BY E. J. TILLYARD. 



457 



or cone of some plant, and this has also caused the transverse tear, which 

 can be seen across veins M4 and Cm in Plate lii., fig. 31, somewhat anterior to 

 this point, and" is indicated by the dark shadow. The distal halves of M4 and 

 Cui are thus shifted upwards out of their proper levels, and at the same time 



n.^lriassagrion a!istral!ense,n.g. et sp.. Restoration of wing, (jee Plate 

 lii., fig. 31.) (x 6.3). Convex veins marked -[-, concave veins — . 



it is probable that, by sliglit longitudinal rucking, the veins M3, Ms and M2 

 have got pushed together more closely, near their origins, than would be the 

 case if the wing were lying flat. The correct positions of these last three veins 

 cannot be exactly restored; but, in the case of M4 and Cu, the former being 

 convex and the latter concave, it is easy to pick up their broken courses, and 

 to restore them as in Text-fig. 77. 



Plate lii., fig. 31 shows this fossil wing with the light so arranged that the 

 main veins are well shown up; consequently, the cross-veins are not well shown, 

 being mostly at right angles to the main veins. Under a moderate power, how- 

 ever, every single cross-vein of this wing can be seen, though they are all of 

 very fine calibre. The only parts which cannot be restored vsdth absolute cer- 

 tainty are those where there has been a break or rucking. In this connection, 

 I desire to emphasize the following points: — 



(1) In the restoration, the origins and basal portions of M3, Ms and M2 

 are probably crowded a little too closely together, owing to the rucking already 

 mentioned. 



(2) It is not absolutely certain that there are only two antenodals; there 

 may be another one closer to the distal end of Se. 



(3) Cu, being a concave vein, is raised up in this reverse impression, and 

 its basal piece, within the petiole, has the actual impression of the vein removed, 

 as often along a ridge; its course, however, seems fairly well indicated, and it 

 is restored in its normal position for Zygoptera. 



(4) The posterior border of the petiole is also missing. But the anal 

 crossing, Ac, can be seen, with faint indications of the origin of lA below it. 

 The restoration is given in the only possible way in which these remnants can 

 be made to fit into the wing-scheme, but must not be taken as being absolutely 

 accurate. 



(5) The breaks along the costa have been filled in by completing the series 

 of postnodals, and by continuing the oblique sides of the pterostigma upwards 

 from the preserved posterior portion along Ri. As this vein can be seen to be 

 strongly thickened below the pterostigma, the assumption that the latter was 

 well chitinised is, I think, justified. 



