REASONING IN A CIRCLE. 339 



"But," it will be replied, "in past eras the same, or 

 similar, organic forms were more widely distributed than 

 now." It may be so ; but the evidence adduced by no 

 means proves it. The argument by which this conclusion 

 is reached, runs a risk of being quoted as an example of 

 reasoning in a circle. As already pointed out, between 

 formations in remote regions there is no means of ascertain* 

 ing equivalence but by fossils. If, then, the contempora- 

 neity of remote formations is concluded from the likeness 

 of their fossils ; how can it be said that similar plants and 

 animals were once more widely distributed, because they 

 are found in contemporaneous strata in remote regions ? 

 Is not the fallacy manifest ? Even supposing there were 

 no such fatal objection as this, the evidence commonly as- 

 signed would still be insufficient. For we must bear in 

 mind that the community of organic remains commonly 

 thought sufficient for inferring correspondence in time, is a 

 very imperfect community. When the compared sedimen- 

 tary beds are far apart, it is scarcely expected that there 

 will be many species common to the two : it is enough if 

 there be discovered a considerable number of common gen- 

 era. Now had it been proved that, throughout geologic 

 time, each genus lived but for a short period — a period 

 measured by a single group of strata — something might be 

 inferred. But what if we learn that many of the same 

 genera continued to exist throughout enormous epochs, 

 measured by several vast systems of strata ? " Among 

 molluscs, the genera Avicida, Modiola, Terebratida, Ijin- 

 gula^ and Orhicula^ are found from the Silurian rocks up- 

 wards to the present day." If, then, between the lowest 

 fossiliferous formations and the most recent, there exists 

 this degree of community ; must we not infer that there 

 will probably often exist a degree of community between 

 ■trata that are far from contemporaneous ? 



Thus the reason in 2: from Avhich it is concluded that 



