430 Professor E. B. Poiilton on 



portion of both hippocoon and tropli07iius than any other. 

 Looking at the six families as a whole, five yield con- 

 cordant results in the vast predominance of cenea, while 

 Family 4 stands apart. And even in this latter cenea is 

 nearly three times as numerous as either of the other 

 forms. 



The immense preponderance of cenea over the other 

 female forms in Natal has been observed in the field 

 (Trans. Ent. Soc. Lond., 1904, pp. 687, 688). It follows 

 from this predominance that for every male which meets 

 and pairs with hippocoon or trophonius in Natal, several will 

 meet and pair with cenea. Cenea ancestry will quickly 

 predominate over that of the other forms in the males, and 

 will also predominate in the other female forms them- 

 selves, while these latter will exert but little influence in 

 the ancestry of cenea. 



The two models of the cenea female form — Amauris 

 alhimacnlata and A. ccJieria — are immensely predominant 

 over any other Danaine butterfly in Natal, and especially 

 over Amauris niavins, L., subsp. clommicanus, Trim., the 

 model of the hipp)Ocoo7i form. Tlie comparative rarity of the 

 trophonius form in all parts of Africa, in spite of the wide- 

 spread abundance of Danaida {Limnas) chrysipyus, L., has 

 already been alluded to by Mr. Roland Trimen, F.R.S., as 

 a difficult problem which awaits solution (Trans. Ent. 

 Soc. Lond., 1904, p. 688 : see also p. 432 of the present 

 memoir where the solution is attempted). In marked 

 contrast with tropJionius, the relative proportion of cenea 

 in Natal certainly follows that of the two Danaines whose 

 pattern it reproduces. 



The interpretation offered above of the state of things 

 proved to exist in Natal derives strong support from an 

 investigation of these proportions in and around Chirinda 

 Forest, Gazaland, in South-Eastern Rhodesia. A very 

 large collection made, almost entirely in 1907, by Mr. C. 

 F. M. Swynnerton, in this locality, has been recently ex- 

 amined in the Hope Department, and the numbers are 

 sufficient to admit of fairly safe conclusions. The investi- 

 gation and tabulation of the collection is still incomplete, 

 and the figures given below will probably be slightly in- 

 creased, but not to an extent which will affect the conclusions 

 here drawn. The collection was made nearly indiscrimin- 

 ately, and allowing for the considerations mentioned on pp. 

 431, 432, the proportions of the larger species, here alone 



