BY R. J. TILLYARD. 399 



the distal endings of some of the tracheje were different from those 

 recorded above; e.g.^ tlie trachea in R.j of the forewing passed 

 down into R4 via a cross-vein. The median trachea of the fore- 

 wing also gave out a short branch to the cubital fork. 



Only two pupje were available for study. There can be little 

 doubt that further minor variations would have been found had 

 a longer series of puppe been examined. 



There is only one conclusion to be come to from this result. 

 It is, that the Mecoptera, as exemplified by the archaic genus 

 Cho7'ista — and, therefore, presumably, by other existing genera — 

 are highly specialised as an Order along the same lines that we 

 find in the Hymenoptera, Trichoptera, and Diptera, viz., by the 

 reduction of their wing-tracheation. The cause of this reduction 

 is almost certainly the same in all four Orders, viz., that the 

 trachese do not grow into the wing-rudiment until the latter has 

 been fully formed, with the venational scheme completely laid 

 down. In such a case, there is no longer any need for a trachea- 

 tional scheme, to guide in the laying down of the venational 

 scheme originally based upon it; the only necessity is to supply 

 the wing with oxygen. Consequently, there will be a tendency 

 to reduce the tracheal supply to the minimum necessary for this 

 purpose; and, also, for the tracheae to cease to follow their 

 original paths, and to take instead the path of least resistance. 

 Both these tendencies are well illustrated by the case here 

 studied. 



The same tendency, with many stages still preserved to us, has 

 been noted already in the Order Plectoptera; an Order which, 

 curiously enough, approaches most closely to the Holometabola, 

 in that it possesses, in its sub-imaginal stage, what appears to 

 have been Nature's first attempt to evolve a true resting-stage or 

 pupa. The pupa of the archaic Mecoptera closely resembles the 

 imago in everything except tlie form of its mouth-parts and the 

 non- expansion of its wings. Thus it only differs from the sub- 

 imago of the May-flies in being unable to fly, and in slightly less 

 mobility of its free appendages. In the Order Plectoptera, there 

 are a number of genera known in which the remaining wing- 



