Heredity on the one hand and variation on the other are the two forces upon which the origin of new species or kinds of plants is based. Both of these work in rather definite ways, some of which are fairly well understood, but many of which are still among the things that scientists are striving to clear up. As the capacity to inherit characters from one generation to another reflects itself in the generally stable conditions which the plant world exhibits, while the capacity to vary is the only source of new forms, it is quite naturally the variations of plants from one generation to another which have been most studied. And the study of variation in plants is not the simple thing we might assume it to be, having in mind only the well-known fact that no two organisms are exactly alike. Wherein do they differ? Are their perhaps temporary or even quite casual differences passed on to their progeny? These and many other questions about variation make it at once the most complicated and often one of the most fruitful subjects of plant research. It is clear enough that with the bewildering variety of different plants in nature it is next to impossible not only to record accurately the amount of variation or its probability of being handed on, but least of all to arrive at any clue as to the origin of that variation. Because of this, and still more because practically no one has ever seen the actual origin of a new species in nature, for we only see the finished product, practically all our knowledge of the laws of variation has been derived from studying cultivated plants. The ease of controlling them and of recording thousands of observations of their characteristics has made the work of the plant breeder, and others who study variation in the vegetable world, much more of an exact science to-day than the mass of often interesting but usually unrelated data that crowd the pages of older botanical literature.

One of the main facts about variation is that it is itself a very variable thing, and the nature of those “fluctuating variations” which are so common in nature well illustrate the point. Within what we know as a species there are many individuals that vary one way or the other from a fairly central, we might almost call them normal, mass of individuals which are typical of that species. Nearly all these forms on the fringe of the species due to the environmental changes and not to changes of hereditary constitution will, if left to themselves, tend through their progeny to become more like the central mass as time goes on, while their position, or some other equally nontypical edge of the fringe, will be taken up by other variants from the average conditions. The amount of this fluctuating variability among plants is beyond calculation, and its action has often been likened to the swinging of a pendulum, which of course spends twice the time passing through the center of its arc, that it does on the limits of it. This very nearly expresses the proportion of fluctuating variants to the mass of typical individuals in many species of plants. In many others, often peculiarly unstable species, the number of individuals at the fringe is very large indeed. Sometimes there may be one or more that do hold their characteristics, in which case we know that they are not true environmental variations but have actually a different constitution. These will be considered presently under another and different sort of variability. But, speaking generally, these fluctuating or environmental variants are merely forms of the species, and, other things being equal, they will not actually originate new species.

It should be emphasized here, and before we go further in our discussion of variation, that species and varieties are after all largely creations of the mind of man rather than the reflection of actuality in nature. When we speak of a “species” it is merely a term which through usage by botanists becomes the symbol of a group of plants more like one another than like anything else. It is obvious that it is therefore necessarily an inaccurate designation of the actual conditions found among plants, which might almost be considered as all belonging to one great group of which, for our convenience in referring to them, we mark off units (families, genera, species, etc.), much as units are marked off on a rule. Species, then, and varieties of plants, notwithstanding the utmost refinement of method used in designating and describing them, and this is historically the most ancient and the most widely developed phase of botanical science, cannot reflect the true conditions, and for a number of reasons. The chief one is that species differ in usually several characters one from another and in large genera there is often a bewildering recombination of characters of the genus in the species belonging to it. Species and varieties are concepts of convenience, nay of absolute necessity, in talking or writing about plants, but hardly expressions of exact truth.

With this in mind we can appreciate the position of those plant breeders who insist that the basis of differences in all plants are the simplest, so-called, factor expressions, which can be isolated and studied with some approach to exactitude in experimental cultures. A factor may be defined as the hereditary determiner or base, which, either singly or in conjunction with other factors, is expressed as a character, such as tallness in peas, or brown eyes in human beings. Such studies have built up a body of information about variation in plants that show it to be of several kinds and with different chances of being passed on from one generation to another.

It was noted in the paragraph before the last that fluctuating variants were sometimes so far off the usual that they might almost be considered distinct forms or varieties. Many such changes appear to be the result of different conditions of the local environment, due to changed conditions rather than to any internal difference in the constitution of the plant itself. A familiar illustration of environmental variation may be seen in lima beans. In any considerable number of plants one often finds smaller and larger pods, either sparsely or well filled with beans. If the beans from the small-podded, few-seeded variants are planted they will produce, apparently quite indiscriminately, large and small podded progeny, just as there will result a mixed progeny if only beans from the well-filled and large-podded kind are sown. In other words the plant fluctuates about a general average which typifies the usual or mass characteristics of the species. One should not, however, regard all variations of the character in lima beans, or any other plant as environmental or fluctuating ones, for some of them may be due to differences in hereditary constitution. And these could only be determined by breeding tests.

Environmental variations are as frequent as the ever-changing conditions of plant growth may determine, and it is common knowledge that such diversity of the environment and the variants resulting from it are extremely frequent.

A much more fruitful source of new forms of plant life results from natural cross-fertilization, which, as we saw in an earlier chapter, is the nearly universal condition in the plant world. If species and varieties can be distinguished only by factor differences, as the plant breeders no doubt correctly insist, it becomes obvious enough that we have in cross-fertilization to consider not alone the factor differences of the pistillate or female, but also of the staminate or male contribution to the union, and how these are reflected in the progeny. Our knowledge of this has practically all been based on work done on cultivated plants under control conditions and it shows some interesting developments which occur from crossing.

If garden peas with, let us say, reddish-purple flowers are crossed with white-flowered ones, the progeny will not be a mixture of these colors but all reddish-purple. If all danger of subsequent cross-fertilization is excluded this first generation of reddish-purple progeny will themselves produce reddish-purple and white progeny in the ratio of three to one. But the extraordinary part of it is that in the third generation all the white and about one-third of the reddish-purple plants will breed true to color. The balance of the reddish-purple plants, which comprise about two-thirds of the second generation, will, if their seeds are germinated, produce colored as against white-flowered progeny in the three-to-one ratio. In other words, these artificial crosses, made by the plant breeder, and this splitting up of hybrids which has been many times verified, are seen to be very fertile causes of the origin of new forms of plant life, if only the factor and character differences in the ancestry be sufficiently complex. With no two plants precisely alike, with cross-fertilization so nearly universal, and with all characters, not a single character or factor expression, as in control conditions likely to be affected by the cross, it may be seen how fruitful a source of new forms this natural crossing may be. It is, in fact, not surprising that plants vary, but that the force of heredity will hold them into such recognizable categories that the red maple, or white ash, or blue cohosh are, with thousands of other species, after all fairly definite designations without which talking and writing about plants would be all but impossible. Some of our most beautiful garden plants have arisen either as the result of natural crossing, or crosses deliberately made by the plant breeder. The scores of forms of the common garden lilac have mostly come about by such crosses, although many other garden plants have arisen by still another kind of variability.

The effects of crossing which have been so briefly noted were not understood, as indeed the cause of them is still unknown, before 1865, when Gregor Mendel, an Augustinian monk, published the results of his work on peas, which furnished the basis for all subsequent work on this kind of variability. His work was neglected until 1900, when what is now known as Mendel’s law, involving the Mendelian ratio already noted, was rediscovered by three independent workers. It is now practically universally accepted as the way in which natural or induced hybrids transmit their characters.

There remains still another type of variability which has been noticed from very early days, and received the name sport, because quite suddenly, from a crop of otherwise similar specimens, one or a few plants showed marked and permanently transmitted differences from the average condition. Such sudden offshoots, which occur rather frequently in many plants, are known as mutants, the process as mutation. Hugo de Vries, a Dutch botanist of world-wide fame, was the chief modern figure who drew attention to mutants, and explained how they differ from fluctuating variants in that while these tend to revert to the average or mass conditions the mutant, once it appeared, held true to type. A well-known example of mutation is the cabbage, brussels sprouts, cauliflower, and kohl-rabi, all of which are sports or mutants from a weedy seaside plant of the mustard family, native in Europe. Since their appearance, hundreds of years ago, they have held their essential characters. If they had been environmental variants they would in all probability have reverted to their weedy ancestor. Hundreds of sports or mutants have been recognized and isolated, so that many of our most valuable garden plants have arisen through this ability of plants to vary in often sudden and rather startling degree. The gardener and horticulturist, from long observation and a keen sight for valuable novelties, have always known that sports are fruitful sources of new forms of plants, but De Vries first scientifically studied them and worked out the principles by which they apparently react. The cause of them is still unknown.