458 Mr. N. A. Vigors on the Natural Affinities 



other would leave us in some doubt as to where the line of de- 

 marcation may be drawn between them, did not the conforma- 

 tion of the foot of the CerthiadcB evidently evince a deviation 

 from the perfect structure of the more typical Scansores, and thus 

 distinguishing them as an aberrant group of the tribe, make it 

 necessary that they should be placed in a separate family. 



The different structure of the foot from that of Ficus thus 

 brings us among the Certhiadce. Here that member is not 

 strictly scansorial. But a similar assistance to what is conferred 

 on the FicidcB in climbing, by the two toes being placed behind, 

 is afforded the Certhiada by the single hind toe being consi- 

 derably longer and more robust than is usual among the Perch- 

 ers. The affinity that brings these birds together in a natural 

 group is thus preserved, not by an identical, but by a similar 

 and equally effective mode of conformation. I feel some diffi- 

 dence, I must confess, in infringing upon the usual and more 

 regular rules of systematic arrangement, by placing this family 

 among the birds with true scansorial feet : but I am influenced 

 by a consideration paramount to that of uniformity of system ; 

 by observing, in fact, that the whole group of climbing birds be- 

 fore us is united by strong natural affinities. And when I perceive 

 a series of natural objects thus united, I draw the conclusion that 

 it is the group which affords the character, and not the character 

 which constitutes the group. — " Scias," says the great master of 

 natural science, " characterem non constituere genus, sed genus 

 characterem : — characterem non fluere e genere, sed genus e cha- 

 ractere : — characterem non esse ut genus fiat, sed ut genus nosca- 

 tur*." — In a group strongly and naturally united as that before 

 us, the evanescence in any of its subordinate subdivisions of any 

 particular character common to the rest, does not prove that 

 the subdivision in which such a character disappears is incon- 



♦ Linn. Phil. Bot. p. 1 19. 



gruous 



