No. 1774. THORAX OF HYMENOPTERASNODGRASS. 39 



base, and the thoracic structure of its various branches would have 

 been much more difficult to decipher. 



The Hymenoptera are usuall}^ given the highest place in the scale 

 of specialization, yet in almost every feature, members of some other 

 order might be placed ahead of them. The mouth parts of the Hem- 

 iptera, the thorax and wings of the Diptera, and the internal organs 

 of many other forms are more specialized than the corresponding 

 parts of the Hymenoptera, while the Tenthredinoidea are certainly 

 more generalized in their adult characters than the lowest members 

 of the Hemiptera, Lepidoptera, or Diptera. These other orders, 

 however, have picked out some one character or group of characters 

 for extreme specialization. The Hymenoptera, on the contrary, 

 have carried nearly all their organs to a high state of perfection and 

 specialization. The mouth parts, the thorax, the legs, the wings, 

 the ovipositor or sting, the alimentary canal, the tracheal system — 

 all constitute a group of specialized organs unparalleled in any other 

 order. Added to this is the high development of their instincts and 

 the great diversity of their habits. Hence, there can be no doubt 

 that the order amply merits the place of honor assigned to it. 



2. GENERAL STRUCTURE OF THE INSECT THORAX. 



It may still be confidently asserted that the thorax of insects consists 

 of three segments, the attempts of various entomologists to make out 

 a contrary case notwithstanding. Those who would elaborate this 

 region of the body into a composite structure of many original seg- 

 ments may be grouped into two classes. The first includes those who 

 look upon each apparent segment as a compound of two or four 

 primitive segments. The second includes those who believe that all 

 but three of the original segments have disappeared, except in some 

 of the lowest insects where their rudiments persist as the interseg- 

 mentalia or little sclerites situated between the normal segments. 

 The theory of the first class of speculators derives the consecutive 

 parts of each definitive segment from a series of coalesced primitive 

 segments; that of the second class leaves each modern segment a 

 unit, and only assumes that there were once a great many more such 

 units present. In the study of insects alone neither of these theories 

 seems to be demanded. There is no necessity for supposing that the 

 parts of any segment are anything more than secondary differentia- 

 tions, or that the intersegmentalia are anything more than secondary 

 products of the principal segments. Embryologists have never 

 discovered more than tliree nietameres in the true thoracic region of 

 any insect. 



In this connection it is interesting to note that both of these theories 

 have been urged principally by myriopodists, or by entomologists 

 who have included the Myriopoda largely in their studies. In fact, 

 both theories are really based on the idea that insects are lineal 



