the law discovered by Mendel, and with this as a solid basis and the brilliant methods of Mendel for example, the further domestication of the species of this fruit ought to go forward in leaps and bounds.
Selection, continued through successive generations, so important in the improvement of field and garden plants, can play but small part in the domestication of the grape. The period between planting and fruiting is so long that progress would be slow indeed were this method
relied upon. Moreover, selection, as a method in breeding, is possible only when plants are bred pure, and it is the experience of grape breeders that in pure breeding this fruit loses in vigor and productiveness and that the variations are exceedingly slight and unstable. Many pure-bred grapes have been raised on the grounds of the New York Agricultural Experiment Station under the eyes of the writer, of which very, very few have surpassed the parent or have shown promise for the practise of selection.
From present knowledge it does not appear probable that new characters are produced in plants by hybridizing. New varieties so origi-
