112 FOSSIL INSECTS OF THE BRITISH COAL MEASURES. 



the presence of a wide interval between the former and the radius, and the wide 

 area occupied by the latter. 



The division of the radius into two unequal branches, with its symmetrical 

 double bifurcation, is also, I venture to think, a feature of primary importance. 



In the wide divergence of the radius and median, this species agrees with 

 Scudder's o-enus Spiloblattina ; but in this case the veins do not converge again to 

 enclose an elongated or oval area, 



Archimylacris, sp. indet. Plate VII, fig. 5. 



1911. Archimylacris, sp. indet., Bolton, Quart. Journ. Geol. Soc, vol. lxvii, p. 163, pi. x, fig. 3. 



Type. — A fragmentary wing, lacking the apex and the base ; Museum of 

 Practical Geology, Jermyn Street (no. 2-1-503). 



Horizon and Locality. — -Base of Upper Coal Measures (Gellideg Level of the 

 Mynyddislwyn Vein) ; near Maes-y-cwminer, Monmouthshire. 



Description. — The specimen is much too fragmentary for any attempt at specific 

 determination. A portion of the distal margin is present, with three branches of 

 the subcosta, These are succeeded by eight straight veins, two at least forking, 

 which belong to the radius and median series. These are followed by an equal 

 number of veins partially hidden by the surface of the integument, which is mueh 

 wrinkled over this area. These, I assume, are parts of the cubitus vein. The 

 wrinkling of the integument over the cubital area is very marked. 



The special interest of this wing-fragment is in its association with portions t)\' 

 a leaf of Cordaites. Scudder and others have commented on the general association 

 of the wings of Blattoids with leaves of Cordaites, but have not. so far as I am 

 aware, drawn attention to a feature which is well shown by this leaf, namely, pits 

 on its surface. I have repeatedly found such depressions on the leaves of Cordaites 

 in the Lancashire Coalfield, and in many cases seen the hollows occupied by the 

 shells of Spirorbis pusillus. The shallow pits on the Cordaites-leaf associated with 

 this fragmentary wing show faint traces of a spiral, similar to the impression 

 of Spirorbis, and such shells were, I think, once attached to it. While the 

 Carboniferous Blattoids may have been wholly phytophagous, we are led by our 

 knowledge of the living Blattids, especially by the common cockroach (Periplaneta 

 orientalis), to assume the contrary, and to regard them as more likely to have been 

 omnivorous, in which case the association of Blattoid remains with the leaves of 

 Cordaites bearing the sedentary Spirorbis is easily understood. The association 

 also lends support to the belief that the Blattoids were semi-aquatic in habit, or 

 lived in marshes and swamps in which decayed vegetation formed a home for 

 Spirorbis. 



