242 FOSSIL INSECT WING, ORDER PAUAMECOPTERA, 



the vein forming the arculus, we have to face the fact that it is 

 present in our new fossil, and more strongly formed than in any 

 other known type within the Panorpoid Orders, either fossil or 

 recent. Now the arculus, in the form in which we find it in this 

 fossil, is typical of the oldest existing Trichoptera and Lepi- 

 doptera. It does not occur m the Mecoptera, with the single 

 exception of the Triassic fossil Stereochoristafl). In the Diptera, 

 the arculus is present, hut of different form from that in the fossil: 

 in the Megaloptera and Planipenniaj it has not yet been shown 

 to occur at all. The conclusion, then, is obvious, that our fossil 

 must be closely related to the archaic types of existing Trichoptera 

 and Lepidoptera. 



(i.) Affinity with the Trichoptera. 

 We may take the forewing of Rhyacophila dorsalis Curtis, 

 (Text-fig. 4) as a good example of the wing of an archaic Caddis- 

 fly. Comparing this closely with the wing of Belmontia, it will 

 readily be seen that the wing of the Caddis-fly can be directly 

 derived from that of the fossil, simply by reduction of certain 

 parts. 



&» Aj*: 



hn 



Ra 



wlwu > Cu lb Cu la 



Text-fig. 4. 

 Khyacophila dorsalis Curtis, forewing, ( x 8). Scotland, recent; Order 

 Trichoptera. For lettering, see Text-fig. 2, p. 238. 



The resemblance between the subcostal veins of the two wings 

 is very striking. Both have the humeral cross- vein, the distal 

 dichotomy, and an oblique vein towards the distal end of the 

 costal space. Only in Belmontia there is an extra veinlet not 

 present in Rhyacophila, and the subcosta of the latter extends to 



