Dahlia Seeds and Seedlings: Selection, Heredity, and Breeding Value

Originally published as The Evolution of the Dahlia: Chapter Twelve (1929)

Original text by J. J. Broomall is in the public domain.

Compilation, editorial notes, introduction, biography, and revised text copyright © 2026 by Steve K. Lloyd. All rights reserved.

In Seeds and Seedlings, Broomall gives his fullest numbered account of how the breeding value of dahlias should be judged through their offspring.

Earlier chapters had considered variation, inheritance, pollination, cultivation, environmental influence, and the difficulty of predicting the results of crosses. Here Broomall brings those subjects together in a practical breeding argument. He begins with variation under cultivation, moves to the enormous scale of seedling production and the rarity of genuine improvement, questions simple applications of Mendelian theory to repeatedly crossed commercial dahlias, and then argues that the value of a parent must be determined empirically from its progeny.

The chapter’s central distinction is between the visible quality of a cultivar and its usefulness as a breeding parent. Broomall reports that one of his finest dahlias produced almost no seedlings worth saving, while another, visibly imperfect variety produced desirable offspring at a much higher rate. Superior appearance, he concludes, does not necessarily imply superior “inheritability.”

He also emphasizes the severity of selection required for continued improvement. As standards rise, flowers that might once have been considered first-class can no longer justify introduction. The chapter closes by extending that principle from his own garden to public judging: neither breeders nor judges should give mediocre seedlings a standing they have not earned.

Its placement after Chapter Nine is editorial rather than authorially specified. Chapter Nine emphasizes the limits of control and prediction in crossing; Chapter Twelve develops the breeder’s empirical response to those limits through large populations, progeny testing, and severe selection.

The chapter appeared in the 1929 Success Dahlia Gardens Catalog.

Plant variation makes possible the improvement of nearly all plants; Nature never makes two plants exactly alike; in most plants growing in their wild state the variation may be so slight as to be unnoticed, but when they are brought under cultivation the variation is often so great that even those who are unacquainted with plant culture will notice the difference.

Probably the greatest cause of increased variation when plants are brought under cultivation is the change in the character and quantity of the plant food which becomes available to the plants under the changed conditions; in their natural state it has been a struggle for existence on a limited amount of nourishment; when an unlimited amount of plant food is supplied it is not surprising that the plants would do something different, indeed, they cannot help but act differently.

If the plant should happen to be what is known as a “gross feeder” (that is a plant that is capable of absorbing nourishment very rapidly) the change in such a plant will be great in proportion.

Now the Dahlia is a gross feeder; I know of no plant that will respond more quickly to good treatment and favorable environment, hence it is no wonder that it has astonished the world with its marvelous improvement.

As to its future I can see no limit to the possible improvement not only of the Dahlia but many other plants, Economic as well as ornamental. I believe the only limit to be our knowledge of the plants, and for that reason I never lose an opportunity of urging upon my fellow men the necessity of taking a greater interest in plant life, and in Nature, the most interesting and essential of all studies.

It may appear to be an exaggeration, but I believe the possible variation of plants to be as limitless as time and space.

Take the Dahlia, for instance: I think a million would be a moderate estimate of the number of seedlings of it grown this year; some commercial growers raise many more than I do because they have more room for growing them. I planted about 8,000 seedlings this year; no two of those seedlings will produce flowers alike.

Do not think that this means that I will be growing 8,000 varieties next year; the only plants that will be saved will be those that appear to be better than or different from varieties already in cultivation; the greater part of them will be discarded and thrown on the junk-pile to make plant food for future use.

In past years I have introduced on the average about three out of each thousand, and eventually many of those will also go into the discard. When it is taken into consideration that ALL of the seedlings must receive the very best care and culture in order to determine their merits, some idea may be had of the amount of work that is necessary to obtain a new variety that is a “Topnotcher,” and what I have said here in regard to the Dahlia will apply more or less to all plants that have been or are being improved by man in co-operation with the forces of Nature.

In highly improved plants very few of the seedlings will be an improvement on the parents; the majority will not be as good as the parents in most cases. There is always a tendency toward retrogression.

A writer in the Bulletin of the America Dahlia Society has considerable to say about the “Mendel Theory” in connection with the question of Heredity of the Dahlia. I have two books on the “Mendel Theory” and as near as I can understand, it is claimed that certain results occur from the second or third generation of crossing plants of different colors, etc.

Now this theory may be all right in some cases, but I do not see how it can apply to the most of the Dahlias we are working with at the present time as many of them are the result of twenty or thirty years of crossing and recrossing; it seems to me to be impossible to determine what generation of crosses we are dealing with.

Incidentally, he speaks of “Pedigreed Dahlia Seed” and gives quite a lot of information that may or may not be of value. Another writer in the Bulletin advocates Year Old Dahlia Seed; it may be that the information given by those writers is valuable but I must confess that I cannot agree with them. Incidentally it should not be out of order to inquire of these gentlemen who have taken so much trouble to enlighten those of us who are more benighted as to how many Prize-winners THEY have introduced.

It may be all right to plant old Dahlia seed but I prefer New seed and the more fresh it can be kept until planting time the better, in my estimation.

While I am unable to understand how the “Mendel Theory” can apply to Dahlia culture at the present time, I am nevertheless a firm believer in the heredity of the Dahlia. I believe it to be by far the most important factor in the improvement of the Dahlia.

I believe it to be the greatest Hindrance, and the greatest Help to Dahlia Growers in their efforts to improve the Dahlia; the greatest hindrance because it upsets all our calculations in our efforts at crossing certain varieties, the greatest help because by taking advantage of the superior qualities of some varieties as Parents we can obtain a greater number of progeny that are desirable.

I have found out by experience that some of our best Dahlias are poor Parents, the seedlings from them being nearly all retrogrades, so that it does not pay to bother with them, while there are other varieties that prove to be much better Parents altho they may not be so fine in some respects as some that are NOT satisfactory Parents.

My experience this year (1928) has more than ever confirmed my belief in this theory:

THAT SOME DAHLIAS ARE MUCH BETTER PARENTS THAN OTHERS BECAUSE OF SUPERIOR INHERITABILITY.

Seed from one of the best Dahlias I have grown did not produce more than one in a thousand that was worth saving, while another variety that was far from faultless produced four or five first-class flowers in a hundred.

This is proof-positive, it seems to me, that the latter variety had the best (blood) in it, that while IT was not the best its Ancestors WERE the Best, hence the superiority of the progeny.

So it seems to me that if we would succeed in improving the Dahlia to the greatest possible perfection, we must study the heredity as shown by the offspring, and select those for parents that give the most satisfactory results, and this can only be determined by careful scrutiny and study.

At a time when thousands of growers, both Amateur and Professional, are growing seedlings in unlimited quantities, too much care cannot be exercised in the selection of those that are to be kept and introduced. It will not hasten the improvement of the Dahlia nor benefit the Grower to flood the market with a lot of mediocre seedlings.

Because of the higher standard of excellence of the Dahlias that have been introduced in recent years we are now obliged to throw away thousands that ten years ago might have been considered first-class. I know from experience that it is sometimes almost heart-breaking to discard a beautiful flower because it may have some fault that would disqualify it in the show room, but we must be careful in selection if we would develop the Dahlia to the highest possible perfection.

Having acted as one of the judges of seedlings at the recent show of the Dahlia Society of Southern California at the Biltmore Hotel in Los Angeles, I want to say that we awarded no prizes in some classes because to have done so would have given some Dahlias a certain standing to which, in the opinion of the judges, they were not entitled, and I am sure the only thing the judges had in mind was THE IMPROVEMENT OF THE DAHLIA, and that is why this Chapter has been written, for the good of the greatest of all garden flowers, THE DAHLIA.

1. Variation under cultivation. Broomall regarded variation as the basis of plant improvement and believed that cultivation increased visible variation by changing the quantity and character of available nourishment. This explanation is preserved as part of his historical understanding of plant variation rather than as a modern genetic or physiological account.

2. Seedling numbers and selection rates. Broomall states that he planted about 8,000 seedlings in 1928 and that in earlier years he introduced approximately three seedlings from every thousand grown, many of which were later discarded. Elsewhere he gives different figures for other stages of selection and commercial survival. These figures describe different stages and should not be combined into a single success rate.

3. “Bulletin of the America Dahlia Society” and the “Mendel Theory.” The historical source prints “Bulletin of the America Dahlia Society,” and that wording is preserved. Broomall’s criticism of the “Mendel Theory” does not amount to a rejection of heredity; he objects to what he understood as simple generational predictions applied to commercial dahlias already shaped by decades of crossing and recrossing. The writers he criticizes have not been independently identified.

4. Visible quality versus breeding value. Broomall’s major practical conclusion is that the finest-looking cultivar is not necessarily the best parent. He reports that one excellent dahlia produced no more than one seedling in a thousand worth saving, while another, visibly imperfect cultivar produced four or five first-class flowers in a hundred. The two cultivars are unnamed and should not be retroactively identified from later sources without direct evidence.

5. “Superior inheritability” and “proof-positive.” Broomall uses “superior inheritability” and “best (blood)” as historical terms for parental breeding value and calls the contrast between his two unnamed parents “proof-positive.” The chapter does not document controlled crosses, equal cohort sizes, standardized scoring, replication, or identical growing conditions, so the comparison records Broomall’s empirical breeding judgment rather than a modern controlled experiment.

6. Southern California judging. Broomall states that he served as a seedling judge at a recent Dahlia Society of Southern California show at the Biltmore Hotel in Los Angeles and that judges withheld awards in some classes. The statement is preserved as printed, but the event details, judging panel, affected classes, and withheld awards remain independently unconfirmed.