CLASSIFICATION OF INSECTS. 59 



system of Echinoderms, the gills of Crustacea, and the 

 aquatic respiratory organs of Vertebrates, though sub- 

 ject to great modifications, have typical forms which 

 are characteristic of large groups. The tracheal sys- 

 tem of insects, on the other hand, is characteristic of 

 the class, and can be compared with the respiratory 

 system in other branches or classes of the animal 

 kingdom, which have a structure specially adapted, 

 like the lungs of Vertebrata or the pulmonary sacs of 

 the land-snails, for breathing in the air. 



Mr. Samuel H. Scudder, whose acquaintance with 

 fossil and living forms has been as extensive and thor- 

 ough as that of any author, considers upon paleonto- 

 logical grounds that the earliest known insects were 

 generalized hexapods with two pairs of equal and sim- 

 ilarly developed wings, represented to-day by the 

 cockroaches. Miall and Denny entertain a similar 

 opinion founded upon the general characters of the 

 larvae as well as the geologic record, and Brauer ^ and 

 Packard notice, what we have observed above, the re- 

 semblances of Lepisma and their larvae. The Ephem- 

 eroptera have, however, an ancient origin, and the 

 history of winged insects is so imperfectly known, the 

 diversity of known forms at the earliest epochs so 

 striking, and the differences of opinion so great, that 

 we cannot assert that any special forms among the 

 more generalized orders represent the main stock of 

 winged insects out of which all the others arose. We 

 have therefore ventured only upon the assumption that 

 there was probably one or more of these ancient stocks 



1 See F, Brauer, Zool. Studien, op. cit. 



