90 BRITISH LEPIDOPTERA,, 



purpose, though armed with the hard material of the pupal jaws. In 

 the Palaeo-Lepidoptera (Eriocraniides) and Trichoptera (Phryganeides) 

 the work is still done by the pupal jaws, nor do the imaginal jaws 

 work them, it having been found that they could be worked by varia- 

 tions of pressure of the parts beneath, and the imaginal jaws became 

 unnecessary. 



We have here, then, the first appearance of imagines (with mandi- 

 bulate larvae) that did not require jaws for this purpose. We have 

 already noted that there are in Hymenoptera, families (Cynipidae, &c.) 

 that needed their jaws for no other purpose, and to which they must 

 be an unnecessary appendage, and, therefore, an inconvenience after- 

 wards. It is easy, therefore, to understand that a large field for 

 further evolution was thus opened. As a matter of fact, there can be 

 little doubt that this opening gave rise to the two large orders of 

 Diptera and Lepidoptera. Up to this point the pupa is always of soft 

 and delicate texture, except as regards the jaws, and the imago is often 

 able to walk and move about freely whilst still encased in the pupal 

 skin. In the Palaeo-Lepidoptera the vermicular movements assist the 

 escape of the pupa (imago in pupal skin), and it is easy to understand 

 that greater density and roughness of the pupal surface, would much 

 assist the emergence, and would be preserved and increased, if, in any 

 case, they appeared. The Eriocraniids have been left us, however, as 

 a very isolated scrap of the Palaeo-Lepidoptera, enough to show us 

 how the interval between the Neuroptera and Lepidoptera was bridged 

 over as regards pupal evolution, but we know of no intermediate stages 

 in existence till we come to the Adelids and Nepticulids. In the 

 Nepticulids the pupal skin is still very soft and delicate, so that it 

 shrivels up a good deal when the imago emerges. Its parts are very 

 feebly soldered together and easily separated, so that one sees it would 

 hardly be impossible for the imago to walk about clothed in it ; all the 

 first six abdominal segments retain freedom of movement. It differs 

 from the Eriocraniid pupa in the loss of pupal jaws, in the gain of 

 some soldering together of parts, and of roughness of pupal surface in 

 the form of minute spines covering the dorsal surfaces of some 

 abdominal segments. In the Adelids, we find, more marked than in 

 the Nepticulids, another feature of the earlier pupae in the Neo-Lepid- 

 optera, viz., a prominent spine or beak on the pupal head. It would, 

 therefore, appear that the pupal jaws have been lost, their function 

 being efficiently replaced by a valved, or otherwise easily opening, 

 cocoon, a beak to the pupa for forcing this, and a roughened surface 

 to enable vermicular action to move the pupa out of the cocoon. 



In the Adelids, which in many respects are the nearest of the Neo- 

 to the Palaeo-Lepidoptera, we find that the delicate texture of the pupal 

 skin, still partially preserved in the Nepticulids, has given place to a 

 more densely chitinous texture. Though the wings, legs, &c, are still 

 soldered together, and to the body, in a rather flimsy manner (in the 

 xower forms), the thorax is now certainly one solid mass, and the 

 terminal abdominal segments form another, whilst the chitinous rings 

 of each are so solid that there is no difficulty in recognising the inter- 

 segmental membrane, where this remains functional, and its solidifica- 

 tion where segments are soldered together. 



In most lepidopterous pupae, and especially in the lower section 

 (Incompletae), there are, more or less obvious, the basal remnants of 



