336 THE BIOLOGY OF INSECTS 



Lepidoptera, mandibles are absent or vestigial in the adult. 

 It is interesting to notice that the caddis-flies (Trichoptera) 

 which have no mandibles when adult, retain these jaws in the 

 pupa which has to bite its way out of the protective '' house " 

 formed by the larva and contracted and closed before 

 pupation. In the great majority of the Lepidoptera the 

 pupa is without mandibles, but in those primitive Micro- 

 pterygidae mentioned above, the pupa has relatively large 

 mandibles, and these jaws are recognisable in a reduced 

 form in the imago. The conditions in these orders suggest, 

 therefore, a development of the more highly organised 

 Lepidoptera from an ancestral stock some characteristics 

 of which are preserved in the Mecoptera, Neuroptera, and 

 Trichoptera. 



That wing-structure is an important guide to the student 

 who endeavours to trace out the relationships between 

 various groups of insects was apprehended by the naturalists 

 who gave more than a century ago to most of the insect 

 orders names suggested by the nature of their wings (Coleo- 

 ptera, *' sheath- winged " ; Leipdo-ptera, *' scale- winged," 

 etc.). We have seen how, in the course of a winged insect's 

 development, air- tubes grow into its wing- rudiments, and 

 prefigure the course of the longitudinal series of supporting 

 wing-nervures — sub-costal, radial, median, cubital, and 

 anal. The work of various students, notably of J. H. 

 Comstock (1918) and R. J. Tillyard (1919), in correlating 

 these wing-nervures in the various orders of insects has 

 emphasised the importance of wing-neuration as a guide to 

 the relationship. In this way a true kinship has been shown 

 to exist between the great majority of the metamorphic 

 orders — the Neuroptera, Mecoptera, Trichoptera, Lepido- 

 ptera, and Diptera — an assemblage which together with 

 certain extinct orders has been distinguished by Tillyard 

 (1918-20) as the '' Panorpoid Complex." 



Some evidence as to the course of evolution of the insect 

 orders afforded by the study of fossils may now be con- 

 sidered. The remains of the oldest insects known to us 

 are of Carboniferous (late Palaeozoic) Age ; a number of 



