REVISION OF STREPSIPTERA PIERCE. 9 



1. Similarity of hexapod larvae to those of Meloidae (and Rhipi- 

 phoridae). 



2. "Metamorphosis nearer to that of Coleoptera" than of any other 

 "elytrophorus order" (quoted from Kirby, 1813). 



3. Same buccal organs in hexapod larvae notwithstanding the dif- 

 ferences of their respective habits. 



4. Variation of buccal organs during metamoi-phosis of Sitaris, as 

 great as differences between these organs in Stylopidae and in Meloidae. 



5. Participation with several abnormal coleopterous groups in the 

 divergence from conventional formularies of organization and dis- 

 tinguished by the same parasitic associations. 



a. Myodites, having rudimentary elytra and radial wing venation. 



b. Macrosiagon, Myodites, in the longitudinal folding of the wings. 



c. Rhipidius, in the eyeless, apterous females. 



d. With the males of all these in the highly developed ramose 

 antenna?. 



e. With Meloe, Sitaris, Zonitis, in the remarkable coarctate pupal 

 metamorphosis and preliminary transformations. 



In the appendix of the present article will be ,found a discussion of 

 the fallacy of using such characters to bring together unallied groups. 

 As to the metamorphosis, Mr. E. A. Schwarz says that it was prob- 

 ably reached by a line of development parallel with that of the 

 Rhipiphorida*, due to somewhat similar parasitic habits, because, he 

 says, the Strepsiptera have "nothing in common with the Coleoptera." 



The resemblance of the triungulinids of the two families Meloidae 

 and Rhipiphoridae and the Strepsiptera is mainly superficial. The 

 meloid hexapods have three claws on the tarsi; the rhipiphorid 

 hexapods have one claw only and a sucker-like pulvillus, and are 

 also provided with a caudal sucker; the strepsipterous hexapods have 

 no tarsal claws, but do have the pulvilli, and instead of the anal 

 sucker are provided with long bristles. 



Since the Thysanura are probably the nearest approach to the 

 primitive insects, it is reasonable to expect that where parasitic 

 habits have brought about a hypermetamorphic type of develop- 

 ment, the first stages should revert to the primitive Campodeoid 

 type and that distant groups with similar habits should exhibit simi- 

 lar forms in these first stages. 



The only genera in the Coleoptera which have been considered close 

 to the st}dopids are Hornia and Leonidea in the Meloidae and Rhipi- 

 dius seiis latum, in the Rhipiphoridae. The two former are very dis- 

 similar in appearance, and the females have legs. The Rhipidiini 

 differ in that the larvae leave the host to pupate ; the females are like 

 the males, but wingless and with 11-jointed filiform antennae. 

 According to Mr. Frederick Knab the females also have eyes, legs, 

 and a well-developed ovipositor. Thus it will be seen that there is 

 no resemblance whatever to the Strepsiptera. 



