The organ called _gizzard_ in the flea, for want of a better name, is, however, remarkable. Its function is not quite certainly known. It is a bulbous expansion in the front of the stomach and situated at the junction of the stomach and the gullet. It contains a multitude of chitinous finger-like processes tapering towards their extremities. From their general arrangement the complete collection of processes would act as an effective sort of valve and prevent the return of the fluids from the stomach. It seems most probable that this is their function. During the life of the flea the stomach is constantly churning its contents. Some valvular arrangement between the stomach and the pharynx would seem to be essential; the pharynx is normally collapsed, as the reader may remember, and its walls are drawn apart by muscles attached to its exterior. When the pharynx is full of blood the muscles relax, the walls collapse like elastic, and the blood is forced into the stomach. In many cases a flea will feed when the stomach is already tensely full of blood; and some sort of valve is therefore needed to prevent regurgitation into the pharynx when the pharyngeal muscles contract and the walls of the pharynx itself are drawn asunder. This valvular arrangement at the anterior end of the flea’s stomach has been minutely studied in connection with recent plague investigations, because there was a theory that fleas carried infection by vomiting the septicæmic blood from their stomachs and so transferred the plague bacillus to the puncture which they made in the skin. But an experiment, which has been tried several times, seems to show that the supposed valve is effective. The stomach of a flea which had recently fed was dissected out intact. As long a portion of rectum as possible was left attached at the hinder end. The gullet having been severed, well in front of the valve, pressure was applied with a blunt tool with the object of forcing the blood through the gullet. The hind aperture of the stomach was, at the same time, closed by pinching up the rectum. The result was that, in no instance, was it possible to force blood through the passage which leads into the gullet. Yet sufficient pressure was applied to burst the stomach.