strata there forming, its remains will disappear; they will 

 be absent from some of the supposed strata ; and will be 

 found in strata higher up. But in whafr shapes will they 

 re-appear? Exposed during the 21,000 years of their slow 

 recession and their slow return, to changing conditions of 

 life, they are likely to have undergone modifications ; and 

 will probably re-apipear with slight differences of constitu- 

 tion and perhaps of form — will be new varieties or perhaps 

 new sub-species. 



To this cause of minor breaks in the succession of or- 

 ganic forms — a cause on which we have dwelt because it 

 has not been taken into account — we must add sundry oth- 

 ers. Besides these periodically-recurring alterations of 

 climate, there are the irregular ones produced by re-distri- 

 butions of land and sea ; and these, sometimes less, some- 

 times greater, in degree, than the rhythmical changes, must, 

 like them, cause in each region the ebb and flow of species ; 

 and consequent breaks, small or large as the case may be, 

 in the palseontological series. Other and more special geo- 

 logical changes must produce other and more local blanks 

 in the succession of fossils. By some inland elevation the 

 natural drainage of a continent is modified ; and instead 

 of the sediment it previously brought down to the sea, a 

 great river begins to bring down sediment unfavourable to 

 various plants and animals living in its delta : wherefore 

 these disappear from the locality, perhaps to re-appear in a 

 changed form after a long epoch. Upheavals or subsiden- 

 ces of shores or sea-bottoms, involving deviations of marine 

 currents, must remove the habitats of many species to 

 which such currents are salutary or injurious ; and further, 

 this re-distribution of currents must alter the places of sed- 

 imentary deposits, and so stop the burying of organic re« 

 mains in some localities, and commence it in others. Had 

 we space, many more such causes of blanks in our paljeon- 

 tological records might be added. But it is needless here 



