MKEE : MOUTH-PARTS OP THE HEMIPTERA. 273 



tract and pull the chitinized piston back. The valve at the 

 lower side of the chamber opens and saliva from adjoining res- 

 ervoirs fills the cavity. As the muscle relaxes the elastic chiti- 

 nous wall resumes its normal position, the valve closes, and 

 the saliva is forced down the salivary canal into the smaller 

 maxillary tube, through which it reaches the food supply. That 

 this action is possible has been demonstrated by actual experi- 

 ment. Although the injector is exceedingly small, under a 

 suitable microscope I have succeeded in watching the move- 

 ment of the elastic parts as the muscle of the injector was pulled 

 back and forth mechanically. In some of the Heteroptera, 

 Geise describes a valve for closing the salivary canal while the 

 saliva is being pumped in from the glands. Wedde does not 

 figure such a structure, and I am unable to find it in the Cicada. 

 It is probable that at that point the canal is thin enough to press 

 together and serve as a valve. In any case the efficiency of the 

 injector would scarcely be affected, for the secretion might easily 

 be forced into the chamber by the pressure of the reservoirs in 

 which it is contained after having been secreted by the salivary 

 glands. 



The saliva of the Hemiptera is alkaline, and has the power 

 of changing starch to sugar. Plateau made this discovery by 

 boiling four of the glands from two Nepas in starch. Glucose 

 was found in the mixture as a product. Primarily, then, the 

 saliva serves as a digestive fluid. The fact that it is forced to 

 the end of the labium, instead of being mixed with the food 

 somewhere in the digestive tract, goes to show that the secre- 

 tion lias other uses. It is undoubtedly forced into the puncture 

 made by the insect in order to increase the flow of sap. It 

 also probably serves as a lubricant for the closely fitting maxillae, 

 since they would need something of the kind, and no other se- 

 cretion is present. 



As soon as a good flow of sap is started, the pharyngeal mus- 

 cles (fig. 12, m, p) contract. By the alternate lowering and 

 raising of the upper fold of the pharynx, a suction-pump is 

 produced, the pipe for which is the pharyngeal canal and the 

 upper maxillary tube into which the canal extends. The lower 

 part of the tubes formed by the inner concave surfaces of the 

 maxillae are probably not air-tight. Suction, then, would not 

 be sufficient to raise the fluid the entire distance from the tip 



