Family and Generic Differentiation. 345 



I am not advocating a policy of sinking or ignoring genera 

 where obvious generic distinctions exist. That would, I 

 tliink, be a very dangerous and regrettable policy. But I do 

 think we ought to have some method of emphasising generic 

 relationships inter se, by way of combining them into groups. 



It may be answered that subfamiiy divisions already do 

 this. But take the case of these five genera of the larger 

 Caciques or Oropendolas. Here we have a very distinct 

 group of the Icteiidfe, well difl'erentiated as regards 

 structural generic characters, as regards specific characters, 

 and as regards habits. For instance, in connection with 

 the last, the Oropendolas only rear one chick, which presents 

 peculiar characteristics, and it is only on very rare occasions 

 that two eggs are found in their very peculiar nests. 



They form, therefore, as I have said, a very distinct group, 

 but, at the same time, it has been generally held by recent 

 systematists that it is impossible to subdivide the family of 

 the Icteridae into subfamilies. 



The obvious or logical conclusion, therefore, is that we 

 ought to more often make use of super-genera by way of 

 expressing or emphasising the near phylogenetic relation- 

 ships of groups of genera. 



That any real progress in this direction is retarded or 

 rendered practically impossible is due to the fact that at 

 present it is only ou very rare occasions and quite sporadi- 

 cally that we find nestlings or immature examples of species 

 represented in our collections. 



Finally, if colour-pattern is going to be applied as a 

 generic factor on anything like a large scale, it will, I am 

 afraid, be fovmd necessary to create many more genera or sub- 

 genera than exist now. But why not, if, and so long as, these 

 genera or subgenera were found to be natural groups, cor- 

 responding, as I have found in many instances they do, 

 with faunal areas or geographical regions or subregions; 

 and if, and so long as, such subgenera or genera are 

 integrated into larger natural groups, genetic phyla, or what- 

 ever term is employed, so that we may be aware of their 

 natural relationships. These minor groups of birds are not 



SEK. X. VOL. Ill, 2 A 



