370 THE TRACHEAL SYSTEM OF THE JMAGO. 
The exact relative share taken by the hypodermis and the 
bulbous enlargement of the trachea in the formation of the 
second spiracle is not known, but analogy renders it probable 
that the vestibule originates from the hypodermis, as well as 
the stigmatic plate. 
Anterior Spiracle.-—The development of the anterior spiracle 
in the larva is not easily studied; but if it is developed in the 
same way as that of the nymph, which is most probable, the 
process must be very nearly as follows. 
The spiracular tracheal trunk is probably developed in the 
usual way, by the proliferation of the peritoneal coat of a 
lateral cutaneous branch of the main trunk, which first appears 
as a solid bud; and the hypodermis on the outer surface of 
this bud is then invaginated. The finger-like processes are 
developed within the invagination. Subsequently the invagi- 
nation is turned inside out and forms the larval stigmatic 
cornu. Hence there is no spiracular sac, and the outer surface 
of the cornu apparently represents it. 
The question as to whether the stigmatic cornua of the 
dipterous larvee and nymphs should be regarded as gills or 
as spiracles, has been the subject of some difference of opinion. 
Weismann considered the anterior spiracular apparatus of 
Corethra to be intermediate between the digitate spiracular 
cornua of the larva in Musca and the tracheal gills of some 
Tipulidae, and Palmén goes even further [153] and says: ‘The 
prothoracic horns of Corethra are identical with those of 
Culex; these are not stigmata but true tracheal gills, which are 
shed with the exuviz, and take no part in the formation of the 
prothoracic spiracles of the imago.’ Whether these respiratory 
appendages are closed as Weismann and Palmén believed, and 
as those of the Blow-fly apparently are in the adult larva, or 
whether they are open and functionally spiracles, as Hurst has 
recently shown them to be in Culex [155]; it is certain that they 
are shed and are not converted into the spiracles of the imago. 
This fact is of interest in relation to Palmén’s discovery 
that the spiracles of the imago in the Ephemeride are de- 
veloped independently of the tracheal gills, and are not, as was 
