230 THE ENTOMOLOGIST S KECOKD. 



The relationship of Philea (irrorella), Cybosia (mGSomella), and 

 Endrosa (aurita). 



By T. A, CHAPMAN, M.D., F.Z.S., F.E.S. 

 The second volume of the British Musewn Catalogue of Phalaenae is 

 as striking a monument of industry on the part of Sir George Hampson 

 as the first. It contains those famihes that are allied to the Arctiadac, 

 viz., the JJtliosianae and Nolanae. Though I believe the reasons 

 against this association of the Nolanae to hold good, my high estimate 

 of the value of the work is not affected. 



Having no special knowledge of these groups to enable me to test 

 the soundness of the classification within the groups, I have looked 

 up the little group of Enclrosa, Philea and Cybosia, with which I 

 happen to be familiar. The first point to strike one is the accuracy of the 

 characters laid down for each genus and species, short and condensed, 

 and implying much work compressed into small space, and giving one 

 some idea of the labour involved in the Avhole volume, and if equal 

 accuracy obtains elsewhere, the value of the work is certainly great. 

 The next point is that these three genera are separated and defined, 

 and their order and phylogeny determined chiefly (apparently) by the 

 one fact that the point of origin of vein 2 of the hindwings varies a 

 little. 



In Philea it is from the middle of cell, in Cyhosia from hetjond the 

 middle of cell, and in Endrosa from irellbeijond the middle of cell. The 

 other characters are of minor importance or very inconstant. It is 

 true, for instance, that vein 8 of the hindwing usually rises nearer the 

 end of cell in Endrosa {aurita) than in Philea (^irrorella), but I have 

 specimens of aurita with it quite as far back as is usual in irrorella. 

 The $ genitalia are almost identical, probably quite so if one took a 

 long series. Those of Cyhosia {mcsomella) are quite different. Eevert- 

 ing to vein 2 of forewing, this also like 8 of hindwing varies much in 

 position and length, theaverage position is asstated, but many specimens 

 overlap. The only constant differences between aurita and irrorella 

 are in the density and length of the hairs and scaling, and a greater 

 breadth of the palpi in aurita. But these are certainly not generic 

 differences, and do not admit of their being separated by Cyhosia 

 which is structurally distinguishable in antennae and various other 

 particulars. 



In the phylogeny given, however, Philea gives rise to Cyhosia, and 

 Cyhosia to Endrosa, and yet Cyhosia is more distinct than are the other 

 two genera. Not only, for instance, are irrorella and aurita so close 

 as to belong really to one genus, possibly distinct from that containing 

 mesoinclla {Cyhosia), but it vfould be possible to say something as to 

 their even being races of one very variable species. The curious 

 auricular (?) development of the 8rd trochanters in PJ)idrosa and Philea, 

 which is wanting in Cyhosia, is not alluded to, but is a strong proof 

 that Cyhosia cannot be wedged between them. 



The Sagacity of Larva of Galleria mellonella (cereana), Linn. 



By E. HAMLYN HAMLYN-HAEEIS, F.E.M.S., F.Z.S., F.E.S., &e. 

 It is a matter of no small interest to note how the larvfe of certain 

 species of insects show a decided instinct for self-preservation. Among 

 many, more especially among imagines, we are in possession of varied 



