70 MR. A. MURRAY OJ* THE GEOGRAPHICAL RELATIONS OF 



of the Coleoptera of Angola, published by ^Erichson in Wieg- 

 mann's 'Archiv,' in 1848, as corrected by Wollaston's removal 

 of the Cape-Verd species, which the death of the collector, 

 Grossbentner, had occasioned to be confounded with the ge- 

 neral Angolan collection, and partly by the collections made 

 by Dr. Welwitsch. The Fellows of the Linnean Society know 

 Dr. "Welwitsch chiefly as a botanist, or, perhaps, through his and 

 M. Morelet's recent work ' On the Mollusca of Angola,' as a 

 conchologist. But his entomological collections are not less 

 admirable in every respect than his botanical; and through his 

 liberality and kindness I have had the advantage of studying in 

 them an amount of material greatly exceeding in extent the 

 collection described by Erichson. From it I am enabled to say 

 that the Brazilian element does not come south into Angola. The 

 type of the Angolan Coleopterous fauna is Caffrarian beyond 

 any question. 



The Brazilian stirps should alone now remain to be treated of; 

 but in speaking of other regions I have already said by anti- 

 cipation every thing that I have to state regarding it. 



There is one point, however, on which I have a general remark 

 to make, which is also specially applicable to it. In Columbia and 

 some of the border-lands nearest the microtypal stream in the 

 Andes we see many fine, rich, glowing metallic species, which, 

 from their size and beauty, we are naturally led to refer to the 

 Brazilian stirps, but which in reality belong to genera which, 

 without doubt, are naturally microtypal, as Harpalus, Philonthus, 

 Xantliolinus, Staphylinus, &c. When we have to determine to 

 which stirps such species beloug, we must discriminate between 

 the natural brilliant elements of the Brazilian fauna and the su- 

 peradded brilliancy developed upon microtypal forms by the 

 special conditions of the locality. Columbia and Ecuador abut 

 on the eastern margin of the microtypal range in Equatorial 

 America, and a modification of the stirps is to be expected there. 

 Mexico, Central America, the West Indies, New Granada, Colum- 

 bia and Ecuador, Cayenne, and in some cases even Venezuela 

 are to be regarded as debatable lands, in which the true character 

 of the stirps to which the species inhabiting them belong is to 

 be determined, not by the place in which they are found, but by 

 a sound consideration of their affinities and distribution elsewhere. 



