102 MOSQUITOES OF NORTH AMERICA 
the wings. The same segmentation into femur, tibia, etc., is recognisable as in 
the adult gnat, but the segments are more nearly equal in the pupa, and the 
joints of the developing and shrinking legs of the future imago soon lose their 
correspondence with those of the pupal cuticle enclosing them. They arise in 
the larva, like other appendages, as folds of epidermis enclosing mesoblastic 
tissues. 
“The abdomen is dorso-ventrally compressed and exceedingly flexible dorso- 
ventrally, though not from side to side. It is the only part of the pupa in which 
the segmentation of the body is readily recognisable, and as I shall very fre- 
quently have to refer to the various segments by number, J shall use the terms 
“first segment, etc., to signify ‘ first segment of the abdomen,’ etc. 
“Nine segments are readily recognised in the abdomen, and the last one, 
though it is probably composed of no less than three condensed and highly modi- 
fied segments, I shall call simply ‘ ninth segment.’ 
“ Hach abdominal segment has a chitinous tergum and sternum, and sete are 
distributed sparingly over them, being almost confined to the hinder parts of 
the terga. The terga and sterna of successive segments are united by soft 
arthrodial membranes. 
“ Of the sete, only one pair need special mention. These are placed on the 
hinder part of the first segment, the base of each being a triangular plate at- 
tached by one angle to a soft membrane, and the distal side of the plate is divided 
into a number of bars which, by repeated division or branching, give rise to about 
one hundred sete all lying in one plane parallel to the median plane of the body. 
Each seta bears a few fine hairs. When at rest, the pupa floats with the tips of 
these sete, and the tips of the respiratory siphons, at the surface of the water, 
and these sete probably assist in maintaining the equilibrium of the animal in 
this position, as well as serving as sensory organs by means of which any dis- 
turbance of the surface is felt. 
“ The eighth segment bears a pair of large fins, thin oval plates about 1.2 mm. 
in length, attached by the narrow end beneath the tergum behind. Each is 
stiffened by a midrib which projects beyond the hinder border of the fin as a 
spine. 
“ Beneath the fins and behind the eighth segment is the ‘ninth segment’ with 
its appendages. ‘Though this region is probably made up of more than one seg- 
ment, its composite nature is not easy to recognise, as the plates supposed in 
other insects to represent the terga and sterna of tenth and eleventh segments 
[see, for instance, Huxley and Miall and Denny] are not developed in the young 
pupa, nor, indeed, is there in any stage any such development of the pupal 
cuticle, though plates developed within as parts of the imaginal cuticle may per- 
haps represent some of these parts. 
“The appendages of the ‘ninth segment’ of the pupa are a pair of blunt 
processes arising below and in front of the anus, and directed backwards below 
the fins. They are much larger in the male than in the female. A pair of ap- 
pendages are already recognisable in this region in sections of the larva, and I 
think even two pairs, but this portion of the larva is particularly difficult to 
cut, and I am not yet certain as to the hinder of the two pairs. Of the existence 
of one pair I have no doubt.” 
Some mention must be made of the modifications of structure which occur in 
the pupe of different species or groups of species. The early writers on the 
biology of Anopheles made much of the difference in the shape of the respiratory 
trumpets between Anopheles and “ Culex.” Their statements appear to have 
remained unchallenged and are still widely quoted. The variations in the shape, 
