Logo for the Journal American Rhododendron Society

Journal American Rhododendron Society

Current Editor:
Dr. Glen Jamieson ars.editor@gmail.com


Volume 7, Number 1
January 1953

DLA Ejournal Home | QBARS Home | Table of Contents for this issue | Search JARS and other ejournals

A Final Discussion on "A Further Discussion of Breeding Methods."
By David G. Leach

        Having once repaid in kind Mr. G. Nearing's acidulous comments. I propose now to extend and conclude my discussion on breeding rhododendrons in a manner which will not obscure the agreements and differences between us. The amateur hybridist can then review the evidence and arrive at his own conclusions for practical use.
        My adversary and I are disagreed on the value of inbreeding. I advocate crossing two rhododendrons which exhibit the virtues to he combined and then self-fertilizing the best of the progeny (or crossing two of them together if they will not accept their own pollen.) Having been professionally trained as a geneticist it is incomprehensible to me how anyone could oppose this method, which is a basic, primary tool of plant breeders the world over. A cursory examination of the cytological process of plant fertilization demonstrates that this is the only way to obtain the potential variability inherent in the original cross. Surely many of the desirable attributes we are seeking are inherited in a recessive manner, and it is equally certain that this is the classic and almost the only practical method of obtaining an expression of these attributes.
        Dr. Hans Kahnus, the distinguished scientist at University College, London, says: "The extreme restriction of the mating system is called inbreeding and it is the most potent method that both the geneticist and the plant . . . breeder have for detecting the presence of recessive genes."
        My opponent maintains that inbreeding an interspecific hybrid results in an "inferior norm of the genus . . . Which is halfway between R. ponticum and nothing at all." I believe that this "inferior norm" is a misleading concept resulting from faulty observation. The average duality of self-fertilized progeny from a primary hybrid rhododendron is generally below that of the parent. This undisputed fact has deceived many amateur plant breeders. But we are not concerned with average duality. and if the hybridist will only grow a reasonably large number of seedlings he will find among them rhododendrons exhibiting the extremes of their heritage. These individuals, comprising a small minority of the total plant population derived from the inbred hybrid. combine the desirable attributes of their grandparents in the manner sought by the breeder. They represent the successful attainment of the goal which was envisioned when the first of the two progressive crosses was made.
        Dr. A. L. Hagedoorn, renowned consulting geneticist and professional plant breeder, say, "Cross-breeding will almost always give hybrids showing showing dominant qualities, even when the corresponding recessive characters are better appreciated: very often hybrid plants are thrown away for this reason whereas further (in)breeding from them would have given the plant breeder what he wanted."
        Obviously, this concept of the "inferior norm" even as descriptive of self-fertilized progeny is erroneous. The inbred seedlings of a primary hybrid can not possibly exhibit anything worse than the quality of the species which were their grandparents. The breeder must not blame the method if he does not exercise good judgment in choosing parents. catawbiense album Glass x griffithianum, inbred, will assuredly not produce any seedlings "halfway between R. ponticum and nothing at all." Even the most casual reader will recognize that iron will not result from a mixture of silver and gold.
        Apparently there is no difference of opinion as to the value of back-crossing except that I wish to stress that this method is the method of choice when only one or two virtues are to be contributed to an otherwise desirable plant; and that these virtues must be inherited in a simple manner, for this method to be efficient. Such things as color characters are simply inherited in that they are controlled by one or two gene pairs. As an example, then, backcrossing would be the logical choice to intensify the color of a red rhododendron. Such things as flower size, on the other hand, are generally controlled by many genes and backcrossing would not be the logical choice of method for a project to increase blossom dimensions.
        There is also no difference of opinion in this debate regarding the desirability of crossing complex hybrid rhododendrons. In my article in the "Bulletin" for July, 1952, I wrote: "I do not advocate crossing complicated hybrids...
        However, my opposition is based on the belief that such crosses are a wasteful and inefficient method of ameliorating rhododendrons, requiring enormous numbers of seedling and an amount of effort out of all proportion to the result which could be duplicated much more easily by the application of established genetic principles.
        I think the crossing of complex hybrids, though I do not advocate it, has produced and can produce good results, and there is abundant evidence to support my contention. My opponent originally denied the success of such crosses, citing once again his concept of the "inferior norm" as the result of mating intricately bred hybrids.
        The difference between our positions, then, is that I believe such crosses, with totally unnecessary effort, can be successful whereas my opponent once again thinks that this method also produces seedlings of an inferior norm "halfway between R. ponticum and nothing at all."
        More recently, however, Mr. Nearing has modified his position very substantially. In the October "Bulletin" he admitted that a heritage of staggering complexity has yielded superb Ghent, Mollis and Kaempferi hybrid azaleas. But, he says, azaleas are different.
        All of us know that azaleas are rhododendrons but we are expected to accept the statement that they are rhododendrons in everything except inheritance. I think all who have followed this discussion will want to know why rhododendrons, if subjected to the same breeding and selection, will not produce results equally as fine as azaleas.
        My opponent has now also conceded that the standard commercial catawba hybrid, which are universally planted throughout the East are in themselves tremendously complicated in their heritage. He admits that these rhododendrons are a mixture of six and possibly seven species intermingled through at least four generations of breeding by Anthony Waterer.
        But there is a reluctance to concede that the most popular and beloved nursery hybrid, for milder climates are also the result of crossing complex hybrids. We are told that the nurseryman who produced them misrepresented their parentage and concealed the facts.
        Fortunately we can also disprove this allegation. 'Britannia', 'Earl of Athlone', 'Borde Hill', 'Armistice Day', 'C. B. Van Nes', 'Countess of Athlone' and the long list of other hybrids mentioned in disproof of his position go back to 1890. when Mr. Otto Schulz of Berlin crossed R. griffithianum with various complicated hybrids which were then popular. In 1896 Mr. J. H. Van Nes, while on a business trip to Germany, purchased 230 seedlings from this group and he brought them back to Boskoop, Holland, but subsequently all except 12 were killed in the winter of 1910-11. These 12 survivors had previously been crossed, in 1896, with 21 different popular and complex nursery hybrid, to produce the extensive list of distinguished rhododendrons which are to this day by far the most widely grown in commerce for benign climates. The records are absolutely indisputable, having been published by J. Blaauw and Company after they bought the celebrated C. B. Van Nes & Sons Nursery at Boskoop. Each clone is minutely documented. 'Britannia', for example, is Van Nes No. 146, Stanley Davis x No. 1. The No. 1 parent is then detailed from the Berlin cross, and the pedigrees of the others are similarly set forth. Readers will be quick to note the absurdity of any claim that these records were falsified when they were made public almost half a century after the rhododendrons were introduced commercially.
        My adversary poses the question of why no outstanding results have been achieved from crossing the common catawbiense hybrid rhododendrons. The answer is, of course, that there have been fine rhododendrons produced in this manner.
        'Meadowbrook', Paul Vossberg's hybrid is a good example to cite in disproof of my opponent's position and it also illustrates the reason for my opposition to crossing complex hybrids. This rhododendron resulted from crossing 'Mrs. C. S. Sargent' with 'Everestianum' To obtain it, many hundreds of seedlings were grown and this was the only worthwhile plant among them. In all probability a very similar hybrid could have been obtained much more easily by another method.
        In any case, Mr. Donald Hardgrove, in "The A. R. S. Yearbook for 1915" gives it his highest rating for a hybrid, three stars, and remarks that it is "very fine, vigorous." He rates it far ahead of such distinguished rhododendrons as 'Essex Scarlet', for example, and better than 'C. B. Van Nes' and 'Cynthia'. It is given a rating equal to 'Souvenir de W. C. Slocock', 'Mother of Pearl', 'Dr. Stocker' and 'Mrs. A. T. de la Mare'. By comparison, 'Pink Pearl' is given only two stars and none of the common commercial catawba rhododendrons are assigned any at all. I have not seen "Meadwobrook" in flower but it must be very fine to have a three star rating.
        The result of this cross between two complex hybrids by a careful and experienced plantsman, Mr. Vossberg, proves that crossing complicated catawbiense hybrids can produce good results; and that such crosses are not "worthless" because of the "inferior norm" concept which my opponent has proposed. Other examples in addition to "Meadowbrook" could be cited but this clone has been the most publicized.
        But enough of this. My opponent and I are agreed, though for quite different reasons, that crossing complicated hybrids is not advocated, and this issue has already received more attention than it deserves.
        Mr. Nearing's formula for creating hardy rhododendrons is a cross of two hybrids involving tender and one hardy species. He proposes (dichroanthum x griersonianum) x (griffithianum x catawbiense) as a good example.
        As a careful review of my comments will show, I proposed this formula for one reason only: because it can not possibly produce rhododendrons of sufficient hardness to be widely useful in the East.
        To produce progeny which are hardy in the sense of being as cold-resistant as the commercial catawba hybrids it would be necessary for hardiness to act in inheritance as a simple dominant in this formula. It has been abundantly demonstrated again and again by every hybridist that rhododendrons are like other plant groups in that hardiness is a quantitative character controlled by many genes, some favorable alleles being dominant and some recessive. No breeder has ever proposed, to my knowledge, that hardiness is a simple dominant.
        If we envision for a moment the process of egg cell fertilization we can see that, having been preceded by reduction division, dichroanthum x griersonianum will contribute one chromosome and griffithianum x catawbiense will contribute the other to make up each pair of chromosomes which the progeny possess. The seed parent, in the proposed formula. can contribute no hardy genes. Only the pollen parent in this cross can do so. Since the contribution of the pollen parent does not dominate hardiness, the effect is interacting with the genes from the pollen parent exerting a positive influence toward greater hardiness; and the genes from the seed parent acting in a negative fashion toward lesser hardiness. The net result is an intermediate. The very best that can be hoped is that the progeny will be in hardiness about halfway between R. catawbiense and (dichroanthum x griersonianum). And that is not nearly hardy enough. I therefore repeat my contention that this formula is an utter waste of time for a breeder who wants to produce rhododendrons hardy for the East.
        Even on a pragmatic basis, hybridists who are familiar with the principles of segregation recognize that only three possibilities of hardiness exist in this cross: catawbiense x dichroanthum; catawbiense x griersonianum; and catawbiense x (dichroanthum x griersonianum). t is already known from actual experience that none of the three, individually, will produce progeny hardy for Eastern conditions and therefore no possible segregation of genes affecting hardiness in Mr. Nearing's proposed cross could possibly result in satisfactory progeny.
        Readers will be well advised instead to learn the lesson which is apparent from Mr. Nearing's cross of (decorum x griffithianum) x 'Kettledrum'. He describes it as his most successful effort.
        Instead of regarding the results as merely "fortunate" we must apply the same principle to crosses which have at least the possibility of yielding satisfactory results. This cross of Mr. Nearing's succeeded by a segregation of the semi-hardy decorum genes with the griffithianum flower size to produce in the intermediate progeny hardy plants bearing pink flowers of impressive proportions.
        The same method, if the principle were only recognized and applied, could be used with equal success in other directions. Since we know that catawbiense x fortunei produces hardy progeny, we can confidently expect good results from such a cross as catawbiense album Glass x (fortunei x campylocarpum). The semi-hardy fortunei genes will combine with the yellow color factor from R. campylocarpum to produce in the intermediate progeny hardy rhododendrons with pale yellow blossoms in size half way between those of catawbiense and of fortunei.
        Since we also know that catawbiense x discolor produces hardy progeny, such a cross as catawbiense album Glass x (discolor x dichroanthum) will result in a segregation of the semi-hardy discolor genes with the orange color factor from R. dichroanthum to produce hardy progeny with good sized flowers of orange hue. If enough seedlings are grown, we can also obtain a measure of the dwarf stature from the dichroanthum heritage in the same hardy progeny with orange colored blossoms.
        This method is the quickest and most direct that could possibly produce completely hardy rhododendrons with blossoms of yellow or orange coloration. The two crosses mentioned and many others utilizing the same principle have already been made here. Some of them will come into flower before long and there is every expectation that among them will appear forms worthy of commercial introduction. Until plants of the right pedigree are available as a means of achieving a more perfect result, there should be a place for fully hardy rhododendrons with blossoms in color ranges not now available.
        The writer fully recognizes, however, that no such easy method will produce the ideal rhododendrons with deep yellow and with true orange blossoms, of dwarf stature and compact growth, with flowers of perfect proportions appearing at the most desirable season, and with the many other worthy attributes which can only be achieved by more intricate hybridization.
        As my adversary has pointed out, what may be a virtue in a rhododendron with one set of characteristics may well be a fault in a hybrid with different specifications. He states that he would welcome the appearance of a plant containing all of the faults of the four species in his proposed formula because he could then expect to find among other seedlings from the same cross all of the good qualities similarly combined in one or more plants. The defect in this reasoning should be apparent.
        A virtue may be determined by genes acting in a dominant manner or by the actual absence of genes which cause the defect. A fault may be manifest through genes acting in a dominant manner or by the mere presence of genes controlling that factor. The contention, therefore, that the proposed formula can result in a redistribution of all characters, or even a majority of them, is obviously untenable. Characteristics controlled by recessive genes would not even be seen in the progeny. Another generation obtained by selfing or by sibling crosses would be required for their outward expression. Except for dominant qualities, the formula presents no opportunity whatever for the appearance of any but characteristics roughly intermediate between the two parents and/or the four grandparents. In this respect it is similar in result to a cross between two species. As every breeder has learned to his sorrow, too few of the desirable characteristics we seek appear as simple dominants.
        I am challenged by my opponent to write the specifications for the rhododendrons which I would like to create through hybridization. In my opinion, all of our readers who have had the interest to follow this discussion very well know the numerous attributes which are desirable in a top quality rhododendron and I refuse to burden them at length with such an obvious listing of rhododendron virtues. Discrimination in evaluating rhododendrons is not a monopoly of anyone, nor is it obscure or difficult.
        The history of plant breeding clearly shows that time and patience will produce any reasonable goal for the hybridist who applies sound genetic principles to his work. The enormous advances in production and quality obtained in rubber, oil-palms and coconuts in Sumatra; in tea and coffee in Java; in grains and grasses in Europe; in tropical tree crops in Africa: in citriculture in South America: and in corn and potatoes in the United States all give testimony to the astounding success of scientific plant breeders with crops of commercial importance. When a disease-resistant plant has been required, the geneticists have promptly produced it. When hardier crops have been requested to extend the range Northward, they have been forthcoming. There are now a hundred strains of hybrid corn. Each bred to order for specific climatic conditions.
        I do maintain that the principles of genetics are universal in their application and that rhododendrons are susceptible to the same direction and control in breeding as are other plant groups. It is perfectly possible, indeed it is a necessity. for the hybridist to decide in advance exactly what he wants before embarking on any breeding project. Mr. O. C. A. Slocock. whose name is a by-word in the world of rhododendrons, says in "The Rhododendron Yearbook 1951-52": "It is necessary, first of all, to decide what one is endeavoring to get and then everything done should be a means to that end." Mr. F. C. Puddle, probably the most successful of living rhododendron hybridists, in the same publication says "I always work on definite lines and never make crosses without a definite object in sight." For my opponent to infer, then, that the breeder can not set his sights on a specific goal with a reasonable prospect of attaining it is obviously not in accordance with the facts in other plant groups or in rhododendrons themselves.
        In concluding my comments I should like to add the hope that readers of the "Bulletin" have been stimulated to a new awareness of the problems and the procedure, of hybridists, all of whom are working sincerely according to their beliefs in an effort to produce new, more beautiful and more useful rhododendrons. Both the mail and direct comments show a lively concern with hybridizing on the part of many hobbyists and I am confident that our debate has served a useful purpose in arousing interest among fanciers who have not previously engaged in this fascinating pastime.


Volume 7, Number 1
January 1953

DLA Ejournal Home | QBARS Home | Table of Contents for this issue | Search JARS and other ejournals