76 FOSSIL INSECTS OF THE BRITISH COAL MEASURES. 



1883) somewhat vaguely placed it, "with a strong degree of probability, in the 

 same general group as some other Palaeozoic wings." This reference can only be 

 to the Palaeodictyoptera. Later he published a restoration of the wing, and classed 

 it with the Protophasmidas. Brongniart at one time regarded it as a Dictyoneuron, 

 and later as a " Neurorthopterou " of the group Sthenaropterida. Still later 

 he placed it as a Neuropteron in the group Platypteridge. Handlirsch considers 

 that the costa is marginal, and that this character, with the sharp inward curve 

 of the anal veins, justifies its inclusion in the Pakeodictyoptera, although he does 

 not attempt to indicate its allies. 



A second species was described by Brongniart as A. lucasi (' Bull. Soc. Amis 

 Sci. Nat. Rouen' [3], aim. xxi, p. 60, 1885), but this throws no light on the 

 genus, nor does it appear to be generically referable to Ardttfopfiln*. 



The only details preserved which can be used in the determination of relation- 

 ship are the spiny outer margin, the great width of the principal veins, the 

 well-developed cross-nervures, and the strongly curved and numerous anal veins. 

 Eveu these are too fragmentary for safe conclusions to be drawn in the absence 

 of other material. 



The general structure of the wing-fragment is to me more suggestive of the 

 Protodonata than of the Palaaodictyoptera, but the presence of well-marked anal 

 veins discounts this view, unless we are prepared to accept the specimen as an 

 early and archaic prototype of the Protodonata. 



Order MIXOTERMITOIDEA, Handlirsch. 



L906. Handlirsch, Proc. U.S. National Museum, vol. xxix, p. 695, and Die Fossilen Insekten, p. 126. 

 1919. Handlirsch, Revision der Paliiozoischen Insekten, p. 26. 



Subcosta much shortened ; radial sector arising close to the base, with 2 — 3 

 branches, of which only one forks. Median long, four-branched and suggestive of 

 the Pakeodictyoptera. Cubitus with 2 — 3 inward branches. Anal vein 3, simple. 

 Cross-nervures strong, wide-spread and regular. 



This is a provisional order established by Handlirsch to include two forms 

 only — Mixotermes luganensis, Sterzel, from the Coal Measures of Saxony, and 

 Geroneura wilsoni, Matthew, from the Carboniferous of St. John, New Brunswick, 

 North America. Both wings show clearly their Pakeodictyopteroid ancestry, but 

 Handlirsch is uncertain whether they should be brought near to the Protorthoptera 

 or to the Perlidae. 



Genus GERONEURA, Matthew. 

 1889. Geroneura, Matthew, Trans. Roy. Soc. Canada, vol. vi. sect, iv, p. 57. 



Generic Characters. — Wing three times as long as wide; apex obtusely rounded; 



