JASA 13 (June 1961): 2-6.
Neutron-Induced Variation of Roses*


*Paper presented at the Fifteenth Annual Convention of the American Scientific Affiliation, Seattle, Washington, August, 1960.
**Dr. Lammerts received his Ph.D. in Genetics from the University of California and is Director of Research for Germains Seed Company, Livermore, California.

Curiosity over the origin of a naturally occurring albino bud sport on a plant of Audie Murphy, on the part of my physicist friends, Philip Livdahl and Harlan Zodtner, led to the consideration of experiments to induce mutation. Since studies on the effect of gamma radiation were already underway at Brookhaven, neutron radiation was an obvious line of attack. The following experiments, begun in the fall of 1957 and continuing until 1959, have given such unusual results as to be worth reporting.

Materials and Methods

Rosebuds of the vigorous Grandiflora rose, Queen Elizabeth, a red Hybrid Tea H56024/39, and Rosa multiflora were cut from bud sticks and placed on wet filter paper, 50 in each Petri dish. Except for a control group of about 100 buds left on bud sticks in Experiment 1, this method of exposing buds was used throughout. The following five experiments were performed:

1. In the fall of 1957, Philip Livdahl, then at Lawrence Radiation Laboratory in Livermore, California (now at Argonne Laboratory near Chicago), used the A-48 linear accelerator bombarding a copper target with 7.5 Mev deuterons. The approximate radiation intensity at the main axis of the beam was estimated to be about 400 rad, with neutrons emitted at a wide angle. Two Petri dishes, each with 50 Queen Elizabeth buds were placed one immediately behind the other. A small bundle of bud sticks having about 100 buds of Queen Elizabeth was placed as near as possible to these Petri dishes as a control, to study possible shielding action of bud stick tissue. All the exposed buds and 50 control unexposed buds were budded into R. multiflora understock the same afternoon immediately following exposure.

2. On March 14, 1958, 50 buds of Golden Rapture #5 were exposed to thermal neutrons from a "water boiler" reactor receiving approximate 1000 rad and budded into R. multiflora understock in the greenhouse.

3. On March 18, 1958, four Petri dishes, each with 25 buds of Golden Rapture #5 were exposed to 400, 600, 1500, and 3000 rads respectively of 14 Mev neutrons from a Cockcroft-Walton accelerator. These were neutrons resulting from deuterons accelerated to 500 kev bombarding a tritium target. The buds were immediately budded into R. multiflora understock in 4" pots in the greenhouse.

4. On April 23, 1959, 50 buds of a red Hybrid Tea H5 5024/39 were placed 11 centimeters from the target of a Van de Graaff accelerator and exposed to about 90 rads of 15 Mev neutrons obtained by the accelerated deuterons impinging on a tritium target. Similarly 4 other petridishes of 50 Queen Elizabeth buds each, were exposed to 90 and 120 rads, and 2 at 180 rads. As a check to test the effect of placing one Petri dish behind the other, an upper dish at 9 centimeters away was exposed to 120 rad. All were budded in R. multiflora understock soon after exposure; i.e., the same afternoon.

5. Finally on July 9, 1959, 3 Petri dishes (150 buds) of H56024/39, 1 Petri dish (50 buds) of Queen Elizabeth, and 50 buds of the single 5-petaled white R. multiflora understock were exposed at 9% cm from thetarget to 15 Mev neutron radiation obtained by using a Van de Graaff accelerator. As a c eck on the "Protective effect" of Petri dishes a "triple piggyback" group of 3 Petri dishes, each with 50 buds of Queen Elizabeth, were placed one above the other so that the dish closest to the target (8 cm) received 500 rad of neutron radiation. These were all budded into R. multiflora understock the same afternoon following radiation exposure. The R. multiflora, single Petri dish of Queen Elizabeth, and all 3 of the H56024/39 Petri dishes at 9Y2 cm. received about 350 rad of neutron radiation. The third Petri dish of the "triple piggy back" group received about 28D rad. One of the H56024/39 dishes was removed after 180 rad of neutron radiation.


Before giving results it should be explained that Queen Elizabeth is a hybrid of the Hybrid Tea rose, Charlotte Armstrong x Floradora, a Florabunda.

Experiment 1—The most striking result of this experiment was the almost complete lack of mutations or even bud retardation from buds left on bud sticks when exposed. By comparison the 50 buds in the Petri dish nearest the target were all retarded and slow to start and very deformed as to foliage and leaf appearance. Gradually many of them became more normal by growth from buds on deformed stems, but even these shoots were abnormal, having "strap" leaves and segments of heavily pigmented tissue. Finally by the spring of 1958, most of the plants had normal shoots which "took over" the growth. The original deformed branches continued to grow, and buds taken from them showed a wide range of variation. About 20 of the plants, however, remained completely changed, both as to color, form, number of flower petals, and appearance of plant and leaves. A thick large rugose leaf type was obtained so unlike Queen Elizabeth that one would never suspect its relationship by induced mutation. Equally unusual was a very narrow strap leaf type, having thin long, pointed buds. Several had flowers about the size of Floradora with 50-60 petals as compared to 20-25 in Queen Elizabeth and were light scarlet in color, All were weaker in growth than Queen Elizabeth and one was so similar, except for its dwarf habit, as to be of possible commercial value, since some gardeners object to the remarkably vigorous growth of Queen Elizabeth.

Even more interesting was the unusual fact that plants from buds in the petri dish above the one next to the target were normal, except for one possible color variation to light scarlet. The buds were but little retarded, as compared to the control buds not exposed to any radiation.

Experiment 2 may briefly be stated as negative, in that no mutations were observed following exposure to thermal neutron radiation.

Experiment 3 was discouraging in that the two most desired mutations, i.e., increased petal number and darker yellow color, were not obtained. Occasional changes, such as reduction of petal number to 5-6 from the usual 20-25 (2 plants), and one plant with slightly narrower leaves were observed in those radiated at 3000 rad. These changes are valueless commercially. Evidently the genetic variability potential of Golden Rapture #5 was not great. Possibly this may be due to the fact that it carries at least 2 factors for pale yellow (YY). Hence, though genetically carrying recessive factors for deep yellow (yy), both dominant factors would have to be inactivated to allow expression of recessive deep yellow as hoped for.

Experiment 4—As the buds of those exposed to lower dosages of the high energy neutron radiation of this experiment began growing, it seemed at first that, even at this low level, we had variability comparable to that of Experiment 1. The following is a summary of observations made June 9, 1959, followed by summary of final results, as of August 18, 1960.

  1. H56024/39-20-25 petaled, rose-red hybrid were radiated at 89 rad. 15 plants were normal in height, i.e., 16"-20". 7 were 7-8", 5 were 2"-4", and 8 buds were just starting. Some were of normal height with lighter anthocyanin pigmentation. One plant has flowers with only 7 petals.

    8/18/60. 34 surviving plants. One very weak plant died after transplanting. Final results were: I plant has 8-10 petals and 2-3 petaloids, and 1 plant is very slender and weak and 12" high compared to normal of 26"-W.

  2. Queen Elizabeth—89 rads exposure. Nine were normal 16"-20", 13 were 6"-8", 20 were 1"-4", 3 buds were just starting, 2 were dormant, and 3 were dead. Some normal plants were darker in anthocyanin pigmentation and all showed malformation areas.

    8/18/60. 22 surviving plants. All were normal Queen Elizabeth plants both as to flower and plant habit.

  3. Queen Elizabeth—178 rads exposure. Ten normal plants were 16"-20", 20 were 6"-8", 16 were V'-V, 3 buds were just starting, 1 was dormant, and 2 buds were dead. Much greater deformity was present, even in those plants normal in height, than among the group receiving only 90 rads. One plant was almost black with anthocyanin. About a dozen were sectoral chimaeras for 50-60 petals; several of these also were light scarlet.

    8/18/60. 28 surviving plants. Six were much weaker than normal, of which 2 had abnormally small leaves and flowers.

  4. Queen Elizabeth—178 rads exposure. Four were normal 16"-20", 22 were 6-8", 19 were 1"-4" high, 3 were dormant, and 4 buds were dead. Like #3, all plants were much deformed, even those normal in height.

    8/18/60. 24 surviving plants. Four were much weaker than normal.

  5. Queen Elizabeth—118 rads exposure. Four were normal 16"-20", 18 were 6'4%, 18 were 1"-4", 4 buds were just starting, 4 were dormant, and I was dead. Much deformity was present in all, seemingly more than in #3 or #4. Some small plants with leaves were not at all like Queen Elizabeth.

    8/18/60. 34 surviving plants. All quite normal in appearance.

  6. Queen Elizabeth—177 rads exposure. The bottom petri dish was nearest target at 9 centimeters. These generally were more affected than any. None were normal in height: 15 were 6"-8" high, 24 were 1"-4”, 3 were just starting, 3 were dormant, and 2 were dead. Many, as they developed, showed sectoral chimaeras for increased petal number, light scarlet, and small-sized flower.

    8/18/60. 37 surviving plants. All were quite normal in appearance.

  7. Queen Elizabeth—118 rads exposure. Upper dish, 11 centimeters from target, was protected by first petri dish. Eight were normal height 16"-20", 26 were 6"-8", 9 were 1"-4", and 7 were dormant buds. Relatively little deformity noted on 4-5 plants affected at all.

    8/18/60. 33 surviving plants. One plant had most of its leaves smaller and yellowish green. Flowers definitely were more scarlet. A normal shoot is now growing vigorously from the base. One plant had a thicker, leathery, dark green foliage and a light scarlet flower. Two very weak plants had narrow, dark green foliage. Three plants had sectoral chimaeras, i.e., stems having flowers much more scarlet than typical of Queen Elizabeth; one of these had over 60 petals, beautifully imbricated. All three stems were weaker than the rest of the plant.

At low-dosage rates evidently only a small percentage of cells in each bud mutates. As the plants continue to grow, normal shoots "take over" and except as noted the final result was a normal Queen Elizabeth plant.

1. Mehlquist, G. A. L., "Inheritance in the Carnation, Dianthus Caryophyllus—1. Inheritance
of Flower Color," Proc. Amer. Soc. Horticultural Science, 37: 1019-1021 (1940).

Generally then the many changes were transitory, i.e., chimaeral. However, by budding from these, some were stabilized and may have horticultural value, even though they are not as vigorous as Queen Elizabeth.

Experiment 5, which was started on July 9, 1959, was comparable to Experiment 4, as regards Queen Elizabeth and H56024/39. In addition, 50 buds of R. multiflora were exposed to 356 rads of neutrons. Since the results are rather similar, they may briefly be summarized as follows:

  1. Queen Elizabeth—50 buds in petri dish 8 cm from target receiving 502 rads of neutrons. Only 4 plants survived which are now quite normal. Ten of the exposed buds started and the plants at first were very delicate and prostrate. But buds from these more normal and finally basal shoots were completely normal. The original prostrate shoots are being budded to study variability.
  2. Queen Elizabeth—50 buds in petri dish 9.5 cm from target, i.e., just above #1 and receiving 356 rads of neutron radiation. 41 plants survived of which 5 remained much weaker than normal. One originally prostrate plant now has normal basal shoot. There were many sectoral chimaeras for such qualities as:
    (a) Scarlet.
    (b) Small flowers similar to Floradora, having many petals.
    (c) Flowers the size of Queen Elizabeth with many petals.
  3. Queen Elizabeth—50 buds in petri dish 11 cm. from target, i.e., above #3 and receiving 265 rads of neutron radiation. 42 plants survived, one with chlorotic leaves and 7 weak plants. Relatively few sectoral chimaeras appeared compared to #2.

Samples 4, 5, and 6,each had 50 buds of H56024/39 and received 356 rads of neutron radiation. Sample #7, also H56024/39 received only 178 rads. There were 49, 50, 46, and 43 surviving plants, all discouragingly normal in appearance. Occasionally flowers with only 10-12 petals were found, but unlike Queen Elizabeth radiated buds, no flowers with increased petal number were found.

The last experiment—petri dish #8—was 9.5 cm from the target and received 356 rads of neutron radiation. It contained 50 buds of a basic species, R. multiflora, having small white flowers and only 5 petals.

Evidently this species was highly resistant to change as there were no variations in flower color, petal number, or size. Of the 47 surviving plants, 6 are very definitely much weaker than normal and one of these is very weak. These plants will be budded to see if this reduced vigor is a stable condition.


Quite evidently by neutron radiation of buds at about 4000 rad we can tremendously increase the rate of mutation almost to the 100 per cent level in terms of mutation per bud radiated, since all buds showed at least sectoral chimaeras for changed tissue.

Horticulturally, the results were encouraging in that we did get several selection with increased petal number, having good bud form and a pleasing change to a light scarlet or what is popularly known as "coral pink" color. Also a dwarf type identical to Queen Elizabeth in bud, open flower form, and color was obtained which may have commercial value in that some gardeners object to the extremely vigorous growth of Queen Elizabeth (to 8 feet high in California gardens). This "buds true" in tests so far made.

Horticulturally, the results were discouraging in that we did not get color variations at least theoretically possible genetically. Thus, though we did get a white variation, the leaves were adversely affected, being semi-chlorotic, i.e., yellowish green in appearance. Possibly bud selection may segregate the color change from the leaf change should the cell tissue still be variable. Perhaps the situation may best be stated by listing the variations we did not get as follows:

  1. 2. Lammerts, Walter E., "The Scientific Basis of Rose
    Breeding," American Rose Annual, 1945, pp. 71-79.
    Change to dominant RI or crimson red color from the recessive r or light scarlet range.
  2. Change to dominant S or dark scarlet such as is characteristic of Floradora. Evidently all variations were to the recessive s or salmon to coral pink (light scarlet of British Horticultural Color Chart).
  3. Change to the large-flower and large petal characteristics of Charlotte Armstrong. All the variations had petals either slightly smaller to much smaller than Queen Elizabeth.
  4. Change to the long bud characteristic of Charlotte Armstrong.2 Long bud is dominant to short bud due to a series of dominant or semi-dominant factors.
  5. Change to climbing type or more vigorous pillar type.
  6. Changes to long cutting stem with flowers occurring singly as in Hybrid Teas. All changes were toward cluster behavior of Floradora, the Floribunda parent.
  7. Increase in leaf size. Queen Elizabeth has a large-sized leaflet fully as large as Charlotte Armstrong. Most variations, even the large rugose one, were toward smaller-sized leaflets.

Quite evidently most mutations were toward the recessive type as exemplified by:

  1. Decrease in size of flower coincident with increase in petal number, i.e., resulting in many smaller petals.
  2. Change to the recessive s or salmon (light scarlet appearance). These mutations were very frequent, i.e., in ratio of about 100 to 1 for change to white.
3. Lammerts, Walter E., "Inheritance of Magenta Red Color
in Roses," American Rose Annual, 1960, pp. 119-125.

It would seem that the dominant M factor3 was frequently inactivated, thus allowing the recessive s factor to show, or simultaneously the dominant S factor was also changed to s.

Compared to Queen Elizabeth, Golden Rapture is much more difficult to affect by radiation. Genetics tests show definitely that it carries recessive factors (yy) for dark yellow, a much desired color. Accordingly it was hoped that neutron radiation would result in mutation to dark yellow. However, quite evidently, much higher dosage rates must be used and probably more buds radiated in order to inactivate both of the dominant YY factors present. Contrariwise, Queen Elizabeth has only one dominant TNT factor which was relatively easy to inactivate, as judged by the frequency of sectoral chimaeras for light scarlet.

Both the red H56024/39 and Golden Rapture are of Hybrid Tea ancestry. Unlike Queen Elizabeth they are not Fl hybrids between a many-petaled, small-flowered Floribunda and a large flowered Hybrid Tea. Accordingly they do not carry factors for many petals. Hence their variability potential in this respect is low. We know that the single, 5-petal condition is recessive. As noted above, both H56024/39 and Golden Rapture mutations to 5 petals were found, as well as intermediate conditions. Mutations increasing the number of dominant factors for increased petalage simply did not occur.

CR In 1931, McClintock studied X-ray induced "mutations" in maize. She reported her tesults in her paper, Cytological observations of deficiencies involving known genes, translocations and an inversion in Zea mays. Missouri Agr. Exp. Sta. Res. Bull. 163, l-30 (1931).
      In her 1983 Nobel Lecture, she summarized briefly, "None of the recessive phenotypes in the examined plants arose from 'gene mutation'. Each reflected loss of a segment of a chromosome that carried the wild-type allele, and X-rays were responsible for inducing these deficiencies."

Finally the stability of the R. multiflora species is highly significant. Again mutations to the dominant, many-petaled condition simply do not occur. Likewise mutations to the A factor, basic to pigment formation, the dominant R factor for crimson red, or the S factor for dark scarlet did not occur. Again it is significant that there were no variations to increased flower size. R. multiflora is a semi-trailing, climbing type of plant. Though the sample is small, it is also significant that no mutations to upright bush type of growth, or recurrent flowering occurred. The conclusion is inescapable that mutations to dominant traits are difficult to effect by radiation.

In other words the success of a radiation experiment depends, not only on dosage rate, but the variability potential of the variety used.

Biologically stated, all changes were toward weaker less viable types, certainly not comparable to Queen Elizabeth in vigor or survival ability. In fact, it was quite clear that as the plants developed, the most viable tissue combination "took over" in that if any normal unchanged cells were present, eventually buds formed from them and they became the main plant. In the case of the 1957 experiment, evidently the 20 plants which remained changed and transmitted their changed traits by budding, were of homogeneous cell type. However, as these grew from buds, one could observe segregation of tissues and the plants grew progressively stronger as the cells with best gene combinations "took over" and became buds growing into uniform-appearing but changed plant type.

Some of these obviously differ from Queen Elizabeth by many factors. Since all are either completely sterile or semi-sterile, it may never be possible to determine their exact genetic make-up. Crosses of four, which showed a low percentage of apparently good pollen back to Queen Elizabeth, do not give much promise of setting seed.

Though it is recognized that many of the mutated types may well be "abnormalities" involving chromosome disintegration and so not be pertinent to the following discussion, many obviously are rather simple changes to the recessive condition. Certainly we have here a technique by which we can observe more “mutations" from one group of 50 neutron-radiated buds than usually is possible in a lifetime of looking at several million plants per year in the field run of plants from unradiated buds.

Are then these changes the sort of thing leading to evolutionary divergence? In this respect the results of neutron radiation of rosebuds checks closely with reports of other radiation work. By far the greater majority of all mutations found are defective. Occasionally as in the rose experiments, some are found which horticulturally are desirable. Even certain ones are recorded in the literature as showing increased resistance to some plant disease. But so far careful inquiry has brought out that as in roses, associated defects make the survival value of the mutant type in a natural environment highly questionable.

It would seem then that the idea that evolution can occur through the accumulation of mutations is simply not borne out by the ever-increasing array of experimental evidence. In fact, it becomes more and more clear that species and even varieties are very complexly integrated units tolerating very little "tinkering" or change. Quite obviously some species such as the chromosomally basic R. multiflora species are particularly resistant to change. Others have a greater variability potential.

Quite obviously the ability of species and varieties to change as they spread over the surface of the earth has been pushed out of all proportion into the idea of evolution. The inability of older naturalists such as Darwin to evaluate this variability potential of species and varieties properly and so to mistake it for evolution or change of one species into another is quite understandable. After all, they had no clear concepts of genetics and even set up genetic postulates completely at variance to what we now know occurs. But the continued interest and belief in evolution, at least in some modified form, on the part of so many modem biologists is puzzling, since the facts of genetic variability and mutation so clearly show that species and varieties have such clearly defined boundaries of variability.

4. Lammerts, Walter E., "Effect of Photoperiod and Temperature on Growth of Embryo-Cultured Peach Seedlings," American Journal of Botany 30, No. 9: 707-711.
CR Here he claims that species and varieties possess great "variability range", a fact widely discussed through much of the 19th and 20th centuries, while also asserting that students of evolution believe that all evolutionary change is slow because it depends on gene mutations that occur along the way. He ignores the question, "How can a species remain relatively stable in phenotype while concealing great genetic variability?"
The closest I've found to Lammerts' view is Calvin Bridges, a fruitfly geneticist (not evolutionist) who traveled to the Soviet Union to say that plant breeding is a waste of time.

Contrariwise, the idea of evolution has had a deadening effect on the minds of many, especially younger and less imaginative research workers, since the very concept tends to make one feel that changes come slowly. Actually, plant breeding experience shows that within the limits of species or variety—variability potentials changes can be effected very rapidly.CR In fact, with the development of modern tools of radiation research, we can in a few months probe the variability range of even such normally slow reproducing species as apples, pears, and peaches. Thus by neutron radiating buds of peach varieties and growing the budlings under continuous light,4 we can get a crop in one year from radiated buds from which we can see the entire range of variability potential. However, we also must be clearly alert to the fact that once the range of variation is reached, further attempts to increase variability are a waste of time.


  1. Data are given showing technique and dosages use to get maximum percentage of mutations from rosebuds by neutron radiation.
  2. Varieties differ strikingly in their variability potential. Thus variations in Queen Elizabeth toward increased petal number, smaller flowers, and change in color from carmine rose (presence of M factor) to coral pink (m) with s factor for light scarlet occur frequently. Change to white occurred only once.
  3. Horticulturally, neutron radiation of buds shows great promise, since it so greatly speeds up mutation.
  4. Biologically, all mutations are defectives, hence any postulate of evolution by mutation is contrary to evidence.
  5. It is observed that the concept of evolution has had a bad influence on research in that it tends to make the scientist work on the theory that nature acts slowly and mutational changes occur slowly. Hence he tends to limit himself as to possibilities of getting results rapidly.


The author wishes to acknowledge the advice and help of Philip Livdahl, Howard Tewes, and Harlan Zodtner in planning and executing the radiation of the rosebuds in these experiments.