( 23 ) 



which required clearing up was the distribution ; for although 

 each tawny species is known to exist in a separate island (or 

 perhaps group of islands in the case of comoraruvi), outside 

 Madagascar, /roSenm and dumetorum have both been stated to 

 occur in Madagascar as well. In settling this point it was 

 convenient to decide with it the precise distribution of the 

 species of Euj^loea in the outlying islands, for some of these 

 have also been affirmed to exist in Madagascar. Upon this 

 subject Monsieur Charles Oberthiir can speak with greater 

 authority than any living naturalist, and he has kindly 

 answered my questions in detail. Writing on April 30th and 

 May 4th, he says : — 



^^ NejJtis dumetorimi. — Tous les exemplaires que j'ai vus 

 viennent de I'ile Bovirbon, Jamais je n'ai vu dumetorimn 

 provenant d'une autre localite. 



'* N. comorarum. — Tous les exemplaires ont ete pris aux 

 Comores ; jamais ailleurs. 



"iV. mayottensis. — J'ai regu seulement un petit nombre 

 pris a Mayotte, Toujours mayottensis." 

 xxxvii] 



Concerning frohenia he writes: — "Jamais Ne})tis froh&nia 

 n'a ete rencontre a Madagascar — du moins d'apres ce que je 

 connais— mais seulement a I'ile Maurice." 



*' Quant aux Euplcea, je crois goudoti speciale a I'ile Boui-bon; 

 eu2)hone, a Maurice ; desjardinsi., a Rodriguez ; et mitra, aux 

 Seychelles. Je possede les 4 especes. Mais jamais je n'ai regu 

 aucun exemplaire de goudoti, ou euphone a Madagascar." 



I am also informed by Mr. H. Grose-Smith, F.E.S., that 

 he has never received any of the following species from 

 Madagascar : — Nej^tis dumetorum, N. frohenia, Ewplota euphone, 

 and E. goiuloti. The erroneous statements as to the occur- 

 rence in Madagascar of certain species of Euplcea and JVepfis 

 confined to the outlying islands, are probably due to Mabille ; 

 but they are unfortunately repeated, although sometimes 

 queried, by Aurivillius. 



Each of the four outlying forms of Keptis may therefore^ 

 be looked upon as the product of its own island, and a very 

 interesting problem of evolutional history is presented to us. 

 One great difficulty in attempting its study was the want of 



