In order to do so it is necessary to explain to you as briefly as possible the present state of biological science and the modern way of considering the problems relating to species.
Modern botany, abandoning the ancient methods which depend more on metaphysics and speculation than on experiments, has given up the idea of discovering the origin of species by the prevalent method of comparison and reasoning. The separation of forms, of varieties and of species, as it is made by systematists, the herbarium specialists, is based on judgment; it depends essentially on the degree of intuition of the botanist who compares and draws conclusions. I do not mean to say here that the methods of this science are conjectural, but I may be permitted to say that it is only an outline of a science, that it is provisional knowledge, a first attempt at classification. More precise methods are necessary in order to resolve serious biological questions. The best representatives of contemporary biological science are much less hurried than their predecessors; they have acquired the conviction that there is no short cut to truth. The scientific highway is paved with difficulties. In this explanation, then, I shall not touch upon the evolutionary speculations of Darwin or others, but shall give my time exclusively to exact data.
Contemporary biology accepts the constancy of types as a well-established fact. It has discovered that this constancy is experimentally demonstrable if the following facts, not known to Darwin and his followers, be taken into account.
Every species in its natural state, and often even in cultivation, includes a large number of forms which were formerly considered variations, but which, analyzed by modern methods, appear to be constant types, all of which taken together form the Linnean species. In order to discover these small constant species which ordinarily live mixed together, it is necessary to segregate them. Vilmorin had already recognized that unequivocal results could not be obtained in the study of variation if one starts with an isolated plant or even with a single seed. A single grain of wheat may be the ancestor of innumerable generations. If these isolated grains, carefully catalogued, be sown separately, it is seen that they give birth to constant races or lines which are called pure, because they are without mixture. To evaluate these lines and differentiate them from other lines, we must not consider the isolated individual, but rather note the character of the descendants as a whole by means of experimental pure cultures. The individuals of the same race, of the same line, may differ very much according to their age, nutrition, position during the embryonic or ontogenic development, but their descendants taken as a whole are identical. In a pure race, the dwarfs as well as the giants give birth
