There are, however, peculiar difficulties in those cases in which, as among the Lepidoptera, the same species is mandibulate as a larva, and suctorial as an imago. From this point of view _Campodea_ and the Collembola (_Podura_, &c.) are peculiarly interesting. There are in insects three principal types of mouth:— First, the mandibulate; Secondly, the suctorial; and Thirdly, that of _Campodea_ and the Collembola generally, in which the mandibles and maxillæ are retracted, but have some freedom of motion, and can be used for biting and chewing soft substances. This type is, in some respects, intermediate between the other two. Assuming that certain representatives of such a type were placed under conditions which made a suctorial mouth advantageous, those individuals in which the mandibles and maxillæ were best calculated to pierce or prick would be favoured by natural selection, and their power of lateral motion would tend to fall into abeyance; while, on the other hand, if masticatory jaws were an advantage, the opposite process would take place. There is yet a third possibility—namely, that during the first portion of life, the power of mastication should be an advantage, and during the second that of suction, or _vice versâ_. A certain kind of food might abound at one season and fail at another; might be suitable for the animal at one age and not at another. Now in such cases we should have two forces acting successively on each individual, and tending to modify the organization of the mouth in different directions. It cannot be denied that the innumerable variations in the mouth-parts of insects have special reference to their mode of life, and are of some advantage to the species in which they occur. Hence, no believer in natural selection can doubt the possibility of the three cases above suggested, the last of which seems to throw some light on the possible origin of species which are mandibulate in one period of life and not in another. Granting then the transition from the one condition to the other, this would no doubt take place contemporaneously with a change of skin. At such times we know that, even when there is no change in form, the softness of the organs temporarily precludes the insect from feeding for a time, as, for instance, in the case of caterpillars. If, however, any considerable change were involved, this period of fasting must be prolonged, and would lead to the existence of a third condition, that of the pupa, intermediate between the other two. Since the acquisition of wings is a more conspicuous change than any relating to the mouth, we are apt to associate with it the existence of a pupa-state: but the case of the Orthoptera (grasshoppers, &c.) is sufficient proof that the development of wings is perfectly compatible with permanent activity; the necessity for prolonged rest is in reality much more intimately connected with the change in the constitution of the mouth, although in