THE PHYSIOLOGY OF RESPIRATION IN INSECTS. 379 
exhausting by means of an air pump, but never succeeded in 
| doing so. It is true that bubbles of air do escape from the 
thoracic spiracles, when insects are put into hot water or even 
spirit ; but in the former case the valves of the spiracles are 
dislocated or ruptured, and in the latter the air, which rises in 
bubbles to the surface, is that which adheres to the exterior of 
the spiracle and fills the vestibule. The tracheal vessels them- 
selves remain full of air for a long time, a fact which is suffi- 
ciently manifest when the insect is dissected after immersion 
for an hour or more. When the waxy secretion is removed 
the efficacy of the external valves is destroyed, and the con- 
tracting muscles open the internal valves. Experiments which 
give rise to solution of the wax and to muscular rigidity, are 
therefore without value; such, for example, as immersing the 
insect in alcohol, or even in hot water. 
The inadequacy of the supposed expiratory movement of the 
body wall to change all the gas in the trachee, or even a large 
proportion of it with each respiratory act, is so manifest, that 
it precludes the acceptance of the expiratory theory of Trevi- 
ranus; if I am right in my conclusion that all or nearly all the 
inspired oxygen must be utilised to account for the metabolic 
activity of a lively insect. 
Passage of Gases through the Integument.—Herman Dewitz 
[159] made direct experiments on the passage of gases through 
the skin of the larva of Smerinthus ocellatus. He used the 
portion, which has no spiracles, to close the mouth of a test 
tube containing air, and immersed the tube in an atmosphere of 
carbon dioxide. He then found that the skin became very convex 
in an hour, showing that carbon dioxide passes in more rapidly 
than air passes out. He also reversed the experiment by 
filling the test tube with carbon dioxide, when the skin became 
deeply concave. In both experiments he observed on perforat- 
ing the skin that equilibrium was restored by a rush of gas, 
which made a loud sound. 
Dewitz does not mention the condition of the skin, whether 
moist or dry, but it was presumably moist. 
There can be no doubt the passage of gas through the 
