BY R. J. TILL YARD. 711 



a side-branch from the older Brachycerous stem; their only 

 known fossil representatives are of Tertiary age. 



The line of ascent of the Parameeoptera leads directly to the 

 common stem of the Trichoptera and Lepidoptera, as has been 

 already shown in a previous paper (29). This common stem 

 probably divided, somewhere in the Trias, into true Trichoptera, 

 in the one hand, and Homoneurous Lepidoptera, of a Jugo- 

 frenata type, on the other. From this latter stem, the present 

 Homoneurous Lepidoptera aie directly derived. The Micro p- 

 terygidae, Eriocraniidae and Mnesarchaeidae are three very re- 

 duced and isolated remnants of the original stem, the true 

 Jugata a somewhat more successful side-branch, reaching a com- 

 paratively high stage of evolutions in the Hepialidae. The 

 Jurassic Palaeontinidae may well have been a specialised off- 

 shoot from the old Homoneurous stem, unconnected in any way 

 with present-day Heteroneurous types, seeing that the method 

 of reduction of the veins of the hindwing is not the same in 

 these fossils as it is in recent Heteroneura. The Heteroneura, 

 representing, at the present day, the great bulk of the Order 

 Lepidoptera, arose from the Homoneura as their archetypic 

 family, the Protocossidae, probably either in the Upper Jurassic 

 or in the Cretaceous; but the only known fossils definitely 

 referable to this Suborder are of Tertiary age. 



The two Neurojjteroid Orders, Megaloptera and Planipennia, 

 are venationally scarcely distinct enough to be retained separ- 

 ately, and are best regarded morphologically as two Suborders 

 of a single Order Neuroptera. The Table shows that the com- 

 mon stem of these two Orders must have possessed a high per- 

 centage of archaic characters, and must have branched off from 

 the original stem of the Complex very early. The recognition 

 marks for fossil wings of these two Orders are the complete 

 series of costal veinlets, and the pectinate form of Rs. The 

 earliest known Megaloptera are found in the Lower Trias, the 

 earliest Planipennia in the Upper Trias. But it is clear that 

 the Archetype of the Megaloptera, at any rate, is more archaic 

 than the Permian Belmontia in possessing a complete series of 

 costal veinlets, and also in its type of cross-vein system. The 

 possession of a distally forked Cu 2 marks the Megaloptera and 

 Planipennia as an offshoot from the main stem of the Complex, 

 rather than from the line of the true Mecoptera and the Paia- 

 trichoptera, which had this vein unbranched. Thus we are led 



