THE PUREST WHITE IN NATURE
STRIKING COLOR CHANGES IN THE WATSONIA
Bulbs are not usually measured by the cord. But I do not know how better to give an idea of my work with the Watsonia than to say that in recent years I have destroyed about eight cords of bulbs of this plant each season. A cord, it will be recalled, is 8 feet long, 4 feet wide, and 4 feet high. If you will picture in your mind's eye eight cords of wood piled together, and recall that the Watsonia bulbs have corresponding bulk in the aggregate, and that each individual bulb is of the size of a small gladiolus, you will gain a fairly clear conception of one of the least satisfactory aspects of the plant experimenter's work. These discarded bulbs, it should be understood, would produce very beautiful flowers. It seems a pity to destroy them, when so many people would be glad to have them for cultivation. But past experience teaches me that I have no alternative in the matter. If I permitted these bulbs to go out they would presently be exploited by someone as "Burbank's Best Watsonia," or under some still more spectacular title, and my reputation as a producer of fine varieties would suffer, as it has many times in the past through similar deceptions. So there is no recourse, after selecting the comparatively small numbers of bulbs that give greatest promise for the carrying on of the experiment, but to destroy all of the remainder, even though, as in the case of the Watsonias, these may number a quarter of a million bulbs of considerable intrinsic merit, representing an enormous amount of labor.
A FLOWER THAT RIVALS THE GLADIOLUS
The Watsonia has been somewhat recently introduced, and has made its way slowly. So it may not be superfluous to tell the general reader that this plant bears a close resemblance to the gladiolus. It is indigenous to South Africa, one species being found also in Madagascar, and is represented by a number of wild species, among which two or three have pre-eminent importance from the standpoint of the horticulturist. Perhaps the similarity of the Watsonia to the familiar gladiolus has interfered with its rapid introduction. Moreover the new plant is somewhat less hardy than the old one, and although it thrives abundantly in the climate of California it cannot as yet be grown satisfactorily in the gardens of the northeastern United States. But there is one quality of the Watsonia, in its perfected varieties, that puts it quite beyond rivalry of the gladiolus. It produces a beautiful snow-white flower. As we have elsewhere noted, there has been, until recently, no truly white gladiolus. But the white of the Watsonia has been characterized as the "whitest white" in nature. In point of quality of its whiteness, this flower is perhaps the only rival of the Shasta daisy. Each of these flowers is of a snowy whiteness, undimmed by the slightest trace of pigment. The original wild forms of Watsonia, to be sure, are not white. On the contrary they are of various hues of red and pink. But there is apparently a spontaneous tendency to produce now and again a white variant, for at least two, and perhaps more, of these have been introduced from South Africa that were probably of independent origin. The white forms that are most familiar under cultivation are so similar that they have been thought to be identical, the origin of one of them being somewhat in doubt. The white Watsonia whose origin is clearly known is descended from a plant discovered about eighty miles from the Cape of Good Hope by Mr. H. W. Arderne, of Cape Town. He took the plant to his garden and in 1892 had a goodly colony of the white flowers under cultivation. In due course they were introduced, and gained a measure of popularity among discriminating horticulturists, chiefly because of the exquisite whiteness of the flower. Meantime, however, the original species has not been neglected, although comparatively little work has been done, in this country at any rate, in the cultivation of any of the Watsonias at the time my experiments commenced. Possibly the flowers would not have been prized but for the introduction of the white, as the others are rather dull and not particularly attractive in color.
HYBRIDIZING THE WATSONIA
I have never been able to determine clearly whether the white variety named W. Ardernei in honor of its discoverer is identical with the variety introduced as W. O'Brieni. They are closely similar, but it seems not to be clearly established as to whether they come from the same stock, although the individuals from which the two races have developed were undoubtedly discovered independently. On receiving the white Watsonia I planted it on a damp piece of sandy land at Sebastopol, but the bulbs did not thrive, and it was two or three years before any one of them bloomed. I learned through experience that the bulbs do not require too moist soil. They thrive in soil that contains a great mass of leaves, and under proper conditions they put out numerous branching stalks, about four feet in height, which for months together are covered with beautiful snow-white flowers, which have, as already stated, the size and much the general appearance of small gladioli. The conditions of soil under which the Watsonia thrives are similar to those required by the gladiolus. As soon as the colony of white Watsonias was fairly established, I began making crossing experiments, using for the cross the reddish pink species and including, a few years later, also a pink variety of the W. Ardernei that was sent out by a Dutch florist. As usual in these experiments, I hybridized with one species after another until in the course of a few seasons we had crossbred forms of multiple ancestry. There were strains of the white Watsonia in them all, but also strains of the reddish and pink species. By 1904 I had a crossbred colony of Watsonias numbering about fifty thousand seedlings. This doubled in the succeeding season, and in recent years the colony has attained the proportion suggested by what has already been said about the elimination of bulbs by the cord. Needless to say there is great variety among these complex crossbred flowers. All of them retain the essential characteristics of bulb and stalk and manner of growth of the Watsonias. But in their size of flower, and in various important characteristics, they show departure from either of the parent forms. Perhaps the most striking individual development is that of a pure white form of Watsonia that has double flowers. This double Watsonia is an unusual flower. The doubling has been brought about, not by the transformation of stamens, as in the case of a double rose or dahlia, but by growing a new circle of petals outside the old ones. This form of doubling, to be sure, is not altogether anomalous. It occasionally takes place in the case of the rose and the carnation, and I have known it to occur with the apple blossom. But it is not very common. It is sometimes spoken of as supernumerary doubling, to distinguish it from the usual type in which each new petal takes the place of a stamen. In addition to the double white Watsonia, the crossbred colony has presented single white ones that have much larger and more open blossoms than the original forms. Also some that grow on much taller stems, and others, on the other hand, that are dwarfed, and of more compact form. Not a few show marked improvement over the original form. If possible they surpass the original in snowy whiteness, and they not only bloom much earlier than the type form, but they are the most persistent bloomers, putting forth flowers almost perpetually throughout the season.
STRANGE FORMS AND ENTRANCING COLORS
But even these snow white members of the colony are surpassed in beauty by some of their associates that show the most remarkable combinations and blendings of colors. The parent forms, as we have seen, are reddish and white, but these blended hybrids present such combinations of colors as I have never seen in any other tribe of flowers except the orchids. It would be impossible to describe with any degree of accuracy the varied hues that these amazing and delightful blossoms present. There are combinations of violet and rosy pink, soft apricot yellows, salmon, nearly pure yellow, yellow shading into pink, deep, dark crimson, light crimson, and purplish tints of many shades. And these various tones and colors are so shaded and blended as to produce an effect which, as I said, can be matched only among the far-famed orchids. To produce races of flowers of such varied and entrancing hues from parent forms that had no exceptional distinction except for the whiteness of one variety, is to experience in full measure the best rewards that await the patient plant experimenter. It chances also that these wonderful blossoms are not only individually delightful, but they are produced in such profusion as is not approached by the uncrossed races of Watsonias. And to cap the climax, these profusely borne and gorgeously colored blossoms are put forth throughout the season, early and late. All in all, then, the new hybrid Watsonia must be given high rank among the aristocrats of the flower garden. They now lack nothing but an element of hardiness that will adapt them to grow in regions of the country where the climate is doubtful and the conditions are less favorable than those that prevail in California. Somewhat earlier, the species Watsonia coccinea was introduced into the combination. It had smaller and more scattered flowers, long and tubular, and it was of doubtful value, and introduced with trepidation. Some of the new hybrid forms presently developed long slender tubes, while the flowers sit close to the main stalk. Some have star-shaped flowers with narrow pointed petals, others have wide rounded petals that give the flowers the appearance nearly of single dahlia blossoms. Still others are of a curious intermediate form-three of the petals being rounded, and three star-shaped. Flowers of the last named type are quite anomalous. Petals of some of the old Watsonias were star-shaped, and others were rounded, but the combination of the two qualities is unique. Among the hybrid seedlings there are some that are only seven or eight inches high, appearing with tufts of wide dark green leaves at a time when others with slender leaves have shot up to a height of eighteen inches or two feet. We have seen similar differences among other hybrid plants. They show at once the diversity of the racial strains within their germ plasm, and the possibility of segregating and recombining traits of different ancestors. There is a corresponding diversity as to the bulbs, and in particular as to the degrees of rapidity with which they multiply. There are varieties that will produce a bushel of bulbs from a single one in a comparatively short time, whereas others multiply very tardily. It is rather curious to note that the bulbs that are the most rapid multipliers are usually the ones that produce the best flowers and bloom most abundantly. In dividing the bulbs of the hybrid seedlings, it is observed that some spread out naturally into bulbs of even size, and are easily pulled apart, like gladioli, thus being multiplied with facility. Others grow together in clusters that must be wrenched apart, breaking the bulb seriously, or else cut with a knife. All these matters are taken into consideration in the selection through which the few are singled out for preservation and the many are destroyed. It is my custom, having selected a certain number for preservation, to cut away nine-tenths of the seed pods in order to strengthen the bulb, thus stimulating the fullest development.
THE CARE OF SEEDLINGS
As the Watsonia is not generally known, it may be worth while to give a few specific directions as to the raising of seedlings of this interesting plant. In general it should be said that, where the climate is suitable, the Watsonia may be raised as easily from the seed as the gladiolus, and the treatment required is altogether similar. My method is to plant the seed in shallow, well drained boxes of sandy soil, as soon as they are ripe in the fall. By March we have, in each box eighteen inches square, perhaps a thousand Watsonias about six inches in height. They thrive very well when planted as thickly as this in the boxes. In the spring the choicest appearing plants are transplanted singly into rows. The ones that are not quite so choice are set out in the mass, by breaking up the soil into squares holding fifty to one hundred plants, all being planted in the open field, and by fall are ready to transplant into rows for testing. A still more rapid method is to sprout the seed in moist sand at the proper season in the fall, just as the rains commence, sowing them quite thickly in drills eight inches apart and an inch deep in sandy soil, half the covering being sawdust. Scatter a few weeds over the surface to keep the winter winds and heavy rains from removing the sawdust. Early in the spring the young Watsonia seedlings come up as lustily as blue grass on a lawn. Those that do not make a strong growth are allowed to stand thickly in the row another year, when they can be removed and planted. For field culture, they should be planted four inches apart in rows four feet apart, being set quite deeply that they may resist the summer droughts. Of course the more careful method first suggested is desirable if we are to raise plants of the finest quality. You also get results a year or two earlier by handling the plants individually. At the same time you insure the production of a plant from each seed, while when the plants are handled in a mass a good many of them no doubt fall by the wayside. When treated according to the first method, many of the plants mature in the second year, and all of them in the third, so that they can be fully tested in that period. Moreover by the third year each bulb has developed quite a nest of bulbs about it, from each of which a new plant may be grown. The results already attained with the Watsonia mark this plant as one that must take high place among favorites of the flower garden. What chiefly remains to be done is to make the bulbs more hardy, so that they are adapted to different conditions of soil and climate. At present the flowers are chiefly grown in California and shipped to the eastern market. But in due course races will be developed that can be grown in the east, and the Watsonia will come to rival the gladiolus and in some respects to outrival it for all the uses to which that flower is adapted. Moreover it will perhaps prove possible, through hybridizing the Watsonia with the gladiolus, to develop new races of plants combining the qualities of each in a way that cannot be definitely predicted. Up to this date, 1914, I have produced a great number, say four of five hundred, hybrids between Gladiolus gandavensis and the Watsonia, in most cases using the pollen of gladiolus (chiefly because it is more abundant and the Watsonia is more certain of seed), but sometimes making the reciprocal cross. Only three or four of these blossomed, largely, perhaps, because a great number of them were very tender and were destroyed by the frost, although older plants of the Watsonia withstood the season. It should be explained that the gladiolus does not withstand the coldest part of our winter any better than the Watsonia; but fortunately the gladiolus does not generally make a winter growth, so it may be left in the ground with less danger. Whether most of these hybrids were made tender by crossing with the gladiolus, or whether some new element came in through the crossing, this experiment, which promised so much, was finally a failure. The seedlings showed the thick stem of the gladiolus. Some of them grew only a few inches, while others grew to great height. Of the hybrids that blossomed, all died next year from gladiolus diseases. It may be of interest to add that double white seedlings from the Watsonia have been produced; also double flowers of other colors, pink, light and dark salmon, and white. We have naturally had occasion, in recent chapters, to pay more attention to the matter of color than to any other single flower quality. For it might almost be said that flowers have been developed for color alone. A certain amount of attention has been given to modifying their forms but this has always been subordinated to the question of modifying their colors. If attempts have been made to increase the size of the flower, and to multiply its petals, the central thought has been to produce a more striking color display. In exceptional cases, notably that of the orchid, anomalies of form add greatly to the interest with which a flower is regarded; but even with the orchids, it is unquestionably the delicate beauty of the coloration, and not merely the grotesqueness of form, that gives the flower popularity. From the standpoint of the plant experimenter, the question of color in the flower is one that has perennial interest. In a very large number of cases new varieties are developed solely along the lines of color variation. We have seen that there are almost endless modifications of color in the same flower, particularly in such variable races as roses and poppies and dahlias, and the case of the Watsonia, which has just come to our attention, illustrates the interest associated with the modification of color even when no other change is involved. The white Watsonia that was discovered among its pink fellows was given a new botanical name, and went forth to conquer the world captioned as a new sub-species. Yet it differed in no obvious regard from myriads of its fellows except in the matter of color. It was pure white, and all of its fellows were reddish pink. That fact gave the one Watsonia distinction among the millions and insured the propagation of its progeny and their migration to the utmost corners of the earth.
EVOLUTION OF COLORED FLOWERS
Something has been said in various preceding chapters as to the philosophy of color variation. The origin of the colored floral envelope has been ascribed to the influence of insects. We have been made aware that the floral envelope was developed as an advertising device to attract insects, that their services may be engaged for the transfer of pollen that is so essential in keeping up the necessary adaptability and vitality of a race of plants. We have been led to infer that the floral envelope is one of the most recent developments in plant evolution, inasmuch as the earlier forms of plant life had no such apparatus and their successors developed it only after the evolution of the insect tribe. And we have doubtless been correct in ascribing the ready variability of the floral envelope to the fact of its relative newness. The stalk and branches and leaves of the plant have persisted, more or less modified in form but essentially unchanged in functions, from the remotest periods, and hence have attained a fixed and determinate arrangement of their hereditary factors that is difficult to disturb. The conspicuous advertising sign that we call a flower has been put forth so recently that it has not attained any such degree of stability. And in particular, the color of the flower is an endowment that, as contrasted with the general structure of the plant, must be thought of as only a thing of yesterday. We are justified in believing that even among the old tribes of plants-those whose primeval forebears have left their remains in the geological strata-the flower is the one structure that has been most subject to variation. And we may doubt whether there is any flower whatever that has not changed its color more or less within comparatively recent times, geologically speaking. Something has been said as to the probable relations of the different primary colors in their various associations in the floral envelope. We have seen that flowers of the same species may vary from deep red to delicate violet, and that it is the commonest thing for a species that is usually gaudily colored to have representatives that are pure white. And it is possible, by a careful survey of the field, to draw conclusions as to the probable sequence of development through which the variously colored flowers have been evolved. In the first place, certain inferences may be drawn from what is known as to the hereditary responses of different flowers, in particular when hybridizing experiments are performed, that at least give clues to the story of the evolution of color. Analogies drawn from the study of the spectrum are also of aid, in connection with these practical observations, in developing theories of the philosophy of flower-coloration, which, if they cannot be said to be definitive, have at least a large element of plausibility and are full of interest.
THE PHYSICAL BASIS OF COLOR
It has been suggested that the earlier forms of vegetation were probably red in color, where now the leaf structures in general are universally green; the basis for this belief being the observed manner of reaction between plants of green foliage and those of red foliage when hybridized, the fact that sea weeds are usually red, and the further fact that young vegetation, such as the buds of trees in the spring, is very generally red in color, the subsequent greenness being due to the development of chlorophyll granules. Just why the chlorophyll granule is green is of course only matter for conjecture. But it is obvious that this is the ideal color for this purpose, otherwise it would not have been so universally adopted. The presumption is that the plant finds it desirable to utilize the short rays of the upper part of the spectrum and the long rays of the lower part-those that stimulate chemical action chiefly, and those that are the greatest conveyors of heat, respectively-and that the intermediate rays producing the color green are not needed, hence are reflected or transmitted without influencing the plant. A possible clue to the reason for this is found in the supposition that the plant needs the short light waves to enable it to carry out its chemical function of transforming water and carbon into sugar, and that this process is facilitated by having the tissues warmed by the long waves of the lower end of the spectrum. It has been calculated that the sun beating on a leaf would raise its temperature to a point that would destroy the protoplasm and kill the leaf outright in a very short time were it not for the transpiration of water from the pores of the leaf, through which the temperature is equalized. In spite of this danger the sunlight is known to be absolutely essential to the carrying on of life processes, but it is obviously desirable to limit the amount of heat as much as possible. So the question of the heating effect of the sun must have a share in determining the color of the floral envelope. A flower that blooms in the open and is exposed to the blazing rays of the sun may advantageously develop a glossy surface, just as a leaf does, in order to reflect the largest possible amount of light; and may in addition take on to advantage such transformation of its tissues as will make them reflect the long heat-bearing waves of the spectrum. Such a flower, interpreted in ordinary language, is red in color-for of course that is the untechnical way of stating the facts that a given object reflects the long rays of light, and absorbs the others. Contrariwise, it would be almost fatal for a blossom of ordinary texture to develop such consistency as to absorb the main bulk of the lightwaves, inasmuch as such a blossom would soon be heated to a dangerous temperature. That, doubtless, is why flowers that are even approximately black are the rarest of all blossoms. On the other hand, a flower that reflects all the rays of light, and hence that appears white in color, is given protection against the heating influence of the sun even though it grows in the open. When we add that white is a conspicuous color, the extreme abundance of white flowers is sufficiently accounted for. It is true that flowers that bloom in the shadow, and particularly those that open in the twilight or at night, are almost universally white, also, but this is sufficiently explained by natural selection, since white flowers are more conspicuous at night than those of any other color. Moreover, it must be recalled that white objects transmit heat less readily than dark ones, so white is not a bad color for a night blooming flower, inasmuch as it conserves the internal heat even if it is not called upon to shut out heat from the sun's rays.
THE SEQUENCE OF COLOR DEVELOPMENT
All this is more or less axiomatic, but the further development of the theory of flower coloration involves a certain amount of assumption, and must be held only as a tentative theory. Briefly stated, the essentials of the theory are that the original or earliest color of the flower was green in imitation of the leaf. All the older or primeval types of plants-palms, pines, cypress, ferns, etc. have green flowers even to this day; in some cases slightly tinged with yellow. It is suggested that the color next developed was blue, the genesis of which involved but a slight modification of the molecular structure of the flower, inasmuch as the light waves that produce blue lie next to the green on one side in the spectrum. The subsequent modifications of color were made in two directions progressively. Some flowers were modified in the direction of the violet end of the spectrum, and others in the direction of the red end of the spectrum. The former were first light blue, then deep blue and indigo--represented among existing plants, let us say, by the larkspur and gentian-and ultimately violet. Flowers modified in the other direction were at first yellow then orange and finally red. Evidence is lacking to answer the question as to which end of the spectrum was reached first-that is to say whether the flowers of violet color or red were first evolved. But possibly the two may have been developed somewhat contemporaneously, and they would thus take their place in the hereditary scale more or less on an equality. In any event, we may be fairly assured that there were blue flowers and yellow ones, and probably also indigo colored flowers and orange ones in existence before there were flowers of pure violet or of deep red. In other words we may feel that the violet-colored flower and the red flower are the newest things in the way of color in the plant world. The time of development of white flowers is more debatable. There is reason to suppose that the white flower owes its whiteness to a combination of the conditions which by themselves would be interpreted as greenish-yellow and as blue respectively. It is knowvn that these are the only pigments that can be compounded to produce the color white. So it is perhaps the safest assumption that white flowers were evolved by the hybridizing of greenish yellow ones and blue ones, and that their origin antedates the development of red flowers or of violet ones. In other cases whiteness is due to air in the cells, and may have been a very recent development. Nor is all this a matter of mere unsupported assumption. The inference that such was the sequence of evolution of the different colors seems logical, inasmuch as it pre-supposes the modification of molecular structure of the flower substance in such a way as to reflect successive rays of light in a graded series-or rather in two graded series, one involving shorter and shorter rays as in the flowers that develop from blue to violet; the other involving longer and longer rays in the series that developed from yellow through orange to red. But an assumption based solely on this plausible analogy would not call for very serious consideration. The real strength of the theory lies in the support given it by the observed relations of the different flower colors when brought together through cross-pollenation of the flowers themselves. It is believed, on independent grounds, that the relations of dominance and recessiveness in Mendelian heredity are determined exclusively, or at least in large part, by the newness or oldness, in an evolutionary sense, of the respective elements that make up a pair of Mendelian factors-referring, it should be understood, to the number of repetitions, not to the mere lapse of time. If this assumption is correct-and there is a large amount of evidence drawn from many fields to support it-then a guide is at hand with which to test the theory of color evolution. Indeed, it is largely through the application of this guide that the theory of color evolution itself has been developed. Making a practical application, it would appear that the color green, as manifested in a flower, is so remote an inheritance that it would be recessive to any and every other color; that blue would stand next in line of recessiveness; and that violet and red would be more or less on a par as colors of pre-eminent dominance. White, according to theory, should be dominant to green and blue, but should itself be recessive or hypostatic when brought in combination with red or with violet. As a corroborative illustration, note that Mr. Burbank's blue poppies, crossed with white poppies, produce only white progeny. It would also appear, that the factors for yellow and for blue, which are really balanced or masked to produce the color white, might be segregated when a white flower is combined with another white flower or with a flower of a different color, white perhaps disappearing altogether and being represented only by its disunited elements. Moreover, we have already seen that where various colors are segregated, two dominant colors such as red and yellow being brought together in the same unit system, the two may neutralize each other and fail of tangible representation; just as the colors gray and black are known to do in the color scheme of the coat of the mouse. The practical working out of the scheme is revealed in numerous cases that we have already examined. Thus union of our yellow poppy with a white one that produced a crimson progeny finds ready explanation when we reflect that yellow is commonly formed by the blending of the pigments red and green, and that white is probably due to the blending of yellow and blue. The combination of the yellow flower and the white one may thus be supposed to have re-segregated the colors in such a way that yellow and blue were grouped to form white which was in turn submerged as a recessive factor when coupled with red; the result being that the progeny were all outwardly red. In a similar way may be explained the result of combining the orange daisy with the white daisy; and in general the multiplex presentation of reds and pinks and yellows and whites in the hybrids of poppies and roses and gladioli and dahlias. The fact that blue so seldom appears is explained by the assumption that it was the first color to be developed, after green itself, and hence that it is recessive to all the other colors. When a blue is brought out as in the case of my blue poppy, it is unearthed, as it were, with difficulty, and represents the bringing forth of a quality that has been submerged from time immemorial. Of course there are numerous flowers-although as we have seen they are relatively rare-that are blue in color. These are races that have either retained the ancestral color unmodified because it served them well in adaptation to their environment; or they are plants in which the recessive blue, which must occasionally appear in the course of hybridizations, was preserved and restored to prominence because it served its purpose better than the other colors, whatever they might be, that had supplanted it. As might be expected, deep or indigo blue flowers are more abundant than light or pure blue ones. It is perhaps not without significance that blue flowers have usually a white counterpart-the bluebell furnishes a familiar example. Blue and white, according to the theory just presented, lie close together in the evolutionary scale. Either will be recessive to red or orange or violet; and it is only flowers from the germ plasm of which these dominant colors have been largely removed that are likely to develop blue or white races. Yet the fact that the white flower carries a strain of yellow is an ever present menace to its whiteness, as it may furnish the basis at any time for variation that will introduce yellow strains which stand a good chance of supplanting the blue and white ones. Some further illustrations of the application of this theory of the evolution of color in flowers will appear in our subsequent studies. For the rest, the reader who is interested in speculations of this character will be able to make application for himself, and to test the theory as to its details, in particular if he enters the fascinating field of plant development.
most inviting field for the amateur, even while they still hold their full attraction for the practiced experimenter, and one can hardly proceed far with flower experiments without becoming interested in the phenomena of color-variation.
This text is from: Luther Burbank: his methods and discoveries and their practical application. Volume 9 Chapter 9