248 SECTION D. ARTHROPODA. 



as to the existence, in former geologic epochs, of forms that may 

 have been ancestral to the existing divisions. It shews a pre- 

 ponderance of evidence that the exopterygotous condition was the 

 primitive one in insects. The facts that the insects of the Palaeozoic 

 epoch appear to have been almost entirely exopterygotic, and that 

 the great majority of existing insects are endopterygotic suggest 

 the probability of the latter being descended from the former, but 

 the origin of Endopterygota remains at present a mystery. It seems 

 clear that Endopterygota could not possibly have been derived from 

 Exopterygota except by intermediate states of Anapterygotism. 



The only group of Endopterygota that we have any real reason 

 for believing to have existed in the Palaeozoic epoch is the Sialid 

 division of Neuroptera planipennia, and the evidence is both scanty 

 and uncertain : there is, too, no reason for supposing this group to 

 have been actually ancestral to the great orders of Endopterygota, 

 Coleoptera, etc. 



In reference to these aetiological questions I may point out that 

 the groups Apterygota and Anapterygota are each of them partly 

 grounded on aetiological hypotheses. 



Grassi was formerly of opinion that the wingless condition of the 

 Apterygota is a truly primitive state ; but he is now willing to admit 

 that it may possibly be derived from a winged condition. Hence 

 taking everything into consideration it seems far from impossible 

 that all insects may have been derived from Exopterygotous forms, 

 the Apterygota directly, the Endopterygota indirectly by means of 

 Anapterygotous forms. But it appears more probable that Col- 

 lembola — if not Thysanura — are truly apterygotous. 



