can't get the money to work with, so for the most part they have to work with ornamentals for the fun of it out of their own pockets.
Bob Schill: Did you say when you cross species with a hybrid you can use the species as the seed parent?
Dr. Mehlquist: I would use the hybrid as the seed parent. If you are crossing two hybrids of varying complexity, use the most complex as the seed parent, but whenever possible make the cross both ways. You will find that sometimes one parent will be far superior as a seed parent to the other, despite what you are thinking, and then go ahead and use that one in the future.
Furthermore, you find a given hybrid is very effective as a seed parent in one combination, and in the next combination gives virtually no seed. This is due to characteristics that are inherent in the hybrids. The only way you can find that out is to follow the simple rules that are laid down in the elementary genetics texts: always make the cross both ways whenever possible.
If you are interested in the world's simplest plant breeding book, read: Practical Plant Breeding by W. J. C. Lawrence [3rd ed. London: George Allen & Unwin, 1951]. Lawrence is a remarkable Englishman who started as a gardener and became a scientist. He wrote, in my opinion, the simplest and best plant breeding book that's ever been written. It was published in England and republished by MacMillan in this country. It gives you very simple, yet adequate, genetic information.
Perhaps in the future we will have to take more cognizance of chromosome numbers in rhododendrons. I have some populations that should be diploid, and I am sure are diploid except for an occasional plant that's obviously polyploid. It has much bigger leaves and it has much heavier flowers. If we have a hailstorm or windstorm, it still stands up, whereas the normal rhododendron flowers do not.
Of course, many rhododendron species are polyploids, but these are allopolyploids. Allopolyploids do not ordinarily show this increase in substance of flowers and leaves that is characteristic of autopolyploids. But that is a subject in itself.
