6 MODERN CLASSIFICATION OF INSECTS. 
From the equally mandibulated Hymenoptera, they are removed by 
the equal size of their wings, by their non-possession of a pungent sting, 
or multivalve ovipositor, and by their maxille and labium not uniting 
to form a tubular apparatus. From the Trichoptera, which are still 
associated with them by the continental entomologists, they are separ~ 
ated by the collar-like neck, pilose, branching-nerved wings, elongated 
cox, and obsolete trophi, which characterise the latter named order. 
This order comprises several well-known families of insects, namely, 
the beautiful dragon flies, the May flies, lace-winged flies, white ants, 
and ant-lion flies ; and derives its name from the Greek, vevpor, a 
nerve, and zrepov, a wing; in allusion to the beautiful net-work with 
which their wings are ornamented, forming a very numerous series of 
cells or areolets, far exceeding in number the cells in the wings of any 
other insect. The order is one of comparatively small extent, being 
far inferior in point of number of species to many of the other orders; 
but the characters, which distinguish even the few families of which it 
is composed, are far more discordant than those of any of the rest, 
there being scarcely a leading characteristic of the order which does 
not meet with an exception; thus, in some genera, the posterior wings 
are either larger or smaller than the anterior, sometimes, as in certain 
genera, as Cloeon, they entirely disappear ; in others, as the female 
Boreus, and one of the species of Atropos, the wings are entirely ob- 
solete ; again, in the male Boreus, they are not membranaceous but 
leathery. The structure of the mouth is very varied; in Libellula it 
assumes an anomalous appearance; and in Ephemera the jaws are en- 
tirely obsolete. The transformations are also equally varied ; indeed, 
Mr. MacLeay states that the essential character of the order is 
varied, the larvee undergoing either an incomplete (Corydalina), obtect 
(Myrmeleonina), subsemicomplete (Libellulina), or semicomplete me- 
tamorphosis (Termitina). Myrmeleon, however, most certainly has 
an incomplete pupa; whilst the transformations of the Libellulide and 
Ephemeridee, which (according to Latreille, as quoted by MacLeay), 
compose the stirps Libellulina, although peculiar and very unlike’each 
other, appear to me to enter into the semicomplete species of trans- 
formation. It isto be regretted that Mr. MacLeay has given no defi- 
nition of the term subsemicomplete, which he proposed for this stirps. 
As to the transformations of the Panorpide, it will be seen, from my 
account of that family, that the theories which Mr. MacLeay enter- 
tained respecting it (Hore Ent. p. 433.) are completely fallacious. 
