10 PROC. ENT. SOC. WASH., VOL. 23, NO. 1, JAN., 1921 



directly to the mandible and not to an apodeme! Yet the 

 retractor (flexor) muscles are attached to a true apodemal arm 

 of the seta. To avoid this difficulty it might be proposed that 

 the sulcus between sclerites A and B is the attenuated apodemal 

 invagination, making the plate (fig. 13, 14, iPAp) arising along 

 its floor from g to the setal base, the true extensor apodeme with 

 its muscles attached to the side walls of the head. Contradict- 

 ing such an assumption, however, is the structure of the plate 

 itself in the head of an immature cicada. Here (fig. 14) there 

 is a very evident cartilage-like shaft extending through the 

 middle of the plate from its upper end into the base of the seta. 

 This shaft has the appearance of being of apodemal origin and 

 suggests that the two wings forming the plate have grown from 

 it, the outer coming in contact and fusing with the floor of the 

 sulcus. The retractor apodeme at this stage has an independent 

 origin in the hypoderm at the base of the seta (fig. 14, IRAp). 



So again we are but led into morphological absurdities. If the 

 plate carrying the insertion of the protractor muscles is a true 

 apodeme it should arise near the base of the seta as does that 

 carrying the retractor muscles, but there is no evidence of its 

 having any such origin (fig. 14) while its distal end is attached 

 to (continuous with) the head plate C at the point g. If g, on 

 the other hand, is its origin, as it should be since the invagination 

 for the anterior arm of the tentorium (fen) is just above it, and 

 we may easily assume that the mandibular base has been moved 

 down, how could it come about that the distal end of a mandibu- 

 lar apodeme should become attached to the mandible! 



Any way we turn, in an effort to make these parts fit the facts 

 of other insects, we encounter these baffling incongruities. 

 Both Muir and Kershaw (1911 a) and Peterson (1915) homolo- 

 gise the Hemipteran setae with the setae of the Thysanoptera, 

 but they do not agree as to which is which, and both portray 

 internal connections in the Thysanoptera which can not be 

 clearly identified either with those of the mandibles of biting 

 insects or with those of the setae of the Hemiptera. The whole 

 problem seems to become only more bewildering the more we 

 study it, and we may well wish that Mecznikov had been right- 

 in fact, in desperation we can almost believe that he must be 

 right, that the Hemipteran setae can not be and are not either 

 mandibles or maxillae. Of course, it may not be necessary to 

 assume that the ancestor of the Hemiptera ever had biting 

 mouth parts of the type possessed by modern mandibulate 

 insects. Their primitive head appendages may have evolved 

 directly into setae while they evolved into mandibles and 

 maxillae in the others. However, if any such gulf exists between 

 the Hemiptera and the mandibulate orders we should expect it 

 to be indicated also in the rest of their organization. 



An interesting comparison may be made in two other orders of 



