Survival of the Unlike (1896)
Liberty Hyde Bailey

V.
VAN MONS AND KNIGHT, AND THE PRODUCTION OF VARIETIES.1

1Address before the Pennsylvania Horticultural Society, Philadelphia, March 17, 1896. Printed under the title, “New Varieties of Plants,” in American Gardening for May 16, 23, 30, June 6 and 20, 1896.

EVERY species of plant which man has cultivated for any considerable length of time has numerous forms or varieties. This simple and undisputable statement represents one of the most significant facts in nature. There are two diverse types of inquiries which may grow out of a contemplation of this fact. One type attempts to collect information concerning the various forms, for the direct and immediate use of the cultivator. It endeavors to ascertain the best varieties for certain purposes or for given conditions. This is a matter of practical agriculture. The other type of inquiry asks why and how these various forms carne to be. It seeks for fundamental truths, that is, for laws or principles; and inasmuch as principles are eternal, so far as we know, it follows that the enunciation of a law is prophecy of progress and destiny.

The type of mind which inquires into the reasons for the existence of these forms is essentially a modern one. It may almost be said to be recent. The inquiry into the nature of garden or domestic varieties of plants and animals is simply a special application of the desire to know the genesis and destiny of that congeries of objects which we call Nature. This desire began to express itself soon after the Restoration of Knowledge, and, upon the side of plants, the earliest distinct record of it is in those naïve, ponderous and cyclopedic works which we call herbals. These writings culminated in the concise and scientific attempts to delineate and classify all natural objects, of which the works of Linnaeus are the historic examples. Linnaeus and his editors worked upon the plan of an inventory of nature, or a species of bookkeeping, and this idea, so far as the vegetable kingdom is concerned, did not die out until the discontinuation of De Candolle's Prodromus after the middle of the present century. This task of enumerating all species of living things was less onerous to the old systematic naturalists than to ourselves, for it was not then supposed that the organic creation is anything like so extensive as we now know it to be. Biberg, writing in 1749 in Linnaeus' "Amoenitates Academic," estimates "the whole sum of the species of living creatures" to be about forty thousand. Of these, twenty thousand were supposed to be vegetables, three thousand worms, twelve thousand insects, two hundred amphibious animals, twenty-six hundred fishes, two thousand birds, and two hundred quadrupeds. We now know that the species of single classes run up into the hundreds of thousands. Of flowering plants, about one hundred and twenty-five thousand accepted species are described, and it is estimated that only about half the existing species are known. The flowerless plants probably far outnumber the flowering plants. Of insects, something like two hundred and fifty thousand are described, and it is probable that less than a tenth of the existing species are known. Riley concludes that "to say that there are ten million species of insects in the world would be, in my judgment, a moderate estimate."

An enumeration of all the known animals and plants of the world almost necessarily aroused a desire to know whence and why they came; hence we find a school of speculative naturalists arising in the latter part of the last century, whose heretical and unholy unrest culminated in the philosophies of Lamarck and Darwin.

This analytical mind could have found its most abundant and suggestive material in the study of cultivated plants, but the old ideas of the entity and immutability of species had taken such firm hold upon men that they could not completely shake off the influences of tradition and habit; therefore, garden plants, being so endlessly variable, could not represent or express natural laws. It is probable, also, that the very variety in cultivated plants was so perplexing as to repel the student. They present a boundless extent of detail which, with no fundamental conception of the method of the unfolding of the vegetable kingdom, meant nothing to the philosopher.

It is significant that all the early attempts to explain the origin of domestic varieties were made for the benefit of the horticulturist, and not primarily to elucidate the genealogy of plants. This remark is worth the making, because it really explains why the early theories failed,—because they conceived, almost uniformly, that a philosophy of the genesis of garden varieties has no necessary connection with the general uplift and dispersion of the vegetable kingdom. Horticulturists were looking for some secret key, some alchemy, by means of which the cultivator could unlock or dissolve the barriers and bring forth plants to his liking. It was a refinement and expansion of the old magic which smote the rock or swung the enchanted wand to disclose to some oracle the secret of the mysteries of nature. It needed the modern analytical mind to grasp the meaning of the forces of nature, to see that there was nothing supernatural, and to pick the kernel of truth from the husk of sophistry. It was in the latter half of this present century that such a mind grasped the entire sweep of organic nature and attempted to discover its meaning in order that the most common man, as well as the oracle, might apprehend the truth and apply it to his own life. This, as I think of it, is the transcendant merit of Darwin. His theories and conclusions may perish, but his life marks an epoch in the habit of thought. All the old notions and traditions, the panorama of nature, the rise of civilization, the destiny of beings and events,—all these are but links or factors in a grand spectacle whose beginning and end are one and whose concerns are every man's.

Van Mons.

With this introduction, you can understand the setting in which the theories of Van Mons and Knight appeared, and we may be able to construct a perspective in which to contemplate them. Jean Baptiste Van Mons was born in Brussels in 1765, and he died in Louvain in 1842. At the age of twenty he became a pharmacist, but at that time the brilliant experiments of Lavoisier and his contemporaries turned the attention of the young student to pure chemistry. With the French occupation of Belgium, he became professor of physics and chemistry in the department of the Dyle, and he gave up his pharmaceutical interests that he might devote himself wholly to science. He received the degree of Doctor of Medicine from the faculty of Paris in 1800, and he was one of the savants of the Institute of France from its formation. He came into correspondence with Lavoisier, Berthollet, Chaptal, Fourcroy, Volta, and other celebrities of the time. In 1817 Van Mons became professor of physics and chemistry in the University of Louvain, where he remained until the last. He was an expert linguist, a profuse correspondent and a facile writer. He published important works upon chemistry and electricity, many of which exerted wide influence.

This is the man who first propounded a complete system or theory of the philosophy of the origination of varieties of cultivated plants. His system was applied particularly to fruit trees, to which he devoted most of his attention, but it was conceived that the principles which he enunciated are of general application in the vegetable kingdom. This system was expounded in various papers, chiefly in his admirable ''Arbres Fruitiers," which was published in 1835 and 1836, in two volumes, but he had conceived the fundamental idea of the propositions as early as his twentieth year, and the system had early become current amongst naturalists. The various characteristic features of the system were brought together and codified by the illustrious Poiteau shortly following Van Mons' death. They are as follows:

  1. A natural species of tree does not vary through its seedlings, in the place in which it is born; so long as it remains in its natal place, it reproduces only plants which resemble itself, or, at most, only sub-varieties.
  2. The causes of variation are a change in soil, climate or temperature.
  3. Whenever a natural species of tree produces one or many varieties, these varieties continue to vary always, if they multiply by means of seeds, without ever being able to return to the primitive form.
  4. The source of all variation, which is transmissible by sowing, resides in the seeds.
  5. The older a variety of fruit or other tree, the less the seedlings vary and the more they tend to return towards the primitive form, without being able ever to reach that state; the younger or newer the variety, the more the seedlings vary, or, as we might say, the better the variations are for the use of man.

Another epitomist expresses Van Mons' theory as follows:

"In sowing the first seeds of a new variety of fruit tree, one obtains trees necessarily variable,—for they cannot escape this condition,—but they are less disposed to return to a wild state than those coming from seeds of an old variety; and as that which tends towards the wild state has less chance of being perfect, as measured by our tastes, than that which remains in the open field of variation [or tends to vary still further], it is, therefore, in the sowing of the first seeds of the most recent varieties of fruit trees that we must hope to find the most perfect varieties for our tastes."

The student will observe that there is little in these statements to challenge controversy, save only the last or fifth law,—that seeds from old varieties tend to give small differences in the seedlings, and that these differences are usually in the direction of inferiority, being reversions toward the primitive type of the species; and that, on the contrary, the seeds of new varieties give wide variations, which are usually in the direction of improvement. It seems that this particular doctrine—to which we shall now restrict the name "Van Mons' theory,"—was not originally deduced from observation, but was a precognition. "The system of Van Mons upon the means of producing the best fruits for the table is not founded upon experience or practice," writes Loiseleur-Deslongchamps; "it is a preconceived idea of the earlier years of the author, which he has endeavored by every means to verify and develop, and which he made the fond child of his imagination." Yet there was some apparent basis for the generalization. Many of the old varieties of fruits seemed to be failing, whilst the new varieties were strong, healthy and productive. There seems at times to be a tendency for old varieties to deteriorate towards some assumed primitive or inferior type. In fact, we hear everywhere at the present day that varieties wear out with age, and we are cited to the disappearance of once cherished forms for proof of the statement. But we are learning to analyze these instances of wearing out, and we seem to find, in every instance, that there is some specific reason for the disappearance of given varieties, and that their loss is not the result. of the operation of a general law. We now know that the Bordeaux mixture is a specific for the so-called running out of the Virgalieu pear, Catawba grape, and other fungus-infested fruits; that change in fashions and demands has stranded varieties of intrinsic merits; that certain varieties have failed because they have been taken into regions to which they were not adapted; that others have passed out because they are difficult to propagate or are wayward growers, so that the nurserymen dislike to handle them, and so on. If we could trace out every case of the disappearance of varieties, we should no doubt find a special and separate reason for each one. On the other hand, we should find various varieties, like the Green Gage plum, the Ribston Pippin apple, Bergamot pear, and others, which have persisted for centuries in undiminished excellence; and everyone must recognize the fact that in the past a variety of apple has rarely gained much prominence until the first or second generation of trees has passed away. It is significant that many of our best fruits are also the oldest, as, for instance, the Baldwin, Greening, Roxbury Russet and King apples, and the Bartlett pear.

There are various other contributory reasons for the founding of this hypothesis of the different behavior of seeds from new and old varieties. It was conceived that new varieties, and particularly young trees of new varieties, are not yet fixed in their characters and are in a state of variation or amelioration. One of the best proofs of this, to Van Mons' mind, seemed to be the experiences of the colonists in America. At first, no famous or notable varieties of fruits appeared. This was conceived to be because the seeds had been taken from old varieties in the mother country; but after a time, through the successive generations of trees coming from these first sowings, there began to appear many excellent varieties of fruits. I am afraid that if this argument could be tested by historical facts, it would be found to rest upon a very slender foundation. The fact that pomologists know of the existence of few meritorious varieties in the early days does not prove that such varieties did not exist; for the fruit plantations were few and scattered, there was little incentive to name and disseminate such good kinds as might have originated,- and there were no books or periodicals, or other public prints, into which accounts of them would be likely to find their way.

Van Mons set out with most commendable vigor to prove his theory, and he continued the work for about half a century. He conceived that the best results were to be obtained by taking the first seeds from wild or spontaneous plants, for the transfer to the new conditions would itself tend to awaken a variation, and the starting point would be a new type or variety. From the first fruits to ripen on any of the seedlings, he saved the seeds and sowed again; and this practice was continued generation after generation with unabated zeal. His experiments were begun in 1785. Thirty-eight years thereafter—in 1823—he had eighty thousand seedling trees in his "Nursery of Fidelity," at Louvain, and he had then begun to distribute cions in many countries. These were sent to America, also, chiefly to the elder Robert Manning, of Massachusetts. These cions were sent out freely under numbers, and were never sold. He gave his attention almost wholly to pears. In 1823, he issued his first catalogue, which contains ten hundred and fifty pears, three-fourths of them bearing names. Of this number, Van Mons himself appears to have originated four hundred and five varieties, two hundred of which were named. Amongst Van Mons' pears are the Diel, Louvain, Frederic of Wurtemburg, Bose, Colmar, Manning's Elizabeth, and many others which are little known in this country.

The theories and work of Van Mons probably exerted the widest and most profound interest and influence of any horticulturist up to his time. He was introduced to the American public by Robert Manning, who received and distributed his new varieties, and who described these novelties in Hovey's "Magazine of Horticulture," and in his own excellent "Book of Fruits," which was published at Salem in 1838. Van Mons' system was first clearly enunciated in this country by the brilliant Andrew Jackson Downing, in the first edition of his "Fruits and Fruit Trees," in 1845; and this outline of the theory has remained unchanged through the many editions and revisions of this work. American horticulturists now know Van Mons only from this historic record in Downing. In England, Van Mons' influence seems to have been comparatively small, owing largely, no doubt, to the overshadowing effect of the contemporaneous work of Thomas Andrew Knight, to whom we shall presently recur. Upon the continent, however, his authority was unbounded. Loiseleur-Deslongchamps, himself a great horticultural authority and yet one who did not subscribe to Van Mons' theories, writes of him: "We have no fear in saying that Van Mons himself accomplished more than had been accomplished since horticulture began; for there had been no labor, so far as I know, that resembles or even approaches it. Pomology owes him the greatest obligations. In fact, it is from his time that we have seen good fruits of all sorts, and principally of pears, multiplied in a most extraordinary manner; and that whatever reproaches one may make against his system (and I do not fear myself to raise objections to it), it is justice to him, which I am glad to grant, to say that there has never been a man who made known such a large number of new and good fruits as Van Mons did." The praise which was everywhere bestowed upon him, and the prodigious labors which he accomplished, seem to place Van Mons in the very front rank of those bolder men who, by the aid of science and philosophy, have contributed to the permanent advancement of the cultivation of plants. Yet one will find that this fame rests more upon a regard for the man and the varieties which he produced, than upon an acceptance of his system. Van Mons was unfortunate in having a theory to prove by means of experiment, rather than in attempting to construct a theory as the result of experiment. He assumed, as most persons do at the present day, that there is some mysterious or hidden means which, if discovered, will enable the operator to produce forthwith and with certainty such plants as he desires. This appears to have been the almost universal type of mind in pre-Darwinian times. Even Loiseleur-Deslongchamps, whilst refusing to accept Van Mons' system, yet writes in 1842 that "we are still ignorant of a positive means by which we can proceed with certainty to produce new fruits of the best quality; it is a mystery of which nature guards the secret." We are now convinced that this attitude of mind is erroneous, and that it is rarely productive of useful results in investigation.

One might think, from the bare statement of his principles, that Van Mons had really anticipated some of the characteristic generalizations of Darwin. One of the most important and inextricable problems now before philosophical naturalists is the source or cause of variations or differences between individuals of any species. There are some thinkers who refer all useful or permanent tendencies towards variation to innate or predisposed inclinations; and there are others, like the Lamarckians and Darwinians, who believe that much, or perhaps all, variation is the result of the conditions or circumstances in which the organism is placed. Van Mons plainly propounds that the causes of variation are change of soil, of climate, or of temperature; but he assumed, in common with most thinkers of his time, that species are essentially distinct and immutable. Therefore, he could not look upon variation as anything more than an incidental feature in nature, and whatever the causes of this variation may be, they are significant only as they explain how the cultivator may manipulate his plants, not as throwing any light upon the evolution of the vegetable forms which cover the earth. It is reasonable to suppose that the origination of new kinds of plants in the garden is but a local or specialized expression of the means of origination of all forms of plants, whether in the garden, field, swamps or woods. I am constantly reminded that horticulturists do not apprehend the fundamental principles of the origination of new varieties simply because they refuse to look at the problem broadly, in the light of evolution, and persist in asking for some short-cut or so-called practical method which they can apply in the garden without testing its probable fitness by comparing it with the means which are operative in the uplifting of the vegetable world. Horticulture has always suffered by being cut off from the studies of scientific men, so that it has grown too much into a mere art, which is not conceived to rest upon the very same fundamental laws, so far as plant-breeding is concerned, as have been and are the slow but mighty forces which have been operating throughout the ages.

You are now wanting to ask how it was that Van Mons obtained such useful results if his system were untrue. It is indisputable that he obtained many very excellent new varieties of pears, and that in some of his series the generations came into bearing earlier and earlier, until, in the fifth generation of certain pears, he was able to secure fruit at three years from the seed. This result was thought to be indubitable proof of his proposition that the first fruits from the newest varieties, —that is, from seedlings,—give the quickest and best results. In the first place, it should be said that the failures were much more numerous than the successes. We are told that he had as many as eighty thousand seedlings growing at one time, but the number of good new varieties which he obtained, whilst aggregating perhaps three or four hundred, was much less than one per cent of the total number of efforts. In the second place, Van Mons' methods of cultivation were such as to hasten precocious fruiting. He conceived the idea—which, unfortunately, is prevalent at the present day—that progress in amelioration of fruits is correlated with an enfeebled or refined condition of the tree. His seedlings were planted close together, and they were kept closely headed-in, in order to lessen their exuberant natural vigor. The seeds were also selected from unripe fruits, a process which is now known to result in more or less enfeeblement of the offspring, and consequently in precocity. In the third place, it must be observed that this increasing precocity and amelioration in the succeeding generations are also due to simple selection, and not to any inherent tendency towards perfection in the first fruits of seedlings. Probably no experimenter in plants has ever given the world more excellent proofs of the value of judicious and repeated selection than Van Mons has; and this single lesson is worth all the arduous labor of his long and useful life. This lesson has now been accepted as one of the canons of horticultural teaching, and It has been strengthened by the experience of every experimenter and every careful cultivator, that the one and the only infallible means of producing better plants is through good care, and judicious and persistent selection.

Joseph Cooper.

Although Van Mons is the leading early apostle of selection for the amelioration of plants, there were other experimenters who had early demonstrated its value. One of these early explorers in the field of plant-breeding was Joseph Cooper, of Gloucester county, New Jersey, who, at the close of last century, had made most suggestive experiments in the improvement of plants, and who apprehended the value of selection more clearly than any other person of his time with whose work I am acquainted. Unfortunately, Joseph Cooper appears to be almost unknown and therefore I have the greater pleasure in introducing him to his posterity; although he had been discovered by the patient search of Darwin, who cites his work to show that selection may accomplish much when it "has been silently carried on in places where it would not have been expected." Darwin, however, did not know the particular paper and experiments of Cooper's to which I am about to refer. This paper is a short letter which was written in 1799, and published in the first volume of the "Memoirs of the Philadelphia Society for Promoting Agriculture." The title given it in the Memoirs is: "Change of seed not necessary to prevent degeneracy; naturalization of plants important caution to secure permanent good quality of plants." The editors say that the paper "has already been published in the United States and in Europe; and has deservedly excited very general attention." It is further explained that "the writer is entitled to every degree of respect, both for his practical knowledge, and integrity of relation. His experience and opinions differ widely from those generally received. The results produced, require the care and attention which few will give. The merit of Mr. Cooper is therefore the greater." Cooper was also a pomologist of note, and was the originator, amongst other things, of the Cooper plum, a seedling of the Orleans, which William Coxe said, in 1817, "is the largest plum I have seen."

Cooper said that he was "greatly embarrassed at the opinion very generally entertained by farmers and gardeners, that changing seeds, roots and plants, to distant places, or different soils or climates, is beneficial to agriculture; such opinion not agreeing with my observations or practice." He deplored the general acceptance of this notion, because "it turns the attention of the husbandman from what appears to me one great object, viz. that of selecting seeds and roots for planting or sowing, from such vegetables as come to the greatest perfection, in the soil which he cultivates." Cooper's experiments were a credit to his time, and they have probably not yet been excelled in this country for simplicity and usefulness. "What induced me to make experiments on the subject," he writes, "was, my observing that all kinds of vegetables were continually varying in their growth, quality, production, and time of maturity. This led me to believe that the great author of nature has so constructed that wonderful machine, if I may be allowed the expression, as to incline every kind of soil and climate to naturalize all kinds of vegetables, that it will produce at any rate, the better to suit them, if the agriculturists will do their part in selecting the most proper seed." For over fifty years he had maintained the "long, warty squash" on the farm "without changing," and he adds that they "are now far preferable to what they were at first." He had also maintained early peas and asparagus in vigorous and pure condition for over half a century. It is significant that both of these, also, had been "improved." He made similarly successful experiments in keeping and improving strains of the potato, for even at that time the complaint was "very general," as he writes, "that potatoes of every kind degenerate." Indian corn was equally plastic in his hands. "For many years past.," he says, in closing, "I have renewed the whole seed of my winter grain, from a single plant which I have observed to be more productive, and of better quality than the rest; a practice which I am satisfied, has been of great use."

*For a somewhat full discussion of the philosophy of the benefits resulting from change of seed, the reader is referred to my handbook upon "Plant-Breeding."

It will thus be seen that Cooper clearly apprehended the value of repeated selection for the amelioration of plants; and finding it so potent, he made the natural error of discouraging the change of seed. For himself, however, he was wholly correct in refusing a change of seed, because his own seed was better than that which he would be likely to secure by exchange; but we now know that while selection is the greater factor, change of seed is also important because it incites variation.* Cooper knew that plants mix by crossing. He had also learned that the character of the entire plant is more important, when choosing seed-parents, than that of the particular fruit from which the seed is taken. "The common method of saving seed corn, by taking the ears from the crib or heap, is attended with two disadvantages, one is, the taking the largest ears, which have generally grown but one on a stalk. This lessens the production; the other is, taking ears which have ripened at different times, which causes the production to do the same." A year or two ago I wrote: "The practice of selecting large ears from a bin of corn, or large melons from the grocer's wagon, is much less efficient in producing large products the following season than the practice of going into the fields and selecting the most uniformly large-fruited parents would be." This remark was drawn from general experience and observation. I had not then read Cooper. I now find that my advice is a hundred years behind time! This is not the only instance in which I seem to have copied Cooper. I have said several times that the seeds of the southern watermelons are almost worthless for the north because they give late fruits, but that the variety may eventually be fitted to our conditions by a constant selection of seeds from the earliest plants. "A striking instance of plants being naturalized," writes Cooper, "happened by Colonel Matlack sending some water melon seed from Georgia, which, he informed me by letter, were of superior quality. Knowing that seed from vegetables which had grown in more southern climates, required a longer summer than what grew here, I gave them the most favourable situation, and used glasses to bring them forward, yet very few ripened to perfection; but finding them to be as excellent in quality as described, I saved seed from those first ripe; and by continuing that practice four or five years, they became as early water melons as I ever had."

With this digression, made for the purpose of introducing a worthy and unappreciated compatriot, and to still further illustrate the early development of the ideas associated with the amelioration of plants, we shall now return to the main course of our narrative.

Knight.

Whilst Van Mons was experimenting in Belgium, another bold and prophetic spirit was pursuing similar studies in England. This was Thomas Andrew Knight, who, in the variety, accuracy, significance and candor of his experiments, stands to the present day without a rival amongst horticulturists. Knight was born in 1759, and died in 1838. He completed his academic studies at Oxford, and soon removed into the country where, as he had intended, he spent the remainder of his days. He established himself at Elton, near the paternal home of Downton. He seems to have been brought into agricultural studies chiefly through the efforts of Sir Joseph Banks, who recommended him to the Board of Agriculture as a fit person to answer correspondents' inquiries. He soon became deeply interested in matters relating to the physiology and amelioration of plants, and entered upon investigations which have now come to be considered amongst the classics of botany. These experiments were concerned with the reasons for the upward growth of stems and the downward growth of roots, the motions of the fluids in plants, the physiology of the wood and bark, the motions of tendrils, and the like. In purely horticultural lines, he took hold of the common perplexities of the time and endeavored to solve them. He made studies touching the best methods of cultivating many plants, and he was amongst the first to make really scientific experiments with the growing of plants under glass. He gave particular attention to physiology and methods of grafting, and appears to have been the first to perfect the method of root-grafting which is now in common use. The activity and variety of his interests during the first third of the century attracted the widest attention, and placed him at the very front of English-speaking horticulturists.

But Knight did his greatest work in the direction of ecological studies, through which he desired to discover the best means of improving plants. He took up the vexed questions of the running out of varieties, and he made great efforts to produce new ones. It will thus be seen that the greatest problems which presented themselves to Knight were exactly those which appealed to Van Mons. But the two men were unlike in temper. Van Mons, as we have seen, projected a general theory and then set out to prove it. Knight, on the contrary, began an inquisitive study of nature, and never arrived at a general theory of the amelioration of plants. It is true that he had hypotheses for some of the minor problems which he undertook, but this is essential to any efficient study. An hypothesis is the line to which the axman works. But these hypotheses of Knight's were never of the dogmatic kind, which apply themselves with unvarying assurance to large classes of facts. One of these hypotheses is worth mentioning here, because it is so closely like that held by Van Mons. He was convinced that all varieties of fruit trees "become subject within no very distant period to the debilities and diseases of old age," and that each variety has a "most productive and eligible" epoch, and that this epoch occurs whilst the variety is still young; or, as Knight expresses it, "the most prolific period is that which immediately succeeds the age of puberty." That is, varieties are strongest and most productive early in their existence, and thence tend to gradually fail. He also maintained that the cions of any seedling tree cannot be "made to produce blossoms or fruit till the original tree has attained its age of puberty," and that the longevity and behavior of any variety are intimately connected with the behavior of the original seedling from which the variety had its birth. These ideas were suggested by experiment and observation, for he tells us that he was first convinced "that each variety possessed its greatest value in its middle age," but certain experiments led him to change his views. This dogmatic hypothesis of the duration of varieties was widely repeated, but it appears, fortunately, never to have exerted great influence, and it is so insignificant in comparison with Knight's greater work that we need not dwell further upon it.

Van Mons was the first horticulturist to boldly exemplify and demonstrate the value of the great principle of repeated selection in the origination of varieties. Knight was the first to show the value of crossing for the same purpose. Koelreuter, at the middle of the last century, had made many suggestive experiments in the crossing of plants, but his studies were concerned with the immediate means and effects of the operation. Sprengel, at the close of the century, had observed some of the wonderful adaptations of flowers to insects, but he did not perceive the meaning of these adaptations to the progress of the vegetable world. Knight was the first to directly undertake the improvement of plants by means of crossing. "New varieties of every species of fruit will generally be better obtained,'' he writes in 1806, "by introducing the farina of one variety of fruit into the blossom of another, than by propagating any from a single kind." He made experiments in crossing which, for extent, variety and importance, would do great credit to any experimenter of the present day, even after we have obtained much definite knowledge of the results of cross-breeding. The varieties of fruits which he raised, largely by means of crossing, were many and important. Amongst those which American horticulturists know are Elton and Black Eagle cherries, Ickworth Impératrice plum and Downton nectarine. He originated many varieties of potatoes, and several of peas, cabbages, pears, strawberries and apples.

The transcendent merit of Knight's studies and experiments lies, however, in the fact that he made them contributions to our knowledge of the general forces and processes of nature, rather than to restrict them to a special application to horticulture. He was one of the pioneers of that inductive type of experiment which reached such a high level in the work of Darwin, and which has come to be a passion in our recent life. In other words, he was a philosopher. In the closing year of the last century, he hinted at the fact that nature employs intrabreeding for the purpose of improving plants and animals; he demonstrated the value of crossing as a means of producing new forms; and he propounded the idea, which is now very generally accepted, that the leading cause of variation, at least in plants, is an excess or modification of food supply.

Retrospect.

Such, then, are the leading features of the attitude of two great horticultural philosophers to the history of the ideas respecting the breeding of plants. The work of these men derives its chief value when it is interpreted by means of the work of Darwin and his successors. We now understand the fundamental nature of cultivated varieties, and we are able to specify many of the reasons why they come and go. The key to the entire subject lies in two propositions, which may be stated as follows:

First. Variation, or the truth that no two living things are alike, is the most important fact in organic nature. This variation is important because we know that it is the starting-point for the making of greater differences.

Second. Variation may be augmented by constantly propagating from the individuals possessing the most pronounced characteristics.

These two facts represent the sum total of the forces with which man has worked from the beginning for the improving of plants and animals. Even barbarians practice selection in the growing of their plants. Any being possessed of the faculty of choice and capable of planting seeds must habitually and necessarily choose from those plants which suit him best. The most ignorant workman in our fields does the same. This unconscious choice of parents, operating slowly during a very long period of time, had so profoundly modified plants and animals that when the descriptive naturalists appeared last century, they were unable to determine the origin of many of these domestic forms. Even at the present day, with all our study of nature and our inquisitive searching into the uttermost parts of the earth, there are more than a hundred domestic species of which we do not positively know the aboriginal forms. The overwhelming majority of ameliorated forms of plants have appeared in just this way,—as the result of half-conscious or even unconscious and unrecorded efforts. The definite breeding of domestic animals began about the middle of last century with Robert Bakewell, and the breeding of plants may be said to have begun with Van Mons and Knight. Even at the present day, the phenomenal amelioration of the chrysanthemum, rose, potato, and other plants has been, for the most part, undirected. They have developed rapidly because variation has been so rapid and so marked.

You now ask me why variation has been so marked of recent years. The question is readily answered: It is because the conditions under which plants have been grown are so varied. Better cultivation, greater attention to training and feeding, the growing of plants in many and unlike, regions and soils and local conditions, the prodigal exchange of seeds and plants between dealers and buyers, crossing,—all these are the agents which tend to make plants more and more various and unlike. Selection of these variations, by means of which they have been intensified and augmented, has also been more universal and more thorough. The greater the number of persons who grow plants, the more diverse are the ideals of selection; and the more variable plants are, the more plastic they are in the hands of the breeder. These new forms have no limit of duration. They do not wear out. But as often as they are grown in new conditions and by different persons, they again vary and are again the subjects of selection. So there is a constant shifting of forms, but the longevity of any form depends, not upon any predestined limitations, but upon the accident of the conditions under which it is grown and bred. The same forces which have brought domestic plants and animals to their present condition, and which are largely responsible for the general uplift of the vegetable kingdom from the beginning, are the ones which, variously modified and refined, must carry the domestic flora and fauna on to the end. What, then, is the mystery of plant-breeding? Only this: Good care, varying circumstances, judicious selection for what you want!