how the amount of this reconstruction is greatest where the divergence between the larval and perfect stages is most marked. Whatever differences of opinion may prevail on points of detail, the general explanation of insect metamorphosis as the result of divergent evolution in the two active stages of the life-story must assuredly be accepted. No other explanation accords with the increasing degree of divergence to be observed as we pass from the lower to the higher insect orders. The successive incidents of the life-story of most insects are largely connected with the acquisition of wings. Wings, and the power of flight wherewith they endow their possessors, are evidently beneficial to the race in giving power of extending the range during the breeding period and thus ensuring a wide distribution of the eggs. In no case are wings fully developed until the closing stage of the insect's life, they are always acquired after hatching or birth. We have already noticed (p. 40) how Sharp (1899) has laid stress on the essential difference between the exopterygote and endopterygote insects, the wing-rudiments of the former growing outwards throughout life while those of the latter remain hidden until the pupal instar. Sharp considers that there is some difficulty in bridging, in thought, the gap between these two methods of wing-growth, and has put forward an ingenious suggestion to meet it (1902). Reference has already been made to insects of various orders in which one sex is wingless, the Vapourer Moth (p. 96) for example, or all the individuals of both sexes are wingless, as the aberrant cockroaches mentioned in Chapter II (p. 15), or certain generations of virgin females are wingless, for example aphids (pp. 18-19) and gall-flies (pp. 94-5). Insects may thus become secondarily wingless, that is to say be manifestly the offspring of winged parents, and such wingless forms may on the other hand give rise to offspring or descendants with well-developed wings. Frequently, as in the case of the aphids, many wingless generations intervene between two winged generations. A striking illustration of this fact is afforded by an aquatic bug, _Velia currens_, commonly to be seen skating over the surface of running water. The adults of Velia are nearly always wingless, but now and then the naturalist meets with a specimen provided with functional wings, the possession of which enables the insect to make its way to a fresh stream. Moreover there are whole orders of parasitic insects, such as the lice and fleas, which, showing clear affinity to orders of winged insects, are believed to be secondarily wingless. These orders are designated by Sharp 'Anapterygota.' And from the analogy of the periodic loss and recovery of wings in various generations of the same species, he has concluded that the gap between the exopterygote and the endopterygote method of development may have been bridged by an anapterygote condition; that the ancestors of those insects with