of judgment, in which he excels all his contemporaries. The best proof of this is to be found in the great success his creations have made, not only in North America, but also in Europe.
His methods of work are the same as those followed by plant breeders in Europe. Secrets he has none, and if he is not willing to demonstrate his cultures to everybody, this must be attributed to the fact that his time is too valuable. There is no fear that any one could 'steal his trade' by merely looking at it. Every one is left free to follow in his path, but without the special disposition for it nobody will succeed, and for simple imitation the entire process is too complicated.
To give an idea of the immensity of his cultures, it is sufficient to cite one instance. When selecting a new kind of blackberry he picked out the best from 60,000 specimens, all in full bearing, dug up the rest and burned them. This is his way of working, not only with one kind of fruit or flower, but with all. The most remarkable trait, however, of his work is that he experiments with as many forms as possible. This method is carried to the highest degree of perfection, and thereby his results are so stupendous that they receive the admiration of the whole world.
However large may be the number of forms subjected to crossing and selecting, this method is in itself limited. Burbank's products are all, with a few exceptions, reproduced not from seed, but by vegetative propagation. Grafts or cuttings, bulbs, shoots or division of roots are the means of multiplication. It is well known that vegetative propagation results in much greater stability than raising from seeds, which often produces degenerate types. Because of this fact, Burbank hardly ever experiments on annual or biennial plants, but confines himself to perennials.
In Burbank's methods selection plays the most important part. To accomplish a good selection, however, the greatest possible degree of variation is a prerequisite. This variation is attained mainly through selection of the starting points and through artificial hybridization. The results are next cultivated on a large scale under environmental conditions which will develop as many differences as possible.
Varieties coming from separate localities differ not only in regard to external characteristics, but their capacity of modification varies considerably, and can often be ascertained only in the special environments of an experimental garden. The greater this power of adaptation the more chances for the experimenter.
As a general rule, it holds true that the results of crossing depend primarily on the selection of varieties used for that purpose. These indicate, so to say, the program, the list of possibilities from which the choice and the combinations have later to be made. Outside of this
