546 THE PAXORPOJD COMPLEX, in., 



(Text-fig. 50) the reduction of the tracheation has only gone so 

 far as to eliminate M and the anal tracheae . In this Order, so far 

 as known, the alar trunk is complete, but the middle portion of 

 it is of very small calibre. The eosto-radial group of tracheae 

 remains well developed except for the loss of M, and there is in 

 most cases a definite costal trachea (C) present, running close up 

 to the anterior border of the wing. 



Much more specialised by the reduction of their pupal wing- 

 tracheation, and, through this character, therefore, not ancestral 

 to any other existing Order, are the Mecoptera and Trichoptera. 

 In the former Order, the anal trunk is incomplete; the costo- 

 radial group sends only one trachea into the wing, viz., R; and 

 the cubito-anal group likewise only sends one, viz., M. If this 

 latter trachea is the original M, then it would suggest that there 

 must once have been a time when the alar trunk was complete, 

 so that M might have migrated along it to the cubito-anal group, 

 with which it is now connected. On the other hand, the basal 

 portion of the cubital vein is much weakened in this Order; and 

 it may be that the trachea grows out into the wing after the 

 venational scheme is laid down. Hence, if the natural course 

 along Cu is closed up, the trachea might be expected to penetrate 

 the media instead. In whatever light we look at it, the condition 

 of the pupal tracheation in this Order must be regarded as highly 

 specialised . 



The most specialised Order of all, as regards the condition of 

 its pupal tracheation, is the Trichoptera. The alar trunk is in- 

 complete, but, even- in so old a group as the Rhyacophilidae, the 

 costo-radial group sends only one trachea into the wing, usually 

 along R, but sometimes apparently quite aimlessly, as if it were 

 content to penetrate anywhere within the sheath; and the cubito- 

 anal group is greatly reduced, sending only one weak trachea 

 into the wing either along Cu or 1A, or aimlessly into the sheath, 

 without any relationship to the venation. 



Owing to the great reduction of the tracheation in these two 

 last Orders, we cannot be sure that they may not have passed 

 through a stage of development, in times past, in which the alar 

 trunk was complete, though weakly formed. All we can assert 

 with confidence is that their present condition is a very highly 

 specialised one, and indicates a line of evolution, for this charac- 

 ter, quite different from that of the three Orders with holo- 

 tracheate wings. 



