XCll 



altitudes in Central Ceram (4600-6000 ft.) is evidently a nearly 

 related species to lemnia, and is again invariably ^ ; the insect, 

 taken at the same altitudes, which {loc. cit. 292) I thought 

 must " surely belong " to Eucharidema apora Prout, " in 

 spite of remarkable differences in venation," bears so closely 

 the same relation to posticigutta as does bicolorata to lemnia 

 that I have now no hesitation whatever in associating the 

 two pairs, the more so as an aberrant $ of posticigutta, which 

 I am exhibiting, retains a vestige of the white cell-spot on the 

 fore-wing above and beneath, and some slight suffusion of 

 orange scales on hind-wing beneath. 



Further confirmation is just to hand in the arrival of a good 

 series of a new race of posticigutta collected by the Pratts 

 on Buru, both sexes differing from the name-type in quite 

 similar directions — reduction or suj)pression of the white, 

 yellow or orange markings. 



Finally, B. anisochrysa Prout (Ann. Mag. Nat. Hist. (8), 

 XX, 127, pi. 7, f. 7, Biak, the type unique) shows on the upper- 

 side some general resemblance to the other ?? of the group, 

 while the hind-wing beneath conserves quite the typical 

 colour-scheme of the B. tricolor (Warr.) group, being orange 

 with irregular black band and black border. 



How, then, did the remarkable sexual dimorphism arise ? 

 I do not think it can be doubted that the occurrence of postici- 

 gutta together with Eucharidema apora — which deceived me 

 into taking them for sexes of a single species — furnishes at least 

 a part of the explanation. 



In the type of Eucharidema trichroa Rothsch. and Jord. 

 (Deutsch. Ent. Zeit. 1907, p. 197, British New Guinea) the 

 sexes are quite alike. Of apora, a race or close ally with 

 greatly reduced red area on hind-wing, the Pratts unfortunately 

 took SS only (11 in all). The very wide divergence of the ?? 

 above considered from the normal Bordeta pattern (which 

 is retained by the Sc^, as also on the underside of ? anisochrysa) 

 and their schematic resemblance to Eucharidema points to 

 a very fine case of mimicry between diurnal Geometridae, 

 though the nucleus of the mimetic association may have to 

 be sought in some commoner species outside this family. 

 Of posticigutta 18 S6 and 18 $^ have been taken. 



