346 Proceedings of the Royal Irkh Academy. 



maxillula of the CoUembola. The exopodite is greatly reduced in 

 some of the thoracic legs, and there is less distinction between the 

 thoracic and abdominal regions than in most Malacostraca. For the 

 ancestors of Insects and Myriapods, we must suppose animals without 

 a carapace and with all the limb-bearing trunk -segments similar. 

 Such forms might conceivably have existed among the earliest Mala- 

 costraca; and Anaspides and its Palaeozoic allies (Palseocaris, etc.) 

 come nearer to the ideal than any other Malacostraca known to 

 zoologists. Nevertheless there is a great difficulty in supposing that 

 an Arthropod with similar body-segments could be developed from 

 one in which differentiation between thorax and abdomen had set in. 

 On the whole, therefore, it is most reasonable to believe that the 

 ancestors of Insects, Millipedes, and Centipedes were an offshoot from 

 the progenitors of the primitive Leptostraca. The head and its 

 appendages show such close correspondence in the Insects and Crus- 

 taceans, that the Tracheate branch must have arisen above the branch 

 that had given origin to the Trilobites, in which there was no differen- 

 tiation between the head- and body -limbs. The result of our inquiry, 

 therefore, is to trace back the Insecta to ancestors that are essentially 

 Crustacea, although Crustacea of a very generalised type. 



It will now be convenient to discuss the meaning of the position 

 of the genital aperture or apertures. In this there is a marked 

 difference between the Insects and the Centipedes, in which the ducts 

 open near the hinder end of the body, and the Millipedes, Symphyla, 

 and Crustaceans, in which they open more or less anteriorly. This 

 divergence has often been considered very radical, only explicable by 

 imagining common ancestors with a large series of paired, segmental, 

 coelomic ducts which served to carry off the germ-cells. Now it is 

 very hard to believe that the primeval Crustaceans, which seem to have 

 been ancestral to all these classes, could have possessed such a very 

 generalised reproductive system as that. It is impossible to suppose, 

 on the other hand, that any sudden shifting of the genital aperture from 

 one region of the body to another could have taken place. 



Two sets of facts, when taken together, seem to give the solution 

 of this problem. There is sometimes not absolute constancy in the 

 position of the apertui'es within the same group. And if the Insects, 

 Crustaceans, and Millipedes be compared together, it is found that 

 the apertures of the Leptostraca and Malacostraca lie on segments of 

 the body situated between the genital segments of the Millipedes on 

 the one hand, and of the Insects on the other. It is found, taking the 

 antennular segment as the first, that, in the Diplopoda, the genital 



