Marshall and Severity—Anatomy of Ranantra fusca, 4-95 
part of the body pass, forward and, just before reaching the 
mesothorax, they in their turn curve towards the sides of the 
body. At this; point each gives) off a, branch which passes for¬ 
ward through the muscles to which they give off a. great many 
small branches. After this branch is given off, the trachea con¬ 
tinue in their course towards the side of the body, but before 
reaching it each divides into two branches, one passing backward 
to join the metathoracic stigma, the other branch passing for¬ 
ward to enter another stigmla situated at the posterior lateral 
edge of the mesothorax. From each of the last stigmata a large 
trachea passes forward along the side of the body and in the 
posterior end of the prothorax joins with the branch we have al¬ 
ready mentioned as passing through the muscles. There is thus 
formed at either side a large dark trachea which passes through 
the prothorax, and at its. anterior end divides into; two branches, 
the inner of which passes forward into' the head, the outer enter¬ 
ing the anterior pair of legs. 
The connection between the trachea of the abdomen and the 
stigmata is not clear. When the stigmata were cut out, cleared 
and mounted, a small trachea could be seen coming from each, 
which undoubtedly led into* the main trunk. 
We have already mentioned that the rectum is filled with air 
and has a hydrostatic function. There are also present in Ra- 
nantm two pairs of air sacs very noticeable when fresh speci¬ 
mens are used. We were unable to trace these to their origin 
and do 1 not know from what part of the; tracheal system they 
arise. Each is a thin walled sac easily collapsable and without 
any resemblance to a trachea. The anterior pair, as far as we 
could make out, arise ventrally in the anterior part of the head 
and extend backward to near the middle of the prothorax. The 
posterior pair we could trace forward to about the middle of the 
prothorax and near its ventral wall. From here they pass back¬ 
ward above the intestine often extending as far as the fourth 
abdominal segment. 
Locy (5) mentions that when Ranantra is preparing for 
flight, these air sacs are filled. As to this we can not say but 
know that all specimens taken from the water had the sacs par¬ 
tially or entirely filled with air. In reaching the surface of the 
water, the* insects rise with the head down, but do not always 
remain in this position. We noticed that many in a short time 
assumed a position very nearly parallel to the surface of the 
