clicking on this icon will always return you to this page

What's new

English Roses
Hyb. Bracteatas
Hyb. Musks
Hyb. Perpetuals
Hybrid Teas
Modern Shrubs
Moss Roses

Site Index


The ABC's of Hybridizing
by Roy Sheppard

The subject of hybridizing has been discussed so frequently in this and other publications that further discussion might appear to be unnecessary. Dr. Nisbet, however, apparently believes otherwise as he has asked for another article on this intriguing sideline of rose growing. He demands, however, that it be non-technical and, wherever possible, confined to words of few syllables. This suggestion implies, and I concur with it, that there are probably many rose lovers who should be enjoying this hobby, but are not because they feel it is a highly complicated procedure requiring specialized training. This is definitely not true as no super-intelligence is required in performing the various operations. Many of our most valuable introductions have been originated by persons who were not specially trained for the work. They knew and loved the material with which they worked and were endowed with sufficient optimism and enthusiasm to counteract their lack of scientific knowledge. There is no reason why you can't do as well. Whether you do or not, you will enjoy yourself and gain much knowledge of roses.

Although the facilities of the amateur plant breeder and his knowledge of the subject may be considerably less than that of the professional, he may luckily discover a combination of varieties that the latter has overlooked and originate an outstanding new rose. Failing in this, he will at least spend many pleasant hours in trying to do so. With enthusiastic anticipation he will watch the development of the seed hip, the growth of the seedling, the unfurling of the first petals and subsequent tracing of its heritage of color, size, petalage, form and stature. Do not expect that your first attempt at rose breeding will result in the creation of a remarkable new variety. Such a result is possible but not a certainty. This is particularly true when working with modern roses, as they are of a very mixed ancestry and results are definitely unpredictable.

The fact that there is a certain amount of speculative uncertainty as to what the result will be adds interest to the work. Frankly, rose breeding is a fascinating game of chance between man and plants and luck is apparently an important element. However it is not all important, as patience, perseverance and the ability to formulate a definite plan or objective are also prime requisites if success is to be achieved. The beginner will invariably commence by planting seeds developed from unknown pollen or by playing hunches and crossing any varieties that occur to him. He will probably have but a vague idea as to the results he hopes, or expects, to obtain. Experience is a valuable teacher and, if this haphazard method does not lead to too many disappointments, no harm has been done. On the other hand, it is much better consciously to set a goal and then select parents carefully in seeking to achieve it. The objective may be reached more easily and in less time, this way.

Your first improvement may be minor, but the experience gained will help with later work. Briefly, have an objective visualize the type of rose you would like to create and then study the material you have for use as parents. Then eliminate those that possess undesirable characteristics. Unfortunately it appears to be far easier to transmit undesirable than desirable characteristics.

If a desired quality is not apparent in a variety the chances are definitely in favor of it not appearing in the offspring. It is not entirely impossible, however, as the gene (a unit of inheritance) for a desired characteristic may be present in the chromosomes of the plant but may not have influenced its external appearance or performance. Perhaps it may assert itself in a later generation but the odds against its doing so are so great that the beginner is rarely justified in making the gamble. The surest and best policy is to cross only those varieties that you are reasonably certain possess the characteristics you wish to combine. Then you have some assurance that the progeny will be endowed with only the best.

Acquired characteristics such as those induced by environment or culture are not transmitted to the offspring, even though Luther Burbank was firmly convinced that they could be. Certain viruses, such as tulip mosaic, may be transmitted but they should be classed as "inherited diseases" rather than characteristics. Do not think that you may combine the good qualities of several parents and create a horticultural "hash" by placing the pollen of several different roses on the stigmas of one. It can't be done, as only the first good pollen to reach the stigmas is effective. This is a common fallacy of the beginner. Although the possibility of obtaining, let us say, a blackspot resistant rose by crossing two extremely susceptible varieties or a deep red rose from two light colored parents is slight, it is not impossible.

Our modern hybrids carry the genes of many ancestors and it is practically impossible to predict the results of any specific cross. A single gene may control blossom color and others leaf shape, stature, disease resistance or the other characteristics which go to make up the individual plant. Due to their minute size no one has ever seen a gene but there are probably thousands in every hybrid rose and each influences the development and inheritance of some part of the plant. The genes for many colors exist in our modern hybrid rose. Some of these are dominant and others are recessive. Crimson Glory, for example, is crimson because the gene, or a combination of genes, for that particular color are present and are dominant over the other genes which the variety carries. Roses bearing blossoms of various other colors are found among its ancestors and it apparently contains genes for practically every color known in roses. Although now recessive, they may become dominant in future crosses involving Crimson Glory. In fact they already have, as this famous rose is the parent of several roses of entirely different colors.

It is generally agreed that a single dominant gene governs leaf texture, disease resistance or susceptibility, plant stature, petalage. The interaction of several genes control vigor, fragrance, thorn structure, rigidity and length of flower stem and shape of flower. In breeding roses you will soon discover that the dominant colors (those which occur most frequently) are those of the wild, or species, roses; pink to rose red. Other colors, such as dark red, scarlet, orange, yellow or white, are recessive to these. Dark yellow and white are also recessive to light yellow and cream.

In early times, a plant that customarily bore five petaled blossoms occasionally produced one of increased petalage. By selection over many years, quite double flowers were eventually produced. When the seedling, that was later named Charlotte Armstrong, produced its first blossom it had but 11 petals and, although the originator, Dr. Lammerts, was somewhat disappointed in the petalage, he admired the color and bud form so greatly that a few plants were budded. Today the blossoms of this well known variety have about 40 petals. The first blooms produced by a seedling are rarely a true criterion of later production. The color rarely changes but the petalage may become greater as the plant becomes more sturdy. There is also a definite possibility that self pollinated seeds obtained from the plant will produce seedlings that more nearly approach your objective.

The real skill required to become a successful hybridist comes not in simply applying the pollen of one blossom onto the stigmas of another but in selecting the proper parents. Unfortunately all roses are not capable of producing seed or supplying functional pollen. Others may produce seed of poor germinative quality or have a positive tendency to pass on undesirable characteristics rather than good ones. The following list of proven seed plants may save you considerable experimentation and many disappointments. They all accept foreign pollen, produce good seed and possess one or more desirable characteristics.

There are many others equally good but some limitations must be placed on the list.


Ami Quinard Golden Scepter Mrs. Paul R. Bosley


Baby Chateau Golden State Narzisse

Blanch Mallerin Granat Natural Flower Guild

Buccaneer Independence Nocturne

',,.California Centennial Juno Ophelia

Charlotte Armstrong Korovo Opera

Charles Gregory Lady Leconfield Poinsettia

Charles Mallerin Lady Margaret Stewart Prima Donna

Circus La Jolla Red Ripples

City of Norwich Li Bures Roundelay

Commonwealth Luna Show Girl


Crimson Glory Masquerade Sir Henry Segrave

Daisy Bud Maureen Thompson Sister Therese


Duchess of Wellington McGredy ' s Pink Souv. De Claudius


Dusky Maiden Michelle Meilland Tallyho


Ena Harkness Mme. Butterfly Valentine

Etoile de HolIande Mme. Nicholas Aussel Valsheda


Fandango Montezuma Velvetier

Flambeau Mrs. Sam McGredy World's Fair


Whether the pollen parent or the seed parent exerts the greatest influence in passing on certain characteristics to the progeny is a much debated question. Professional hybridists disagree on this matter and the consensus of opinion tends to the belief that no definite recommendations can, or should, be made. Reciprocal crosses should be made whenever possible. That is, make each cross both ways. First put "B" pollen on "A" variety, and then "A" pollen on "B." In some instances roses that originated as sports will give better results if used as pollen parents rather than seed parents. In explanation I might state that, in sporting, either the entire cell structure or only the outer layer of the plant may be changed. Since the pollen grains are derived from a more superficial layer than the embryo sacs the mutant gene is more likely to be present in the former than in the latter.

Few roses are entirely sterile but many are partially so. Quite a few of the early Floribundas rarely set seed or produce functional pollen, as they are triploids resulting from crosses of the tetraploid Hybrid Teas and diploid Polyanthas. However, the most seemingly sterile of all roses may bear an occasional seed. If you wish a long-shot gamble apply the pollen of such a rose on a Hybrid Tea that has proven its worth as a seed parent. The progeny may be a tetraploid that will be extremely fertile. This procedure has been effective in adding to Hybrid Chinas, Bourbons, Floribundas, and other groups.

Another thought that deserves mention is that continued inbreeding of closely related varieties results in lack of vigor. A fairly sturdy progeny may be expected, however, when roses of the same class but of different lines of breeding are crossed. For example, Charlotte Armstrong, which is comparatively vigorous, resulted from a cross of two very distinct Hybrid Teas: Sister Therese and Crimson Glory. The fact that hybrids are often more vigorous than either parent may be attributed to a rather interesting trait of nature called hybrid vigor, (heterosis). This well established principle of genetics results from the union of complementary factors for vigor. As like begets like and it is far easier to transfer weakness to the offspring than strength, the breeder should, whenever possible, choose only the most vigorous varieties as parents. In instances where a weak growing variety must be used it is sometimes advantageous to use the weaker as the seed bearing parent. Choose not only the best of the newer varieties, but also select a few of the old standbys whose value over a period of years has been proven by their continued popularity.

If the beginner is of an investigative trend of mind he undoubtedly wonders what transpires when he applies the pollen of one variety to the stigmas of another, how the characteristics of the two varieties intermingle and how these two entirely different sets of characteristics are incorporated in one seed that might produce a plant that resembled one parent, both parents or, possibly, neither parent. If we are to understand the principles of cross-fertilization we must first familiarize ourselves with the structure of the flower, its organs and the names by which they are known. The structure of the rose is possibly more elaborate than that of some other flower, but its organs are clearly defined.

Having picked a well developed rose, let us begin our study of its anatomy by first removing the petals and sepals so that they will not interfere with our observation of the more important sex organs. Then use a sharp knife or razor blade to cut the remaining portion longitudinally to the center of the stem. We then find, exposed to the eye, the various organs of the flower in which the rose breeder is, at present, interested. The stamens, or male organs, are usually quite numerous and form a complete circle around the pistils, or female organs. At the tip of each stamen will be found a small bilobed expansion or knob which is referred to as the anther. The minute golden grains of pollen, which are the male element (or sperm), are produced in the anthers and are discharged as they ripen. The pistil may consist of one or many thread-like styles each of which terminates in a somewhat enlarged knob known as the stigma. Each style has at its base and attached to, it the female cell, ovule or unfertilized egg. This ovule with others is enclosed in the ovary which is encased in a relatively large green cylindrical body known as the hip or fruit.

As we examine the structure of the hip we can not fail to be impressed with the fact that nature uses many precautionary methods to protect the important organs of perpetuation of the plant. Pollen applied to the stigmas, when they are in a receptive condition, is held in close contact by the sticky substance on the surface of the stigmas until the tiny pollen grains absorb sufficient moisture to begin growth. At this stage a slender thread-like "drill," known as the pollen tube, bores its way down through the center of the style, in search of a tiny egg in the ovary. In the tip of this is a tiny germ called the male gamete, (a reproductive cell), which escapes into the ovule as the "drill" enters it and quickly locates and unites with the unfertilized egg.

As the sole purpose of the flower was to insure reproduction of the plant on which it was borne and this purpose has now been served, the stigmas and styles soon assume a dried or withered appearance. The cells formed by the fusion of the male gametes and eggs grow, divide, redivide, and eventually develop into tiny embryos. Upon reaching a certain stage their development is arrested, they become dormant, and food in the form of starch, sugar, fat and other substances is packed around them; and hard shells are formed to protect them. The seeds in the hip have now attained maturity and may be harvested.

We are certain that the seed. formed by the union of the male gamete and the egg of the mother plant should contain characteristics of both plants, but the proportion in which they exist is problematical. One desirable characteristic may be dominant while another may be recessive. We quite often find that a hybrid may evidence characteristics that are not clearly defined in either parent. In some instances the issue may very closely resemble one parent and show none of the characteristics of the other, while in other cases they may closely resemble a forbear of either parent. Occasionally they will combine the defects of both parents, a good quality of one and a poor quality of the other or the good qualities of both. All of these tend to make hybridizing an exceedingly interesting pastime and, although many disappointments may be encountered, there is always the hope, and possibility, that your objective may be suddenly reached.

If the offspring of a seemingly desirable cross may appear to be inferior to either parent do not lose hope. Such a plant should not be discarded, as it may be used as one of the parents for another cross. It contains the genes for the characteristics you desire and they may assert themselves later. On close observation a seedling may exhibit a characteristic that is not apparent in either parent. This variation from type may be noted in the color or petalage of the bloom or in habit of growth. Many experiments have proven that there is a greater possibility of obtaining desirable recombinations of the characteristics of the parent plants in the second gereration than in the first. It is quite possible to obtain choice seedlings from mediocre first generation hybrids and seeds of these plants should be collected and sown.

The actual mechanics of rose breeding, from pollination to seedling care, consists of several operations. None require a great amount of mental or physical effort and each offers an opportunity to pleasantly anticipate what the ultimate results will be. I will endeavor to simplify the various operations by discussion, each in the order that they are performed. The first step, of course, is to choose the male and female parents that you contemplate using. From the former cut a flower bud (with about a one inch stem) that is almost ready to open and with the aid of a pair of tweezers remove the sepals and petals. Put the remaining brush-like portion in a small container (jar lids are suitable) and place it in a warm room but not in direct sunlight. Within a short time the anthers will discharge their pollen and you have a ready made brush with which to apply it to the stigmas of the seed bearing parent. Personally I prefer this method of pollen application, as it wastes less pollen and avoids the necessity of sterilizing camel hair brushes when a pollen change is to be made. Sterilization of brushes is effected by dipping them in alcohol and permitting them to dry thoroughly before they are again used. This seemingly is a waste of time and effort when nature herself applies an equally acceptable brush. However if you prefer to use a camel hair brush simply remove the anthers and place then in a receptacle until they ripen. If a scarcity of blossoms exists the latter method has one advantage as the bud from which the anthers (the male organs) were removed may also be used as a seed bearing parent. Nature is extremely extravagant in some of the material she supplies and enough pollen is usually discharged to pollinate several blossoms. It retains its potency as long as it remains dry, but is worthless otherwise. In fact it may be kept for several months if wrapped in cellophane packet and placed in a closed jar containing a small amount of calcium chloride. A handful of this material can usually be obtained from your highway department for the asking, as they purchase it in large quantities. So much for the pollen parent.

The seed bearing parent is prepared at approximately the same time you collect the pollen bearing anthers and the bud selected should be of about the same degree of development. Simply remove the sepals (this assists in locating the hip, if the identifying tag is lost), the petals and all the stamens. After this operation of "emasculation" has been performed, cover the remaining portion of the bud with a glassine, or paper, bag to prevent unwanted pollen from reaching the stigmas. Glassine is preferable to cellophane, as considerable heat is generated under the latter and this may cause injury to the stigmas. Incidentally, don't feed the seed bearing plant too generously, as a well fed plant is somewhat reluctant to set seed. The inherent tendency of all plant life to perpetuate its kind is evident in the freer production of seed by a hungry plant than by a complacent, well fed one.

There is a difference of opinion as to the best time of day to pollinize. Some hybridists insist it should be done in the morning, others at noon, while a third group insists that best results are obtained if the pollen is applied in late afternoon. After several years of this work I am firmly convinced that the time of day is of small importance if the stigmas are in a receptive stage of development. The proper conditions may exist for from an hour or so to a day or more and is dependent, both upon the type, or variety, of rose and weather conditions. It is evidenced to the eye by the appearance of a wet or glutinous substance on the surface of the stigmas. If this is apparent, the time has arrived to apply the pollen but if the stigmas have a dry appearance they are either immature or too ripe. In the latter event discard the bud; in the former, postpone the operation until you are certain the stigmas are receptive. The time of year in which pollenation must be performed is far more important than the time of day.

Be sure that you make your crosses early enough so that the seeds will ripen before cold weather. In northern Ohio (hardiness zone V) pollinization. is begun as early in the season as possible and ceases about the first of July. The hips commence to ripen, about the middle of September and are harvested as soon as they begin to lose their normal "growinggreen" color or the stem begins to darken. Extremely premature fall usually denotes that the seeds are hollow and therefore valueless. Germination is somewhat more rapid, and better, if the seeds are not permitted to become too ripe or are not dried unnecessarily between removal from the hip and planting. Both of these apparently cause the seed shell to become extremely hard and delays, or prevents, the emergence of the cotyledons.

Previous to harvesting the seeds, some thought should be given to the media in which they are to be germinated. Experiments along this line are mentioned in the 1954 American Rose Annual and observations since then confirm the thought that none is superior to a mixture of equal portions (by bulk) of soil, sharp sand and sieved peat moss. The seedlings experience no difficulty in breaking through this friable moisture retentive mixture and may be removed without injury to the roots or adjacent seeds. As dampening off is an ever present problem in seed germination, the soil and sand should be sterilized before mixing with the peat moss. This may be accomplished by baking, or by diluting two and a half tablespoonfuls of formaldehyde in a pint of water and mixing this thoroughly in a bushel of the soil and sand combination. Cover for three or four days so that the formaldehyde may permeate the mixture and then expose to the air until all odor of the formaldehyde has disappeared. It is then safe to combine it with the peat moss and use.

The seeds are sown to a depth of about three sixteenths of an inch in flats or pots and these are watered and placed in a cool but not freezing location during the germination period. Germination is thought to occur best at temperatures near 41 degrees. This temperature is difficult to maintain consistently in a cold frame or greenhouse and a variance of several degrees, one way or another, does not appear to decrease the rapidity of germination to any noticeable extent. A fruit cellar usually offers the best possibilities as a location for the seed containers, as a rather uniform temperature may be expected there. The seedlings, of course should be transplanted into individual pots and exposed to light as soon as possible after they emerge. They require light and greater heat than that provided by a fruit cellar to develop properly. They also require a richer soil than that in which the seeds germinated. During the short days of winter, artificial light will accelerate growth and that is of extreme importance to those who live where the growing season is short and who do not own a greenhouse. The seedlings should be planted in the garden as soon as the ground can he prepared and the danger of hard frosts is past. They should be fed and watered generously. To assure survival during their first winter out of doors, the seedlings should be forced to maximum growth before midsummer and then hardened off, by withholding feeding and excessive watering. If the preceding procedure appears to be impractical under your conditions, excellent, but less rapid results, may be obtained by planting the seeds in a cold frame. By this method they will not germinate until spring, however, and part of the fun in breeding is derived from watching the seedlings develop during the cold winter months. In fact some of the house grown seedlings may bloom before the outdoor planted seeds germinate.

A few words of caution should be inserted at this point; field mice are extremely fond of rose seeds and, whether planted indoors or out, every precaution should be taken to guard against them.

Germination often begins as soon as one month after planting (if planted indoors) and will continue spasmodically until the advent of warm weather. When it ceases, the media should be dried off, as moisture is unnecessary during the summer and causes the soil to become unnecessarily compact and mold to form on its surface. If flats are to be retained for the second year, watering should be resumed about the first of October. Approximately 85% of the total germination will occur during the first winter, about 10% during the next and remaining 5% over a period of three to five years. Dr. Van Fleet retained his flats for seven years, but space limitations prevent many of us from following suit. In fact only those flats that contain seeds of particularly difficult crosses or seeds from parents that have proven to be slow germinators are kept for more than two years. Use your own discretion in

this matter. The influence of the pollen parent is apparent not only in the time required for germination, but also in the season of ripening, the number of seeds the hip contains and, to some extent, the size of both the seeds and hip. This fact is apparent to anyone who has worked extensively with the species roses. In other words, do not discard seeds of a cross that have not germinated because other seeds borne by the same seed parent have. There may still be hope if the pollen parent of that particular cross is a slow germinator.

Dampening-off is a perpetual problem to all those who grow plants from seeds. If both the germinating media and the potting soil have been properly sterilized, sanitation has been practiced and watering is done early in the day, there is little possibility of it affecting your seedlings. Blackspot may occasionally appear on a rose seedling, but sanitation, morning watering and an occasional dusting with a known control will usually lick it. Mildew is a more likely possibility, but it too may be prevented by providing adequate ventilation and heat to reduce humidity during periods of low temperatures. A light dusting with a good sulphur dust or frequent syringing of the plants with water will keep it in check. Mildex at the rate of one level teaspoon to two gallons of water with about one teaspoon of Dreft or other detergent added is remarkably effective in the eradication of mildew should other methods fail. Aphids occasionally appear on indoor grown seedlings but are easily destroyed by picking or a nicotine spray.

An occasional seedling may give all the appearance of being sturdy and normal but suddenly die. Or it may be constitutionally weak and lack vigor at all stages of growth. After eliminating the possibility that these conditions may have been caused by disease, pests or injury we might assume that a "conflict" of genes is the cause. In this instance a certain incompatibility exists that prevents the normal development of the seedling although seed germination was apparently normal.

When your seedling produces its first bloom, study it critically. If you believe sincerely it is superior to any existing variety ask the opinion of others. If they agree (you may be understandably prejudiced) send a bloom to your favorite rose nurseryman and ask his opinion. He will be frank, as he is not interested in introducing a rose of mediocre quality but is continually on the lookout for those that are distinctive. Although he may not show enthusiasm you will at least derive some satisfaction from the fact that the seedling is the result of your efforts and that there is no other rose exactly like it.

Protection during the first winter presents somewhat of a problem in the colder areas of our country, as a one-year seedling lacks the cold resistance of an older plant. Satisfactory results have been obtained in our garden (where 10-12 degrees below zero is the expected minimum) by placing a heavy mat of straw over each seedling. This is applied following a few heavy frosts and removed about March 15th. If your seedlings are not too numerous, they could be mounded with earth or, if you are particularly hopeful of a particular one, pot it up and place it in a cold frame or other cool location over winter.

Undoubtedly rose breeding is somewhat of a gamble, but a legitimate one, in which you may win big stakes and can suffer no loss. The only possibility of loss is that of time and I assure you this will be lost enjoyably. Perseverance and optimism are necessary requisites of the successful hybridist. Though many have failed to accomplish a certain cross you may succeed and it's a lot of fun trying. R. spinossissima altaica for example, has intrigued hybridists for many years as it possesses several qualities that commend it as a parent. Several well qualified breeders have endeavored to combine its extreme hardiness and good plant habits with the remotant qualities of our modern roses. To the best of the writer's knowledge the combination has been successfully effected but twice. These two exceptions to the usual failures are the white Karl Forester created by Kordes, (Frau Karl Druschki X R. s. altaica) and our own Golden Wings (Sister Therese X R. s. altaica). Both are hardy and both repeat generously. I'll confess that several hundred similar crosses were made before Golden Wings resulted and I assume that Kordes also failed many times before he achieved success.

Before concluding this discussion I feel it is necessary to stress the importance of keeping an accurate record of all crosses. Such a record, supplemented with a copy of Modern Roses IV, will increase interest as it will enable you to trace the heritage of each seedling. It will also be of invaluable help in making future crosses involving the seedling. The record should begin at the time of pollinizing and should consist of the parentage, dates of seed harvesting, planting and first germination and seedling performance. The parentage is usually first recorded by attaching a tag bearing the name of the seed and pollen parents just below the hip when the cross is made. The seed bearing parent should appear first and the customary procedure is to record it thusly -Buccaneer X Goldilocks. Dangling tags such as those used in clothing stores were once considered suitable for this purpose but the string often breaks and the tag is blown away. The string may also cause injury to the stem in windy locations. A one half by three inch strip of paper pinned around the stem, with writing denoting the parentage on the inside, has proven very satisfactory. The paper should be hard surfaced so as not to deteriorate too rapidly and, if brightly colored it is less liable to be overlooked among the foliage. We have found that red or orange are desirable colors. As the seeds are planted, the cross is customarily given a number and this, with the other data, is recorded in a book. The number is associated with the seedling until it is discarded or proves worthy of a name.

In spite of progress made by hybridizers in recent years, there is an unlimited field for the improvement of garden roses. Even the most ardent rosarian will admit this fact, for no known rose is perfect; none has all the qualities we desire. We have yet to gain the rose that is a heavy and continuous producer of well formed, fragrant blossoms of rich color, has good habit of growth, is disease resistant and is dependably hardy without protection in all sections of the country. All of these characteristics are represented in one or more known roses but are combined in no one variety. Seemingly their combination is possible a goal toward which the amateur hybridizer should strive. Many of the plants that beautify our gardens are the result of the efforts of the amateur breeder whose main compensation has been the pleasure the work has afforded him. Some few have been financially rewarded, but the keen anticipation one enjoys while watching the growth of the seedling and the development of its first blossom, be it worthwhile or useless, is reward enough for most of us. I wish you success.

Back to propagation articles

This website made possible by a grant from the Uncommon Rose

There have been  hits on this page. WebCounter does the stats.
Original photographs and site content © Paul Barden 1996-2003