1926] Tilly ard’s Work on Insect Phylogeny 93 
lepidopterous and dipterous types. An interesting backward 
extension of the series afforded by the discovery of true Mecop- 
tera in the Kansan Lower Permian places the origin of holome- 
taboly back at least half a geological period. The Upper Car- 
boniferous Metropator , a fossil from the earliest horizon yet known 
to furnish insect remains, is now believed to be a true Mecopteron, 
leading to the possibility that complete metamorphosis may 
have evolved even earlier and that the ancestors of the present 
Panorpoid Holometabola may have been not merely Mecopte- 
roid but actually Mecopterous. 
Perhaps the most interesting link in the phylogeny of the 
Neuroptera is afforded by the Lower Permian Protomerope, in 
which the strong series of costal veinlets, the form of Sc and the 
abundant branching of Rs and of M lead at once to the condition 
exhibited by primitive Neuroptera. 
The Australian fossils threw no light on the relationships of 
the Hymenoptera. There were indications that this order was 
related to those composing the “Panorpoid complex/’ but no 
definite venational types from which the hymenopterous con- 
dition could be derived. It remained for the Kansan Lower 
Permian to supply more definite information as to the origin of 
this order. In the beautifully preserved fossils for which has 
been founded the new order, Protohymenoptera Tillyard, the 
venation and texture are distinctly Hymenopteroid and yet 
show, especially in the number and position of the cross-veins, 
some evidence of Mecopteroid relationships. The divergences 
are, however, greater than the resemblances and the Proto- 
hymenoptera, with their supposed derivatives the Hymenoptera, 
are believed to have sprung from another stem than that which 
gave rise to all the other Holometabola. If, as the ecturer 
suggested, the hitherto enigmatical Sycopteron symmetricum 
Bolton from Commentry, France, belong also to the Protohy- 
menoptera, then the origin of the Hymenoptera must be put 
back to the Upper Carboniferous. 
The acceptance of Protohymen and its relatives as near to or 
identical with the ancestors of the Hymenoptera leads inevitably 
to the replacement of the complicated MacGi livray (1906) in- 
terpretation of the wing-venation in this order by a much 
