and Dilated Trachec in Insects. 491 
and elongated, and mainly assist in inducing a rush of blood into these struc- 
tures, which are thus expanded on the sides of the new pupa as the rudi- 
mentary wings. This elongation of the small trunks at the sides of the lon- 
gitudinal tracheæ in the thorax, relieves them of a portion of that tension 
which results from the powerful respiratory efforts of the insect in effecting 
its change; and which, with a tendency to enlarge by the natural forces of 
' growth in these structures, results in the dilatation first of the trachec at 
the base of the abdomen, and afterwards of those of the thorax, and the sides 
of the abdomen, into distinct sacs. This is the manner in which the air-sacs 
are formed in all insects. After the main trunks have become dilated their 
ramifications also are enlarged in like manner, and this enlargement con- 
tinues from the time when the insect enters its pupa, to that of its appearance 
in the perfect state. 
But although we are enabled to show the manner in which the vesicles are 
formed, it is difficult to give direct proof of their immediate use. It is assumed 
from the facts I have mentioned, that they are most numerous and large in 
volant insects, and entirely absent in apterous, that they are connected with 
the power of flight; but in what way, has never yet, so far as I am aware, been 
actually proved. It was assumed by John Hunter that they are for the pur- 
pose of enabling the insect to alter the specific gravity of its body,—a view 
which I have myself long ago adopted,—but the facts on which the opinion is 
founded are only those of anatomy, and not of direct observation or experi- 
ment, which are required to confirm it. Anatomical facts alone, as will be 
seen in this instance, although the safest guides, are not always sufficient. 
Thus we find that a vesicular structure of the organs of respiration exists in 
the whole of the air-breathing Vertebrata. The lungs of Man and the Mam- 
malia aré formed of multitudes of vesicles, which are more numerous than in 
any other class; those of Birds exhibit a like condition, but in these the vesi- 
cles are distributed more extensively over the whole body. A vesicular con- 
dition of the respiratory organs exists also in the Reptiles and Amphibia, but 
far less extensively than in Birds. Thus each of these classes agrees with in- 
sects in possessing a vesicular form of the respiratory organs. The whole of 
them also, excepting the Mammalia, agree in the fact that the vesicles are 
parietal, that the tracheal structures are dilated in different parts of their 
