BY R. J. TILLYARD. 655 



found in these three Orders, togetlier with tlie very definite 

 evolutionary trend towards reduction. 



(3) A second line of descent is less definitely indicated, com- 

 prising the Megaloptera and Planipennia, in which the evolu- 

 tionarv tendency runs towards early complete suppression of 

 the inacrotrichia upon the membrane of the wing, and a later 

 invasion of the macrotrichia from the main veins on to the cross- 

 veins, culminating in the Kaphidioidea on the one hand, and in 

 the higher Planipennia on the other. Both Orders may well 

 have been derived from an early type of Protomecopteron, but 

 neither of them can be derived from the other. For the oldest 

 Planipennia have the largest microtrichia, while the oldest 

 Megaloptera still retain the macrotrichia upon the membrane, a 

 €ondition which is not to be found within the Planipennia. 



(4) As the Trichoptera show a more archaic condition of the 

 macrotrichia upon the membrane than do most of the Mecoptera, 

 they can only be derived either from the very base of this latter 

 Order, or from the preceding Protomecopterous type. 



(5) In the Lepidoptera there exist certain types that show 

 portion of the archedictyon still present in an aphantoneuric 

 condition: while, in many others, the pattern of the archedictyon 

 is well preserved. Botli the oldest Trichoptera and the oldest 

 Mecoptera are in advance of these types. Hence we can only 

 trace back the Lepidoptera to tlie Protomecopterous or some 

 similar, extinct type, and may not derive them either from the 

 ^lecoptera or from the Trichoptera. 



Tliough the conclusions to be drawn from the structures 

 studied in this Part are somewhat indefinite in thejnselves, they 

 will be found to be of considerable value wlien taken in conjunc- 

 tion with the rest of the evidence. Also, in the study of the 

 Phylogenies of the families of each separate Order, (which lies 

 outside the actual scope of this paper), a great deal of vevy valu- 

 able evidence may be gathered from the Wing-trichiation. 



We may conclude this Part by giving, in taljular form, the 

 principal characters of the Wing-trichiation for the different 

 Orders — 



