72 S. IT. SCTJDDER ON PALAEOZOIC COCKROACHES. 



while in Etobl. anaglyptica it is only simply forked ; the division of the externomedian 

 vein commences farther from the base in Etobl. oenusta, and the apical portion of the inter- 

 ne-median vein does not have a subarborescent distribution as in Etobl. anaglyptica. In the 

 form of the wings and the distribution of the scapular branches, it shows a certain likeness 

 to Etobl. mantidioides, but its much greater size and more extensive mediastinal area read- 

 ily separate it from that species. The points in which it differs from the only other Amer- 

 ican species of the genus will be found mentioned under Etobl. Lesquereuxii. 



A single specimen only has been found, which lacks the base and tip, and has a ragged 

 inner margin. It is doubtful whether the dotted line in PI. 6, tig. 12, by which I have in- 

 dicated the supposed outline of the base, is correct, for the curves of all the veins would 

 seem to indicate that too much is represented as lost ; but as this would represent an anal 

 area of unusual brevity, I have only indicated in the measurements given above the possi- 

 bility of an error in my delineation. 



It comes from Frog Bayou, Arkansas, and was obtained by Prof. Leo Lesquereux in black 

 carbonaceous shale, with broken fragments of plants, overlying the thin seam of coal be- 

 tween the millstone grit and the subcarboniferous limestones ; and if the period of deposi- 

 tion of the millstone grit was the same in the eastern and western coal deposits (see the 

 introduction), is the oldest, as it was the first discovered of the American fossil cockroaches. 



Etoblattina mantidioides. PL 3, fig. 8. (See also the figure on the opposite p;ige.) 



Blatta sp. Kirkby, Geol. Mag., iv, 380, pi. IT, fig. G, 7. 

 Blattidlum mantidioides Gold., Faun, saraep. foss., ii, 20. 



The single known specimen of this species is composed of only the basal half or more of a 

 front wing, so that it is impossible to give its shape with any certainty, or to be sure of the 

 limit between the scapular and externomedian areas. The costal margin is regular and 

 rather gently arcuate, and the inner margin, beyond the basal curve, is straight. This is 

 inaccurately given in our plate as curved like the costal margin, and the terminal portion 

 has therefore been incorrectly restored; it is far more probable that the shape of the wing 

 was much as in Etobl. carbonaria, and a corrected figure is therefore given on the opposite 

 page. The veins originate above the middle of the base of the wing, and have a gentle 

 basal curve. The mediastinal vein is very short, almost reaching the end of the frag- 

 ment, which certainly does not represent more than one-half of the costal border ; it runs 

 parallel to and not distant from this border, the area occupying less than a fourth of the 

 breadth of the wing; it emits five or six simple or forked, parallel, oblique branches. 

 Between the mediastinal and internomedian areas, near to the base of the wing, are three 

 veins; whether the middle of these belongs to the upper or lower — I. e., the scapular or 

 externomedian — cannot be told from the broken specimen ; it appears, however, to branch 

 from the scapular, and this I deem to be the most probable relation of this vein, although 

 it is otherwise indicated on the plate by the mark at the border; for on first study its 

 approximation, toward the end of the fragment, to the internomedian vein seemed to 

 render this its probable relation, and to make me then conclude that its basal union with 

 the scapular was onl} r apparent; subsequent study, especially in comparison with the species 

 to which it appears most nearly allied, has since made the opposite appear the truth; and 

 while one cannot be certain of one's ground with so imperfect a fragment, the comparison 

 of this fragment with the more perfect relics of the species near which I have now placed 

 it will convince any one, I think, that its true affinities are here, and that the middle main 



