77 [Vol. XXXV. 



much alike in all, though the colour in the different species 

 included varies greatly. It wouM be easy to divide this 

 group into five or six sections on colour alone, but I do not 

 think any advantage would be gained by doing so. Colour, 

 in ray opinion, is of no importance generally, but colour- 

 pattern is, and I therefore keep the whole of these species 

 together in one genus — Munarcha. There is no reason why 

 a genus should not include a hundred species or more, pro- 

 vided they constitute a well-defined group. By splitting up 

 groups of species into endless unnecessary genera, we are 

 only overburdening our memories with useless names and 

 rendering classification more difficult, thus retarding what 

 we are trying to advance — the study of ornithology. 



Mr. W. L. ScLATER : I have got very little to add to 

 the discussion. My views are almost entirely those of 

 Mr. Ogilvie-Grant. I think genera are a matter of con- 

 venience more than anything else, and a genus is entirely 

 a human conception, and does not exist in Nature at all — • 

 it is purely artificial. We can probably define a species and 

 a subspecies, but a genus is merely a number of species put 

 together for our own convenience. 



As regards Dr. Lowe's views that generic characters 

 should be based on colour-pattern, I must say I agree with 

 him. I think colour-pattern is often a very ancient and 

 deep-seated character, and obviously colour-pattern must be 

 a much more primitive character than the relative lengths 

 of the tarsus and the middle toe or the relative width and 

 length of the bill. These characters are easily modified 

 by external circumstances, and you cannot regard these 

 characters as more deep-seated than colour-pattern. 



Mr. Pycraft's views about the necessity of studying the 

 anatomical characters, the skeleton and the anatomy of the 

 soft parts of the birds, are absolutely true, but I do not 

 think they are, as a rule, of use in generic separation. These 

 characters are surely those which are of importance for 

 family distinction. You must always recollect Birds are a 

 group in which evolution has worked its will to a great 

 extent. We have an enormous number of slightly varying 



