MR. TRIMEN ON MIMETIC ANALOGIES AMONG AFRICAN BUTTERFLIES. 511 



the insular race constant to one pattern, and not differing greatly from the male ; while 

 the African female presents four distinct forms (besides certain intermediate varieties), 

 not one of which resembles the male. It would appear reasonable to argue from this 

 that Madagascar was the original starting-point of this type of Fcrpillo, and that the 

 harder and more complex conditions of African life, causing a severer persecution, had 

 occasioned a necessity for the less active, and perhaps, as now, scarcer ? to assiune the 

 protective colouring and outline of the surrounding Dcmaidce. Yet the very wide 

 dispersion of this butterfly over the continent seems rather to indicate that the original 

 form of Ilerope was of African derivation, and at one time had extended to Madagascar, 

 possibly before that region became insulated, but that since that period, during slowly- 

 changing conditions of life, natural selection has induced the elimination in Africa of 

 aU the pale, conspicuous females of the male coloration, only preserving those that 

 more or less resembled the protected DcmaidcB, — wliile in Madagascar the female, in 

 the absence of any keenly persecuting agency, has retained the form and colour possessed 

 by the first immigrants from the continent. In the broad black costal bar of the fore 

 wings which distinguishes the female in Madagascar, regarded in relation to the hind- 

 marginal black border, it is not difficult to recognize the material upon which natural 

 selection might gradually work, to the ultimate production of a Danaidiform butterfly 

 like Hippocoon or even Cenea ; and it is remarkable that, in all the African forms of the 

 female, an oblique, narrow, whitish marking remains near the extremity of the discoidal 

 cell of the fore wings, in a position exactly corresponding to the outer border of the costal 

 bar, as if to record, with the other pale spots and markings, how the black of the margins 

 had gained upon the ground-colour as the process of increasing resemblance to Danais 

 was slowly wrought out. 



Returning to the subject, from which this has been so lengthy a digression, it is worthy 

 of note that the mimicking Diaclema above described seems only to occur at Natal, and 

 coi'rectly copies the variety of Danais Echena which is there prevalent, viz. that which 

 has all the spots of the fore wings white. 



4. Danais Niavius, Linn. (Tab. XLII. fig. 6.) 



Danais Niavius, Syst. Nat. ii. p. 76G. uo. 109 (17G7) ; Cram. Pap. Exot. t. 2. figg. E, G. 



This is an abundant butterfly in Tropical Western Africa ; but the only special localities 

 that I have found recorded for it are Sierra Leone, Ashanti, and Angola. In the two 

 former of these districts occur two very accvu-ate imitators of Niavius, viz. Diaclema 

 Anthedon, Doubl., and the prevalent West-African form of the ? Pupilio Merope (P. 

 Sippocoon, Fab.). The Papilio has also been received from Calabar. 



It is to this striking case of mimicry that Boisduval refers in the passage wliich I have 

 quoted at the head of this paper. He mentions, it is true, Diadema dttbia ; but this is 

 owing to the confusion that has prevailed regarding the closely allied mimetic Diadema, 

 D. Anthedon being the species concerned, and being easUy distinguished by the very 

 large inner-marginal white patch, and broad subapical bar of the fore Avings. Cramer, 

 as Prof. Westwood has pointed out (Arc. Ent. i. p. 152), figured the Papilio as the $ 

 Niavius, in his Plate 234. fig. A ; and Palisot do Beauvois subsequently did the same 



