I20 TYPES OF ANIMAL LIFE 



torn up by the roots. Some of our own modest shrubs 

 we have come to believe merit the same fate, though we 

 have not to answer for much such arboriculture, on 

 account of our having from the first believed in and 

 called attention to the " independent origin of different 

 structures."'* 



It was at one time believed rather widely that the eft 

 group sprung, through the labyrinthodonts, from certain 

 air-breathing fishes, and, in turn, gave rise to frogs on 

 the one band and to beasts and reptiles on the other. 

 This may be all ver}^ true, but as yet we must regard it 

 as a mere speculation, without any sufficient evidence of 

 its truth to hinder us keeping quite '' an open mind " 

 about it. We have already seen, with respect to the 

 opossum's order and class, how two difterent and con- 

 tradictory hypotheses may be suggested by one set of 

 facts. As to the above-mentioned belief we do not think 

 there can be any reasonable doubt about its first article 

 — that batrachians sprang from fishes. What kind of 

 fishes those were, however, we do not know. It is also 

 most probable that frogs did also spring from eft-like 

 creatures, but the utter absence to his time of any 

 discovered links between the two is very remarkable. 



But the once asserted direct afiinity between batrachians 

 and beasts we do not at all believe in, or in any affinity 

 between existing bntrachians and reptiles, though very 

 probably the first reptiles spi'ang directly from some 

 ancestor or collateral relative of the labyrinthodonts. 

 The resemblance of frogs to tortoises and terrapins has 

 often been remarked, and it is remarkable. It is clearly, 

 however, but an instance of the independent origin of 

 similar structures, and a direct descent of tortoises from 

 frogs is quite incredible. It is none the less interesting 



* In our ■' Genesis of Species," 1870. 



