Finally, if we now mention the _history of the development of animals_, we shall have to postpone to the next section the consideration of the most essential facts furnished by this science; for the individual development of animals is a process which could speak not only for a descent of the species, but also for a descent of them through gradual development. But where, as in the present section, we treat the descent theory apart from the evolution theory, we have also to think of the possibility that the species or groups of species are not originated through gradual development, but nevertheless do originate through descent--namely, in leaps through metamorphosis of germs or a heterogenetic generation; and for such an idea we find confirmation in the {74} observation of the history of development of animals, which we call _change of generation_ or _metagenesis_.[4] By this is meant the following phenomenon: Certain animals, as the salpa and doliolum of the order of the tunicata, as well as certain mites and many tape-worms, produce offspring which are wholly dissimilar to the mother stock. These offspring have the capacity of reproducing themselves--if not by sexual means, as at the first generation, still by the formation of sprouts; and it is only the animals originated by the second generation (with many species, even those by the third) which return again to the form of the first generation. The plant-lice transmit themselves through six, seven, even ten generations by means of sprouts, until a generation appears which lays eggs. Now it is indeed true that the change of generation forms a circle in which the form of the last generation always returns to that of the first, and therefore leaves the species, as species, wholly unchanged. But it is nevertheless a process which shows that the natural law of an identity between generator and product, observed in other relations, is not without exception; and if we once have reason to suppose that the generation of new species took place in past periods of the globe, but has ceased in the present, such processes in the single period open to our direct observation--namely, the present (in which, however, according to our knowledge, the species remain constant)--are {75} nevertheless hints worthy of notice. For they refer us to ways in which in those former times, when certainly new species did originate, this formation of species might possibly have taken place.