318 Observations upon the Affinities 



lent only to some of the divisions of Heterogenes : yet I con- 

 fess my inability to reduce the series of the latter to groups 

 of a corresponding degree of value ; as a continuity of suc- 

 cession, (though not an intergradation of species), prevails ge- 

 nerally, which effectually resists all attempts at separation, 

 except into a considerable number of divisions. 



Thus, although no passage exists from the motmots into 

 the kingfishers, or into the toucans, and although the king- 

 fishers and toucans differ materially, yet the interposition of 

 the motmots so connects those other two groups, that they 

 cannot be ranged with either of them to the exclusion of the 

 other ; and the toucans, in like manner, combine the charac- 

 ters of the motmots and touracos, &c. Again, the hornbills 

 and hoopoes accord in their alimentary organs with the roll- 

 ers, bee-eaters, and kingfishers, but not in the skeleton ; while 

 the todies and jacamars approximate them in the skeleton but 

 not in the alimentary organs, &c. This much is clear, that 

 the structure of the foot, and even of the bill in many instan- 

 ces, is of considerably less importance than is commonly 

 deemed. 



In my succeeding communications, I will bring together 

 all the information that I have collected, relative to each group 

 of the Heterogenes respectively : but I am not prepared to en- 

 ter into similar details on the Cantrices, the component groups 

 of which being of inferior value, are less obviously different. 

 Moreover, ornithologists have written less on their differential 

 characters, than on those which they have conceived to link 

 the different families together ; which latter being often very 

 superficial, that is to say, consisting merely in a slight resem- 

 blance in the shape of the beak, while everything else is over- 

 looked, as in the instances of a tit, (Falcunculus), being placed 

 with the shrikes, a fringillidous bird, (Calamophilus, which 

 is most nearly allied to Estrilda), with the tits, &c. &c, the 

 overweening consequence which has been attached to such 

 unimportant adaptive characters, has induced an amount of 

 confusion which requires no little study to disentangle. If 

 naturalists, before offering an opinion on the affinities of 

 groups, were but to consider well the totality of characters be- 

 longing to each of those which they suspect to be allied, were 

 but to study all their points of difference, as eagerly as they 

 catch at those of similarity, the progress of scientific ornitho- 

 logy would be greatly accelerated. To exemplify my mean- 

 ing, let me ask upon what data the Anthus Richardi has been 

 deemed to connect the pipits with the larks ? If we trace the 

 progressive changes of these two genera, we find that the larks 

 are clad at first in a very peculiar mottled nestling plumage, 



