324 Transactions South African Philosophical Society, [vol. xiii. 



This division, like all divisions resting on one or two characters 

 only, is open to grave objections. The dilatation of apex of the basal 

 joint of the hind tarsi is found occasionally in the female only 

 (Micrelaphinis) ; a slender form like Homalothyrea, the general 

 appearance of which is that of Leucocelis, or a depressed one like 

 Tephrcea, will have to be interpolated in the vicinity of Anoplochilus 

 or Psacadoptera, with which they have no relationship whatever, &c. 

 On the other hand, in the second division we are led to include, as 

 Kolbe himself has done (Die Kafer Deuts. Ost. Afrikas, 1897), the 

 Cremastochilides, the habits and facies of which are so very 

 peculiar, with the Ceratorrhinides, Schizorrhinides , &c. This is a 

 most unnatural grouping. 



But if we associate with the two characters mentioned by Kolbe 

 others, such as (a) the presence or absence of an antecoxal pro- 

 sternal vertical process, and (b) the shape of the elytral costae, when 

 they exist, and especially the sculpture of the first two, or more, 

 dorsal intervals, we find that, in so far as the African Cetonini are 

 concerned, the arrangement which I propose brings together species 

 and genera the affinities of which are made evident by close examina- 

 tion. I may add that I believe that this grouping will prove equally 

 natural when applied to the Cetonini of the whole world. 



A linear arrangement is never a perfect one ; thus, for instance, 

 Elaphinis is greatly removed from Micrelaphinis, and Clinteroides 

 from Pseudoclinteria, of which they are undoubtedly near relatives. 

 It is owing in the first case to Micrelaphinis having non-geminate 

 elytra, striae, and a broadly dilated basal joint of the hind tarsi in 

 the 2 , two characters lacking in Elaphinis. In the case of the 

 other two genera which have in common a scutellum covered by 

 the base of the prothorax, only one character is resorted to for 

 separating them, viz., the non-dilatation, in Pseudoclinteria, of the 

 basal hind tarsal joint, this character being strongly in evidence in 

 Clinteroides; although the arrangement I propose relegates these 

 genera in two different sub-divisions, it should not be inferred that 

 their affinities are impaired thereby. It is only an instance of the 

 homogeneity of the Cetonini ; others are Gametis (Cetonides) 

 and Didcinea (Goliathides) ; Discopeltis (Cetonides) and Glaucocelis 

 (Leucocelites), &c. 



Key to the Sub-Tribes. 



A 2 . Presternum without an ante-coxal vertical process. 



Spiracle of last abdominal segment plane * Cetonides. 



One species of Mausoleopsis excepted. 



