100 MOSQUITOES OF NORTH AMERICA 
“The pupa does not eat. It breathes air through the apertures at the ends 
of its siphons. It floats, thorax uppermost, by virtue of a large air cavity lying 
under the hinder part of the thorax and the anterior part of the abdomen. This 
cavity is bounded in front by the legs, at the sides by the wings, and below by the 
mouth-parts. It extends up at each side of the first segment of the abdomen, 
where it is covered by the halteres, and into this part of the cavity at each side 
opens a large stigma, held open by the fairly well-developed cuticular lining 
(‘intima’), and guarded near its entrance by numerous spines. These two 
stigmata belong to the first abdominal segment, and put the air-cavity just 
described into direct communication with the tracheal system. As already men- 
tioned, I regard this cavity and these stigmata as being mainly, if not ex- 
clusively, hydrostatic in function, serving not only to make the pupa float when 
at rest, but to make it float in a definite position, with the thorax uppermost and 
the apertures of the siphons at the surface of the water.” 
There is no special necessity for dealing here with the internal anatomy of the 
pupa, as will be understood from the following statement by Hurst: 
“ As to the anatomy of the pupa, it is only necessary now to state that at the 
beginning of pupal life the internal arrangements are those of the larva; at the 
end of that period they are those of the imago.” 
Hurst has given an excellent description of the external conditions in the pupa 
of Culez, which we quote herewith. We must, however, call attention to one dis- 
crepancy: The region called by him the “ prothorax” is in reality the meso- 
thorax. De Meijere, in an extensive paper on the anterior spiracles of dipterous 
pup, considers the anterior thoracic spiracles and their appendages as pro- 
thoracic, which, of course, also involves the region in question. A careful ex- 
amination of the external conditions leads us to differ. In dipterous larve the 
spiracles in question are prothoracic, in the imago they occupy a position which 
can not be designated as such with certainty. Hurst’s description of the pupa 
of Culex is as follows: 
“The head is broad from side to side; the epicranium has a well-marked 
median groove ; the clypeus, broad above, is gradually narrowed below, and con- 
tinued without any distinct line of demarcation into the labrum. At the sides 
are a pair of compound eyes, to be regarded rather as the rudiments of the eyes 
of the future gnat than as the visual organs of the pupa itself. .. . During 
pupal life they increase in size till they almost encircle the head. Corneal facets 
are never formed in the pupal cuticle, but beneath it the convex facets of the 
imaginal cornea are formed during pupal life. 
“ Behind the compound eye, on each side of the head, is an ocellus with fully- 
developed lens, etc. In the youngest pupe it is separated by a small interval 
from the compound eye; but the growth of the latter obliterates this interval, 
and the ocellus is in the older pupe not readily distinguishable except in sec- 
tions. The statement, found in systematic works, that the Tipulariz are devoid 
of ocelli is, however, not strictly true; in Culex, at least, they are well developed, 
though, as they abut upon the compound eye, they are in the imago so incon- 
spicuous that they may easily be over-looked. 
“Tn the mouth-parts, the labrum, epipharynx, mandible, maxille with their 
palps, labium and hypo- and epipharynx are present, though the two last can 
only be seen on dissection. 
“ Of their mode of origin in the larva I as yet know nothing. At the time of 
escape of the pupa from the larval cuticle they are of the full size, which is con- 
