Notes on the derivation of winged insects through several lines etc. 271 
In this connection it may be stated that the Apteryguta are no 
more to be regarded as degenerate Pterygota, than Amphioxus is to 
be regarded as a vertebrate; and the fact that certain highly specialized 
(or even degenerate) features may be found in any otherwise very 
primitive organism throughout the animal kingdom, should surely be 
known to those who claim that all apterygote insects are degenerate 
forms. Furthermore, the remarkable similarities of structure in the 
more primitive members of the apterygote and pterygote insects are 
too profound and far-reaching to be explained merely as the result of 
a parallelism, or a convergence of development in the two groups. 
From the foregoing facts, it would seem reasonable to regard 
tke more primitive members of the Apterygota as the nearest living 
representatives of the ancestors of the Pterygota — that is to say, they 
have change the least from the original ancestral condition — and 
a comparative study of the insects belonging to the various groups 
included in the lines of descent described above, will furnish valuable 
clues as to the derivation of Pterygota from wingless ancestors. 
It has been already pointed out that there are several separate 
and distinet lines of development leading from the apterygote forms to 
the winged insects. On this account, there is no force to the objection 
raised to the Dicelluro-Dermaptera line of descent on the ground that 
the more primitive types of fore wings, the segmented caudal appen- 
dages, etc., of other winged insects, could not be derived from the 
hightly specialized fore wings, forceps-like caudal appendages, etc., of 
the earwigs; for if there are several lines of descent from the Aptery- 
gota to the Pterygota, it is self-evident that all other winged insects 
were not derived from the Japyx-like forms through the earwigs, as 
this objection would imply. 
The Japyx-like forms were themselves derived from ancestors 
having segmented caudal appendages instead of forceps — an ancestral 
condition which has been retained in such Dicellura as Projapyx. On 
this account the occurrence of segmented caudal appendages in the 
immature stages of certain Dermaptera (such as Dyscritina) is to be 
regarded as a case of „atavism“, and the same may be said of similar 
structures in the Coleoptera larvae. It is merely to be expected that 
immature forms would retain certain primitive characters, even though 
these were lost in the adult condition, since this state of affairs occurs 
everywhere in animal kingdom. Instead of being an argument against 
the Dicelluro-Dermaptera line of descent, the fact of the presence of 
segmented caudal appendages in certain immature Dermaptera would 
therefore seem rather to be an argument in favor of it. 
Allowing for differences of adaptation in the two groups, it would 
be an easy matter to derive the trophi and thoracie structures of the 
Dicellura and Dermaptera from a common ancestral type, and the 
marked similarity between the caudal appendages of the two groups, 
is too profound to be laid to coincidence. The logical inference then, 
is that they sprang from a common ancestry. 
Closely allied to the Dicelluro-Dermaptera line, is the Dicelluro- 
Coleoptera line, leading from the Japyx-like forms to the Coleoptera. 
The similarity in structure between certain of the Dicellura and the 
laryae of such beetles as Cucujus and Pyrochroa is so striking, that it 
