Luther Burbank: Methods and Discoveries (1914) 7:239-257

The Camassia—Will It Supplant The Potato?
And Other Tubers Of Value For Food

A Bed of Camassias
The camassia is, as a glance at this picture will show, a plant having no small measure of interest as a producer of ornamental flowers. But it has also an edible bulb; and Mr. Burbank is endeavoring to improve this bulb in size and quality. The measure of success already attained is shown in succeeding pictures.

FOR the most part plants are cultivated for a single quality. If a plant produces beautiful flowers, we do not usually demand that it shall also produce valuable fruit. We do not ask that a plant which produces a valuable fruit like the tomato shall also produce tubers like the potato. It is only by accident rather than by special design or selective breeding on the part of man that a certain number of plants, notably members of the rose family, produce beautiful blossoms and delicious fruits as well.

The apple-tree in full bloom is indeed a beautiful object, but the apple would probably be raised quite as generally as it is if its blossoms were altogether unattractive. The Japanese, to be sure, have developed the blossoms of their fruits, but in so doing they have usually neglected the quality of the fruit itself.

And as to garden vegetables, about the only member of the clan that is cultivated for its flower as well as for its edible product is the Pink Chive that I have recently developed.

There exists a tribe of plants, however, of which we have hitherto made no mention, that possesses qualities of flower-bearing of a high order, combined with the capacity to produce roots of such quality of edibility as to suggest competition with our best tuber bearers, including the potato itself.

These plants are certain wild members of the lily family that have no colloquial name except that given them by the Indians; a name that has been variously transcribed as Quamash and Camass. From this name the botanist has developed the generic title Camassia. The not altogether unappropriate name of wild hyacinth is sometimes given the species that grows in the Eastern United States.

But it will be most convenient in speaking of the tribe to adopt the generic name of Camassia, in lieu of a better.

Individual Camassia Blossoms
It will be seen that the camassia has a really attractive flower. This is an unusual qualification for a plant that it making bid for a place in the vegetable garden. There are several species of camassia available for purposes of the plant developer.

Hybrid Camassia
The picture shows color variation in one of Mr. Burbank's hybrid camassias. Mr. Burbank intends to retain and accentuate the good flowering qualities of the camassia, even while giving especial attention to its bulb. This makes the process of selection somewhat complicated, but it presents no insuperable difficulties, as the reader who has followed Mr. Burbank's experiments in other fields is aware.

The Wild Camassia
This is one of the wild camassias that furnish material for Mr. Burbank's hybridizing experiments. A comparison with earlier pictures will suggest the measure of improvement that has already been attained in connection with the flower. Succeeding pictures will show the response that has been made by the bulb.

The various species of camassia grow wild in rich moist meadows and along small streams. All the species are hardy. The leaves of the plant are usually lance-shaped, about three-quarters of an inch in width, and of length varying according to the fertility of the soil, usually from eight to sixteen inches. The flower stalk in ordinary soil varies with the different species from eighteen inches to nearly four feet in height.

The flowers are usually purple, blue, or white. In some of the new hybrid species the color has changed to rose, and in others it inclines toward crimson.

All the camassias are bulbous, of course, like other members of the lily family. But there is a great difference in the size of the bulbs among the different wild species, and, as will appear presently, there is enormous variation when the different species are hybridized.


My experiments on a large scale with the Camassia have been carried out for more than twenty years, and have included work with five species.

So far as I am aware, no one had undertaken to improve any of these until my experiments were instituted, about 1890.

My first work was done with a species known as Camassia Leichtlinii, which grows abundantly on Vancouver Island. Considered as a flowering plant this is the finest of the native varieties. It grows almost altogether in crevasses of rocks, but it produces very attractive large, deep purple flowers, with wide petals. First the attempt was made to improve the flower, and I introduced a good many years ago a modified variety of the species that was somewhat dwarfed as to leaf and stem but in which the flowers had been much enlarged, the petals broadened, and the color changed to a dark blue.

As my experiments continued, however, my interest in the camassia increased, and I began to give attention to the bulb of the plant as well as to the flower.

I began working with another species, the Camassia Cusickii, which has relatively large bulbs; and with another of the well-known nature species, Camassia esculenta, the bulbs of which are much smaller but of recognized edible quality.

Most of my work in hybridizing and selective breeding has been done with the three species just named, but I have also raised somewhat extensively two other species, known as C. Howellii and C. Fraseri, as well as a great number of wild varieties of all the different species from British America, Washington, Oregon, California, and Nevada.

From the outset individual plants were selected of each species and variety that were the best I could obtain. Here, as so often elsewhere, I was enabled to produce considerable improvement merely by selecting individual plants that showed the most desirable qualities of flower and bulb, destroying the inferior ones. From the outset careful attention was paid both to the flowers and to the bulbs, as I desired to produce plants that would be ornaments in the flower garden and at the same time would grow enormous bulbs that would make them valuable acquisitions to the vegetable garden.

Having secured the best representatives of each species and variety by selection, I began an extensive series of hybridizing experiments.

I found it a relatively simple matter to hybridize the different camassias. All the species seemed to combine quite readily.

The characteristics shown by the hybrids are those that experience with other plants led one to expect. In the first generation, there is relative fixity, and the greater or less dominance of one parent or the other. In the second generation, the hybrids break up into numerous forms, varying widely as to color of leaves, height of stalk, and size of flowers, as well as in form and size and quality of bulbs.

Some of these hybrids of the second generation produced bulbs smaller than those of their progenitors.

But others grew bulbs of enormous size. Even to one who is accustomed to observe the striking variations that are produced through hybridization, it was surprising to see the extraordinary impetus given to the bulbs of the camassia by the union of different species.

The bulbs of the common edible species, C. esculenta, are relatively insignificant, usually growing about one-half to three-quarters of an inch in diameter. The C. Cusickii produces the largest bulb of all, but it is large only in a relative sense, being usually little over an inch in diameter and two inches in length.

But among the second generation hybrids were some that produced bulbs three and a half inches across and four or even five inches in length.

The difference was about that between an English walnut and a large turnip.

In viewing these gigantic bulbs, sprung thus from dwarf ancestors, one was reminded of the gigantic hybrid walnut trees that came of the union of Persian and California walnuts; of the mammoth Phenomenal Berry; of the Giant Amaryllis; and of sundry other hybrids that were stimulated to excessive growth of stem or fruit or flower by the union of parents of just the right degree of affinity.


Meantime I had taken pains to cross dark flowers with dark flowers, and white ones with white, and pink with pink, wherever possible, so as to intensify the various types.

As already noted, there is a pronounced tendency to variation even among the wild species, all the camassias sometimes producing pale greenish, almost white, flowers. These, if grown from seed and carefully selected, can be changed to snowy white. Some of the variations secured bear flowers that are truly white, while others that are called white are really of a pale greenish hue. The seedlings of these greenish white ones tend ordinarily to produce blossoms that revert to the pale blue color of the species from which they were derived.

So the production of a truly white camassia required continued selection—a process of gradual intensification.

But of course hybridizing facilitated this process. It also gave opportunity for selection with regard to flowers having broad petals—narrowness of petal being one of the original defects of the camassia as a flower. Moreover, a number of extra petals have been added in some cases, and it is only a matter of time until double camassias will be produced.

All along the line, then, the flowers of the camassias have been improved by selecting from among the best of the hybrids. Twenty thousand bulbs have been under observation at the same time, and improvement has been rapid.

A Contrast in Bulbs
Here is one of Mr. Burbank's improved camassia bulbs, contrasted with the bulb of its wild progenitors. It will be seen that the hybrid has attained relatively gigantic size, while holding pretty closely to the general type of its parents. Of course attention is given constantly to the edible qualities of the bulb, no less than to size.

In the end, the camassia will prove to be an ornamental plant of distinct value, highly prized for its flowers.

But it will be prized also for its bulb, which, in the developed and selected hybrids, is assuming satisfactory proportions, as already pointed out, and which has undoubted food value, surpassing the potato even, both as to nutriment and flavor. And of course the work of development in this direction is only at its beginning. The results already attained justify the expectation that the bulbs of the developed camassias will be of really notable size, constituting a garden vegetable of very exceptional food value.

The wild camassias generally produce but few offsets. But some of the hybrid ones not only produce numerous offsets, but tend to divide like the garlic, sometimes making five or six enormous bulbs in a season. Of course this habit has been carefully encouraged among the seedlings, as this rapid multiplication will be of obvious importance when the camassias are grown either for bulbs or for flowers.

I have also successfully hybridized some of the camassias with certain of their relatives, the squills (genus Scilla), of which I have imported many species from South America. The two tribes hybridize readily. The hybrids showed conspicuous changes in the bulb. The outside covering of the bulb of the squill is whitish, while that of the camassia is usually darker. The hybrids showed more compact bulbs of a lighter color than those of their maternal parent, the camassia.

But there are all gradations in the bulbs as to color and other qualities.

I have worked very extensively with the squills, but with reference solely to the development of the flowers, with results that will be outlined in another connection. Here I refer to them only as suggesting that these plants may be of value in introducing new qualities into the strains of hybrid camassias, stimulating further variation, and thus giving opportunity for betterment both of bulb and flower.

It is too soon to predict just what place these improved camassias may take in the vegetable garden. But the experiments have progressed far enough to show that the species has hitherto unrecognized possibilities.

Meantime a plant that is almost equally attractive from the standpoint of florist and market gardener is an anomaly that must make wide appeal to the horticulturist.

A Wide Range of Variation
Here are camassia bulbs small, large, and medium; thick, thin, and spindle shaped. Obviously there is plenty of material for selective breeding. Where such variation occurs, still greater variation will be shown in some of the crossbred descendants. Moreover the flavor of one, the form of another, and the size of a third may be combined in a single descendant; and all the good qualities may then be accentuated through selective line breeding. The camassia promises to furnish a very valuable addition to the rather small list of edible bulbs.

There are twenty or more species of plants belonging to the lily family wild along the Pacific Coast that make up a group which the botanist classifies under the generic name Brodiaea.

There are allied plants in South America, regarding the precise classification of which there is some difference of opinion. But for the purpose of the horticulturist the entire group may be ranked under the name of Brodiaea. The plants have not been extensively cultivated until recently, and they have received no popular name.

The different species vary greatly in form, size, and arrangement of the flower. The color of the flower is usually either blue or rose or purple, though sometimes white. There is also a crimsonflowered climbing species, known as Brodiaea volubilis, which somewhat rarely becomes white.


I have crossed this climbing species with the species known as Brodiaea capitata, and with various others. Some of these crosses produce most beautiful flowers intermediate between the parents. Unfortunately the best hybrids were destroyed by gophers before I had opportunity to save the seed. The interest of the brodiaeas in the present connection hinges on the fact that the plants have bulbs or corms that when cooked are very acceptable as food. Several of the species, especially the Brodiaea lactea, are relished by the Indians, and are often dug and eaten by children. The bulbs of some species contain a very high percentage of starch, probably greater than that of the potato.

I have worked on the Brodiaea lactea to increase the size of the bulbs. When growing wild the bulbs are only about half an inch in diameter. By selective breeding, varieties have been originated that will produce bulbs two inches or more in diameter. The plants can be grown almost as thickly as lawn grass, and it is probable that the yield per acre of the bulbs could be made to equal a good crop of potatoes.

In developing the brodiaea for this purpose, it would be well to search carefully for bulbs that grow to unusual size in the wild state—there is considerable variation in this regard.

The brodiaea is well worth cultivating for its flowers alone, and it would appear that the plant offers possibilities of combining flower-production with the production of a valuable food. Unfortunately, however, there is a complementary relation between the seed and the bulbs, and in order to secure bulbs of the largest size, it is necessary to remove the seed stalk before blossoming time.

Whether cultivated for flowers or for bulbs, the brodiaeas are very interesting plants that give great promise of improvement under the hands of some careful experimenter.

It is a little difficult to cross them. I have produced many hybrids, however, and it is said that occasional hybrids are found where two wild species are growing in the same neighborhood. They all bear seed abundantly, though it takes three, four, or even five years from seed before they bloom.

Some Almost-Edibles
Here we have a group of Watsonia and lily bulbs, some of which are approaching the stage of edibility. Lily bulbs for the most part are provided with a bitter principle, to protect them from insects and various subterranean marauders; but some species are much less bitter than others, and it would doubtless be possible to get rid of the bitter principle in time, by careful selective breeding; and an edible bulb might result. There are numberless undeveloped possibilities of improvement among these lines among the bulbous plants.

They grow by thousands on each square yard of ground, appearing almost as thick as grass on a well-kept lawn.

In the same species there is a good deal of variation in the form and size of the flower. On the heights of the Sierras, the Brodiaea lactea grows only a few inches high, whereas in the valleys it grows to a height of eighteen inches or two feet.

Along the alluvial creek banks Brodiaea Laxa grows very large and tall, with handsome clusters, while on the mountain sides it is dwarfed.

Even plants of the same species in the same locality vary widely as to size of flower.

Brodiaea capitata grows abundantly along the roadsides, and especially in grain fields. It blooms and produces seed before the grain is cut. Brodiaea terrestris has a stem so short that the flowers almost rest on the ground. The blossom is just the color of a blue violet, and the clusters may be mistaken for violets at a little distance.

In other localities the Brodiaea terrestris bears flowers some of which have a white stripe. Sometimes half the blossom may be white, the other part deep blue. Sometimes five or six blossoms will be blue, and a single one white. In other cases the proportions are reversed.

I have not observed any in the wild state that could be called pure white, but by cultivation and selection pure white varieties have been produced.

I have worked extensively on the Brodiaea capitata, the species just mentioned as growing in the wheat fields. On a poor dry soil this plant grows about two feet in height, and on long, straight, slender, wiry stems. But on good soil, especially in the wheat fields, it sometimes grows to the height of three or four feet, or even more, bearing a much larger cluster of blossoms.

Blossoms of the Brodiaea
The brodiaea is another wild plant Mr. Barbank has brought into the garden, and which he is educating with an eye to both its blossoms and its bulb. Some of the wild brodiaeas are edible, and Mr. Burbank expects to improve on their qualities through hybridization and selection. Some rather notable first steps in this direction have already been taken. It will be seen that the brodiaea has an interesting, even if not a very spectacular flower.

In looking over a field of brodiaeas of this species, one may expect to find one in ten thousand, or perhaps one in twenty thousand that is almost white. Seedlings raised from these produce a variety of flowers, white, pale or dark blue, and striped; with a constant tendency to revert to the blue when first taken under cultivation.

By selection and re-selection I have produced strains which invariably come white, and by the same process have produced varieties with flowers twice as large as the ordinary, also making the flower-head larger, and the plant a much more rapid multiplier from the bulbs.

From all this it will appear that the brodiaea is a very interesting plant. As already suggested, it well deserves the attention of some careful experimenter, who might develop certain strains for flowers and others for bulbs. Plants that are of interest both to the lover of flowers and to the vegetable gardener have exceptional claims on the plant developer.


There are two allied tribes of plants known as Bloomeria and Brevoortia, respectively, that are closely related to the brodiaea, and each of which is of interest .

The brevoortia is usually called the Floral Firecracker, from its green, crimson, and yellow flowers. I have grown these plants extensively from seed, to produce new varieties, but the experiments were carried out only to the extent of increasing the yellow and crimson colors.

I have grown the Bloomeria aurea extensively, and have made minor improvements in it through selection. The plant has a bulb like the brodiaea, growing deep in the earth in dry, sandy places. In the wild state the stalks vary in height, and there is also a slight variation in the color of the flower. So there is opportunity for selective breeding. Moreover, I judge from physiological characteristics that the plant should cross readily with the brodiaea, although I have not attempted to make the cross.

It is almost certain that improved varieties might be obtained by hybridization.

There is a bulbous plant called Alstroemeria, that is botanically related to the Amaryllis rather than to the true lilies, which offers possibilities of tuber improvement. The plants are natives of Western South America. I grew seedlings and hybrids by the ten thousand for several years, and became convinced that if the roots and tops could be taught to grow in a more compact form this would become a very popular flower, and perhaps also a very valuable food plant, as the roots are sometimes eaten and are quite palatable and nutritious. I have worked on the species known as A. Chilensis, A. pulchella, and A. Brasiliense, and subsequently on a large number of new species from Chile.

The Chilean Alstromeria
Yet another plant that combines the qualities of beautg of flower with those of a possibly edible root. In this case, as will be seen, the wild flower has notable claims to beauty. Under the hand of the plant developer, it will probably become an exceptionally handsome flower; and the bulb also gives good promise of proving adaptable.

Another Type of Alstromeria
There are various races of alstromeria, and Mr. Burbank is experimenting with as many of them as he can secure. He will follow his usual method of hybridizing, followed by selective line breeding. He is rather sanguine as to the outcome in the case of the alstromeria. The experiments, however, have not yet reached a definite stage.

A great variety of colors and combinations occur in the hybrid forms that may be fixed by selection. I am endeavoring to obtain a more hardy strain with improved flowers and more compact growth.

At one time I crossed plants of this genus with the California lily (Lilium pardalinum) and had several hybrids, but the root and the bulb did not make a good combination. The plants bloomed one year, then died. The hybrid blossom was smaller than that of the lily, and it resembled that of both parents in being speckled and in its combination of colors. The hybrids that blossomed produced no seed.

The long, slender, white tubers of the Alstroemeria Chilensis are edible. This plant grows in very dry soil, and is peculiarly adapted to some of the California soils and climates. It is at present too tender for growth in the Eastern United States, but it is possible that through hybridization and selection it may be rendered hardy, and in that event this may become a valuable garden vegetable.


The lilies and their allies are not the only wild plants with bulbs or roots that are edible and susceptible of improvement.

On the contrary there are several plants of different families that offer noteworthy possibilities in this direction.

There are, for example, tuberous varieties of the genus Carum, relatives of the caraway, that grow on the Pacific Coast, especially toward the Northwest, the roots of which are relished by the Indians.

One species in particular, called the epau potato, is dug in great quantities in the fall and stored for winter use. The roots are small, almost like those of the Ranunculus (cowslip, etc.), and are similar in form to the roots of the dahlia, though much smaller. They have a sweet, aromatic, and pleasant flavor. In different localities they vary a good deal in size and quality. There are places where the plant grows almost like grass, so that hardly a shovelful of dirt can be turned over without exposing numerous roots.

When brought under cultivation, the epau potato appears susceptible to the influences of its new surroundings. The roots increase greatly in size and in succulence.

I have gathered the seeds and roots of this plant, and have from time to time had seeds sent me from many localities, during the past fifteen years. The best seeds came from Idaho. Plants grown from seed sent from Idaho developed into herbs four feet in height, producing roots three to four times as large as most of the California Carums, and seeding quite as abundantly.

I have been able by selecting individual roots to improve the species known as Carum gairdneri quite rapidly. I have observed that when the blossoms are removed, so that no plant energy is required for the production of seeds, the roots are much larger. This is an interesting compensatory effect that illustrates the close correlation between the different parts of a plant, and in particular the reliance of the roots on the leaf system.

There are, as already stated, several species and numerous varieties of the plant that could be used for hybridizing purposes, and doubtless the tendency to variation could thus be accentuated.

A very large number of plants can be grown on a small piece of ground, and if the roots could be developed even to one fourth the size of those of the carrot, this would prove a very valuable addition to the list of garden plants. The roots are not only nutritious, but they have exquisite flavor even when raw; and they are improved by cooking.

I think the plant very well worthy of improvement and general cultivation.

The lilies and their allies are not the only wild plants with bulbs or roots that are edible and susceptible of improvementhere are many noteworthy possibilities in this direction.