NO. 2242. MORPHOLOGY OF THE 8TREP8IPTEBA— PIERCE. 417 



oedeagiis differs very greatly between genera and slightly between 

 species (pi. 64, fig. 8; pi. 65, fig. 10; pi. 72, fig. 7; pi. 74, fig. 4; pi. 75, 

 fig. 6 ; pi. 76, fig. 4 ; pi. 77, fig. 8 ; pi. 78, figs. 5, 6, 9, 12) . 



The tenth segment bears the anal pore. The eighth segment is in 

 some Halictophagidae (Pentozoe, Pyrilloxenos, Delphacixenos, Pen- 

 tacladocera) produced beneath the ninth segment (pi. 75, fig. 4; 

 pi. 78, fig. 3). Otherwise, it is normal, ringlike. 



CLASSIFICATION. 



BEASONS FOB CONSIDERING THE STEEPSIPTERA AN ORDER. 



Argument based on rules of establishing an order. 



In 1813 Kirby set down an excellent set of four rules for the estab- 

 lishment of an order of insects, to which the present writer added a 

 fifth and its converse in Bulletin 66. Taking these rules one by one 

 we may consider the evidence supporting the contention that the 

 Strepsiptera must be considered an order. 



Rule I. When an insect in its perfect state combines the characters 

 of two or more orders {unless it he deemed advisable to place it in 

 an order hy itself), it should arrange with those whose metamor- 

 phosis is the same. 



The Strepsiptera do not combine the characters of any two or more 

 orders, being easily distinguishable in either sex from all other 

 insects. They have the usual parts belonging to the insect anatomy 

 with certain exceptions. Some of the peculiarities of structure have 

 counter parts in other orders, such as the flabellate antennae, the 

 oedeagus, the ensiform mandibles, the elongate trochanters. They do 

 not conform in type of metamorphosis with any other order, although 

 certain features of the metamorphosis have counterparts in other 

 orders. For instance, we find viviparous reproduction occurring here 

 and there in other orders, but none showing it as a constant type; 

 we find hexapod first larvae and legless later larvae in various fami- 

 lies of Coleoptera ; we find pupation in a puparium or last larval skin 

 in Diptera and rarely in Coleoptera; we find a similar pupa in 

 Hymenoptera; but we do not find any order in which the entire 

 Strepsipterous type of metamorphosis is duplicated. Hence, on the 

 basis of Rule I, we are obliged to consider the Strepsiptera an order. 



Rule II. When an insect possesses the characteristics of one order 

 and the metamorphosis of another., in this case it should follow the 

 characters. 



The Strepsiptera do not fit this premise in any way. There is there- 

 fore no reason under Rule II for aligning them with any other order. 



Rule III. Where an insect exhibits the metamorphosis of an order., 

 or of a section of it but none of its characters nor those of any other 

 3343— 19— Proc.N.M.vol..54 28 



