346 THE BIOLOGY OF INSECTS 



throws us back on surmise as to how, in the course of racial 

 history, the wing as an organ of flight was developed out of 

 a thoracic outgrowth less than a wing, such, for example, as 

 the prothoracic " winglets " of some Palaeodictyopteran. 

 Two surmises may be mentioned. C. Gegenbaur (1871) 

 suggested that wings are modified tracheal gills, indicating 

 an originally aquatic manner of hfe among primitive insects ; 

 this view is inadmissible because the typical insectan air- 

 tube system is adapted for atmospheric breathing and all 

 branchial modifications are clearly secondary. A. S. 

 Packard (1898), G. C. Crampton (1916), and others have 

 supposed that the thoracic outgrovi1;hs, early in the history 

 of primitive insects, were of service as parachutes and that 

 a capacity for gliding motion preceded the power of true 

 flight. It may have been so, but the subject is one as to 

 which it is wise to admit ignorance. The problem how 

 wings arose in the course of racial history is one of those 

 problems of evolutionary adaptation which involves some 

 of the questions of method to be discussed in the remainder 

 of the chapter. 



The class of Insects belongs, as we have seen (Chap. I, 

 p. 2), to the great phylum of the Arthropoda, and a dis- 

 cussion of the course of insect evolution would be incom- 

 plete without some reference to the relationship of insects 

 to other classes of that phylum. As this question is mainly 

 morphological it cannot be treated in detail in the present 

 volume. It is clear, however, that apart from their dis- 

 tinctive power of flight, insects are adapted for their 

 life relations — feeding, sensation, response, movement — in 

 essentially the same way as are members of the other 

 arthropodous classes. More than once (see J. S. Kingsley, 

 1894, and A. S. Packard, 1898) the suggestion has been 

 made that the Arthropoda should not be considered a natural 

 group, but regarded as an assemblage of classes which 

 originated independently of each other from segmented 

 worms (Annelida). The distinctive features of the 

 Arthropoda are, however, too characteristic and remarkable 

 to be explicable as the result of convergence, as E. R. 



