JARS v60n2 - Nothing New Under the Sun, Part II

Nothing New Under the Sun, Part II
Donald L. Craig
Centreville, Nova Scotia
Canada

In Part I of "Nothing New Under the Sun," Winter 2006, a letter from Clement Gray Bowers was reprinted along with comments. Part II includes a letter from Wendell H. Camp, a lecture by Dr. Bowers and comments.

Wendell H. Camp: Comments by M.J. Harvey
"Of Camp," Don Craig says, "we know little..." Maybe my memories will help a little. In 1959 or 1960 when I was a graduate student at Durham University, England, we had a visit from W.H. Camp. At the time the Botany Department, under the genial guidance of Prof. D.H. Valentine, was well known for its work on genetics and experimental taxonomy. My own project was a cytogenetic study of the evolution of a group of violets in which polyploidy had played a part. During my three years at Durham there was a series of visits from scientists who gave lectures and stayed for a greater or lesser time. As I remember, Wendell Camp stayed for about a week and had time to interact with the half dozen graduate students in the department at the time. The topic on which I remember him lecture was on polyploidy in the blueberries ( Vaccinium ). Since I was studying polyploidy we talked about the topic.
Camp was an impressive person. Quite large physically, he had a deep voice which he used to effect. He had no difficulty in expressing an opinion on almost any subject. I have always had this image of Camp as the archtypical "Big American" (this is a dated British concept from World War II). He liked to be the centre of attention. By contrast I was a quiet, somewhat shy English kid and looking back I was his perfect foil. I remember discussing my problem of getting my day-length-sensitive violets to flower. He grew violets. "When I want my violets to flower I set the dogs on the (light) timer to short days and they flower." Since we had no such growth chambers at Durham I was impressed. He enjoyed that.
The problem with Camp was that in his eagerness to impress, he frequently resorted to hyperbole - to allow his ideas to run ahead of the facts. I think that in his letter the statement "various of the rhododendrons are highly apomictic" comes under this category. I have myself, after twenty-five years of making hybrids, not come across any evidence that rhododendrons are anything but sexually reproducing. In the other letter Bower states, "I suspect apomixis in some plants of R. catawbiense ." My comment is that "suspect" to a scientist is a suspect word and I await definitive proof. None has been forthcoming.
As a cytogeneticist I should comment that the ease of making hybrids is not affected by differences in ploidy levels between the parents. This is common mythology, but what is being referred to is the undesirability of getting odd-ploid plants. For instance crossing a diploid with 26 chromosomes with a tetraploid (with 52) will give triploid offspring with 39 chromosomes. Triploids are notorious for being completely sterile or producing extremely few seeds, while being in many cases quite desirable as specimens. A good example of this is R. decorum cv. 'Gargantua', a triploid with 39 chromosomes - beautiful plant but useless for breeding.
There may indeed be occasional barriers to producing hybrids between two particular plants but this is related to incompatibility barriers which are genetic, not related to arithmetic differences between chromosome numbers.
I do not understand the remark "maximum largely lacks genes for anthocyan synthesis..." This may refer to the usual pale pink or white flowers of most maximum plants but must predate the discovery of the burgundy-coloured flowers produced on (some branches) of the Mt. Mitchell form. I am attempting to investigate the genetics behind the inheritance of this colour and its sporadic appearance and delayed appearance in some of its hybrids. I hope to have a partial report within a year.

Letter to A.W.S. Hunter from W.H. Camp

Mr. A.W.S. Hunter, Principal Horticulturist
Cytogenetics
Central Experimental Station
Department of Agriculture Ottawa, Canada
Aug. 31, 1953

My dear Mr. Hunter:
Sorry to have delayed answering your letter of 11 August, but my secretary has been on vacation and things have rather piled up.
Glad to note that you are contemplating going into a breeding program in Rhododendron , in the broad sense. I wish that we might be able to sit down together and talk over some of the things. However, being unable to do so at the present time - and correspondence being what it is - about all I can do is to make certain statements. If these are not clear, then you can fire back at me for amplification and further explanation.
Naturally you will build up your own program as you go along - and following the lines which develop. I have been "playing around" with the whole Ericlean complex for a good many years, and so have rather definite ideas regarding its potentialities and practical possibilities. Obviously, I cannot put myself exactly in your place; therefore all I will do is outline what I would do if I were in Ottawa and faced with a program of breeding in the genus Rhododendron .
Naturally, in such a program I'd have a little R. canadense about, but merely to salve conscience. In the northern part of its range it does come out onto what might be classed as "upland" areas; in fact, Rhodora is almost a door-yard "weed" in parts of Nova Scotia, especially near the coast. But then let us remember that the Nova Scotian coast is a fog-drenched area, and "upland" there can be physically about as wet as a midland bog. For the most part, after you had done something with it, you would be still faced with the fact that its progeny would likely require a constantly high water table, or at least abundant moisture in the soil at all times. Therefore, to my way of thinking, it is not a "primary" source of genes for ornamental material, since lawns and shrub borders usually do not furnish the desired moisture conditions.
Furthermore, Rhodora is a tetraploid, so far as our counts indicate. Its nearest relative, R. vaseyi , of the southern mountains, is a diploid - and so far we have yet to see a really successful triploid produced in the genus. In brief, so far as the members of the Sub-series Canadense is concerned, you would seem to be blocked from further breeding, using Rhodora as a parent, except by the accident of the production of a hybrid using non-reduced canadense [Hunter had scratched this out and written vaseyi ] pollen. Clonal selections of value easily could be made - for Rhodora is somewhat variable - but, as I said, after you had them you still would have a plant which required damper sites for success than the usual shrub border or home planting.
Inter-Series crosses have been made in Rhododendron , but usually these are of scientific interest only, or maintained by the avid "collectors" who like to boast of "rarities"; only rarely are they of real horticultural value. Inter-subseries crosses are more easily made (and I think should be tried more). But I keep coming back to my original thesis; as a sound horticultural object, anything with too many genes of R. canadense in it likely would be a difficult subject because of its moisture requirements.
Because of the above, I'd personally concentrate on various of the rhododendrons and azaleas, rather than on Rhodora. You mention using R. maximum . It is a good plant, but has several bad characteristics. In the first place, it tends to get large and "leggy" in age. Furthermore, it has the rather unfortunate habit of putting its flowers out after the advent of the new foliage, thereby hiding the bloom. A lot of people have thought of R. maximum as being more hardy than R. catawbiense . Apparently this is because maximum occurs naturally into New England, whereas catawbiense occurs only from Virginia southward. However, I have tramped many a mile in the rhododendron country in the southern Appalachians, and it occurs primarily at the high and more exposed elevations where the temperatures are actually much lower, and sudden fluctuations of temperature far more severe. Therefore, for my money catawbiense is the more hardy plant. The fact that it stands well in parts of New England when planted there seems to prove my contention. Furthermore, catawbiense makes a much better shrub than maximum and also puts out its flowers early enough for a real show. Bean, in his valuable Trees and Shrubs Hardy in the British Isles, classes catawbiense as the most valuable evergreen shrub ever to be introduced into England; certainly it is in the parental line of every really hardy rhododendron clone produced that stands into New England.
By all means use maximum in your crossings, but personally I'd concentrate on catawbiense hybrids, because of their much better horticultural habits. Furthermore, although the color of wild catawbiense may not be too satisfactory, the point is that it already does have genes for anthocyan synthesis, which maximum largely lacks. Therefore, you already are a couple of jumps ahead in your breeding program.
With the true rhododendrons, I certainly would give smirnowii a whirl. It is reasonably hardy. It hybridizes with the catawbiense group of hybrids, and so gives a good base for jumping off in a breeding program. Aside from its floriferous habit, with showy terminal clusters, the thing I like about smirnowii is the heavy under-face tomentum of its leaves. It is so heavy that it actually deters the usual rhododendron pests, like lace-wing fly, etc. From what I can gather, and based on incompletely controlled pollinations, this is a recessive character; therefore you will probably have to depend on the F2 and back crosses to recover the leaf pubescence (but then it is in these combinations and segregations where one would get the interesting colors anyway, so one will not be behind in a breeding program - it just will take a larger series of plants to achieve success).
Frankly, I have been trying to get my hands on a good series of smirnowii , but so far have had little success. Don Wyman, at the Arnold Arboretum, has a few plants and possibly might be able to furnish seed (although there is a likelihood that it would be open pollinated and therefore possibly mixed with the catawbiense genes from adjacent plants).
For the azaleas, I'd head right for R. roseum ¹ as a basic stock. This has been sadly neglected as a parent in breeding. It is upland, and therefore likely to be able to withstand summer drought better than most that so far have bemused. The sad feature of the Ghent hybrids is that, basically, they have been compounded from lowland swamp forms, and further selected in mild European climates under conditions in nurseries where there is ample soil moisture at all times. As a result, but few of the Ghents are what I call sound horticultural objects, except under the very best of conditions.
And now for R. roseum . Actually, it hybridizes with nudiflorum ² ; therefore a stand of material may be quite variable. R. nudiflorum may be quite roseum-like, and roseum may be quite nudiflorum-like. But from your standpoint this wouldn't matter so much at first. The real point is to get your breeding plants from sites where they are subject to summer droughts. Also, we have to face the simple biological fact that roseum is not the same thing at the ends of its range. In brief, in your case, I'd probably not bother much at first with roseum from the middle to upper southern Appalachians, or the material sold as "roseum" from even lower altitudes in our south - it just isn't good roseum for your purpose; in fact, but little of the stuff sold as "roseum" is that at all! Instead, I'd make every effort to get good material from (for example) the Green Mountains of Vermont. Of course this would have to be collected from the wild, and this would mean that you would have to do a little scouting and take the plants back with you. As you know, they move rather easily at flowering time, and if they already have shoved out leaves it is a simple matter to cut the plants back, for they will sprout again and even set buds for the next year in many instances.
Using good hardy parent stock of known source of both catawbiense and roseum , you then will have your basic materials to start a breeding program. Naturally you will wish to go on from there. This will entail bringing together horticultural clones of merits as well as other material; probably collected from the wild. For ease in breeding, it is likely the homoploids would be best; therefore, you will want to consult the list of counts published by Janaki Ammal, et al., in the R.H.S. Rhododendron yearbook for 1950, our most recent compilation.
For breeding work on the genus in Ottawa, I strongly suspect that you will need a fair sized lath house to hold some of the less hardy breeding stock and also to take care of the large number of seedlings which will be coming through. This ought to have special protection on the north and west sides to further cut down the effect of drying cold winds. Furthermore, it would be best to have a cool greenhouse, even of modest dimensions. Here you could keep other stock which is definitely tender at Ottawa, but which possesses valuable genes. Such a house would have heat available (and if possible thermostatically or otherwise carefully controlled). Thus some of the semi-tender sorts can be easily kept, the temperature lowered sufficiently so that they will harden up and then break dormancy satisfactorily, but yet not subjected to really low temperatures. This might also be accomplished in a series of deep pits, but I have always felt that rhododendrons prefer the cool greenhouse rather than the pit; the deciduous azaleas don't seem to mind too much. (Actually it is my hunch that the pits tend to overheat, if not carefully watched, much more than the larger greenhouse, and that this as much as anything is the source of the trouble.) However, if a cool greenhouse isn't available, then I'd certainly settle for a few deep pits and a good gardener to watch them, on sunny days especially!
The temptation would be to plant all materials in the ground. Actually I'd do this with about half the breeding stock. But if possible, I'd like to put the other half in stout wooden tubs that can be moved about. The flowering season of the various materials are quite different and one sometimes likes to force (or hold back) flowering a bit on breeding stock. However the nice thing about the pollen of the genus is that it stores easily if properly handled. Therefore, the pollen from early bloomers can be held for the later sorts. Even so, it is difficult to hold pollen more than a few months (at least I wouldn't feel too comfortable about it); therefore in reciprocal crosses it is nice to have both plants in flower at the same time, and that is where the tubbed plants come in handy. Also, one sometimes wishes to move breeders about a little, and a tubbed plant can be moved. One need not train them into shapely bushes, but can selectively prune them to easily handled specimens, perhaps a little leggy, but with shapes which will permit a larger number to be handled in a lath house, cool house, frames, or in the open. Of course, tubbed plants have to be watched with much greater care, and the lack of a single watering will sometimes spell disaster to a breeding program through loss of parent stock. But with half your breeding stock in the ground and the other half of the same clones tubbed, you ought to make out in an emergency.
Lastly, I would turn specifically to breeding in the Sub-series Luteum. Here is where I will toss the ball to someone else. Some years ago, Henry Skinner, then curator of the Morris Arboretum, came to me regarding the possibility of a study of our native American azaleas. We talked the whole situation over, and turned over to him my ideas on the group for what they were worth. Skinner went on from there and did his doctoral dissertation under my direction on this general topic. Today Skinner knows more about the American azaleas than any person alive. Furthermore, he has a card file of nearly every specimen of azalea ever collected in the wild in the eastern half of the continent, the card with a full analysis of the plant's characters. These are on a Key Sort system, so that he can get out information on any material he desires by character or geographical locality. His thesis (as yet unpublished) dealt primarily with the southern, early pink-flowered complex, but in dealing with it he had to go far afield in his studies. Furthermore, Skinner for many years has been breeding for good characters as well as hardiness in the mollis complex (also of the Luteum sub-series), and so knows many of its problems. I suggest that you get in touch with Skinner, for he doubtless can give you further tips on procedure. He may now be addressed as follows:
Dr. Henry T. Skinner, Director National Arboretum
U.S. Department of Agriculture
Washington 25, D.C.
It is likely that I haven't answered all your questions - and probably I've raised more than I answered - but the unanswered questions that nobody knows are the interesting ones in plant breeding. As a final note, I'd add that Skinner brought back about 500 living plants, mainly from the southern states. Only part of these have been worked over yet. However, it is quite obvious that polyploidy is far more abundant in our native materials than we had supposed (a further complication in breeding), that what we have long thought to be the same "species" contain both diploids and tetraploids, and that no current taxonomic treatments of the native azaleas (until Skinner gets done with them!) have more than a hint of reality. In brief, if you attack your problem from the standpoint of the development of good horticultural material, you will get farther than if you worry about which "species" you are crossing. Also, it is becoming patently obvious that various of the rhododendrons are highly apomictic - and this always gums up as breeding program. Whether there is apomixy in the azaleas is not yet known, although I have sometimes suspected it among the polyploid forms.
Anyway, good luck to you in your program.

Sincerely yours,
W.H. Camp
The Academy of Natural Sciences of Philadelphia
19th and The Parkway
Philadelphia, 3 PA.

Comments by John Weagle
(Some specific comments to the text. Comments in italics.)
Camp could easily have ruled the common R. canadense out entirely but he knew a good plant and this azalea is indeed worth mentioning. It is highly prized in the U.K. yet inexplicably still neglected in North America, lamentably even here where it is native. It was a surprise to me that Camp knew the coast of Nova Scotia and its climate. Locally it will tolerate ordinary garden conditions; elsewhere it is not so easy and requires the moisture Camp suggests as being crucial to its long-term well-being. Low elevation heat could well hasten its demise.
"However, I have tramped many a mile in the rhododendron country in the southern Appalachians, and it ( R. catawbiense ) occurs primarily at the high and more exposed elevations where the temperatures are actually much lower, and sudden fluctuations of temperature far more severe." A good observation, most northerly is not necessarily the hardiest!
"In the first place, it ( R. maximum ) tends to get large and 'leggy' in age. Furthermore, it has the rather unfortunate habit of putting its flowers out after the advent of the new foliage, thereby hiding the bloom." While I have not followed maximum hybrids closely surely these "defects" could be "cured" in several generations.
"I certainly would give smirnowii a whirl...Frankly, I have been trying to get my hands on a good series of smirnowii , but so far have had little success." The smirnowii recommendation of course, should have had a caveat as many smirnowii are not terribly bud hardy and many more are simply awful, gangly plants. In 1953 this variation in its hardiness was probably not clearly understood. Some 50 years after Camp complains of the same problem, I too would not know where to find seed of a thoroughly hardy, good-looking smirnowii selection. Our smirnowii are scraggly with loose trusses. Warren Berg's and Powell Huie's R. smirnowiis (are they the same?) are compact and outstanding foliage plants but their climates tell us little of their usefulness in breeding for very hardy hybrids. Rad Pike's smirnowii which produced Kentville Research Station's beautiful 'Bellefontaine' ( smirnowii x fortunei ) surely must have been a good one. Dr. Craig's notes state, "Pike's records of the New Hampshire rhododendron and azalea breeding programme states that this smirnowii came via Reefe Point Gardens, Bar Harbour, Maine, second generation in Maine from plants from Edinburgh Botanic Gardens, Edinburgh Scotland, and the fortunei from hand-pollinated hybrid seed from Reefe Point Gardens. Female parent from seed from Edinburgh Botanic Gardens." If that smirnowii is still around it should be propagated without delay.
"From what I can gather, and based on incompletely controlled pollinations, this is a recessive character; therefore you will probably have to depend on the F2 and back crosses to recover the leaf pubescence (but then it is in these combinations and segregations where one would get the interesting colors anyway, so one will not be behind in a breeding program - it just will take a larger series of plants to achieve success)." Wonderful advice and yet to this day how many of us stop at the F1, name the plant and sit back quite pleased with our new hybrid? Shame!
Of Camp's azalea comments I can only say that here in Atlantic Canada we regrettably have not yet assembled a good collection of the selected forms of the native Eastern azalea species and they do so well here, a lamentable situation. One can only wonder what became of Skinner's Key Sort system with all that invaluable information. This requires some follow-up.

Lecture presented to the N. Y. State Nurserymen's Association at Ithaca, N. Y., August 11, 1949.
By Clement G. Bowers, Ph.D.,
Research Associates Cornell University, Ithaca, N. Y.

Modern Rhododendrons and Azaleas
No woody plants are more spectacular than those found in the genus Rhododendron , which includes azaleas. If they do not rank in first place as commercial nursery materials it is due to two things: (1) lack of quick propagation methods, and (2) lack of adaptation to all climates and soils. Progress is being made in both these directions. There is no question as to the possibilities which these plants possess in regions favorable to their growth such as those prevailing in England and on our West Coast. It is a far more difficult task to produce similar results in places like our North-east. Yet here in New York and New England is where we need them most and where the best market exists. So any progress that can be made along such lines is important.
It was recently my privilege to see some of the best things in the rhododendron worlds first on a trip to the West Coast and this spring in England where I attended an international conference on rhododendrons. I have also seen some excellent new hybrid material, both rhododendrons and azaleas in Boskoop, Holland, and the Eastern U. S. While much of this is of a nature that can never be duplicated outdoors in New York State it presents a good view of the ultimate possibilities inherent in this remarkable group of plants. I shall give you a graphic report on this mainly by means of color photographs since it is quite impossible to convey an adequate description of such material without actually seeing it.
To those of you who are familiar with only those sorts which we have been in the habit of growing here for the last 70 or 80 years the new developments among the evergreen rhododendrons of England should prove quite startling. While we have been forced to cling to the Catawba hybrids and their kin the British breeders divorced this group many years ago and have developed several other races some of which, to be sure, bear a bit of Catawba or ponticum "blood" but mostly based upon such groups as the Fortunei series, which are large, fragrant and loose, and the yellow campylocarpum , the bright red thomsonii forms from the Himalayas and (just recently) the brilliant griersonianum hybrids, the peach-colored dichroanthum progeny and the various neriiflorum allies. These may all be said to be somewhat akin to our R. catawbiense and R. maximum forms but, in their best races, quite free from the all-pervading lilac or bluish-crimson tinge that characterizes all Catawba hybrids. In addition to these there has been great development among the so-called lepidote sorts comparable to R. carolinianum ³ in the medium-sized shrub range with spectacular bell-shaped orange-tinted flowers like 'Lady Chamberlain' (a hybrid of the Cinnabarinum series), the various Triflorum species (which almost cover the spectrum in color) and a whole group of excellent dwarfish and rock garden species. One interesting species is R. augustinii which in a very few superior individuals shows flowers which are more nearly blue than one would ever expect to find in the genus. It is not deep blue as sometimes described but is a sort of grayish blue or light periwinkle blue. And don't be deceived: this blue occurs in only a few fine individuals while the others are unattractive washy lavender. So if, for any reason, you should be seeking this blue form, get only the clone which is definitely blue and don't expect to get it from seedlings.
This brings to mind another matter of importance, namely: the fact that among the new classes, the individual or clone is often of much more horticultural importance than the species to which it belongs. With natural species from the Orient, variation is very great among seedlings. Add to this the accidental crossing that occurs in gardens and you will find additional divergence from the main type. Moreover, the British gardeners have been following a practice, which we hope may eventually be outlawed. All progeny resulting from the crossing of two species is given one name and all the brother and sister seedlings, no matter how variable, bear this same name. Thus, a cross between R. griersonianum and R. dichroanthum is called by British gardeners 'Fabia' and all the seedlings of this combination are called 'Fabia', too, although some are red and others are orange, with considerable variation in size and excellence. The only safe way is to depend only upon clones, which are grafted or layered plants of proven excellence.
These new things, however worthy, do not offer much for direct utilization in the Northeastern United States. Almost all the new species were tried out, as they appeared, by Sargent, Wilson and Judd at the Arnold Arboretum, and while many that failed in such a rigorous climate have succeeded in sheltered places on Long Island and elsewhere, it is only the rare individuals that show evidence of adequate hardiness. Careful nursing has made it possible to grow a number of promising things here in upstate New York, but until clones or races are developed that are reliably hardy they cannot be recommended as sufficiently "foolproof " for good commercial use.
A good deal might be said about the relation of rhododendrons and azaleas to site, situation and climate, so that it is always risky to lay down general rules. More definitely than roses, the specific kinds of rhododendrons and azaleas must be geared to the special conditions of the locality where they are grown. Thus there cannot possibly be a list of "best" rhododendrons for this country in general, or even for England. Merit ratings apply only to behavior in a definite place, and so, to a certain extent do hardiness measurements as these appear in the rhododendron literature. Owing to the prevalence of late spring frosts in England (a killing frost in London this May 12th!), the British growers measure hardiness by resistance of the buds and leaves to frost which is quite a different thing from plant hardiness required of species in New York sub-zero weather. Thus, some species marked "C" in England may prove to be hardier in America than those having the "B" rating. I look forward to the time when we will list our plants as suitable for limited regions only; for example, those which are recommended for Central New York and New England, as distinguished from other groups which are recommended for New York City and Long Island and Philadelphia, the mid-Atlantic States, the lower South, the far West, etc., each region being given its own list of "best" plants.
There is no questioning the merit in England of the myriad of new rhododendrons and their hybrids that have been getting Awards of Merit and First Class Certificates of the Royal Horticultural Society, but, in my opinion, comparatively few of them are going to be of any direct value to us in the East. A few, however, are going to be invaluable in improving our hardy races through hybridization, and this is a project upon which we have embarked. Of the 900-odd species now recognized in the genus, however, this may fairly be said: Not too many are worthy of horticultural attention. There are scores of species, distinct perhaps, but all so uninteresting and unprepossessing that no one but a collector will want to bother with them after their luster of novelty has worn off. The whole group is in need of further study and it seems reasonable to predict that a reclassification will ultimately reduce the number of recognized species to considerably less, a good many being finally listed as mere geographical or natural hybrid forms.
We are going to be more and more conscious of new hybrid introductions within this genus, for we are bound to hear of them, and it will always be good to remember that there is generally some flub-dub over a new thing which may not actually be too significant. At the same time, we always need to be alert to hidden virtues which some otherwise unremarkable plants might possess, such as disease resistance or hardiness. Just as R. carolinianum does very poorly in England, so many of their superior sorts do poorly here, and it will take a great deal of testing and breeding, right here in the East, to finally develop new races and strains "tailored" to fit local conditions. But this is a good motto to adopt for the breeding of any of our woody plants if they are to be durable and good, for we can never expect breeders in England, California or other distant places to do a thoroughly satisfactory job for us where conditions are so different. It is also worthy of note that some plants which look unpromising actually make superior parents. Only experience and study-book records can reveal these.
While we are awaiting better plants in the future, a few new sorts are in the offing and worthy of note. The Dexter hybrids were produced on Cape Cod and belong to the Fortunei series. They are promising for the Metropolitan region and Philadelphia, with some possibility of extension elsewhere. Some of the flowers range up toward six inches in diameter - huge, lily-like things in loose trusses with delightful fragrance, mostly blooming with R. carolinianum in May and almost as good as the British races. They lack dark, rich colors, but have many fine delicate tints, notably yellowish and apricot tinges, with a freedom from purple. I believe several good forms will be introduced soon under clonal names, and these will be better than most seedlings. Certain other species and hybrids are now under trial in various places, some of which appear promising and which, in time, will be described by enthusiastic sponsors. Unfortunately, stocks multiply slowly and few statements are safe until more widespread observation of novelties is attained. Besides these, there are a few good things which are obviously hardy, such as new white clones of the Catawba rhododendron, azalea vaseyi and certain others of our native species. One of the best is a late red azalea, doubtless allied to R. calendulaceum but differing in later bloom and a fine round flower truss with darker red flowers, called cumberlandense . This is hardy in Central New York.
We need to make a reappraisal of our native deciduous azaleas. While those belonging to the Obtusum subseries ⁴ are making big displays around New York City and are highly regarded because they propagate easily from cuttings (I refer to such as amoena ⁵ , 'Hinodegiri', mucronatum and others), we cannot be sure of any upstate except R. kaempferi , which kills to the ground here, and R. poukhanense . The new Glenn Dale hybrid azaleas are marvelous, but again cannot be expected to be hardier than their hardiest parent, R. kaempferi , with many varieties less hardy. For the very cold regions, there are no azaleas left but the natives. The virtues of these, however, have not been fully recognized nor realized. If you are selling ericaceous plants in subzero climates, such as that which prevails here, I cannot urge you too strongly to raise seedlings from the best forms of R. roseum , R. calendulaceum , R. arborescens , R. vaseyi , R. atlanticum , R. viscosum and possibly hybridize inter-se among these species. You will find that some of these are far more satisfactory than evergreen rhododendrons, and very spectacular when in bloom. It is important that they be on their own roots. Many of the Ghent hybrids, too, will probably prove perfectly hardy when on their own roots which in grafted form have not appeared satisfactory. In England a very superior group of Ghent Hybrids, called the Knap Hill hybrids, of which an improved strain is now being bred by Slocock of the Goldsworth Nurseries, surpasses anything I have ever seen in this line and are worthy of a trial in this country. If they do not possess too much "blood" of the Californian R. occidentale some of them may be good for sub-zero climates.
The oriental deciduous azaleas, the Mollis hybrids and R. japonicum , a close relative from Japan, are beautiful things with larger flowers than the preceding and fairly hardy in some of their forms, of which the japonicum is the hardiest. My observation is that, while many will survive sub-zero weather if on their own roots, they are not so hardy as the native deciduous azaleas and the Ghents. I am also of the opinion that the old wood becomes tender and the plants need to be cut back and allowed to break again from the base occasionally in order to renew vigorous wood. Persons from colder climes going south return with enthusiasm for Southern azaleas, but are unable to grow them at home. If some of these very hardy sorts were better known and available I believe a ready-made market exists which would make them very popular.

Gray Bowers, Ph.D.

Comments on Modern Rhododendrons and Azaleas Lecture by John Weagle
(Some specific comments to the texts. Comments in italics)
"If they do not rank in first place as commercial nursery materials it is due to two things: (1) lack of quick propagation methods." How times have changed. Propagation methods have greatly improved and rhododendrons are still more expensive than spiraeas. The tissue-culturing of rhodos and azaleas, even with its inherent problems, has at least made the proliferation of many new cultivars very easy. While the price of a rhododendron is relatively high it is certainly not excessive.
"While much of this is of a nature that can never be duplicated outdoors in New York State it presents a good view of the ultimate possibilities inherent in this remarkable group of plants." The extreme summer heat and steam of the eastern states present problems we thankfully don't have to concern ourselves with when breeding rhodos on the Scotian coast - we can merely focus on vigour, cold hardiness, wind tolerance, insect resistance, an off-sync growing period and wildly fluctuating temperature adaptation! Breeding for heat tolerance may take a few extra generations to overcome for those so afflicted. The UK still stands as the model of aspirations for all of us.
If the super-hardy brachycarpum (as subsp. tigerstedtii ). had been recognized it could have been used to cross with everything delectable and tender - great possibilities for our coastal areas. While many of these primary crosses would not be finished plants they could be used in next generation combinations. The brachycarpum (as Roseum Group) could have been used for dwarfs. We have lost a lot of time by neglecting this species.
"Their best races quite free from the all-pervading lilac or bluish-crimson tinge that characterizes all Catawba hybrids." This state of affairs was to change as breeders started using white forms of R. catawbiense . And then reports from the 1990s told of first generation crosses using type R. catawbiense that threw clear colour untainted by this so-called dreaded lilac. Steele's cross of R. catawbiense 'Roan Mountain' x 'Rosa Stevenson'* comes to mind as the cross produced yellows and pale salmon. Was the summer heat of the eastern USA knocking out these colours in favour of the progeny with heat tolerant lilac genes? One wonders.
"There has been great development among the so-called lepidote sorts comparable to R. carolinianum in the medium-sized shrub range with spectacular bell-shaped orange-tinted flowers like 'Lady Chamberlain' (a hybrid of the Cinnabarinum series), the various Triflorum species (which almost cover the spectrum in color) and a whole group of excellent dwarfish and rock-garden species." We now know that the limitation of the lepidotes was summer heat intolerance more so than a lack of hardiness. Sadly few hybridizers in North America were to utilize the dwarf lepidote species to create cold hardy hybrids. We can now thank Glendoick for their efforts.
"Of the 900-odd species now recognized in the genus, however, this may fairly be said: Not too many are worthy of horticultural attention. There are scores of species, distinct perhaps, but all so uninteresting and unprepossessing that no one but a collector will want to bother with them after their luster of novelty has worn off." I think Bowers would be amazed how interest in species has grown. In twenty years the public will demand many of these traits in garden centres.
"We need to make a reappraisal of our native deciduous azaleas." This is an excellent recommendation. Rhododendron bakeri and R. calendulaceum are especially rare in the trade. For Canadians the market is too small for someone to produce them and make a living. Furthermore our only sources of selected forms of azalea species are in USA areas with Japanese beetle. Damaging insecticide drenches are required. We therefore must grow them from all too rare seed.
"I cannot urge you too strongly to raise seedlings from the best forms of R. roseum , R. calendulaceum , R. arborescens , R. vaseyi , R. atlanticum , R. viscosum and possibly hybridize inter se among these species." This should have been apocalyptic for Eastern rhodo enthusiasts! Our gardens are full of hybrid azaleas plagued with mildew. These native species are both immune and exquisite in their best forms.
"With natural species from the Orient, variation is very great among the seedlings." Thanks to the P. and K. Cox and other inveterate plant hunters this point is finally being recognized in the West. The species cannot be neatly button-holed based on a few plants.
"Thus there cannot possibly be a list of 'best' rhododendrons for this country in general, or even for England." This was certainly a surprise and very true. In the last few years the ARS has addressed this issue by publishing lists of regional best performers. How many plants were killed to come up with these lists? One shudders to think! The performer's list for our Atlantic Chapter is problematic. The range of climates in our region is so vast that the next logical step is to come up with lists within the chapters.
"It will take a great deal of testing and breeding, right here in the East, to finally develop new races and strains 'tailored' to fit local conditions." Again, if you want to grow a species at the edge, go out and track down the highest altitude or most northerly seed source you can find and grow a patch of seedlings. By cold-killing a good percentage of them the survivors may very well be dependable. No one else will do it for you; if they do the plant may well be okay for them 2 km. away and perhaps not for you.
In paragraph nine Bowers makes some predictions for the Dexter hybrid - Timely comments as it was to take another twelve years for the Dexter Research Committee to objectively look at the Dexter group. Yes the Dexter colours are indeed all very similar, pale and subtle for the most part.
Regarding evergreen azaleas Gable certainly recognized the hardy attributes of poukhanense and kaempferi. He had combined these two in his 'Big Joe'* and then went on to the F2 which produced 'Springtime'*. These two hybrids are quite satisfactory here and appear to be hardier than either parents thanks to Gable's field testing. Yet both have been neglected somewhat in this area both as garden plants and as parents to develop hardy hybrids. R. kiusianum , probably the toughest of the lot, was still to be introduced and yet to this day few have crossed it with the more desirable borderline named evergreen sorts.
"I am also of the opinion that the old wood becomes tender and the plants need to be cut back and allowed to break again from the base occasionally in order to renew vigorous wood." Now this is a bit puzzling to me, the "becomes tender" bit. I see it as "loses steambecomes senescent." Still this old wood does finally give up the ghost with what appears to be bark split. Perhaps it does become "tender." These azaleas must be grown vigorously with regular feedings and copious annual applications of mulch. Old sluggish wood (typified here by a covering of dense lichen growth and little shoot extension) should indeed be cut out on a regular basis, just as you would forsythia. Nevertheless it was good to see this seldom mentioned point made.

Comments by the late August Kehr (20 March 2001)
It was somewhat akin to turning back the pages of time to read the letters of Clement Bowers and W.H. Camp written in 1953. Yet this half-century old correspondence could easily be undistinguishable in their content from similar letters written only yesterday by some modern day plant breeder. There is still a vital need today of knowledge of inheritance of important genetically controlled plant characters. It is probable that in his initial letter, that was sent out in August 11, 1953, Dr. A.W.S. Hunter was seeking guidelines on breeding for improved very hardy cultivars and, especially, genetic answers. The responses of both Clement Bowers and W.H. Camp appear to be attempts to provide answers.
There is a real reason why the knowledge of the genetics of rhododendrons and azaleas was primitive in the early days, and still is today. It is because one can count on one hand the number of truly genetic studies in the genus Rhododendron . Without doing research on the literature, I can recall only a single good one, that done by Dr. J. Heursal (The Institute of Ornamental Plant Breeding, B-9230 - Melle, Belgium) on the colour inheritance in evergreen azaleas.
Plant breeders are not geneticists, and geneticists are not the reverse. There is a world of difference between plant hybridizing and genetic studies. Hence plant breeders will never discover much of value to an understanding of genetics, while geneticists will, only by pure accident, develop a plant worth the while to name and register. Let us discuss the reasons for these differences.
A plant breeder will chose as parental plants only those that he believes will result in improvement in horticultural characteristics and will select only parental plants that he thinks will best meet those objectives; while a geneticist is totally uninterested in the horticultural potentials of the parent plants he selects, but he chooses only those that differ in some recognizable genetic characters. The geneticist will grow out a population, and afterwards carefully will note the nature of each and every plant in order to calculate mathematically the number in every category of characteristics that he can discern. The plant breeder has no interest in numbers of plants in the progenies, and will happily discard roughly 95-99% that do not meet his needs. This same plant breeder is not always careful to avoid contamination by stray pollens, and such "stray pollen" may actually enhance his chances of reaching his objective. I have often noted the mistaken attempts of some avid plant breeder (emulating a geneticist) to do all he could to avoid "stray pollen," as though it would ruin his breeding objectives. For a geneticist, on the other hand, to have any stray pollen contaminate his crosses could completely ruin his results, and he must do all he can to totally prevent any pollen contamination. In brief, the objectives, methods, plant selection of the geneticists and breeders are diametrically opposite. Is it any wonder that plant breeders will seldom come up with genetic information worth putting into print? This may be a point overlooked by both W.H. Camp and Clement Bowers. They were understandably bewildered about such things as lilac coloured plants appearing in the progenies of pure white forms of R. catawbiense , just as are plant breeders today often likewise bewildered. The answer lies in the fact that there are some recessive genes that suppress all colour in plants that receive these genes from both parents. For example Joe Gable noted that 'Indica Alba' ⁶ is a fine pure white flowered evergreen azalea, but when it was crossed with any other coloured azalea, produced 75-100% purple offspring. Although Joe noted these results he could not give an explanation for them. In the same manner the pure white forms of R. catawbiense can carry the same recessive genes which will hide lilac colours. Such colours are present in the germ plasm, but are not expressed until the genetics are right. It was the same hidden genes that played havoc with my own attempts to develop a yellow flowered evergreen azalea. After many years I developed a breeding line, called 415-1, that had both the recessive colour pattern, along with no hidden colour genes other than the genes which give cream or yellow colours. Breeding line 415-1 is a tetraploid, and crosses readily as a pollen parent on yellow forms of R. calendulaceum , the southern native deciduous azalea species, as well as many diploids used as seed parents. Please note that the yellow plants are used as the seed parent, because most yellows are that colour as the result of carrying the yellow colour as discrete yellow plastid bodies in the cytoplasm of the ovules. Most geneticists now believe that the plant pollen cells have no cytoplasm, and therefore cannot transmit the yellow plastids to their progenies. Unfortunately in the development of breeding line 415-1 new genes "popped up" that resulted in plants with non-functional ovules and were female sterile but male fertile. The inheritance of the female sterile gene is baffling to say the least, and has made the programme of developing a yellow flowered evergreen azalea more difficult. However, breeding line 415-1 is still key to success in this project and can be used as a pollen parent with deep yellow seed parents (which are thought to carry the yellow plastid bodies in their cytoplasm). To anyone who wishes to develop yellow flowered evergreen azaleas, I suggest crosses such as R. keiskei , R. austrinum , deep yellow hybrids, and deep yellow forms of R. calendulaceum be used as seed parents, with pollen of 415-1. The tools for such a programme are now available as the clone exists in several gardens in the Southeastern Chapter. To obtain cuttings only interested breeders should contact president Patricia Phillips.
Bowers made crosses of R. catawbiense x R. calendulaceum , a cross no one since has made to my knowledge. Perhaps, we need to try harder. He also suspects apomixis exists in the genus Rhododendron . I have never found any credible evidence that it does indeed occur. Rhododendron maximum matures its pollen long before the flower opens, and it is probable that flowers are self pollinated in the bud stage, thereby giving the illusion of apomixis. He also alludes to the work on polyploidy by E.K. Janaki Ammal. I tried repeatedly, including once while in India, to learn her methods of chromosome counts, to no avail. She never replied even when I sent the postage. Furthermore, there are many examples of both diploid and polyploid forms existing within the same species. For example, R. occidentale is reported to be a diploid species, yet recent counts of that species indicate hexaploid types appear in nature. Both diploid and tetraploid of the native American azalea species are also known. We are hindered even in 2001 by lack of good cytological studies in Rhododendron, as well as the lack of good genetical studies. Thus, unlike Camp's statement that triploid forms did not exist in his experience, it is very probable that he actually developed some (which were sterile), even in 1953; even today there is not a hybrid with R. canadense (other than x fraseri ⁷ ) as he suggested, not even with R. vaseyi .
Camp seemed to favour R. catawbiense over R. maximum as a parent. It is interesting that the discussions of the relative merits of these two species is still viable fifty years later. He seems to be "right on" with his methods of developing hardy types. John Weagle's suggestion of using the Tigerstedtii form of R. brachycarpum , tops for hardiness in the large leaved rhododendrons, is better yet. I certainly concur fully with Camp's evaluation of Henry Skinner's research on native azaleas, which fifty years later is unmatched in its accuracy and completeness, just as it was in Camp's time.
It is a stimulating experience to read both of these half-century-old letters. It is good to know what happened in the past so that we can move ahead. Knowing past experiences helps us to avoid some of the pitfalls of the present.

Chromosome Numbers and Polyploidy in Rhododendrons by M.J. Harvey
My other comment is concerned with the subject of chromosome numbers and polyploidy which is touched on in several places. I am horrified at the widespread lack of knowledge of polyploidy: what it is, and how it affects rhododendron breeding. Articles in the ARS Journal frequently contain misunderstandings on the topic.
A polyploid is a plant with one or more additional sets of chromosomes compared with the diploid which contains two sets (one from each parent). In Rhododendron , cells with 26 chromosomes represent the diploid number - the egg and the sperm having contributed 13 each. The number 13 is regarded as the base number in the series which is often abbreviated 'x'. Thus cells with 52 chromosomes are said to be tetraploid (4x), those with 78 are hexapod (6x), and 156 (in R. manipurense ⁸ ) is 12x. Uneven levels are possible especially triploids (3x with three sets per cell). As already mentioned these latter are usually sterile or have greatly reduced fertility.
Over my career I must have counted the chromosomes numbers of several thousand plants. Not one of these has been a rhododendron although I have tried. The problem with Rhododendron is a technical one. Not only do they have very small chromosomes but they tend to be "sticky." That is, the chromosomes clump together making it impossible or very difficult to distinguish the individual units. The difficulty is not confined to me. Augie Kehr could not get counts and David Leach had great difficulty. For photographs of rhododendron chromosomes see Leach's article "The Nature of Evidence for Hybridity in Rhododendron yakushimanum," JARS, Vol. 42, Fall 1988, p. 208. When I was being taught cytology the standard was that you had to have a minimum of six cells in which the chromosomes could be counted unambiguously in order to publish a number. This is a very difficult standard to meet in the case of Rhododendron .
I mention the above because in 1950 a team of researchers published a magnificent list of the chromosome numbers of 369 Rhododendro n species and varieties. This list is more than four times greater that the sum of all others counts in the genus and is the foundation of our knowledge of rhododendron chromosome numbers. I once mentioned this to the head of chromosome research at the Jodrell laboratory in Kew Gardens where the research was done and he admitted to being puzzled as to how it had been accomplished. We speculated as to whether pollen grain or stomatal size had been used to estimate the ploidy - the volume of a cell is related to the number of chromosomes in it.
Now in science, as in most professions other than politics, dog does not eat dog. You do not criticise the work of another scientist unless you have really definite proof. Thus I have kept quiet about my unease regarding Rhododendron chromosome numbers. I never discussed the matter with Augie Kehr so I was surprised to see his remarks of his difficulties of getting in touch with Dr. Janaki Ammal who led the original project, to apparently ask her the same questions that had occurred to me and others.
I used to tell my students that the most concise definition of science was that it is the study of repeatable natural phenomena. This cuts out anecdotal stories and miracles. The emphasis on repeatability which is at the root of science refers to an experiment being in theory repeatable by other people in other places. If it isn't repeatable then it doesn't come within the purview of science. We are dangerously near that cutoff point in the case of the Rhododendron chromosome numbers. I would like to see a review of this work preferably using one of the now available methods of measuring the amount of DNA in a cell, thus avoiding the difficulties of separating chromosomes for examination under a light microscope.

Comments by Peter Tigerstedt
(Some specific comments to the texts. Comments in italics).
Letter (Camp): The discussion on maximum and catawbiense is interesting to me and especially the note: "A lot of people have thought of maximum as being more hardy than R. catawbiense . Apparently this is because R. maximum occurs naturally into New England, whereas catawbiense occurs only from Virginia southward...it occurs primarily at the high and more exposed elevations where the temperatures are actually much lower, and sudden fluctuations of temperature far more severe." This is a good example of the fact that the south-north gradient is not conclusive for hardiness. In fact large fluctuations are typical of transitional climates between continental and maritime and in such pockets one may often find very pronounced cold hardiness. This transition is also typical for Finland between the maritime Atlantic and the continental Siberia.
Lecture (Bowers): "This brings to mind another matter of importance, namely: the fact that among the new classes, the individual or clone is often much more horticultural importance that the species to which it belongs. With natural species from the Orient, variation is very great among seedlings. Add to this the accidental crossing that occurs in gardens and you will find additional divergence from the main type." This conclusion in his speech contains almost the whole scientific background to our work in Finland. Firstly it emphasizes within species variation (although he talks more about flower colors than hardiness) and secondly it notes the unpredictable variation that you may get by collecting seeds in gardens including many species of rhododendrons or perhaps even hybrids. In our effort to breed for extreme hardiness, we have put out thousands of new hybrids that contain R. brachycarpum Tigerstedtii Group as the maternal component, in common gardens. Thus we think, that a breeding program that started with hand-pollinated material can now be continued in the longer perspective on the bas is of a large "breeding population" from which you may procure seed that gives your recombinant and unexpected progenies that may become new cultivars, or you may use the "breeding orchard" for another round of hand pollination. In fact, I have often wandered around in the breeding orchard with its several thousand plants in flower and thought that such a population with variation in flowering time, color, growth habit and foliage is a much more interesting and beautiful plantation than one consisting of single clones registered as cultivars. Our modern plant breeders' rights and royalties of course call for clonal cultivars, but a genetically variable population is a much more living collection where you can see in front of you the importance of the genes.
"Thus, some species marked 'C' in England may prove to be hardier in America than those having 'B' rating. I look forward to the time when we will list our plants as suitable for limited regions only." Bowers comes close to what we have always argued: plant hardiness zones are only indicative, the local conditions are the conclusive. The international plant breeding institutes around the world (CGIAR) have increasingly adopted the term "ecoregion" to specify an ecological niche, regionally specified. Using gene banks and other genetic resources, they now compose "breeding populations" eco-regionally adapted to the climate. This has also been termed pre-breeding or genetic enhancement. I visualize our breeding orchards in Finland for rhododendrons as being typical examples of such long-term activities. Such activities are of course very long lasting and often outside the scope of an amateur or private breeder. However, we have been able to place our pre-breeding populations in public parks (city parks mainly) where they serve the double purpose as enjoyable to the public and useful to the breeder.

* Name is not registered.

¹ Now R. canescens .
² Now R. periclymenoides .
³ Now R. minus var. minus Carolinianum Group.
⁴ Now Obtusum Group.
⁵ Now R. 'Amoenum'.
⁶ Now 'Indicum Album', a syn. of R. mucronatum .
⁷ Now Fraseri Group.
⁸ Now R. maddenii ssp. crassum.