LHI 



side of two fairly large occelli in the centre of each of the hind-wings a third 

 appears, being confluent with one of the former. The latter phenomenon is 

 one of evolutional change of form not of infrequent occurrence where the ocelli 

 are strongly developed and often to be observed on the lower surface of 

 Lycaenidae, which has been noted as such by Prof Dr. Couvoisier of Basle 

 in his paper " lleber Zeic/nmiigs-Abcrra/ionrn bci Lycaeniden" {Zcihchrift fi'ir 

 Wissenschaftliche Insektenbiologie, igof). This confluence of ocelli may plainly 

 be observed in many Ypthima species ; in some specimens they may thus be 

 seen as two ocelli joined side by side, being already united where they are in 

 contact ; in more advanced stages they are amalgamated into one ocellus which 

 is still elongate in shape, however, while next this becomes more and more 

 rounded in outline until finally a single large ocellus results. This does not, 

 however, contain a single, but two nuclear points, the nuclei of the two original 

 ocelli; which, therefore, demonstrate the fact of this confluence. Not always, 

 however, is the presence of two white dots in such an ocellus due to this 

 origin. In Neorina Chrisna Westw. a large ocellus may be observed on 

 the apical portion of the fore-wings on the upper surface. This ocellus is of 

 an intense black with yellow border and sometimes with a white nuclear dot in 

 the centre of the black. Now two larger white dots invariably occur, not in 

 the centre, but in the upper and lower part of the yellow border. Consequently 

 they are certainly not nuclei. Moreover, they are not placed within the ocellus 

 proper, but why such white scales should appear just in these places remains 

 unsolved. The difference in the occurrence of these ocelli in butterflies of the 

 genus MoRPHO, quoted by Bateson, is also similar in character. The species 

 of this genus mentioned by him are closely allied, and they are, therefore, 

 subjected to the same evolutionary processes ; in which connection sometimes 

 a few more ocelli occur in one individual than in another. 



These individual differences in size or number of ocelli occur also to a 

 considerable extent in the Java Satyridae, such as Mycalesis Perseus F. ; this 

 is most conspicuous in Cyllo Leda L., so common in that island, and a careful 

 study of this butterfly clearly demonstrates that in this case they are purely 

 phases of colour development, and that of some unlimited or irregular so-called 

 variability there can be no question. In Cyllo Leda L. the lower surface 

 exhibits on the fore-wings 4 and on the hind-wings 6 ocelli which I have indi- 

 cated with numerals in the accompanying illustration. Of those on the fore- 

 wings n^. 3 is the largest and the others are mostly very small. On the 

 hind-wings n''. 5 is the largest, n*'. 7 and n". g being next in size and the 

 remaining three considerably smaller; n**. 10 is sometimes duplicated and n". 8 

 mostly placed further towards the outer margin than the others. Individual 



