University of California. 



[Vol. i. 



stitution of one species for another (of course following the line of 

 evolution), rather than a transformation of one species into another. 

 The same thing is observed in geographical distribution of species. 

 In fact, there is great similarity in the laws of geographical and geo- 

 logical distribution of species, — the distribution in space and the 

 distribution in time. Critical periods with their unconformities in 

 the latter, correspond to great physical barriers in the former, for in 

 both they separate sharply distinct faunas and floras. On the other 

 hand, conformity with its gradual changes of species in the one cor- 

 responds with gradual changes of physical conditions in latitude and 

 corresponding changes in species when there are no barriers. In 

 both cases the changes of species are gradual, indeed, but usually, 

 though not always, by substitution rather than by transformation. 

 In the case of geographical distribution we easily account for this 

 by the dropping out of intermediate forms in the struggle for life in 

 all cases, and by migrations in many cases. How shall we account 

 for it in the geological distribution ? 



Of course, in perfectly conformable strata, in case of change of 

 material of the strata, we account for it largely by local migrations 

 through change of bottoms. But the same is often true even in a 

 continuous unchanged stratum. Of course, again we may attribute 

 it, as Darwin, and many since Darwin have done, to the extreme 

 fragmentariness of the geological record. But even when the 

 record is complete, as it sometimes is, and the number of fossil 

 species is fairly comparable to the number of living species in simi- 

 lar area and time, still the same fact is observed. There must be 

 some other cause inherent in the very nature of the process of evo- 

 lution itself. On this most difficult subject I throw out the follow- 

 ing suggestions: — 



I have already said that while the forces and laws of nature are 

 uniform in their operation, yet phenomena, being usually under the 

 influence of two opposite forces, one tending to change, the other 

 to permanency, the one progressive, the other conservative, are 

 nearly always more or less paroxysmal. Resistance at first prevails 

 and there is little or no change; but forces of change are meanwhile 

 accumulating until finally resistance gives way and conspicuous 

 changes take place rapidly. This might, if necessary, be illustrated 



