A few years after the appearance of Fritz Müller's work _Für Darwin_, Haeckel elaborated Müller's idea, and applied it in a much more comprehensive manner. He formulated it under the name of 'the fundamental biogenetic law,' and then he used this 'law' to deduce from the ontogeny of animals, and more particularly of Man, the paths of evolution along which our modern species have passed in the course of the earth's history. In doing so the greatest caution was necessary, since ontogeny is not an actual unaltered recapitulation of the phylogeny, but an 'abridged' and in most cases--in my own belief, in all cases--_a greatly modified recapitulation_. Therefore we cannot simply accept each ontogenetic stage as an ancestral stage, but must take into consideration all the facts supplied to us by other departments of biological inquiry which afford help in the decision of such questions, especially those brought to light by comparative morphology and by the whole range of comparative embryology. Haeckel was quite well aware of this difficulty, and repeatedly emphasized it by laying stress on the fact that a 'blurring' of the phyletic stages of development had arisen through the abridgement of the phylogeny in the ontogeny, and a 'falsification' of it through the secondary adaptation of individual ontogenetic stages to new conditions of life. He therefore distinguished between 'Palingenesis,' that is, simple though abridged repetition of the ancestral history, and 'Cœnogenesis,' that is, modification of the racial history by later adaptation of a few or many stages to new conditions of life. As an example of cœnogenetic modification, I may cite the pupæ of butterflies. Since these can neither feed nor move from one spot, they can at no time have been mature forms, and cannot, therefore, represent independent ancestors of our modern Lepidoptera; they have originated through the constantly increasing difference between the structure of the caterpillar and that of the moth or butterfly. Originally, that is, among the oldest flying insects, the mature animal could be gradually prepared within the larva as it grew, so that finally nothing was necessary but a single moult to set free the wings, which had in the meantime been growing underneath the skin, and to allow the perfect insect to emerge, complete in all its parts. This is the case even now with the grasshoppers and crickets. In these forms the larval mode of life differs very little, if at all, from that of the perfect insect, and the main difference between the two is the absence of wings in the larva. But when the perfect insect adapted itself to conditions of life quite different from the larval conditions, as was the case with the nectar-sucking bees and butterflies adapted entirely for flight, while the larvæ were still adapted exclusively to an abundant diet of leaves and other parts of plants, and to a very inactive life upon plants, the