RESULTS OF NUMERICAL EQUALITY IN TRIBES. 233 



a race. If, then, all the fossil Pachydermata were 

 alive, and were incorporated, according to their 

 affinities, with those now living, the contents of the 

 whole group would probably be augmented to four 

 or five times its present number ; and those chasms, 

 which now appear so wide, would be proportionably 

 lessened ; nay, it is highly probable they would not 

 be greater, in proportion, than those between the 

 different genera of the parrots. But, secondly, let 

 us suppose that it was essential to the symmetry 

 and harmony of nature, that all her groups of the 

 same rank and value should contain pretty nearly 

 the same number of species, and that their numerical 

 contents should be proportionate to their value. 

 What, in the present instance, would be the result ? 

 The tribe of Scansores, or climbing birds, includes 

 the parrots ; and, upon a rough estimate, certainly 

 contains between four and five hundred species. 

 We know, by induction, that this tribe is equivalent 

 to that of the pachydermatous quadrupeds. Now, 

 if these tribes were as equal in their contents, as they 

 are in their rank, more than half the earth would be 

 overrun with monsters. Elephants would be as 

 common as flies ; we should have to reckon not ttvo, 

 but perhaps two hundred species. All the large 

 rivers would be almost choked with hippopotami. 

 Rhinoceroses would swarm in the woods, in herds 

 of thousands, as the parrots do now in the forests of 

 America. And huge megatheri, perhaps of a hun- 

 dred species, would attack a forest, and strip it of 

 its verdure in a few clays. The world, in fact, 

 would be filled, as it once was, with monstrous 

 animals ; and man would find no resting place in it. 



