322 Deutsch. Ent. Zeitschi-. 1909. 



2. Antennae crassae, segmentis, 4 — 6 latissimis, valde traiis- 



versis. (Corpus apterum) 3. Karscliiellinae. 



2. 2. Antennae setaceae, segmentis 4 — 6 haud valde transversis. 



5. Pijgidicraninae. 



Subfaraily 1 . An a taelinae. 

 This subfaniily includes two isolated monotypic genera, Ana- 

 taelia Bol. , from the Canaries , with smooth mesonotum , and 

 Challia Burr, from Corea, with keeled mesonotum. In the struc- 

 ture of the antennae and sternal plates, these two genera approach 

 the Labiduridae, but their general form and appearance justifies 

 their position in the Pygidicranidae. 



Subfamily 2 . Di p laty i n a e. 



This subfamily contains the single genus Diplatys Serv. as 

 Nannopygia Dohrn, certainly, and Cylindrogaster Stäl, probably, must 

 fall into it. It is a well characterised homogenous group , but 

 the discimination of the species is subtle. The difficulty is in- 

 creased by the fact that in addition to the usual sexual characters, 

 the form of the pronotum varies with the sex. It cannot be 

 repeated too often that it is not only useless, but harmful, to 

 describe females alone, as without the male it is quite impossible 

 to ränge any species in its correct position. 



Here , more than in any other group of earwigs , colour is 

 not only untrustworthy, but even misleading. 



The Diplatyidae are probably primitive forms. The post- 

 embryonic development of the forceps from segmented caudal setae 

 is a most striking feature. 



Of the nine species known by de Bormans , several must 

 be sunk , bat recent discoveries have raised the number to over 

 two dozen. 



Subfamily 3 . Kar s chiellinae. 



This group was granted the rank of a family be Verhoeff, 

 but it has close affinities with the Pygidicrajiinae, in spite of many 

 very striking and peculiar features. It includes the two genera 

 KarscMella and Bormansia, by Verhoeff, with two and three species 

 respectively, all confined to tropical Africa. 



The very thick antennae (see flg. 2) and the structure of the 

 sternal plates separate this subfamily very sharply. 



The larvae of Bormansia have segmented caudal setae , as in 

 Diplatys. 



Subfamily 4. Py gidi c r a n i n a e. 

 This subfamily has recently been revised by me (Ann. Mag. 

 N. H. [8] II, p. 382 [1908]). It includes Tagalina Dohrn, Pygidi- 



