346 The Pedigree of Insects 



the Ametabola and the beetle into the Holometabola. 

 And the facts of insect development (Chapter II) 

 clearly indicate, as we have seen, that complete meta- 

 morphosis with a resting pupal stage is an advance 

 on direct growth throughout which the insect moves 

 and feeds. If then a bug's jaws are more highly 

 specialised than a beetle's, the latter's life-history is 

 more highly specialised than the former's. 



It is clear then that different classes of facts often 

 point to divergent conclusions, and that a careful 

 weighing of all available evidence is needful if we 

 are not to be misled. Is the nature of the life-history 

 a more important character than the form of the 

 jaws ? The examination of the characters presented 

 by insects within a single order seems to show us 

 that it is. Among the Hymenoptera for instance 

 there is very considerable difference in jaw-structure, 

 but every member of the order passes through a com- 

 plete metamorphosis with a free pupal stage. There- 

 fore, in the list of orders given in Chapter IV, the 

 arrangement founded on life-histories was followed, 

 the Ametabola (lowest) coming first, the Hemi- 

 metabola next, and the Holometabola (highest) last. 

 But no linear arrangement can represent truly relation- 

 ships best symbolised by the branching of a tree, 

 any more than a railway time-table, wherein the 

 names of the stations are arranged in vertical columns, 

 can indicate the geography of a line. A shoot arising 

 low down on a branch may send out twigs which 

 overtop the lower twigs of a shoot whose origin is 

 higher. Such a conception may symbolise the difficulty 

 presented by the comparison of the beetle and the 

 bug. 



Collembola and Thysanura. — It has been already 

 mentioned (p. 164) that the Collembola and Thysanura 

 have been separated from all other insects as Aptery- 



