simple as possible, because the hybridization is carried on on such a large scale. First the stamens of the flowers to be crossed have to be removed. This is usually done while the flower is in bud and the stamens close together. One circular cut only is sufficient. Care must, of course, be exercised so as not to hurt the pistil. Next protection against insects has to be provided for, as otherwise pollen might be transferred from other flowers and the expected results spoiled. In scientific experiments a great deal of attention is paid to this, and the flowers are carefully enclosed in cases of metal gauze or in especially prepared paper bags, so that no insects can reach them. In practical plant breeding this would, however, be too cumbersome. By the circular cut mentioned not only are the stamens cut through, but the corolla is also removed, and the flowers are consequently not so conspicuous and do not attract the insects, except where there is fragrance. The majority of Burbank's improved fruit trees belong to the first category. In practical work the visit of a single insect is not so much feared, because all the mischief it may do in bringing the pollen is to produce a valueless hybrid. This can later be destroyed. Besides, the insect may come too late to bring about any result. But there is also a possibility that a new and good hybrid may be produced. The application of any cover is, therefore, entirely out of the question. This is the reason why unexpected results of such practical work are never entirely free from the suspicion that they are due to accidental introduction of pollen. Such results, therefore, do not enable one to draw reliable scientific conclusions.
Burbank's method is to collect the pollen required for these crossings on watch-glasses, as it keeps fresh for about a week. With these glasses he goes to the plants he wants to pollinate and applies with his finger tip a little of the pollen on the stigma. This is, as a rule, not yet ripe, but the pollen adheres to it until it matures. Fecundation thus begins at the time the stigma becomes glutinous, which lessens the possibility of other pollen being introduced.
I wish now to consider one of the most remarkable features of Burbank's work, the immense scale upon which it is conducted. This is the best plan for obtaining the most variations in a short time. He starts thousands of seedlings for each hybrid, and when the culture admits and the interest requires it, this number is increased to 50,000 or 60,000. In order to give an idea of the significance of these figures and of the work they imply, Burbank shows in one of his catalogues an autodafé of hybrid raspberries and blackberries. For the purpose of getting a hybrid with larger berries and bigger bunches he cultivated 65,000 seedlings until they blossomed and were in full bearing. A few dozens were selected, and the balance, heavily loaded with fruit, were dug up and gathered in a pile, which was then reduced to ashes. And
