THE PROBLEMS OF TRANSFORMATION 265 



must have already at that period become marked, and the 

 origin of the order must be sought, like that of the mayfly 

 stock, among those most primitive exopterygote insects of 

 Devonian or perhaps even Silurian age, of which no fossil 

 remains have yet been discovered. 



With regard to the third order of exopterygote insects with 

 aquatic gill-breathing larvae the Plecoptera or stone-flies 

 we have little direct evidence from fossils, but a few specimens 

 of Jurassic age can be referred to the living family Nemouridae, 

 and nymphs as well as adults are represented among these 

 remains. In the stone-flies there is a much closer likeness 

 between imago and larva than in either the mayflies or dragon- 

 flies, and they may probably be regarded as a direct offshoot 

 from the Palaeodictyoptera. 



In all these three orders of exopterygote insects with aquatic 

 larvae, shown by the evidence of fossils to be not very closely 

 akin to each other, we find that the larvae breathe dissolved 

 air by means of gills diversely specialized in each order, the 

 gills of the stone-fly larvae being branching tufts on the thorax, 

 of the mayfly grubs modified and flattened serial abdominal 

 appendages, of the dragon-fly grubs modified terminal ab- 

 dominal appendages or outgrowths from the wall of the hind- 

 gut. These divergences prove conclusively that the aquatic 

 habit of the larvae has been independently acquired in the 

 three orders, while the combination of the gills with air-tubes 

 a type of breathing organs almost exclusively associated with 

 aerial or terrestrial life shows that it is during the preparatory 

 stages of the life-history that these insects are in foreign en- 

 vironment. The air is their native element, and the wonderful 

 aquatic larval adaptations are secondary. 



The remaining large and important exopterygote order, 

 the Hemiptera, stands apart to a great degree, at the present 

 day from the other orders. Its two sub-orders, the Heterop- 

 tera and the Homoptera, are sharply distinguished from each 

 other not only in structure but in life-history, for while young 

 Heteroptera resemble their parents in general body-form, 

 most families of Homoptera afford examples of change during 

 the period of growth. It is of especial interest, therefore, 

 to find in a famous fossil (Eugereori) from the Lower Permian 

 of Oldenburg, the characteristic piercing and sucking jaws of 



