®lj^ ®. 1. ItU mtbrara 5Jnrtli (Earoltna ^tate CToUrgp SB123 B82 V.7 This book is due on the date indicated below and is subject to a fine of FIVE CENTS a day thereafter. =£2 "^if^^W \v -Jf Y^M^'C'-'^^^m?)^ WORK LARGE SIZE AND CO^MPACT GROWTH A quality sought by Mr. Burbank in the development of the gladiolus was the habit of comjjact growth on the stem, so that the flowers should be solidly massed, instead of being scat- tered along the stem. The brilliant blossoms pictured here indicate striding success in the matter of compact growth, as well as in the increased size and symmetrical form of the flowers themselves. iLn-.^iiiAiKo DAVID STARR JOKPAX N' COMT HOW PLANTS ARE TRAINED TO WORK FOR MAN BY LUTHER BURBANK Sc.D ^ FLO W E R S VOLUME VII EIGHT VOLUMES • ILLUSTRATED PREFATORY NOTE BY DAVID STARR JORDAN P. F. COLLIER & SON COMPANY NEW YORK Copyright, 191-i By The Luther Burbank Society All rights resei-ved Copyright, 1914 By The Luther Burbank Society Entered at Stationers' Hall, London. All rights reserved Copyright, 1915 By The Luther Burbank Society Entered at Stationers" Hall, London All rights reserved Copyright, 1921 By p. F. Collier & Son Company man-ufacturkd in u s. a. CONTENTS PAGE Teaching the Gladiolus New Habits 7 The Canna and the Calla .... 33 The Purest White in Nature ... 63 How TO Obtain Variation Among Flowers 95 Improvements in the Much Improved J.RIS 117 The Tigridia and Some Interesting Hybrids ^^^ Four Common Flowers and Their Improvement Ig3 Everlasting Flowers and Some Com- mon Exotics 185 The Hybrid Larkspur and Other Transformations 211 Ornamental Palms and Climbing ^^NEs 237 Vol. 7— Bur. - 1 A 2 CONTENTS PAGE Lawns and Their Beautification . . 263 The Field and Flower Garden . . . 305 The Almond and Its Improvement . . 323 Walnuts and Other Experiments . . 347 LIST OF ILLUSTRATIONS Large Size and Compact Growth Frontis'piece PAGE A New Giant Gladiolus 12 One of Ten Thousand Variations . . 18 An Unusual Color in the Gladiolus . 24 A Sample Gandavensis Primulinus Hybrid 30 A Seedling Canna 36 Another Seedling Canna 42 Giant and Dwarf Callas 48 The Lemon Giant Calla 52 The Siberian Lily 58 Some of My New Seedling Watsonias 66 A Spectacular Iris 80 A Luxurious Type 90 Selected Chilean Iris 102 A Japanese Iris 112 3 4 LIST OF ILLUSTRATIONS PAGE Seedling Japanese Iris 124 A New Evening Primrose — The America 134 Hybrid Tigridias 144 Another Hybrid Tigridia 148 Seedling Tigridias 152 A Blue Tigridia 158 BuRBANK Verbenas 166 More Burbank Hybrid Verbenas . . 170 One of the Fragrant Ones 174 Australian Star Flower 188 A Plant of Australian Star Flowee . 192 A Hybrid Crinum 196 A Hybrid Everblooming Poppy . . . 202 Still Another Hybrid Poppy .... 206 Improved Hybrid Larkspur 214 More Hybrid Larkspurs 218 Educating the Calendula 222 A Calendula of Real Distinction . . 226 The Variegated Negundo 232 Flowers of the Tecoma 258 LIST OF ILLUSTRATIONS 5 PAGE Spanish Broom 268 A Fritillaria 274 dierama pulcherrima 280 Larkspurs with Wonderful Coloring . 286 A Beautiful Hybrid Larkspur . . . 292 The Bottle-Brush Bush 298 A Selected Night-Blooming Cereus . 312 The Almond and Its Relatives . . . 326 Selected Almonds 330 Some Mammoth Specimens 334 Meats of Selected Almonds .... 338 Almonds Grown in Peaches .... 342 Hybrid Walnuts 358 The Royal Walnut 372 TEACHING THE GLADIOLUS NEW HABITS The Mutability of Species THE history of the growth of ideas shows some curious paradoxes. As a minor illus- tration, it may be recalled that an English clergyman was doing his best — and a very good best it was — to build up evidence of the mutability of natural species at a time when it was rankest heresy to suggest that species are mutable. The clergyman in question was the Honorable and Reverend Dr. William Herbert, Dean of Manchester. His work was carried out in the early decades of the nineteenth century. He was a horticulturist of great skill, and he labored assiduously with many plants. And among those with which he attained conspicuous and striking results that seemed to belie the botanical beliefs of the period, was the plant now familiar in every garden as the gladiolus. The time when the important work of this cler- ical amateur was carried out was one in which 7 8 LUTHER BURBANK such men as Erasmus Darwin, the poet Goethe, and the French biologist Lamarck were advocat- ing the idea of the mutability of species. And no doubt the Rev. Herbert had some of their theories in mind as he went about his plant ex- periments in the gardens of the Manchester Deanery. Yet in the main he was probably quite uncon- scious of the full significance of the experiments that he was performing. The particular experiments that are of interest to us in the present connection are those in which he hybridized one species of gladiolus with an- other, and in so doing not only produced new races of gladiolus, but proved to his own satis- faction that these new races were altogether fertile. Almost half a century later Charles Darwin in his "Origin of Species" had occasion to quote the opinion of the Rev. Herbert, based on his ex- periences with this flower and several others, to the effect that hybrids are not necessarily sterile — a point that was still ardentty in debate. He even cites Herbert as having claimed that the hj'brids gained in fertility over the original species — a fact which Herbert himself regarded as being "a strange truth," but regarding which Darwin, writing with fuller knowledge, asserts THE GLADIOLUS 9 that it was by no means so strange as it would appear. To be sure, nothing revolutionary came directly from the reverend horticulturist's ex- periments. He produced interesting new varie- ties of flowers, but the theoretical bearings of his work were doubtless quite ignored by his fellow clergymen and, indeed, as I have already suggested, were probably only vaguely realized by himself. Yet as we look back on this work now, from the new point of vantage that Darwin gave us, we can see that the work of this amateur horti- culturist must have had its share in disturbing the ideas of at least some of the persons to whose attention it came, and in preparing the way for the new view of the flexibility of species that now seems so much a matter of course that we can hardly realize how revolutionary it seemed to our forbears of two generations ago. A demonstration made with a plant that grows in everybody's garden has force that comes home to us more cogently than records of any num- ber of observations of animals and plants of tropical forests and South Sea archipelagoes. And a number of new species of plants, gladioli among others, that the Dean of Manchester cre- ated by hybridizing old ones made their waj'^ into 10 LUTHER BURBANK the gardens of Europe, and gave their message, we may be sure, here and there to a receptive mind in substantiation of the disputed evolution- ary doctrine, which, even before the publication of Darwin's "Origin of Species," was exciting the interest of the thoughtful. Other Gladiolus Hybrids Before the gladiolus made its full conquest of the popular gardens, however, it was further im- proved by other gardeners, both in England and in continental Europe. The species that the Rev. Herbert had crossed were the showy Gladiolus cardinalis and the smaller but more free-flowering Gladiolus hlandus. Subsequently he crossed a number of other species, and produced races of great beauty and fertility. But a race produced by Mr. Col- ville at Chelsea in 1823, by fertilizing the form known as Gladiolus cristus with the pollen of Gladiolus cardinalis gained additional popu- larity. It was not until 1837, however, that the form was originated which was to make actual con- quest of gardens throughout Europe, and pres- ently to attain corresponding popularity even in America. This new form which became the parent from which most modern varieties of THE GLADIOLUS 11 gladiolus have been developed was raised in 1839 by M. Bodinghaus, gardener to the Due d'Aren- berg of Enghien. Like the other hybridizers, he used Gladiolus cardinalis for one parent form, the other parent being a species known as Gla- diolus psittacinus. We have seen that the cardinalis was used by the earlier hybridizers. It appears that the j^sit- tacinus was also used in hybridizing experiments by the Dean of Manchester. But either he did not make the precise cross that was now made by the Belgian gardener, or the strains he used were somewhat variant; for the hybrid now pro- duced had qualities that gave it a new appeal to flower lovers in general, and in particular made it a flower of such easy cultivation and such striking appearance as to make a strong bid for popularity among amateurs. It gained such vogue as to be thought of every- where not only as a distinct species but as repre- senting a type form of the race of gladioli. It was named gandavensis, from Gand (Ghent), the place of its origin. It is believed, however, that the form of gladio- lus that came to be known everywhere as the gandavensis has in its racial strains the blood of many other species beside the original parents. It is almost certain, for example, that the strain A NEW GIANT GLADIOLUS This is a variety selected for size of the individual jioiioer, rather than for arrangement of flowers on the stalk. It illustrates the accentuation of size to about its present liviit. (One-half life size.) THE GLADIOLUS 13 cf G. oppositiflorus accounts for the modifica- tions of form and for the introduction of a tendency to produce white flowers; and that strains of G. blandus and G. ramosus have also been introduced. In a word, the form of gladiolus that came to be familiar everywhere under the name ganda- vensis is not merely a hybrid, but a hybrid that probably carries the racial strains of at least four or five species, and possibly of many more than that. All of which is essential to an understand- ing of the later developments of the race of gladioli. For when we come to investigate the pedigrees of the chief races of gladiolus that are now found in our gardens, we learn that, practically with- out exception, they are hj^brids that carry the gandavensis strain among others, and hence are multiple hybrids, the precise lineage of which is too intricate for tracing. It is this fact that accounts for the wide range of variation as to form and color that charac- terizes the gladioli of our gardens. For the hybrid races have practically supplanted the original species everywhere. The same thing is of course largely true of most other cultivated flowers, and it is altogether 14 LUTHER BURBANK true of the cultivated fruits and vegetables. As regards a large proportion of these, the culti- vated varieties have not only supplanted the original species, but no definite record remains of the original species themselves. The case of the gladiolus differs, and gains added interest, in that the original species were brought from southern Africa to Europe only a little more than a century ago. The development of the new hybrid races under cultivation, and the elimi- nation of the parent forms by their improved descendants, has taken place in so comparatively short a time that its chief steps are matter of record, as we have seen. So the story of the gladiolus has elements of educational interest for the plant developer that are quite lacking in many of the cultivated plants which attained relative perfection at an earlier period. Early Work at Santa Rosa There are a few species of gladiolus that are native to Europe and Asia, but the ones that were chiefly used by the early hybridizers came from South Africa, as already related. Doubtless this fact was not without signifi- cance in determining the results of the work of the early cultivators. We have seen illustrated more than once the effect of transplanting a THE GLADIOLUS 15 plant to new soils, and in particular of transport- ing it from one hemisphere to the other. We cannot doubt, then, that the change in the seasons and in the soils and climatic condi- tions in general had a share in promoting the variability of the gladiolus when brought to Europe, although, as we have seen, the tangible Stimulus to variation was given through the now familiar method of hybridization. And, by the same token, we may suppose that when the gladiolus was finally brought to Cali- fornia, shifted thus halfway roimd the globe from its new home in Europe, there was an added stimulus given, urging the plant to still further modifications of habit, and supplying yet other elements of variation with which the plant devel- oper might work. At all events the gladioli in my gardens at Santa Rosa and Sebastopol have proved respon- sive and adaptable. And further modifications have been produced in the much modified flower that add greatly to the value of what was from the outset one of the most popular of ornamental plants. I began work with the gladiolus about the year 1882, starting with the gandavensis hybrid, the origin of which has already been described. 16 LUTHER BURBANK At that time there was no great interest taken in America in growing gladiolus seedlings, but I was able to secure a large number of the best types of gandavensis, and also obtained bulbs of about a dozen of the natural species. The material was obtained not alone from American growers and the cultivators of Europe, but also directly from South Africa. From the outset experiments were begun on a comprehensive scale, raising the gladioli by the half acre and acre on my Sebastopol place. The first fault observed in the gladiolus was that the blooms would not stand our California sunshine. Under the glare of the California sun, the blooms would wither in a single day, sometimes in a single hour. Other serious defects were that the stalks were too slender, and the flowers too far apart on the stalk. Moreover, the flowers were small, they were illy arranged on the stalks, giving an untidy appearance to the plants; and often they were only half open when at their best. The colors of many varieties, on the other hand, were fine, it being evident that selection had been made largely for color, by some at least of the earlier experimenters. ]My first object, then, was to remedy the defects just mentioned rather than to modify the color of the gladioli. THE GLADIOLUS 17 In particular I sought, while improving the stalks and the arrangement of flowers on the stalks, to make the petals of the individ- ual blossoms stand out flat and in regular sequence. The work progressed along the lines of hybrid- izing and selection with which the reader is already familiar, introducing strains of the long- neglected natural species to give added virility and stimulate still further variation, thus provid- ing materials for selection. Growing the plants by the acre, abundant material for choice was at hand and my usual method of choosing only a few of the very best representatives of the difl'erent forms that seemed worth developing, destroying the rest, was rigidly exercised. In the course of a few years there were de- veloped varieties which were introduced with new names, and which made their way every- where, and were highly prized by gardeners throughout the United States. Doubtless the most interesting development in this early period was the form named the Cali- fornia. This was a really magnificent semi- double variety which not only excelled in the form and size and color of the individual blos- soms, but which had the added peculiarity of ONE OF TEN THOUSAND VARIATIONS The arrangement of color shades and color patterns in the newer gladioli is very remarkable and could never have been believed possible except by its actual demonstration in evidence, {One-third life size.) THE GLADIOLUS 19 bearing the blossoms all around the stalk like a hyacinth, instead of merely on one side of the stalk as had been customary with all other varieties of gladiolus. Even at the present time, although the varie- ties of gladiolus have been subject to rapid de- velopment within the past few years, I recall the California as one of the most beautiful flowers of the family. Unfortunately this variety was lost, probably by freezing, along with the entire stock of other gladioli, by a Philadelphia dealer to whom it was sold. My gladiolus colony progressed admirably, and the new forms attained a degree of virility that made it no more difficult to raise them than to raise potatoes; indeed, much less difficult, in- asmuch as the gladiolus bulbs in California do not require to be dug or stored, but continue their growth throughout the year. The only object in digging them is to divide and separate them for multiplication. The forms of the plants, and the manner of bearing, as well as the shape and arrangement of the blossoms, improved year by year, and the new varieties of gladiolus came to be well known to dealers throughout the coimtry, and were still under process of development when an unex- 20 LUTHER BURBANK pected complication put an end, for the time being, to my further work with this plant. War with the Gopher The complication manifested itself in the dis- covery that entire rows of the gladiolus bulbs had been eaten by pocket gophers, which had tunneled their way into the grounds, and, boring beneath the gladiolus beds, had feasted on the bulbs, destroying large numbers of them ( mostly during the dormant season) before I discovered the presence of the marauders. The plants do not wither at once even when the bulbs are greatly injured, or in the dormant season totally destroyed. So long rows were destroyed before I knew the necessity of combat- ing the enemy. The attempts to exterminate the pests were at first so unsuccessful that I presently decided to give up the gladiolus colony altogether. I sold the entire lot to an amateur Canadian horticul- turist, ]Mr. H. H. Groff, a banker, of Simcoe, Ontario, and for a good many years my experi- ments with the gladiolus were not renewed. Meantime, every effort was made to extermi- nate the pestiferous gophers, whose depredations were of course not confined to the gladiolus, and through which I suffered an annual loss of cer- THE GLADIOLUS 21 tainly not less than a thousand dollars year after year. Not alone with the gladiolus but with other bulbs it seemed that the animals took special de- light in attacking the choicest plants. And the question of their destruction became finally a very urgent one. Numerous methods of combating the pests were tested. A double box trap set in gopher holes was cumbersome and not very effective. An awkward iron trap was supposed to catch the gopher when he poked his nose against the trig- ger, but missed fire or failed to score a hit oftener than otherwise. One form of trap after another was tried and given up. Attempts to smoke out the animals proved ineffective, as the gopher in- stantly builds a wall to shut out the smoke. Bisulphid of carbon, which gives off a poison- ous, heavy gas, was tested with equal lack of success. About the only resource was the use of poison, commonly called strychnin, placed on a piece of apple, potato, or carrot, combined with the use of a wire trap, in the hope that if one failed the other might prove effective. But in spite of all these methods the gophers multiplied, mostly from neighboring fields, where their dam- age to ordinary farm crops was not so marked. A few years ago, however, a gopher gun was 22 LUTHER BURBANK invented that practically solved the problem. This consisted of a trap so arranged that when the gopher pokes his nose against the trigger a charge of powder explodes beneath the animal, killing him instantly by concussion. This device proved more effective than all others. Sometimes thirty-five or forty gophers were destroyed in a day about the borders of my gardens. And in a short time the gophers were so nearly exterminated that they ceased to be a serious pest. When these old enemies of the bulbous plants were thus finally subjugated, after years of effort, I determined to take up again the culti- vation of the gladiolus. In the meantime, the gladiolus had been much under cultivation elsewhere, and its general and special qualities had been greatly improved. But there remained plenty of modifications that could be made to advantage, and in starting a new series of experiments I had no difficulty in discovering faults to be remedied. Recent Work with the Gladiolus One of the modifications, to which reference has already been made, had to do with the arrangement of the flowers on the stalk. My success in developing a race having the flowers THE GLADIOLUS 23 arranged on all sides of the stalks has already been referred to. In taking up a new series of experiments, the attempt was made to improve on the earlier varieties, by shortening the stems of the flowers, so that they were compelled to ar- range themselves more compactly around it; by insuring regularity of placement; and by diver- sifying the plant arrangements in various ways. Some forms were developed having two ranks of flowers, one on either side of the stem. Other races were developed with flowers in four ranks; yet others with flowers in a spiral. Meantime the individual blossoms were en- larged in size, and their stems shortened, so that, when grown in a spiral about the stem, they crowd one another, making practically a solid mass of petals. The contrast in appearance of a stem of gladiolus flowers arranged on this new plant with the old form in which the blossoms grew only on one side of the stem, or at most on opposite sides, is very striking. Attention was given also to the modifications of the form of the individual flowers. In one form, petals were developed that are broad and ruffled so that they overlap, and thus give the appearance of a double flower. In another form the tendency of the anthers to turn to petals was AN UNUSUAL COLOR IN THE GLADIOLUS This is a rather unusual color. Notice the flat, open flozvcr — a quality prized in the gladiolus. {One-third life size.) THE GLADIOLUS 25 accentuated through selection until double vari- eties were produced; and in some cases the extra petals were added without affecting the stamens. In yet another form, and one that I person- ally admire most, two flowers appear to be fused into one, so that twelve petals are presented in- stead of six. The variety was fixed so that the flowers on every stalk come in the same way, constituting a double flower of an unusual type. Particular attention was also paid to the de- velopment of regularity of petal in the case of the double gladiolus flowers. Irregularity of petals may be attractive in such flowers as the rose and the carnation, but with the gladiolus the double blossoms are less beautiful than the single ones, unless the petals are very regular. There was no great difficulty, however, in making the petals regular, as well as increasing their number by selection. Studies in Color In the newer series of experiments, especial heed was given, also, to the matter of color vari- ation, seeking for clear and brilliant colors of varying shades. The blending of shades, and the arrangement of lines, dots, and edges of 26 LUTHER BURBANK different color on the petals were all carefully taken into account. There is opportunity for skill in the blending of different shades in a flower of such diversity of color as the gladiolus, far greater than the painter's skill in applying colors to the canvas. One learns that there are certain combina- tions that will produce disagreeable colors, whereas others will result in new shades of exceptional brilliancy. The characteristics of each flower to be selected for combination must be carefully studied. If, for example, we cross a yellow gladiolus with a white one, we are likely to get a dingy white that is by no means agreeable, though not by any means necessarily so. The cross of a pale pink with a white form is likely to give us a still paler pink, which would not be regarded as an improvement. Again, from the blending of two nearly white strains, you may even get dark colors in unpredictable combinations. By studying the combinations, however, and making rigid selection among the seedlings, it will be discovered that there are certain tenden- cies that tend to be dominant and that as a rule may be expected to repeat themselves in the THE GLADIOLUS 27 hybrid offspring", overshadowing the less fixed colors. Still the races of gladioli are so blended, and their color factors so mixed, that one may confidently expect to find new and interesting combinations among any large lot of hybrid seedlings. Indeed, it is not necessary to make new crosses in order to get interesting new types, since, as we have seen, all cultivated races of gladiolus are hybrids that carry many racial strains, and hence manifest the tendency to vary that we have seen everywhere manifested by hybrids in second and later generations. It is of course possible to exert a powerful directive influence through selecting parents for crossing, and further direction may be given by selection among the seedlings for any given color or combination of colors. So new races with unique color combinations may be very readily developed, and, even if these are not at once fixed, they can of course be propagated in- definitely by bulb multiplication. Here, as with other plants, all forms grown from the offshoots of a bulb will reproduce the qualities of the parent form, and a new race may thus be multi- plied indefinitely. Notwithstanding the great diversity of colors of the different races of gladiolus, there have 28 LUTHER BURBANK until recently been no good gladioli that could accurately be described as pure white. Where so-called white varieties have appeared, they have a dinginess that suggests the presence of an underlying color; also there are spots, stripes, or featherings of other colors, especially on the lower petals. That the hereditary factors for color are really present in these so-called white flowers is demonstrated by the fact, alreadj'' noted, that in crossing two of these we may pro- duce varieties that bear colored flowers. But the fact that these crosses of white gladi- oli produce flowers showing a great diversity of color, suggests obviously the possibility of select- ing among these offspring, in the second genera- tion, some that contain only the hereditary fac- tors for whiteness, and by rigid selection I have produced a race of white gladioli which, when further perfected, will constitute an interesting acquisition. Already these are partially fixed. Other growers of gladioli have observed the same fact. Already the white gladioli breed fairly true, and further selections, with reference to the per- fection and fixation of the type, will give us a race of white gladioli that will meet the approval of the public. But here as elsewhere there is danger that in selecting for one quality other THE GLADIOLUS 29 qualities will be neglected, so that the flowers are not kept up to their best standard. Until quite lately there has been no blue color in gladioli, any more than in the poppy, except, perhaps, submerged in combination with some of the darker colors. And for this reason, it has been found by all growers of the plants far more difficult to produce a blue flower than any other color, and until quite recently nothing approaching the really blue gladiolus had been produced. The first blue ones introduced were in reality more purple than blue. Nearly all hybrid vari- eties have shown lines of pale or smoky blue at times. The first gladiolus that could really be called blue was the one sent out from Europe under the name of Madame Hulot. This had a small flower, and in other respects resembled the older gladioli — a dark purplish blue in color. By crossing this with white varieties of large size, pale blue with extra large fine flowers were pro- duced. Two years ago one appeared, of very large size, and perfect in all respects with a true blue color. The crossing of the gladiolus presents no dif- ficulties. It is merely necessary to cover the three-parted stigmas with pollen of the desired A SAMPLE GANDAVENSIS PRIMULINUS HYBRID The gladiolus having this type of spike has departed rather strikingly from the traditions of its tribe as to arrangement of the flowers along the stem. But there is something strikingly attractive about the manner of cluster- ing adopted by this variety, and we may feel sure that this specimen will be among those selected and allowed to go to seed, in the interests of future gen- erations, (Not quite one-half life size.) THE GLADIOLUS 31 parent so thickly that bees and humming birds cannot interfere with the experiment. In working on a large scale, it is convenient to place rows of different forms that one wishes to hybridize side by side, so that pollen may be readily transferred from one row to another, in walking along the rows each forenoon when the stigmas are receptive. Also this arrangement allows the hybridizing to be carried out by the humming birds which are always aids in the fer- tilization of these tubular flowers. Here, as in most other experiments, I have found that the results of the reciprocal cross are the same; it makes no difference which parent is the staminate and which the pistillate member. So the seed from the contiguous rows of gladioli thus hybridized may be saved in a single lot. New crosses and rigid selection are giving larger flowers, brighter colors, more compact stalks, and a tendency to multiply more rapidly from the bulblets — and especially with greater freedom from disease. The propensity to revert toward the original type of the wild species — small flowers, long slender stalks, closed blooms, dull coloring, narrow leaves, and poor constitu- tion— is being subordinated as the selection is carried through successive generations. 32 LUTHER BURBANK And while there will be no metamorphosis in the essential characteristics of this beautiful and popular flower, further modifications of detail that are of no small practical significance may confidently be expected. We may well suppose that when the gladiolus was finally brought to California, having been shipped first from South Africa to Europe and then from Europe halfway round the globe, there was an added stimidus, urging the plant to still further modifications of habit, and supplying yet other elements of variation, such as form the basis of all plant development work. Vol. 7 — Bur. THE CANNA AND THE CALLA And Some Interesting Work ^vith These So-Called Lilies THE first canna of my own production is the one named the Tarrytown. This canna took the grand gold medal, at the Pan- American Exposition at Buffalo, as the best canna exhibited at that time. There were large numbers in competition. In addition to receiving the gold medal, the Tarrj^town was given special mention as being the most profusely blooming canna ever seen. This canna is a rich brilliant crimson in color, and is rather dwarf in size, standing not higher than three feet. Instead of producing a single offshoot from each stalk or at most three or foiu*, as a good many even of the better varieties of canna did at the time when this was produced, the Tarrytown grows from six to nine offshoots of the main stalk. Thus it makes a splendid and highly effective display. 33 « Vol. 7— Bur. ••* 34 LUTHER BURBANK The individual flowers are of good substance, enduring the sun well. After the blossoms fade, the petals drop to the ground. This is a special feature for which careful selection has been made, as many cannas tend to hold shriveled blossoms, thus having an untidy appearance. Origin of the New Canna The Tarrytown canna was developed from the type known as the Crozy canna, hybridized with a native species of the Florida swamps known as Canna flaccida, a plant with extremely large flowers of pure lemon yellow. The Crozy canna was a well-known horticul- tural variety, developed in somewhat recent years, which differs from the varieties that were previously in vogue in that its flowers are very large, notably attractive, and of varying colors. Until the cannas of the Crozy type were devel- oped, this plant was prized chiefly for its foliage, the flowers being rather insignificant. But the Crozy canna has large flowers, to casual inspec- tion similar to those of the gladiolus. The Florida species (C flaccida) that was used to hybridize with the Crozy, has very fine large petals, but the flowers are not lasting. But it blended well with the other type, and intro- THE CANNA AND CALLA 35 duced an element of variability that facilitated selection and development along the lines similar to those characterizing the perfected Tarrytown ; also the Burbank, Austria, and Italia, since introduced. The Crozy canna is itself a hybrid, one of the parents being a form known as Canna iridiflora, a tall plant with long, dark green leaves, and with a long drooping panicle of rich crimson flowers. I have experimented with this form, but have never known it to produce seed. The new cross that I effected between the Crozy hybrid and the native Florida species, brought together strains widely diversified. The tendency to variation was very obvious even in the first generation, as might have been expected considering that one of the parents was itself a hybrid. From the same hybrid strain it was possible to select a number of plants showing individual peculiarities that seemed worthy of perpetuation. The qualities developed in the Tarrytown have already been outlined. Another race developed simultaneously, through a different series of selections, differed very markedly, in particular as regards the character of the flower, which took on so characteristic a form, and colors of such elusive quality, as to merit the name of Orchid- A SEEDLING CANNA While the carina does not vary from seed nearly as much as the gladioli and dahlias, yet the variation is sufficient to make the growing of seedlings interesting. THE CANNA AND CALLA 37 flowered Canna. It chanced that experhnenters in Italy produced simultaneously and quite independently a race of cannas having closely similar qualities. The best of my cannas of this type was intro- duced under the name of The Burbank. This plant rather closely resembles a variety known as the "Austria," which was introduced about the same time from Europe. The Bur- bank, however, is somewhat larger, and has thicker and more rubberlike foliage; and its flower is slightly less crimson in the throat. Working with the Canna The cross-fertilization of the canna should seemingly present no particular difliculties to anyone who studies the mechanism of the flower. The stamens have a petallike appearance, and the pollen mass could not be transported by the bee or any other small insect. Large moths may carry it from one flower to another, but the usual poUenizer of the canna, in this country, is the humming bird. The hand-pollenizer may readily enough de- tach the pollen mass, and transfer it to the stigma of another flower. But it does not follow that hj^bridizing is easy. In fact, I found it exceedingly difficult, especially 38 LUTHER BURBANK when attempting to cross the ordinary canna with the Florida species ah-eady mentioned. I worked for eight years with that purpose in view before succeeding. And even then the seedlings were greatly lacking in fecundity, producing very little seed, notwithstanding the fact that cannas in general usually produce seed abun- dantly in California. The infecundity of the canna hybrids suggests that the species in question are almost at the limits of affinity. But the seeds produced, al- though few in number, were some of them fertile, and the hybrid progeny showed possibilities of development, as already suggested. Most of the later generations, however, are almost or quite sterile, refusing to seed. The chief difficulty in growing seedlings of the canna is to insure the germination of the seed. The familiar name "Indian-shot plant" by which the canna was first known suggests the character of its seeds, which are not unlike small bullets in appearance and in hardness of texture. The old plan of germinating the seeds used to be to file off part of the thick shell, in order that the seed might absorb moisture. This works very well, but can hardly be applied on a large scale. My own method has been to disinfect the canna seed with a solution of bluestone (sulphate of THE CANNA AND CALLA 39 copper), and place them in coarse gravel, taking pains to pour water through the gravel at fre- quent intervals. Under these circumstances, the seed is less likely to decay through attacks of fungous pests than if planted in the soil. In the coarse, clean, sterilized gravel, a high percentage of the seed will come up in a few months. The porosity of the gravel, giving free access to air, is also an element that is advantageous to the seed of the canna. Seed treated in this way will germinate at a relatively low temperature; but germination is facilitated if the heat is kept between sixty and seventy degrees. As soon as the seedlings appear, they are trans- planted thinly into boxes where they are allowed to stand until May, when they are planted in the field and cultivated like other crops. A large proportion of the seedlings will prove worthless. The weeding out of the first year is done readily, but selection in the second year re- quires skill, to judge as to which plants are wor- thy of preservation. Beyond that, of course, the usual process of selection through several years will be carried out along the lines of the desired modification at which the experimenter is aiming. The objects that the experimenter may advan- tageously bear in mind in developing new cannas 40 LUTHER BURBANK include hardiness, the production of a double flower, and the production of a white flower, among others? In California the canna may be left out-of- doors over winter; indeed it does much better when so treated than when the bulbs are lifted and stored. In the northeastern States, it is necessary to dig the roots and store them where they will not be subject to too low a temperature. It will be of advantage to develop the canna to a stage of hardiness that would enable it to be treated as an ordinary perennial, leaving the roots in the ground and only dividing them now and again for purposes of propagation. Still this might require more work than is worth giving to the task, inasmuch as the canna is already grown far to the north, and the work of digging and storing the bulbs is not excessive. A double canna would certainly be a novelty and one that is worth working for. The same is true of a pure white canna. By hybridizing and careful selection, it should be possible to develop this novelty, judging from analogy with other flowers. Of course it is possible to increase the size of the flower, and to produce other color variations along the line of recent developments. Most important of all, the flower should be made THE CANNA AND CALLA 41 more lasting. Just now white cannas of very good quality are appearing and every desirable quality in plant and flower are being brought forth from year to year. Improved Callas (Richardias) The manner of production of my fragrant calla was described in an earlier chapter. It will be recalled that this anomaly was pro- duced through selection from an individual found among a large company. The question of odor and its variation in flowers was further discussed in a recent chapter. There appears to be, in point of fact, as wide a range of variation among flowers in the matter of odor as in regard to color. But inasmuch as most selective experiments have been made with reference to color and quite without regard to the matter of odor, the cultivated plants have natu- rally developed along the lines of color variation, and even those that were originally fragrant have in many cases lost their perfume. Improving Essential Oils In recent years, however, much more attention has been paid to this matter. In particular, the studies of the chemistry of essential oils, with reference to the production of artificial substi- ANOTHER SEEDLING CANNA This is a very pleasing variety of carina, the characteristics of which are well revealed in the jncture opposite. The blossoms are individually of large size, and are borne in profusion. THE CANNA AND CALLA 43 tutes for the natural ones, has given clues that the plant developer is beginning to take up. I have been requested, for example, to im- prove the clove and the cinnamon, as well as the coifee plant, in the production of races having a higher percentage of the various essential oils for which they are prized. Coffee, as everyone knows, depends very largely on its aroma and fragrance, and it has been found that these may be greatly modified according to the soil in which the plant is grown. The fragrant qualities are often greatly intensi- fied when the plant is grown on volcanic soil and at a high altitude. It is known that various spices differ markedly. In the same way the quality of alkaloids, such as caffeine and quinine, may vary in the same species under different conditions of soil and climate. There is a species of coffee that is practically without caffeine ; but this has little aroma. It has been proposed to combine it with the Arabian coffee and it may be possible to produce a coffee without caffeine — which may or may not be popular. Among garden plants that are prized for their aromatic quality, the thymes vary widely in the amount and quality of their essential oils. The notable variation in the odor of the calla, which gave the scented variety, is duplicated in U LUTHER BURBANK a good many species of plants. The individuals even of the wild species vary, some of them having a really delightful fragrance, and some none at all. In crossing the different individuals, you may accentuate the perfume, add one ele- ment of fragrance to another; or, on the other hand, you may make such a combination that the two aromas seem to neutralize each other, pro- ducing an odorless hybrid. The plant developer who works with these anomalies in mind, paying heed to the fragrance of his flowers as well as to their other qualities, is almost certain to produce varieties that will be appreciated, for, as already suggested, the per- fume of the flower and the flavors of foods are nowadays receiving more attention than formerly. New Giant Callas I have introduced four main varieties of calla in addition to the calla Fragrance. This work began largely with raising seed- lings for the trade, from the form of calla known as Richardia alho-macidata, a dwarf variety with spotted leaves that was at that time very popular. The leaves of this plant bore attractive white or transparent markings on the bright green sur- face. The flower was white, with a brown tinge THE CANNA AND CALLA 45 at the base, and in the original form was insignificant. Among these almost numberless seedlings ap- peared, now and then, a golden variety, but this proved difficult to fix, although very handsome and attractive. Presently I secured another variety of calla known as the Pride of the Congo, Richardia ha- stata. This is a much stronger grower than the other variety and has pale yellowish flowers larger than those of the albo-maculata. Many seedlings were raised from this variety on my Sebastopol farm, and developed by selection until it produced very large bulbs. Then the two species were crossed, using our own hybridized golden variety of the R. albo- maculata and the highly developed varieties of R. hastata. The cross was made reciprocally as usual, and here as elsewhere it appeared to make no difference which was the staminate and which the pistillate parent. The hybrids vary considerably as to bulb, plant, and flower — much more so than either parent species when raised from an uncrossed seed. And among the hybrids there were some plants that produced enormous bulbs, sometimes eight or ten inches in diameter and weighing from two to six pounds each. The plants that 46 LUTHER BURBANK grew from these bulbs were of large size and bore blossoms that were of much brighter yellow than those of either parent. This plant was introduced under the name of the Giant Calla, a name later changed to Lemon Giant. Subsequently I obtained a number of other species of calla, including those known as R, Elliottiana, R. Pentlandi, R. melanoleuca, R. Nelsonii, and R. Rehmanni. These were all hy- bridized with one another, and with the species that previously was in hand. Among these complex hybrids were plants that were very unique in form and foliage and flower. The blossoms varied in color not only in the different hybrid plants, but sometimes an individual blossom would be partly deep purple, partly deep yellow, and in part almost white. Sometimes the colors were mottled or striped, but usually the purple color appeared in the throat of the flower. The purple is apparently a combined inherit- ance from the Elliottiana, Rehmanni and mel- anoleuca; and the hastata also has a faint touch of it. The yellow is heritage from hastata and Elliottiana. These plants varied as much in size as in qual- ity of flower. Some of them grew three and a THE CANNA AND CALLA 47 half feet in height, others only eight or ten inches. In some cases the foliage and stalks were smooth and in others actually hairy, covered with soft excrescences of thornlike appearance. Some of the hybrids were very easy to raise, but most of them quite difficult. Among the freak forms that appeared in this hybrid colony were plants bearing double and even triple flowers, and others in which the flowers and leaves were combined in the most curious manner. Of course the so-called flower of the calla is a modified leaf that has not alto- gether lost the leafiike form and manner of growth. So the reversion through which the flowers become still more leaflike in these mixed hybrids was perhaps not altogether surprising. But the particular manifestations of the tend- ency to reversion were most astonishing. Other New Varieties Among the hybrids that departed less mark- edly from the calla traditions were some that bore flowers of a splendid deep yellow, and that had all desirable qualities of easy multiplication and abundant blooming. Some of these have a purple spot low down in the throat, others are a pure yellow, not dissimi- lar in appearance to my early varieties. GIANT AND DWARF CALLAS We have developed the flower of the calla in both directions, producing both giants and dwarfs. The one at the left in this picture is perhaps the smallest calla ever developed. Its spathe measures only two inches across, while the blossom of the new giant measures eight inches across. IHWWI^' H)!ll|l