62 SUBFAMILIES CYPH A LEIN.fi AND CNODALONIN^H, 



Some of the genera (e.g., Chartopteryx, Prophanes, and Cyclo- 

 phanes) present a feature which seems to have escaped notice, in 

 the excavated, or strongly depressed central part of the apical 

 segment of the abdomen. Most of the genera, including the 

 above three, also Paraphanes, Hemicyclus, Oremasis, and Cypha- 

 leus, show strong sexual characters, viz., (l)the protuberant 

 ovipositor of the female, (2) the strongly enlarged three basal 

 joints of the front tarsi, and, to a less extent, of the intermediate 

 tarsi. The ovipositor has not been mentioned, so far as I am 

 aware, by other writers, and should not be mistaken for the male 

 organ. It is linear-lanceolate, slightly enlarged towards the 

 apex, sulcate or convex on its upper surface, bifid at its ex- 

 tremity, with two small linear appendages, and a few hairs near 

 apex. It has been customary for writers on this group to insert 

 an apology for adding a new genus. In adding four new genera 

 to the present list, the author would rather apologise for not 

 adding more, since some of the existing genera, especially Cypha- 

 leus and Chartopteryx, contain species of strikingly different 

 facies. The possession of more material, however, is necessary 

 before an author can sacrifice or mutilate rare specimens for dis- 

 section; and until this is done, it is better to include such doubt- 

 ful species under existing genera, if their salient characters render 

 the classification suitable. As a partial compensation, two of 

 the existing genera are omitted, Tetraphyllus as being a genus of 

 the Subfamily Cnodalonince, recorded only from Madagascar; the 

 two Australian species T. Reaumuri Caste!., and T. sumptuosus 

 Hope, belonging to other genera,' vide infra): while Decialma Pasc, 

 = Olistluena Erichs., and must be sunk. Pascoe had evident doubts 

 on this subject, when proposing the genus Decialma. The slight 

 difference in the antennae may readily be explained by an error of 

 observation, and the great difficulty in estimating the ratio of 

 length to breadth of small antennal joints {vide note on Decialma, 

 infra). There is very little distinction between Rectus and Olis- 

 thcena, save in the wider form and more widely separated eyes of 

 the former. For the present, I would retain Hectus until inter- 

 mediate forms appear. Pascoe's Table of Genera( Ann. Mag. Nat. 

 Hist (4.) iii., 1869, p.288) is misleading, in placing Prophanes 



