A review, however, of the facts of hypermetamorphism, particularly the life-history of Mantispa, throws much light on the subject, since it is very probable that the supernumerary stages and marked changes of form characterizing them are due to changes of environment, of habits, and of food, causes which have exerted such a profound influence on organic beings throughout all time. Besides these, as the result of changes in the environment and nature of the food, we have the results brought about by the use or disuse of structures brought into existence by the action of stimuli from without, the class of insects abounding in examples of temporary structures which perform a certain function, and then disappear.
Again, if the origin of a hypermetamorphosis can thus be explained, it follows that normal metamorphosis is most probably due to changes of habitat, of seasons, of food, and to accelerated growth resulting from the approach of sexual maturity.
The following facts and conclusions appear to be well established:—
1. The apterous insects (Synaptera) are ametabolous, only the winged insects undergoing a metamorphosis.
2. The complete metamorphosis was not inherited from the primitive ancestor of all insects, but acquired at a later period (F. Müller). The eruciform type is a secondary, adaptive form, derived from the earlier, campodeoid type of larva.
3. The earliest, most primitive pterygote insects passed through only a slight metamorphosis. In other words, as soon as the wings were evolved and insects became adapted to live or take refuge in a new medium, the air, at the approach of the period of adult life, with the ripening or perfection of the reproductive organs, a metamorphosis began to take place, and the number of species greatly multiplied. On the other hand, the Arachnida and Myriopoda, in which as a rule there is no metamorphosis, being confined to a creeping life, with no change of medium, remained poor in number of species.
4. At first the nymphs mainly differed from the adults in lacking wings, though having the same habits; in holometabolous insects, the larva became adapted to entirely different habits and environments, so that in Hymenoptera, and especially Diptera, the larva became remarkably unlike the imago.
5. Until the Mesozoic age, or late in the Carboniferous period, there were, so far as we now know, only ametabolous and heterometabolous insects, and these orders (Orthoptera, Dermaptera, Hemiptera, Plectoptera, Odonata, and Neuroptera) were not numerically rich in genera and species, while since early Mesozoic times geological extinction has reduced their numbers.
6. During the Mesozoic age, and since then, the number of species, genera, families, and orders has greatly increased, and insects have become more and more holometabolous. The orders of Coleoptera, Lepidoptera, Hymenoptera, and Diptera are many fold greater in number of species and variety of form than the heterometabolous orders.
The rapid increase in the number and variety of types of insects evidently is correlated with the profound geological changes which took place at the end of the Paleozoic age, involving the appearance of larger continental masses, or a greater land area, thus opening new regions for settlement. Also the origin of flowering plants at about this time undoubtedly had much to do with the genesis of new adaptive structures, such as the changes in the mouth-parts and wings.