PEAR BREEDING
Valuable Group of Fruit Trees Lends Itself Readily to Cross-Pollination, and Valuable Results Are Possible—Considerable Time Needed—Methods of Operation

[1916. Journal of Heredity. October. 7(10):435-442.]

   HYBRIDS are easily made in the pear genus (in which many botanists now include the apple, quince, medlar, etc). and many of these hybrids offer valuable commercial possibilities. As trees of this kind are accessible to nearly every one in the temperate regions of the world, it is a particularly good group for amateur plant-breeders to work with, provided they are not in a hurry.
   The European pear (Pyrus communis) is found throughout southern Europe and Asia, being indigenous as far east as Kashmir. It has been cultivated since the dawn of history, and selection of the best. with such cross-pollination as has been done by nature, seems to be responsible for the production of such superb pears as those which are known in Europe as the beurre type, and which are represented in the United States by the Bartlett.
THE BARTLETT PEAR AND ITS ANCESTOR
The wild pear which grows throughout the southern parts of Europe and western Asia was brought under cultivation in prehistoric times, and has evolved into the delicious fruit of the Bartlett type, a small specimen of which is shown below. The wild fruit, shown above, is not actually the Pyrus communis which is the reputed ancestor of the Bartlett, but the Snow Pear (P. nivalis), the fruit of which is practically identical. The leaves differ, but the differences are so slight that some botanists have considered the Snow Pear nothing more than a variety of P. communis, a good photograph of which was not obtainable. Photograph (natural size) from the U. S. Department of Agriculture. (Fig. 1.)

   Pears of this kind were among the first fruit trees imported by the early settlers in North America. Some time before the middle of the nineteenth century a very distinct species, the Chinese Sand Pear (P. sinensis [changed to P. serotina and now P. pyrifolia -ASC]) was introduced. The fruit of this species is considered good by the Chinese and Japanese, but it is hard and lacking in flavor, and full of stone—cells. When one's teeth crunch into these, it is not difficult to understand why this variety got the name Sand Pear.

   Peter Kieffer, of Roxborough, Philadelphia, an Alsatian gardener who died in 1890, grew some of these Sand Pears, according to L. H. Bailey, “and sold the seedlings as ornamental trees, for this species is of very distinct and handsome growth and the fruit is ornamental and fragrant. Alongside the Sand Pears were Bartletts, Among one of the batches of seedlings from the Sand Pear he noticed a plant with different foliage, and this he saved. Its fruit was found to be superior to the Sand Pear, and it was introduced as the Kieffer. The Kieffer pear is now very popular in many parts of the country because of its great vigor, healthfulness, productiveness and the keeping quality of its fruit. In point of quality the fruit is distinctly inferior, but it meets the demands of the market and is an excellent fruit for canning." [This opinion changed over time. Some better alternatives were identified in a Georgia study.
   There is no doubt that the Kieffer, which is second only to the Bartlett in popularity in the United States, is a hybrid of the Bartlett and the Sand Pear. While the circumstances of its origin, as told by Bailey, would not be conclusive on that point, there are plenty of other crosses of a similar nature, many of which are in the trade. The Le Conte is perhaps the best known of these hybrids, after the Kieffer.

THE KIEFFER, A VALUABLE CHANCE HYBRID
The popular Kieffer pear, a fine specimen of which is here shown, natural size, originated as an accidental cross in a Philadelphia nursery half a century or more ago. Its parents were the Bartlett pear and the poor Chinese Sand pear shown in Fig. 2. The Le Conte and several other less-known varieties have come from the same sort of hybridization. Although many epicures have condemned the quality of the Kieffer, it holds its place as one of the most widely grown varieties in the United States.“ Photograph from the U.S. Department of Agriculture. (Fig. 3.)

ROOM FOR NEW VARIETIES

   One who undertakes pear-breeding need not expect to produce, at the first attempt, a hybrid that will supplant the Kieffer, although many connoisseurs have declared that it would not be hard to surpass the quality of that popular variety. There is need of varieties that will he hardy and disease-resistant, and that will extend the fruiting season, and any one who undertakes the task of pear-breeding intelligently and persistently has a chance to succeed.1
   “In the improvement of any fruit," says W. R. Ballard of the Maryland Experiment Station,2 “an intimate knowledge of varieties is essential to intelligent effort. Such information should include not only the adaptability of varieties to climatic conditions, habit of growth. season, productiveness, character and quality of fruit, but should also include as far as possible information concerning disease resistance and historical data in regard to parentage. The great majority of varieties grown in this country are undoubtedly of hybrid origin. This is due to the promiscuous cross-pollination carried on by bees and other insects. This fact makes the value of a variety for breeding purposes somewhat problematical since many characters show up in the progeny which cannot readily be detected in the parents This hybrid nature, however. works to the advantage of the breeder of orchard fruits for wide variation occurs in the first generation, so that he does not have to wait until the second generation for such results, as would be the case were he working with pure strains. Investigations here and at other stations indicate that the greatest improvement is to be secured by using for breeding purposes only the best of the varieties now in existence. Only occasionally is it necessary to depart from this rule in order to utilize some particular character of an otherwise inferior variety.
   “Since pollination can be done only during a comparatively short blooming, period, it is essential to be thoroughly prepared for work when the time comes, There is more or less variation in the time of blossoming of different varieties, although in abnormal seasons all varieties have been known to bloom at about the same time. Advantage can often be taken of this Variation by using the early bloomers as the male parent. This enables one to collect pollen in workable quantities before the buds of the later varieties are ready to be pollinated. Weather conditions are also more likely to be favorable for pollinating work at this later period.

MAKING POLLINATIONS
   “Normally the anthers burst and the pollen is shed shortly after the blossom opens This usually takes place early in the day. In cross-pollination it was formerly customary to break off recently opened blossoms and brush the stigmas of the emasculated buds with the opening anthers. There was always danger in this method that bees or other insects had already visited the flowers leaving a deposit of foreign pollen in the blossom used for pollinating. When the purpose was simply to get variation without regard to parentage, this method worked well in a small way but where a study is to be made of the laws of inheritance more accurate methods must be followed.
   “A simple method used by the author to insure a good supply of pollen is to gather a number of buds which are just ready to open but not far enough advanced to allow the entrance of insects. This would usually be about a day before they would normally open. By grasping the petals between thumb and forefinger, they can easily be rubbed off, together with anthers. These are then collected on a newspaper and after the loose petals and other waste materials have been sifted out, the anthers are transferred to glass dishes (petrie dishes are excellent for this purpose) and placed in a moderately warm dry atmosphere. The anthers will soon burst open releasing the pollen grains. When the pollen becomes dry it is protected from moisture until needed. Handled in this way it may be kept for several days, although it is always advisable to use it as soon after being collected as convenient. Should doubt arise as to it[s] viability, a test of its germination power is made by placing a small quantity of pollen in a 2% solution of sugar and keeping it at a moderate temperature. If upon examination under a microscope after a few hours, the pollen grains are found to have germinated and sent out their pollen tubes, it is usually safe to use the remaining pollen for crossing. Each kind of pollen is carefully labelled in order that no uncertainty may arise as to its source.
   “In order to effect a cross the buds to be pollinated should first be emasculated so that self-fertilization may not take place. The following method has been used by the writer for several years with pears and apples with excellent results. The bud is held firmly between the thumb and forefinger of the left hand. Emasculation is then accomplished expeditiously by inserting a well-sharpened scalpel into the side of a bud just below the calyx. Then by giving it a quick upward jerk, the whole corolla is removed, carrying with it all the anthers. Pear buds are especially easy to work in this manner, but in the apple the abundant pubescence soon dulls the scalpel necessitating frequent sharpening for the best results.

FEW BUDS REMOVED
   “Fruit buds of the pear and apple are borne in clusters of six or more. Ordinarily only one or two of these buds set fruit and come to maturity In hybridizing work the question arises as to the number of buds in the cluster which should be emasculated and pollinated. The central bud develops previously to the others so that it is safer to remove it. The best results have followed the working of the majority of the buds in the cluster, since the emasculation can be done in about the length of time it would take to remove the extra buds, and pollination requires very little longer time. Sometimes several clusters are close together and may be enclosed in a single bag. As many as eleven fruits of the Seckel pear have been secured under one sack in this way.
   “Until recently it has been customary to emasculate the buds, cover at once with a paper sack and leave for a day or so until the stigmatic surface reaches a receptive condition before pollination is done. The writer has found that equally good results are secured by pollinating at the time of emasculation, provided this operation is delayed until the buds are nearly ready to open. Pollen will remain on the stigma in good condition until it can germinate in the nectar which is secreted on the stigmatic surface. This eliminates the extra work of removing the sacks for pollination and replacing them. A camel's hair brush is frequently recommended for applying the pollen but the writer has found the tip of the finger to answer the purpose more satisfactorily."
   After pollination the buds are covered with manilla paper sacks and carefully labelled. Time is saved by giving the labels simply a serial number for identification. Other data of the cross are kept in a notebook. In recording the cross it is customary to write the name of the female parent first, followed by that of the male parent, e. g., Seckel x Kieffer.

SETTING OF THE FRUIT
   “Not much is yet known about the ”sexual affinities of our cultivated varieties of pears and apples. Unfavorable seasonal conditions and some unknown factors, probably of a physiological nature, make this subject a confusing one. The writer has often noted trees loaded with fruit blossoms which set an extremely small amount of fruit. notwithstanding the fact that what are generally considered favorable weather conditions for pollination prevailed. Such failures do not appear to be due to lack of pollen, for even where buds are thoroughly pollinated artificially, the percentage of fruit set seems to correspond with that of the tree as a whole. It is true that certain crosses generally give better results than do others. Pears generally set a higher percentage of fruit than apples whether pollinated naturally or artificially. Of the total number of pear buds pollinated in our work, the. percentage which set fruit was 24.2, while that of the total number of apple buds pollinated was 9.9. The variation in the different crosses is quite noticeable, some consistently failing to set fruit, while others set well. One of the most remarkable sets of fruit was recorded in the spring of 1910 when 651 fruits of the Seckel pear set, out of a total number of 807 buds pollinated with Kieffer pollen. This amounts to about 80.6%. Over a series of years, however, the average drops to 37.8%.
   ”After the June drop the paper sacks are removed and a record made of the number of fruit set. The fruit is then covered with bags made of mosquito netting which allows a good development of the fruit and prevents loss, should it mature and fall off before being picked. In cases where specimens have fallen out of the sacks accidentally, the scar; which results from removing the calyx in emasculation often furnishes a convenient mark of identification. A few instances were noted where the development in the paper sacks was hindered by infestations of aphids. Protected from their natural enemies the aphids multiplied enormously until the sacks contained great masses of the insects.

GATHERING THE SEED
   “As the fruit ripens it is gathered and the seed removed when fully matured. The number of seeds produced by different varieties varies considerably. a fact which may often be turned to advantage by choosing prolific varieties as female parents.
   "The seeds are kept in a dry place until time to plant. After trying various methods the writer has found that one of the most satisfactory ways of handling the seed is to plant them in thumb pots or flats in light well-drained soil and place them in a cold frame early in the fall. There is always a certain percentage of loss of pots from freezing and bursting so that flats are more economical To prevent getting the seed from the different crosses mixed, the greatest care must be exercised. No one, who has not worked with things of this sort, can appreciate how easy it is to get the labels or seeds misplaced.
   “Early in March the seeds are removed to the greenhouse and are given gentle heat. Germination usually takes place rapidly under such conditions. It is essential to water with great care at this stage to avoid loss from damping-off fungi. Slugs seem to be extremely fond of the young seedlings and they should be guarded against. A trail of some such material as soot, lime or kainit surrounding the pots or boxes makes a good deterrent. Seedlings in flats are potted off after the first two or three true leaves appear and are then transplanted to the nursery row when danger of frost is over.

ELIMINATION OF UNDESIRABLES
   “Since fruit trees are grown primarily for their fruit, it would be highly desirable from an economical standpoint to eliminate undesirable seedlings at an early age. After considerable correspondence with plant breeders, the writer reached the conclusion that, in the present state of our knowledge of the subject, no satisfactory basis of elimination has yet been worked out. In view of the importance of determining whether there are correlated characters which can be used for that purpose, it seems desirable to grow all seedlings up to the fruiting age in order that tree and fruit characters may be compared. The prevailing opinion seems to be that the best results are obtained from seedlings showing a close resemblance to the smooth appearance of our improved forms. In regard to this point, Joe A. Burton, who is in charge of the apple breeding work of the Indiana State Horticultural Society, states that he inquired of a prominent plant breeder if anything could be done in selection.
Mr. Burton writes:
          The following is his reply: “Prominent buds, large, smooth, regular. glossy leaves, large leaf stems, short. distances between buds and a compact sturdy look, are the best indications of a good apple among seedlings.” I was greatly pleased with this information because it coincided with what I and my friends already believed. So on one occasion when the Horticultural Board of the State, all experienced horticulturists, visited the Experimental Orchard, I asked them to select each a tree that he thought most promising. One tree all agreed would surely be grand. As they came into bearing nearly every tree was a sweet variety, and the special tree a very small worthless variety. Not a selected tree was of any worth. I had refrained from cutting out trees that I was sure were worthless, because I had had no experience. It was well that I did. We did not know how to select. . . . Consider a few well-known varieties. Grimes and Rambo are thorny. Rambo is especially promising in bud appearance. Benoni, the king of summer varieties, is very unpromising both in tree and bud. Jonathan grows so straggling that nurserymen don't like to grow it.

   “At best it is a long time to wait for seedling trees to come into bearing. Various expedients have been tried to shorten this period. A number of seedlings were top-worked on dwarf trees but the results were not encouraging. The top-working entailed a good deal of labor, and where several kinds were placed on the same tree it was almost impossible to keep track of them. Besides this many of the seed lings fruited almost as early on their own roots.
   “Girdling was another method tried. The first plan used was to remove a ring of bark but some of the wounds did not heal readily ind later it was found more satisfactory to check the flow of sap by wrapping the branch with a strand of fine wire. When this was done on trees 5 or 6 years old about the last of June very good results were secured. One drawback to girdling is the production of numerous water sprouts just below the point girdled. These water sprouts are very susceptible to the fireblight. Summer pruning helps somewhat in controlling them, but it must be done with care or it will only serve to spread the blight around. Frequently the first blossoms appear on the terminal shoots so it seems desirable not to prune these back severely. Even with these expedients, however, one can hardly expect any very great results in breeding work with pears and apples in less than ten or twelve years."

[FOOTNOTES & REFERENCES:]
1 A cross between the pear and quince was described by Dr. L. Trabut in the last issue of this Journal. The Office of Seed and Plant Introduction, U. S. Department of Agriculture, Washington, D. C., has brought in many wild species of Pyrus for breeders, and any experimenter could probably secure specimens from it; but as is pointed out in the present article, more is to be expected by crossing varieties that have already been improved. The wild species are of value mainly as stocks for grafting, a Chinese species known as Pyrus davidiana [This is now thought to be from the genus Stranv&ligae;sia. -ASC] having proved particularly valuable by its hardiness and vigor.
2 Ballard, W.R. Methods and Problems in Pear and Apple Breeding. Maryland Agricultural Experiment Station, Bulletin No. 196, College Park, MD., April, 1916.