TYPE TBACHEATA. 521 



Development and Affinities of the Insecta. — The early stages of Insect 

 development cannot be discussed here, belonging more properly to text- 

 books of Embryology, but mention should be made of the remarkable phe- 

 nomenon which occurs during the transformation from the pupal to the 

 imaginal conditions in those forms whose metamorphosis is complete. In 

 describing the development of the Acarina it was pointed out that during 

 the transition from one stage to the next a histolysis and subsequent regen- 

 eration of certain parts of the body occurred. In the holometabolic Insects 

 the same process occurs during the pupa stage, the larval hypodermis, 

 the majority of the muscles, and the entire digestive tract and its appen- 

 dages undergoing degeneration, and being absorbed and digested by the 

 blood-corpuscles, the parts being formed anew from patches of cells present 

 in the larva and known as imaginal disks. The histolysis and regenera- 

 tion proceed pari passu, so that the identity of the various organs is pre- 

 served throughout the process. The imaginal discs are to be regarded as 

 portions of the original anlagen of the various organs which have re- 

 mained during larval life in an embryonic condition, springing into activity 

 and completing their development during the pupal stage. 



As regards the afBnities of the various orders of the Pterygota, it may 

 be pointed out that the frequent occurrence of Thysanuriform larvae indi- 

 cates a descent from Apterygote ancestors, and those orders which present 

 larvae of a wormlike or maggotlike form are in all probability the most 

 highly specialized. It is in these cases that the complete metamorphosis 

 occurs, and it is self-evident that the gradual and incomplete metamor- 

 phoses are more primitive than the complete. Indeed all metamorphosis 

 depends upon the differences in habit and structure of the larva and imago, 

 and becomes more and more complete according as the larvae and imagines 

 depart more and more widely from the Thysanuriform type of structure. 

 Consequently it may be concluded that those forms are the most primitive 

 which retain most perfectly both in the larva and imago the Thysanurid 

 characters. These are found most perfectly in the Dermaptera, to which 

 both in the adult and larval stages the Oorrodentia (so far as they have 

 not become modified by parasitism) and the Orthoptera seem closely re- 

 lated, and it is interesting to note in this connection that the earliest In- 

 sects known from the Palaeozoic rocks seem to have been closely related to 

 the recent group of the Orthoptera. 



Another order which has retained Thysanuran characters in the larva, 

 though the imagines are more highly specialized than are those of the 

 Dermaptera, is that of the Thysanoptera, whose habits and mouth-parts 

 indicate affinities with the Ehynchota, these two orders together forming 

 a second group traceable back to- the primitive Pterygota. 



A third group starts with the Ephemeridae, which lead up to the Odo- 

 nata, the larvae of the latter having, however, become greatly specialized, 

 the resemblances being most marked in the adults, and are indicated by 

 the character of the wings and by the mouth-parts. More distantly re- 



