? 
o4 Bulletin of the Brooklyn Entomological Society. Vol. XIV 
marked tendency toward the development of metamorphosis and 
some which do not, one of the derived groups may exhibit a 
marked tendency toward the development of metamorphosis, 
while the other either does not exhibit this tendency or shows 
but slight indications of such a tendency. 
In order to illustrate the above cited principle by concrete ex- 
amples, it is necessary first to clearly understand the grouping 
of the higher insects on the basis of comparative morphology, 
and to bear in mind the types of insects composing the ancestral 
group from which these higher forms are descended. The in- 
sects comprising the ancestral group, which approximates as 
nearly as any the types of insects giving rise to the higher forms, 
have been placed in a superorder called the plecopteroid super- 
order, composed of the Plecoptera, Embiidee (s. 1.), Dermaptera, 
Coleoptera and their allies. The coleopterous representatives of 
this ancestral group have developed a well-marked metamorpho- 
sis, while the embiid and dermapterous representatives of the 
group have not, although the Plecoptera themselves, which .are 
among the most primitive members of the superorder, exhibit 
a tendency for the immature stages to differ in form from the 
adults (a condition also apparent in such primitive insects as the 
Odonata, etc.). The higher insects were developed from ances- 
tral types which I think would have been placed in the above- 
mentioned ancestral superorder, containing some forms which 
exhibit a well marked metamorphosis, and some which do not. 
These higher forms, which are extremely closely related, may be 
divided into a holometabolous neuropteroid superorder and a 
“non-metabolous ” psocoid superorder. The psocoid superorder 
(but few of whose members exhibit traces of metamorphosis) 
is composed of the Psocide (s. 1.), the Mallophaga, the Ano- 
pleura (or Pediculide, s. 1.), the Hemiptera, the Homoptera, and 
their allies. (The Thysanoptera might be included in this group, 
as is possibly the case with the Strepsiptera, but I have as yet 
been unable to definitely determine this point.) The neurop- 
teroid superorder (which seems to be a holometabolous one) 
comprises the Neuroptera, Hymenoptera, Mecoptera, Diptera, 
Siphonaptera (Pulicide, s. 1.), Trichoptera, Lepidoptera and 
their allies. We thus have two derived superorders (one of 
