Commentary: Dr. Clement Bowers' Letter
"Common observations of everybody's results seem to indicate that the "wild" or lilac color of Rhododendron catawbiense, present in varying amount in almost all of the hardy hybrids, is very hard to get rid of. I have found that even the white clones (most of which are pinkish in bud), such as 'Catawbiense Album', 'Album Elegans' and 'Boule de Niege' when inter-crossed give a preponderance of purplish or lilac colored seedlings in the F1." (Dr. Clement Gray Bowers, quoted in the Journal American Rhododendron Society, Vol. 60, No.1, Winter 2006.)
This is not only very understandable, but predictable. It begins with understanding that visual pigments are complex molecules. Since we know what the plant has to work with are sunlight, carbon dioxide, water, and some minerals in the soil, e.g., phosphates, sulfates, magnesium, iron, calcium, potassium, etc., it is obvious that a great deal of biochemical synthesis is going on.
What genes do, and it is all they do, is specify the sequence of amino acids in proteins. Many, if not most, of these proteins are enzymes, the primary biosynthesizers in any organism. Each enzyme takes a particular molecule or two and makes a small change. Different enzymes might join two molecules, split one in two, add an atom here or there, pull off an atom or even just an electron, or just twist the molecule in a certain way to change its shape. The biochemical synthesis process that builds all organisms is then a multitude of these enzymes, each taking molecules that represent their "raw materials" and modifying them to produce new molecules.
Directing our attention to eventual pigment molecules, and simplifying things tremendously, the process can be likened to a "bucket brigade" of enzymes, each produced by a specific gene, taking simpler molecules and modifying them into slightly more complex molecules. The key here is that this is a stepwise process of a great many steps. The genes of our ancestral R. ponticum in particular, known as the "wild type," produce a purple pigment at the end of the process.
Over the ages there may be a variety of genetic mutations which occur. Each mutation produces a change in a specific enzyme. This may be inconsequential to the behavior of the enzyme and everything proceeds as before. It may mean that the enzyme produces a similar but slightly different molecule from the original raw materials, which I will come back to. Or it may mean that this protein intended to be an enzyme can either no longer do anything with the molecules it finds as raw materials or produces molecules the next step can no longer use, breaking the sequence of the entire process. The result is no pigment, and white rhodies, or roses, cats, and chickens White is very common in a wide variety or organisms.
For purposes of discussion, suppose there were only ten steps in pigment synthesis. It may be that one of the white R. ponticum has a "broken" mutation at step seven. Perhaps another is broken at step eight. Both are white; synthesis cannot proceed to completion. But when crossed, the progeny get a chromosome bearing the gene for a "broken" enzyme at step seven from one parent, matched by the other parent with a normal gene at step seven. Pigment synthesis can now proceed through step seven. The complementary thing happens with step eight. Consequently the hybrid has enzymes for production of normal pigment and is purple. This is the classic "di-hybrid" cross done by Mendel with his peas in the 1850s.
In our rhodies there is not one biosynthetic sequence to pigment, but two, producing dissimilar pigments, flavones and anthocyanins. The flavones are generally yellows and oranges, anthocyanins, purples and reds. A critical examination of our orangey elepidotes confirms this. Most are mixtures of yellow and pink as compared to true orange Exbury azaleas. True yellows like 'Crest' and R. wardii will likely be found to have genes and enzymes (all it takes is one) which do not allow anthocyanin synthesis to proceed to completion. Likewise, lavender R. ponticum probably does not complete synthesis of flavones. Rhododendron dichroanthum and the like produce both. (Separating and demonstrating the combined pigments with paper chromatography would be the perfect subject for a high-school science fair project.)
Do not think that genes are "missing." They exist in alternative forms, called alleles, which specify enzymes that participate in a particular synthesis sequence, change them somewhat, or impede them. Alleles and enzymes, which make slightly different molecules yet ones that can be used by the next step in the synthesis, are how pigments of different hue are produced. (Anthocyanins are sensitive to the pH of plant tissues, which can influence hue.)
To return to the previous example, it's apparent that both of the white R. catawbiense of Bowers' comment share the same alleles of genes that keeps yellow flavones from being synthesized, but have different alleles of genes that produce purple. When crossed, that mismatch allows pigment to be produced. A cross of a 'Catawbiense Album' on a white variant from the yellow side of the family, perhaps R. wardii var. puralbum, functions in exactly the same way as Bowers observed, albeit involving genes which participate in both anthocyanin and flavone pigment synthesis pathways. Colored progeny could show purple and yellow pigments, even the effect of flavone synthesis mutations accumulating in the R. ponticum family, sight unseen.
All the genes of peas that Mendel used demonstrated "dominance." In crosses of purple peas on white peas, all the F1 were purple. In this case we can see the enzyme as a catalyst - one gene from either parent produces all the enzyme needed. But in snapdragons and four-o-clocks the gene for crimson red is not dominant. F1 hybrids are intermediate pink. The ratio of F2 is one red, two pink, and one white. Bowers' observation that some purple tint is hard to eliminate in ponticums may paradoxically be evidence of some dominance of purple, or even that it isn't dominant. Careful study is needed.
So far this has been about the genetics of pigmentation, in keeping with Dr. Bowers' comment. However, essentially the same genetic factors apply to issues of plant habit. Mendel discovered genes for tall and short peas. Many miniature roses have hybrid tea ancestry. It's not so easy to identify "broken" synthesis processes in plant habit as in pigment, though miniaturization may reflect missing enzymes. Rhododendron calostrotum ssp. keleticum comes to mind. In any event, much of plant habit is under nuclear genetic control, but not all. Chloroplasts, hence some aspects of photosynthesis and leaf pigments, descend from the seed parent.
Then there's fragrance and indumentum. Even more fun! It all begins with careful observation and the sort of experimentation Mendel began a century and a half ago.
I've used the words "likely," "probably" and "could" because most of our hybridizing has been with the goal of producing a more beautiful plant and flower, not in identifying the genes in our rhodies. I fear that many times open-pollinated seed from a garden with a variety of species has been distributed as though it were seed from the pure species of the mother plant, or the fact that it was "o.p." was forgotten. Hybrids now are so "heterozygous," and some plants identified as species are so doubtful, it's difficult to know where to begin getting the basic genetic information we need. And yet, observations such as Bowers' can be surprisingly revealing. Clear knowledge of the genetics of rhodies can make our hybridizing much less of a shot in the dark.