476 Mr. G. A. K. Marshall on 



Ihicjil and not clirys'qjpus, a few mimic both, while the 

 great majority mimic chrysippns alone. Again, liitgii 

 stands alone among Danaina, while the pattern of chrysip- 

 pm is closely related to that of several other species and 

 genera, such as t^alatiird. The distribution of llugii can 

 be understood by supposing the desert region of Somali- 

 land to bo its centre. From this point it radiates, 

 towards the east becoming gradually rarer, although it is 

 well known in Karachi, and Colonel Yerbury even captured 

 a single specimen in Ceylon, towards the south finally 

 disappearing in South Africa, towards the Nile Valley, 

 here also probably disappearing towards Central Africa. 

 Chrysippus, or at least forms with a black-and-white apex to 

 fore-wings, on the other hand, occur over the whole vast 

 range of the species with the exception of certain parts 

 of Somaliland.* The strongest argument is, however, one 

 which is developed at the end of the section, because wide 

 conclusions of great interest spring from it (see jjp. 4S2-484). 



The white-hind-winged alcijjjms was for a long time a 

 great difficulty to me, but Mr. Marshall's suggestion (see 

 p. 479) that it is a further development in the direction of 

 still more efficient warning colours than the type-form 

 seems to me to be sound, especially considering its distri- 

 bution in the abundant life of the tropical West Coast, 

 and considering the fact, of which Mr. Marshall assures 

 me, that it is much more conspicuous on the wing. 



I have for a long time thought that this great develop- 

 ment of white, combined with the darkening of the 

 fulvous ground-colour so common in African specimens 

 and marked in alcippus, may indicate an incipient synapo- 

 sematic approach to the black-and-white Danaina of the 

 genus Aiiuvm'is, and the large black-and-white Acnvina} 

 of the genus riancma. 



It is in favour of this view that the darkened ground- 

 colour appears to be a recent development, although it has 

 arisen in the Ethiopian region — tlie ancestral, home of 

 the species, if we may judge by the mucli larger number 

 of mimics which resemble it in this part of its range. I 

 believe the lighter ground-colour of clirysippits in India 

 and, with certain exceptions, the East generally, formerly 



* Consult Dr. A. G. Butler's map of the distribution of the forms 

 of rhriisip2>f"^, ^^^ Plate XfiVI, accompanying his paper in Proc. 

 Zoo). Soc. Lond., 1884, p. 478 ; also hy the same author, Proc. 

 Zool. Soc. Lond., 1885, p. 75G. 



