Part ae ps Se ee a 
Aes se Se 
THE METAMORPHOSIS OF FLIES. 719 
entirely new histological form-elements arise ; the masses of nuclei 
which fill the body-cavity of the pupa with a compact mass, whose 
` brood of cells, multiplying through endogenesis, arrange them- 
selves into strings and thus form the primitive germs of the tra- 
chew and in fact the muscles. 
Indeed it is difficult to tell out of what material the muscles of the 
wings of the Muscidæ should arise, if not out of the descendants of 
the masses of nuclei. But in Corethra we find the corresponding 
muscles already indicated in the embryo in the form of fine fila- 
ments; but how could this be the case in the Muscidae, where the 
points of attachment of such filaments, the hypodermis of the 
thorax, are not yet present in the larva? And the same conclusion 
follows in regard to the tracheal system of the imago, which in 
Corethra is indicated beforehand, the entire division of the body 
of the imago answering to the corresponding segments of the 
larva— which in Musca can only arise when these regions are 
formed, i. e., in the course of the pupal life. 
Thus in every relation a much less morphological connection ex- 
ists between larva and imago in the Muscids, than in the Tipulids, 
and in this fact is to be sought the reason for the relation of the 
duration of larval life to the long pupal stage.* 
In Sarcophaga it requires eighteen days from the pupation to the 
exclusion of the fly, in Corethra only three. However, on the other 
nd the life of the larva of Sarcophaga lasts only eight days, 
in Corethra three or more weeks, and this difference cannot alone 
be referred to the wholly opposite mode of nourishment of the two 
æ, which makes it possible in the Musca larva to suffer a great 
mass of food to pass in a few days through its intestine, while the 
Corethra living by robbery only slowly procures food. The evi- 
dence lies in the very long interval, which separates the last moult- 
_ Ing of the larva from the pupation in Corethra. It is surely not 
Unimportant or adventitious, but will be met with in cases where 
the origination and completion of the appendages of the imago 
must take place within this period. 
The pupa of Corethra manifestly differs in morphological . as 
aune ana ee 
_* Evidently it is only the relative, not the absolute duration of the pupa-stage, which 
onside: i i 
vary in one and the same species between wide limits; 
