4 BULLETIN 66, UNITED STATES NATIONAL MUSEUM. 



opterist. Doctor Ashmead believed that there was some connection 

 with the Dermoptera. Frederick Knab has shown the author that 

 there is no resemblance to Bhipidius, the blattid parasite. 



SYSTEMATIC POSITION. 



Although for the past forty years the majority of English and 

 American writers have placed these parasites in the Coleoptera, 

 they seem to have done so without regard to the many good argu- 

 ments and proofs brought forward on the Continent, winch would 

 place them as an independent order. 



Before entering into a discussion of the fallacies which have 

 crept into various classifications, or of proving the position of the 

 group, the writer desires to lay emphasis upon the excellent rules 

 for forming an order (in Insecta) as set down by Kirby (1813). 

 These rules seem still to be perfectly valid: 



Rule I. When an insect in its perfect state combines the characters of two or more 

 orders (unless it be deemed advisable to place it in an order by itself), it should arrange 

 ivith those whose metamorphosis is the same. 



Rule II. When an insect possesses the characters of one order and the metamorphosis of 

 another, in this case it should follow the characters. 



On this rule it may be observed that, since the perfect state is the 

 grand consummation of the insect to which all other states are sub- 

 ordinate and subserve, this state therefore ought to be the principal 

 regulator of its station. 



Rule III. Where an insect exhibits the metamorphosis oj an order, or of a section of 

 it but none of its characters, nor those of any other order, it should not on that account 

 be arranged in such order, but, on the contrary, form a distinct one. 



Rule IV. Where the genera which compose an order have invariably one kind of meta- 

 morphosis, no insects that vary from it in that circumstance should be placed in it, unless 

 they exhibit a perfect agreement with it in characters. 



Kirby after formulating his rules proceeded to apply them to the 

 Strepsiptera, and as many of his points are excellent they have been 

 used by the writer as the framework for the following paragraphs. 



Metamorphosis, though not the regulating character, nevertheless is 

 one of the most prominent of all ordinal characters. There are a very 

 few types of metamorphosis, but the Strepsiptera stand alone as 

 typifying a life history, the most complex of all. They are invariably 

 hypermetamorphic, and endoparasitic throughout life, the only free 

 stages, being the hexapod triungulinids or first larvae and the adult 

 winged males, and in neither of these stages are they known to 

 take food. The metamorphosis begins with the larviparous pro- 

 duction of free living young, which are conveyed by various means 

 to the larvae of their future hosts. These hexapods after beginning 

 the parasitic existence distend and become grub like, and each 

 succeeding molt makes the females more degenerate, while the males 

 undergo a transformation of specialization. Alimentation seems to 



