Transactions of the Illinois State Horticultural Society, 32: 62-72 (1898)



There has been a great deal of discussion upon this question, and very different opinions have been held as to the nature of the influence upon the parts united by graftage. That there are effects of one upon the other no one disputes. That in the case of certain kinds, there is a well recognized and very characteristic result is also well known. A Bartlett pear upon quince roots is a distinct thing in one sense from the same pear on its own roots; and a still better example is a variety of apple like the Yellow Belleflower on ordinary apple roots, and the same on Paradise stock. It is now thoroughly well understood that May or Early Richmond cherries are much more productive, at least in our rich, western soils, on Morello stock, than when worked on Mahaleb seedlings. There seems to be conclusive evidence that somewhat tender varieties of apples resist much more effectually the rigors of winter when top-worked on certain hardy stocks. It has long been understood that wine differed in quantity and quality when, from a given variety, it has been in part produced from grafted vines, and from others not grafted, other things being alike.

*The Nursery Book, p. 74.

Prof. Bailey summarized the mutual influence of stock and cion in this way*

  1. Grafting may modify the stature of a plant (Dwarfing especially.)
  2. May be the means of adapting plants to adverse soils.
  3. May be the means of adapting plants to adverse climate.
  4. May correct a poor habit (Canada Red apple usually top-worked.)
  5. May make plants bear fruit at an earlier age.
  6. Often modifies the season of ripening.
  7. Often augments fruitfulness.
  8. Often delays degeneration of varieties.
  9. Sometimes increases the size of fruit.
  10. May result in a modification of color, of foliage, flowers, or fruit.
  11. May influence the flavor of fruit.
*Transactions Illinois State Horticultural Society, 1887, p. 76.

It was the privilege of the writer to read a paper before the society in 1887,* in which it was admitted that many such effects were well known, brought about by the peculiar effect of a certain stock upon a given variety inserted by grafting or budding. It was there claimed, however, that the reciprocal influences of stock and cion could all be referred to such changes as were connected with nutrition, and that no inherent change was produced in the qualities or character of either stock or cion by the union. Somewhat in opposition to this last, two apparent exceptions were cited: "First, the uniformity of roots of grafted apples, etc., in nursery rows; and second, the fact that cions taken from plants with variegated foliage and grafted upon green-leaved stocks of the same species, so modify the stock as to cause it to put forth leaves showing the characteristic variation." In regard to these it was suggested that the first may be explained by the rooting of the cion itself, and the latter is rather a case of inoculation of a disease.

The general conclusions then expressed are those commonly accepted by those who have by information and experience the best right to form opinions upon the subject. While very marked and practically important modifications are produced by graftage, these modifications are not deep-seated and do not change the inner nature of the stock or cion. A pear may be dwarfed and otherwise modified in growth and fruitage by a quince stock, but a cion from such dwarfed tree inserted on pear stock afterward shows nothing of its previous quince association. The fruit of the dwarf on quince may indeed differ in size and flavor from that of the same kind on its own roots, but it has nothing of the quince flavor. It is a difference produced by unequal development or of maturing, such a difference, for instance, as may be brought about by the manner of ripening the fruit off the tree, rather than any actual union of the quince and pear juices or substances. In other words the influence of the stock on the cion is such an influence as may come from differences in soil and climate. The wine from the same kind of grape is said to be abundantly different when the vine is grown in different geographical districts. We are all well aware of the fact that some apples are characteristically better when grown in some localities than in others; but we do not assume that because a Ben Davis is usually of better quality when grown in Hancock county than when produced in Vermilion county of our state, that cions from these Hancock trees would bear better fruit in Vermilion. It is a question of sun and soil and air-not an inherent change in the constitution and quality of the tree.

All this seemed reasonably conclusive and substantially correct before much was known of the real process of vegetable growth, or at least this was the conclusion of observant persons who knew little or nothing of the method of growth of plants. It is, however, strongly supported by the information revealed by the microscope, and now commonly accepted as the universal rule of cell multiplication. We all understand that the entire body of every plant is made up of distinct cells or anatomical units, that (so far as we are now concerned each of these has a closed cell wall or outer coat, and an inclosed substance—in the living state—called protoplasm. Imbedded in the latter an essential constituent exists, known as the nucleus. Now growth takes place through a multiplication of these anatomical units, and this again, solely by a process of self-division of pre-existing cells or units. The nucleus first divides by a regular method, which has been very closely studied, and which has been found to be wonderfully uniform in procedure in all growth of animal as well as plant structures. After the division of the nucleus into two equal parts, each half having all the structural elements of the original whole and of the other half, a cell-wall is formed between the separate parts, or rather between the two new nuclei. Thus two cells are formed out of one. Each of the two new cells is part and parcel of the old one, and is endowed with all the peculiar qualities, all the distinctive characteristics of the original cell. Whatever it was, these two now are. They may, indeed, owing to some special circumstance, differ in size or outward form, but they are marvelously the same in everything that gives them specific character and separates them in kind from all others. We say they are chips off the old block.

All this remains true when a cion is caused to grow on another living plant. The cut surfaces are simply placed in contact. Cell-division takes place in both structures. The newly formed contiguous cell-walls adhere, but none of them lose their identity. They do not change in inherent character on account of the contact. They do not fuse or blend. Those of the cion simply grow as they might have done in a florist's cutting bench. They absorb from their neighbor of the stock the water pumped up by the latter from the soil, but they gain nothing of the inner or essential characteristics of the stock through the contact, or through the transference of water. These latter cannot and do not dissolve and escape from the home limitations. The life principle is associated only with solid or semi-solid condition of matter, never with liquids, and it is by the material growth, and by this only, that is by continuance of material substance touched with the mysterious, dominating power of vitality that characteristics are transmitted. It is in this way, and in this way only, that life begets life, and that inheritance is possible. In sexual union there is an actual coalescence of two cell-nuclei, and thus a blending of two sets of characteristics; but there is no trustworthy evidence that this ever takes place in such union as succeeds graftage. If it does really occur in rare instances, we have a graft hybrid as the result, not the ordinary combination of stock and cion.

Thus practical experience and scientific investigation combine to prove that there is no essential modification produced on the cion by the stock, or by the stock on the cion whereby anything more than a superficial impress is made, like that which comes from soil or from climatic peculiarities. The effects observed are mostly such as are connected with nutrition. If a Hyslop crab or a Winesap cion does apparently impress itself on seedlings stock so that the roots show a marked peculiarity, even this may be explained upon the latter principle; for it is well known that, in the case of most plants, roots run out in a long, straggling manner in soils deficient nutrient materials, while in specially favorable soils they take a short, bushy-branched form. That is, they temporarily adapt themselves to the special conditions.

Such modification may indeed be of great practical consequence, and the proper selection of stock and cion may be worthy of the best attention that can be given to it; but up to our day the opinion has become firmly established that graftage does not introduce any such change as would produce or tend to produce a new variety capable of self or of artificial perpetuation.

All this might have been written years ago. Indeed, in one form or another, it has often found similar expression. Careful inspection of what has been said upon the other side seems to show that writers have not understood the question at issue, or have accepted the facts with a meaning different from that placed upon them by others. If there were nothing more to be said, you would require from me an apology for intruding upon your time and attention. You have no need to discuss thoroughly settled matters, however important they may be. There are live questions in abundance, which may rightfully claim right of way. Is there, therefore, any excuse for my topic? Let us see.

*Comptes Rendus de L'Academie des Sciences, Tome 118, p. 992.

The French Academy of Science is one of the most notable organizations of the kind in the world. Its transactions are justly esteemed as a record of noteworthy accomplishments, and anything appearing in the semi-annual volumes properly commands the attention of all interested in the special subjects discussed. As indicated by its name, the academy is organized in the interests of general science, and its membership includes only men of repute who have made themselves worthy by their work of the honor of election. In the volume of proceedings for 1894* there appears a paper by M. Daniel upon the Creation of New Varieties by Means of the Graft. This title is sufficiently startling, but when we turn to his conclusions at the end of the article we find that we have not missed the intended meaning of the headlines. The summary is as follows: (1) Hybridization by the graft is possible for certain herbaceous plants which we can make acquire new alimentary qualities by placing them upon plants which are their superiors in this respect, and by sowing the seeds produced by the graft. (2) The impression produced upon the graft and its seeds is more or less profound, according to the plants grafted. It appears so far more particularly marked in plants of the family of crucifers (cabbage, turnips, etc.)

*Ibid. Tome 127, p. 183.

In the transactions of the same body for July, 1898,* there is another paper by the same experimenter, entitled, Amelioration of the Wild Carrot by Grafting it Upon the Cultivated Carrot. Here the conclusion reached is stated as follows: "This experiment shows (1) the existence of the influence of the stock upon the posterity of the graft. This influence causes not only variation in general, but a mixture more or less complete, according to the plants, of the characters of the stock and of the graft. of the kind that in the descendants of the latter there is produced a defection recalling that of cross-breeding and of hybridization, and (2) the possibility of improving wild plants by graftage, followed by seeding and an intelligent selection of the products thus obtained."

There can be no doubt that a modification is here meant wholly different from that of the superficial kind acknowledged above. The mixture of the stock and cion is such as enters into the very life and essence of the latter, for it is perpetuated in the seed. There must therefore be something incorrect in our former generalization, if we accept the accuracy of these later experiments and interpretations. We have shown that no such mixture has or can occur, then we come face to face with the actual fact of occurrence. In order that all may judge for themselves what M. Daniel claims to have done, a translation from the French is appended of the principle parts of the two articles of which the summaries have been given above.


Can we by graftage, followed by seeding, create new varieties? The question is not new under different wording-graft hybrids; influence of stock on cion, etc.; but has not yet received a very precise or satisfactory answer. This is because we have dealt with trees in which the production of seed and their growth is slow. This is not, however, the case in operating upon annual or biennial plants.

Here are the results of the first experiments I have made in this line. They enable me to say that we can obtain definite results by working in determined conditions. In grafting turnips upon the wild hedge garlic [a mustard-like plant] I have obtained seed which furnished me with plants showing a marked return to the wild type. I have concluded that to improve a plant by a graftage, followed by seeding, it should be placed upon a stock which is superior in the qualities which we desire. In my numerous experiments upon cruciferous and edible leguminous plants, I used seed produced by ingraftments in these conditions.

Knight's Wrinkled pea upon the field bean (Faba vulgaris) has not given apparent modifications, as the two kinds are about the same in vigor. In the beans the stock suffer a considerable reduction in size. The seed which they furnish give birth to plants inferior in size to those of the same variety of beans not grafted. Here, then, is a manifest tendency of the seeds to reproduce the effects of the cion.

But it is especially with cruciferous plants that I have observed in an absolutely certain manner the transmission of variation due to the graft. I have sown in neighboring squares the seeds of the hedge garlic not grafted, as tests, and seed of this plant from grafts on Kale (Chouvert.) The first have shown the ordinary wild characters. They had stools of six to ten stems, 6.5 centimeters high, a principle root 2 c.m. thick, and 1.5 to 2 c.m. long, little branched; the leaves were yellowish green, rather distant, and had a strong odor of garlic. In all the stools from grafted plants there were from fifteen to twenty-five stems about 4 c.m. high, greener and more tender than the preceding; a principal root .3 c.m. thick, at least 30 c.m. long, vigorous and abundantly branched; the leaves near together, giving to the plants a very stocky aspect, they were greener and somewhat wrinkled as those of the kale, and with less characteristic odor than had the wild plants, evidently partaking of the qualities of the kale and of the hedge garlic. Upon the whole the seed of the grafted hedge garlic have reproduced the characters of the stock, in the diminished size, in the stocky aspect, in less odor, and in the larger development of the assimilating, to correspond with the absorbing apparatus.

These are the exterior differences. The internal ones are equally pronounced. The roots of the grafted plants are less woody, the medullary parenchyma [pith] less thickened, the arcs of schlerenchyma of the bark are not as in the wild plant, the vascular cylinders are reduced, while the layers of inner and outer bark are on the contrary much augmented. These differences are apparent to the naked eye. The stems of these plants are more tender, not so rich in ligneous tissue, but with more chlorophyll, and the pith has not cavities as in the non-grafted hedge garlic. As to the leaves, aside from the greater abundance of chlorophyll, they present no modification in number, disposition, or in the thickness of the layers of parenchyma.

From seeds of turnips (Navet a collet rose) grafted on cabbage (Chou de Montaque) I have obtained very characteristic results. All the seeds produced turnips somewhat less in size than the tests from non-grafted plants; but with a special taste, recalling at once that of the cabbage and of the turnip. I kept for seed production a good number of specimens. At the time of flowering some remained short, forming heads, flowering poorly and thus exaggerating the characters impressed upon them by the graftage; others gave at the same time normal and shortened branches; and finally others had only normal branches.

Then follows the conclusions given above, "That hybridization by the graft is possible," etc.


I have shown in a previous communication, and in other articles, that we can improve the cabbage by uniting through graftage two properly chosen varieties and afterward sowing the seed produced by the stock; that we can make a wild plant vary, as the hedge garlic, by grafting upon a cultivated plant of a different genus, as the garden kale.

Following these researches, I desired to see what would happen in grafting a wild plant upon stock of a thoroughly established, through long culture, race of plants. I therefore grafted the wild carrot (Daucus Carota) upon the medium red carrot (Carotte rouge demi-longue), a well-known garden variety.

We know that the two plants differ in size, their hairiness (villosite), their color, and in the diameter of their roots. The leaves of the wild carrot are in general more or less horizontally spread out, of a glaucus green, and are covered with hairs; the roots are white, slender, about 1 centimeter in diameter. The medium red carrot has erect leaves, less hairy, and of a more intense green color, the roots are red and large, attaining a thickness of 6 c.m. and more.

When grafted, the stock-the red carrot-with its wild carrot top, developed normally in thickness of root. Seed matured in abundance, each one and a half to two times as large as the tests, and with spines much more developed. I sowed the seed in the month of March of the wild carrot, as tests, side by side with those from the grafts, other conditions being equal.

Here in tabular form are the results:

Wild carrot grafted. Wild carrot tests:
  • With 30 germinations I obtained
    2 seedlings with three entire cotyledons.
    2 seedlings with three cotyledons, one of which was bifid.
    1 seedling with two cotyledons, one of which was bifid.
    1 seedling with only one cotyledon.
  • The plants from the above were larger, greener, less hairy, and were in these respects more or less clearly intermediate between the red and the wild carrot.
  • Eight of these plants shot forth seed stalks directly.
  • Of the rest, about a dozen had one swollen root, attaining a diameter between 1.5 and 2.5 c.m. [one inch]; the color continued white.
  • With 30 germinations and more I found only seedlings with two entire cotyledons.
  • The test plants had normal vegetation. The leaves were medium size and resembled perfectly the ordinary wild carrot.
  • None shot up seed stalks.
  • Roots normal, attaining in the best specimens a maximum thickness of .8 c.m. [one-third inch.]

This comparison shows four interesting differences: (1) The high proportion of anomalies in the number and the form of cotyledons, a common thing in certain umbelliferous plants, but not so marked; (2), the frequency of the shooting forth of a seed stalk, an accident common enough with the cultivated carrots, but rare in wild plants; (3), the characteristic change in the aspect, the color, and the hairiness of the leaves, acquired characteristics of the stock; and (4), the appearance of a notable tubercuization, recalling that of the stock, but without change of color of the root.

The conclusions, beginning: "This experiment shows," etc., have been given above. And thus ends the quotation, though the rendering is not close enough for quotation marks.

The whole work as reported tends to show that profound modifications may occur in grafted plants, and, that there may be mutual influences of the stock and cion which reach the essential characteristics of the plants. As subversive to all previous in information as this is, we must accept the facts, and will do well to repeat the experiments as well as add others of like kind. It is possible that further knowledge will harmonize what now seems to be inextricable contradiction.

Periam—May it be possible that Mr. Berbank [Burbank] has been using the graft system in his wonderful productions, in connection with fertilization?

Prof. Burrill—Yes, if this is true.

Periam—You find it very difficult even to fertilize between two species?

Prof. Burrill—That is often the case.

Periam—Mr. Chairman, I think this is as important a scientific question as has been given us for years; not merely the reacting of the paper, but the conclusions that may by continued investigations be arrived at.

Prof. Burrill—Perhaps I ought to say, Mr. Chairman, that some of the most reputable horticultural papers in the world have discussed these experiments of Daniel, apparently without recognizing the fact that he has made a contribution absolutely at variance with things that were known before. The Gardener's Chronicle, for instance, of London, has reference to these experiments, and quotes them without hesitating, apparently, at all to give them full credence, and I suppose there has been proved now by direct experiment what people had been suspecting all along. 1 can't help but think there is something marvelously out of joint. It may be these things are all right, and that we have been all wrong. Surely we ought to recognize there is a mysterious conflict of opinion.

Periam—We know or believe nature does that very thing in exceptional cases, as shown in mixed genera.

Reihl—Mr. Chairman: I have listened very closely to the reading of that paper, and I must say, while I can't state that those things are not so, yet they are so contrary to everything that I have read or heard, or to my reasoning in that line, that I cannot accept it as true. And I hardly think that those experiments were carried on in the right way. It has seemed to me in listening to the results that are reported, that there is running all through that a possibility that there might have been sexual cross-fertilization there. Now, if this gentleman will go to work and carry his experiment on in the same line, but in another way than he has done and will bring us such results, then I will believe that there is something in it. For instance, if he will graft a tomato on a potato, which has been done, so that the seeds will produce plants that will make tubers on the roots, then I will believe there is something in it, and until he does, I will not.

Periam—And yet, Mr. Reihl, you will remember that the tomato is a very close relation of the potato.

Reihl—Certainly, or else the tomato wouldn't grow on the potato vine. But the point I am driving at is that if he can produce potatoes on a plant grown from the seed of that tomato grafted on the potato vine, then I will admit that there has been something accomplished outside from cross-fertilization, because that never yet has been done. We have never produced a plant that would naturally produce potatoes and tomatoes on the same plant. But if he can do it in this way, then I will think he has done something in the line which that paper indicates; otherwise, I don't swallow that.

W. S. Emrich—I would ask if anyone has ever succeeded in having tomatoes mature when grafted on potatoes.

Reihl—Yes, that has been done. My father did it fifty years ago.

Emrich—I have seen a plum grafted on to an apricot, ripen. It was two inches in diameter, probably double the size of the fruit on the original tree.