fiy R. J. TILLYARb. <^47 



longitudinal stria (Text-tig.31/c.). Such scales may be found in 

 ([uitc a number of isolated genera within the Order. Further 

 broadening leads to the addition of a second, third, or even a 

 fourth parallel stria. The best developed scales known to me in 

 this Order are those of the pecuhar genus Plpcfrofrirf<ufi(Text-i\g. 

 'U,/)), which always show three or four stria\ 



Certain groups of genera show a tendency to the reduction of 

 the hairiness of the wings, which is evidently of a secondary 

 nature. In the subfamily Macro neii tat iiixf, this culminates in 

 the exolution of a numbei- of forms with the wings almost devoid 

 of macrotrichia. 



In conclusion, the Archetype of this Order must evidently 

 have had a wing-trichiation closely resembling that of Ehya- 

 cojjhila, i.e., with the archedictyon absent, its macrotrichia pre- 

 sent uj^on the membrane, and already differing considerably, in 

 their smaller size, and slighter, more flexible build, from the 

 larger and more bristle-like macrotrichia of the veins, Macro- 

 trichia were absent from the cross-veins, and the microtrichia 

 were already reduced to a small size. 



Order LEPIDOPTEllA. 



(Plate Ixviii., fig.ll, and Text-figs. 21, 32-34). 



In this large Order, it will only be necessary to study fully 

 the most archaic types; in the higher families, only the general 

 trend of the evolutionary effort need be considered. 



We take, then, first of all, the three families of the Suborder 

 Homoneura, viz., the Micropterygidce, Prototheoridoi, and Hepi- 

 alidce. In all (jf these, microtrichia are present, but exceedingly 

 minute, never exceeding 2/a in length. They have been called 

 (i.ndexp, in this Order, and an attempt has been made to class 

 together all those families, in which they occur, as the "Aculeate" 

 Lepidoptera. They occur in a number of the older families of 

 the Heteroneura, as well as in the Homoneura. Hence it will 

 be seen that this method of classification is faulty, and should 

 not be persisted in. In all the higher families, microtrichia are 

 no longer present; but it is evident that their loss has been 

 effected along many lines of advance; and that an association of 



