BY R. J. TILLYARD. 633 



they are found to be quite free from these hairs. Hence they 

 show no evidence whatever of liaving been developed ])y straight- 

 ening out of any portion of the archedictyon. If they had been 

 so developed, it is clear that they nuist have carried macrotrichia, 

 which they would receive from the archedictyon itself. 



It will be seen, from the fossil Ao'chipanorpa, that the macro- 

 trichia occurred originally only upon the main veins and their 

 branches, including the true veinlets, and upon the archedictyon. 

 Cross-veins and membrane of the wing are alike free from them. 

 When, however, the archedictyon is abscjrbed into the membrane 

 (as it is on the point of being, in this fossil), it is evident that 

 the macrotrichia that originally occurred upon the archedictyon 

 nuist then become seated upon the membrane. This will explain 

 their appearance upon the membrane in many Orders of Holo- 

 metabola. But, if this explanation be correct, then the arrange- 

 ment of the macrotrichia might be expected to show some signs 

 of the original design of the archedictyon. If we draw the posi- 

 tions of a number of macrotrichia in any of the older Holometa- 

 bola, we shall see that this is the case. For this, it will only be 

 necessary to refer again to Text-fig. 2 2, in which it will be seen 

 that the macrotrichia below the main vein there figured tend to 

 lie along oblique lines similar to the courses of the venules running 

 out from the main veins in the fossil wing shown in Text-fig. 19; 

 and, like them, are connected by shorter cross-branches, indicated 

 by the presence of macrotrichia lying in intermediate positions. 

 (See also Text-figs. 21, 26, 30, 34). 



Further evidence as to how the macrotrichia first appeared 

 upon the membrane of the wing may be obtained by a careful 

 study of the pterostigmatic region in the Planipennia. Text-fig. 

 20 shows this region in Micromm^ tasmanicn Walk. ' The series 

 of costal veinlets, that fills the costal space in all Planipennia, 

 originally extended into the pterostigmatic region. Here there 

 is a gradual thickening of the membrane taking place, with a 

 correspondingly gradual absorption of the veinlets; all stages of 

 this process can still be seen going on by studying various species 

 in the families Hemerohiidoi, Chrysojndce, and Manfispidce. As 



