"division of labor"? These words do not give the exact idea of evolution, such as we conceive it. Wherever there is division of labor, there is _association_ and also _convergence_ of effort. Now, the evolution we are speaking of is never achieved by means of association, but by _dissociation_; it never tends toward convergence, but toward _divergence_ of efforts. The harmony between terms that are mutually complementary in certain points is not, in our opinion, produced, in course of progress, by a reciprocal adaptation; on the contrary, it is complete only at the start. It arises from an original identity, from the fact that the evolutionary process, splaying out like a sheaf, sunders, in proportion to their simultaneous growth, terms which at first completed each other so well that they coalesced. Now, the elements into which a tendency splits up are far from possessing the same importance, or, above all, the same power to evolve. We have just distinguished three different kingdoms, if one may so express it, in the organized world. While the first comprises only microorganisms which have remained in the rudimentary state, animals and vegetables have taken their flight toward very lofty fortunes. Such, indeed, is generally the case when a tendency divides. Among the divergent developments to which it gives rise, some go on indefinitely, others come more or less quickly to the end of their tether. These latter do not issue directly from the primitive tendency, but from one of the elements into which it has divided; they are residual developments made and left behind on the way by some truly elementary tendency which continues to evolve. Now, these truly elementary tendencies, we think, bear a mark by which they may be recognized.