32 SCUDDER ON THE DEVONIAN 



the Jurassic. Examples of the latter are Gerephemera, which has a multiplicity of simple 

 parallel veins, next the costal margin of the wing, such as no other insect, ancient or 

 modern, is known to possess ; find Xenoneura, where the relationship of the internomedian 

 branches to each other and to the rest of the wing is altogether abnormal. If too, the 

 concentric ridges, formerly interpreted by me as possibly representing a stridulating organ, 

 should eventually be proved an actual part of the wing, we should have here a structure 

 which has never since been repeated even in any modified form. 



10. They shoiv a remarkahle variety of structure, iiidicating an abundance of insect life 

 at that epocli. This is the more noticeable from their belonging to a single type of forms, 

 as stated under the seventh head, where we have seen that their neuration does not 

 accord -with the commoner type of wing structure found in palaeozoic insects.-^ These 

 six wings exhibit a divei'sity of neuration quite as great as is found among the 

 hundred or more species of the carboniferous epoch ; in some, such as Platephemera, the 

 structure is very simple ; in others, like Homothetus and Xenoneura, it is somewhat 

 complicated ; some of the wings, as Platephemera and Gerephemera, are reticulated ; the 

 others possess only transverse cross veins more or less distinct and direct. No two wings 

 can be referred to the same family, unless Dyscritus belongs with Homothetus — a point 

 which cannot be determined from the great imperfection of the former. This compels us 

 to admit the strong probability of an abundant insect fauna at that epoch ; although many 

 palaeozoic localities can boast a greater diversity of insect types, if we look upon their 

 general structure as developed in after ages, not one in the world has produced wings 

 exhibiting in themselves a wider diversity of neuration ; for the neuration of the Palaeo- 

 dictyoptera is not more essentially distinct from that of the Palaeoblattariae or of the 

 ancient Termitina, than that of Platephemera or Gerephemera on the one hand is from 

 that of Homothetus or of Xenoneura on the other. Unconsciously, perhaps, we allow our 

 knowledge of existing types and their joast history to modify our appreciation of 

 distinctions between ancient forms. For while we can plainly see in the Palaeoblattariae 

 the progenitors of living insects of one order, and in other ancient types the ancestors of 

 living representatives of another order ; were we unfamiliar with the divergence of these 

 orders in modern times, we should not think of separating ordinally their ancestors of the 

 carboniferous epoch. It may easily be seen, then, how it is possible to find in these 

 devonian insects — all Neuroptera or neuropterous Palaeodictyoptera — a diversity of wing 

 structure greater than is found in the cai'boniferous representatives of the modern 

 Neuroptera, Orthoptera and Hemiptera. 



11. The devonian insects also differ remarkably from all other known types, ancient or 

 modern; and some of them ajypear to be even more complicated than their nearest living 

 allies. With the exception of Platephemera, not one of them can be referred to any 

 family of insects previously known, living or fossil ; and even Platephemera, as shown 

 above, differs strikingly from all other members of the family in which it is placed, both 

 in general neuration and in reticulation ; to a greater degree even than the most aberrant 

 genera of that family do from the normal type. This same genus is also more compli- 

 cated in wing structure than its modern allies; the reticulation of the wing in certain 



1 Cf. Mem. Bost. Soc. Nat. Hist., Ill, 19, note 1. 



