446 THE ZOOLOGIST. 



That the two groups of birds comprising Fringilla (Chaffinches, 

 Bramblings, &c), and probably some others, on the one hand, 

 and the various and akin genera (Linnets, Canaries, &c.) men- 

 tioned above on the other, having progressed along more or less 

 parallel lines and retaining similar structural form, are classed 

 as equal genera amongst the Fringillince ; but that one at least 

 of the latter group, Ligurinus (Greenfinches), having through 

 evolutionary processes, perhaps somewhat recently, developed a 

 specialized powerful beak, has been erroneouly separated from 

 its relatives and classed with others in a separate subfamily 

 (Coccothraustince). The genealogical tree being presumably 

 something like the following : — 



Common ancestor"' gradual divergence parallel development 



Fringilla. 

 Acanthis. 



Ligurinus. 



The question that occurs to my mind is, Are we to base our 

 classification on visible structural similarities, or invisible germ 

 relationship ? Until I can find any reason at present un- 

 known to me, it appears to me preferable to hold to the latter 

 alternative. 



In the genus Francolinus, again, we have another very 

 diversiform group; although split by some naturalists into 

 different genera, they are generally, and probably incorrectly, 

 classed as one. It seems to contain forms which, from what I 

 could observe of their natural habits, appear to approximate, 

 some to Pcrdix, some to Caccabis, and some superficially to 

 Gallus. While some of the forms more resemble in their habits 

 our Partridge t (Perdix), others as F. adspersus (Scleroptera of 

 Gurney approach in this respect the French Eed-legged Caccabis, 

 of which they seem to be the trans-Saharal representative; 

 they have the build and robust shape of this genus, and their 

 carriage in running is similar. F. pileatus, on the other hand, 



