Additional Observations On Breeding Rhododendrons
David G. Leach
There appeared in the July, 1951 issue of the A.R.S. Bulletin an article describing the breeding of rhododendrons from the classic genetic point of view which I wrote at the request of the Editor. The appearance of this article has resulted in a number of inquiries and excerpts from it have been reprinted elsewhere.
Since there exists such an interest in the breeding of rhododendrons I am contributing a continuation of the original article to complete the techniques which might be usefully employed by hybridists engaged in work with this genus.
In cases where it is desired to add only, one or two characteristics (such as an improved truss, a different color, a more dwarf stature) to an otherwise desirable variety, backcrossing is the most efficient procedure to follow. This technique is much more efficient in such situations because it greatly reduces the number of seedlings which must be grown to secure the recovery in one plant of the desired combination of virtues.
Let us suppose that we have a variety which is desirable in most respects but which lacks one or two qualities the addition of which would make it a much - better and more useful plant. By crossing it with another variety which possesses the desirable characteristics lacking in the clone to be improved, we will secure progeny which represent some advancement for the virtues we seek but generally the seedlings will be intermediate in character. To attain the goal which is sought, the next step is to backcross the intermediate seedling which best exhibits the desirable qualities, to the parent to which we wish to add them. Among the resulting seedlings there is a good chance that there will appear an individual similar to the original grandparent but with the addition of the desirable characteristics which we wished to contribute to it. If the virtues we are seeking are controlled in inheritance by five factor pairs, for example, it is necessary to grow only 32 seedlings to secure our improved individual in "pure" form; whereas it would be necessary to grow 1,024 seedlings, as a theoretical calculation, to obtain the improved individual by inbreeding one of the intermediate F-1 progeny of the two original parents.
The foregoing assumes that the one or two desirable characteristics are dominant in the manner in which they are inherited. Frequently it is necessary to backcross yet a second time to the clop we wish to improve, or a third time. But if the virtues we wish to contribute to our original clop are inherited in a recessive manner so that our ideal plant does not occur in any of the progeny resulting from the first backcross, we are then reduced once again to inbreeding one of the backcross progeny (or crossing two of them if they will not accept their own pollen) to obtain the single individual with the desired combination of virtues.
The British, with a longer tradition of rhododendron breeding, are rapidly advancing in the application of genetic principles to the creation of new and valuable clones. Two famous hybrids created by backcrossing are 'Lady Chamberlain' [cinnabarinum var. roylei x (cinnabarinum var. roylei x maddenii)]; and 'Marcia' [campylocarpum x (campylocarpum x fortunei)].
A final suggested method of breeding is the isolation of pure lines and their subsequent merging to produce plants of superior horticultural value. This technique is used primarily for species which bloom quickly from seed; or in a commercial sense for the production of seeds which will yield a superior and rather uniform strain of seedlings. The procedure involves the raising of more generations, though perhaps no more seedlings in total number.
On an elementary plane, some of the azalea specialists producing plants for greenhouse forcing have already made use of the principle involved in this breeding procedure. Each of two different species which has been geographically isolated will be relatively "pure" in the genes they carry. If the two species are crossed in cultivation, the resulting hybrids will carry a complement of the genes common to each parent; but in addition they will inherit all of the genes for which either of the parents was "pure." The result will be that the hybrids are more likely to have valuable dominant qualities than inbred offspring of either species, and they will be much more uniform in those qualities, depending on the degree of purity of the parents. Thus the azalea specialists have found combinations of species which, when crossed, produce beautiful large-flowered hybrid plants. The same cross is repeated,, over and over again, as new plants for greenhouse forcing are required from year to year.
This principle, carried to its logical conclusion, could probably result in still greater improvement, not only of azaleas for forcing, but of any species or combination of them which do not require too long to flower from seed.
The important thing to remember in this sort of breeding is that it is the combination of the right sets of genes which produce the superior plants. It is the result of the genetic difference between two different strains.
Obviously then, the way to most fully accomplish an ideal combination of two different sets of genes is to establish two or more "lines" of plants which will accept their own pollen, and to inbreed them through several generations, in the meantime discarding the plants with evident faults as they appear in each generation and choosing for further inbreeding the most vigorous individuals.
It is one thing to inbreed a plant just once to secure among the progeny a recombination of characters, but it is quite another to inbreed repeatedly for successive generations. Each line which is inbred is reduced in its variability to an astonishing extent-in every succeeding generation the number of genes for which the plant is still "impure" is cut in half. Degeneration from such continued inbreeding is inevitable in rhododendrons and azaleas. We must not look for certain good qualities in one line of inbred plants, and other virtues in the other lines because all such good qualities do not depend each on a separate gene, but rather on the correct combination of the right sets of genes.
Now what happens when the lines which have been inbred and selected through a number of generations are crossed? Generally a tremendous burst of hybrid vigor results, and the right combination of the required sets of genes is restored so that the plants containing them are superior to the parent stock with which we started. The crossing of the inbred lines is the final step in a breeding project of this type. It has been phenomenally successful in many plant groups-in corn, for example, the hybrid strains of which have revolutionized American agriculture.
The question naturally arises: why should the final hybrids as reconstituted from the purified inbred lines be superior to plants from the original non-inbred strain?
Two reasons are suggested. The original non-inbred plants were probably "impure" for several of the genes which help to make for quality, and as a result a certain percentage of their cross-bred progeny would have had undesirable recessive traits. By inbreeding pure lines, and selecting from the progeny, these undesirable recessive traits were weeded out as they occurred in pure form and were manifested in the plants of succeeding generations.
Due to the purification and dropping of genes the inbred lines may not have a full complement of the desirable genes, but they are "pure" for the desirable genes they do carry. The crossing of pure inbred lines restores the full complement of genes which determine quality, and all of the plants from such a cross carry the desired combination of genes. In other words, this method of breeding not only improves the quality of the plants, but makes them uniform in their good quality. It would be an excellent way to produce a fine strain or race of rhododendrons or azaleas which would not need to be propagated vegetatively. Additional plants of uniform quality could be produced again and again merely by repeating the final cross between the inbred lines.