1921] Crampton: Sclerites of the Head 93 
the mouthparts of the two groups. The character of the structures 
bearing the label “pg” in Fig. 77 of a Hymenopteron is somewhat 
suggestive of the condition found in the Neuroptera (Fig. 81), and 
the tendency for the cardo, “‘car,’”’ of Fig. 77 to straighten out in line 
with the stipes, “‘sti,’”’ cccurring in the Hymenoptera is also found in 
certain Neuroptera. On the other hand, the tendency to lose the lacinia, 
and the character of the galea, ‘‘ga,’’ of the Hymenopteron shown in 
Fig. 77 are features suggestive of the Mecoptera, such as the one shown 
in Fig. 84. In fact, the Mecoptera are related to both Neuroptera and 
Hymenoptera, and have apparently inherited tendencies from both 
sources. 
In many respects, the Hymenoptera are fully as primitive as the 
Neuroptera, and in certain features are even more primitive than the 
Neuroptera (e. g., nature of the genitalia, retention of cerci, etc.), and 
I am inclined to seek for the types ancestral to the Hymenoptera among 
the forms related to the Isoptera (with the Psocida and Zcraptera) 
on the one hand, and to the Coleoptera (with the Dermaptera) on the 
other. The mouthparts, however, do not furnish as instructive evidence 
as might be desired,. since even in such primitive Hymenoptera as the 
one shown in Fig. 80, the mouthparts are quite highly modified. Even 
the larval mouthparts are very disappointing in this respect in the 
Hymenoptera, since they do not bear a striking resemblance to the 
mouthparts of any of the larve here figured, although one can detect 
a slight suggestion of affinities with the Coleoptera, on the cne side, 
and with the Lepidoptera on the other. The Mecopteron shcwn in 
Fig. 19 is also disappointingly unlike any of the other insects figured, 
though it has a few features suggestive of affinities with the Trichoptera. 
It certainly is much more specialized than the primitive Neuroptera 
studied, and unless Dr. Tillyard can find some Mecoptercn larva of 
a far more primitive character, he is not justified in assuming that the 
Mecopterous type is more primitive than the Neuropterous one (taking 
the group as a whole). 
The evidence offered by the head structures is in full accord with the 
former contention (Crampton, 1920) that the Coleoptera are anatcm- 
ically intermediate between the Dermaptera on the one side and the 
Neuroptera on the other, and this may have some bearing upon the 
question of the origin of the Neuroptera. Of the two groups (Coleoptera 
and Neuroptera) the Coleoptera are clearly the more primitive, with the 
exception of the feature of the highly modified fore wings. Other very 
lowly organized insects such as certain Blattide have fore wings quite 
as highly modified as the most primitive Coleoptera, however, and the 
fore wings of the Dermaptera are even more highly modified than those 
of the primitive Coleoptera, so that this feature is of no great importance 
in determining the relative primitiveness of a group of insects as a 
whole. The occurrence of the peculiar sclerite, ‘“‘pst,’”’ found only in 
the maxillz of Coleoptera (Fig. 4) and Dermaptera (Fig. 7), in addition 
to other features of resemblance in the’ two groups (e. g., the nature of 
the antennal segments, the segmentation of the cerci in certain immature 
Coleoptera and Dermaptera, the character of the thoracic terga, atc.) 
