38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I42 



transverse anterior and posterior parts variously developed in differ- 

 ent species and supported by props (a, p) on the sternal arms. This 

 structure has been fully described by Denis (1928), who called it a 

 "tentorium." However, it has no resemblance to either the thysanuran 

 or the pterygote tentorium, and moreover, as shown by Tuxen ( 1952) , 

 it is a mesodermal tissue soluble in caustics and lactic acid. This 

 collembolan "tentorium" is thus more nearly comparable to the endo- 

 sternum of Chelicerata. Since it gives attachment to the ventral 

 muscles of the mouth parts, it functionally serves the same purpose 

 as the chelicerate endosternum and the tentorium of the ectognathous 

 insects. In Protura there is no corresponding superstructure on the 

 maxillary sternal arms. 



Incidentally it may be observed that, in the possession of a cuticu- 

 lar tentorium, the ectognathous hexapods appear to be more closely 

 related to the chilopods and symphylans than to the entognathous 

 hexapods. 



VII. THEORETICAL CONSIDERATIONS 



The accumulation of knowledge does more than simply add new 

 facts to old ones ; it changes our ideas about the accepted facts. This 

 is particularly true in our study of insect anatomy and our morpho- 

 logical interpretation of the structural facts. We have now become 

 involved in interpretations and theories that never occurred to the 

 earlier entomologists, and hence they could write their descriptions 

 of insect anatomy in a more direct and simple manner than we can. 

 Furthermore, since morphology (the science of form) is a product 

 of our brains, and our brains are not standardized, we are now per- 

 plexed with opposing theories that purport to explain the same set of 

 facts in different ways. Our descriptive matter, therefore, has be- 

 come so mixed with argumentation that the facts often seem less im- 

 portant than the theoretical discussions about them. In particular, 

 some modern theories of insect head segmentation are so opposed to 

 all our former ideas as to make the insect head seem so complex that 

 it is hard to visualize how it ever got that way in its evolution. Even 

 the embryo appears to be unable to recapitulate its evolution accord- 

 ing to these theories, and adheres to old-fashioned ways of develop- 

 ment. Of course, it is always possible that theories do not represent 

 the truth. So in this study of the insect head we must critically re- 

 examine not only the evidence, but also inductions made even from 

 correctly observed facts. 



The following discussions will be concerned with theoretical ques- 



