23 



aborted. The result is that, owing to this change of habit and environ- 

 ment from those of its active ancestors, it changes its form, and the 

 fully-grown larva becomes cylindrical, with small slender legs, and 

 owing to the partial disuse of its jaws acquires a small round head. 

 Other modifications afterwards take place, and with the result, as it 

 appears, that a change of habit and an abundance of food close at 

 hand have brought about the change of a campodeoid into an eruciform 

 type of larva. In the meloid and stylopoid larvae, as has been noticed, 

 the first stage is campodeoid, the eruciform larva becoming apparently 

 developed subsequently, owing to its changed conditions and as a 

 result of their parasitism. In the final stages of these larvae, as also in 

 certain hymenopterous and dipterous forms, the larvae become vermi- 

 form. 



It was at one time considered that the periodical moulting of larvae 

 was due to the fact that the skin had become, as it were, too small for 

 the rapidly growing insect, and that a larger skin was necessary for it. 

 In spite of the rapidity of growth, however, the skin is really an 

 elastic membrane when shed, its elasticity, however, limited by the 

 hardness of the chitinous deposit. The chitin by which the skin is 

 hardened is now looked upon as an excretory substance, and the 

 phenomenon of moulting is considered by many to be merely a means 

 of getting rid of 'the accumulated waste materials between two succes- 

 sive moults. During the process of moulting a fluid circulates 

 between the old and new skins, and it is this that enables the larva, as 

 it were, to slip off the old skin, much as a finger is pulled from a 

 glove. Not only is the integument shed with its hairs, setae, and 

 other armatures, but also the linings of such internal organs as had an 

 original ectodermal origin ; thus the larva sheds portions of the lining 

 of the trachea and the alimentary canal. In the apodous larvae 

 of Hymenoptera that are reared in cells by nurse-bees the delicate 

 skin breaks away in shreds, portions of the old skin remaining adhe- 

 rent to the new skin long after the greater part of it has been got rid 

 of. The new skin appears to be formed by the secretion of a structure- 

 less chitinous layer from the hypodermic cells, the formation of the 

 new layer drawing off the nutrition from the old skin, so that it dries 

 considerably ; the secretion of a surface fluid on the new skin helps 

 more completely to separate them. 



The " pupa " is a term w r hich, as we have suggested, should only 

 be applied to the inactive stage of holometabolous insects. The 

 typical lepidopterous pupa has the appendages more or less closely 

 folded about the body and soldered to the integument. Such a pupa 

 was called by Linne a " pupa obtecta." On the other hand, the pupae 

 of certain Lepidoptera, such as Eriocrania, Neptiatla, Cochlidon, 

 Atitlirocera, etc., have the appendages free from the body ; they are 

 known as "pupae liberae," or "pupae incompletae," dependent on the 

 amount of freedom exhibited. The pupae of Neuroptera, Mecop- 

 tera, and Trichoptera are in this manner " pupae liberae." When the 

 pupa is enclosed in the old larval skin, as in certain dipterous pupae, 



