448 HOW FAR STRATA MAY BE KNOWN BY FOSSILS. 



hypothesis we have supposed the conditions of the sea-bed to remain 

 essentially the same in the locality R ; but it is obvious that depres- 

 sion might produce in that region deep-sea conditions, and cause the 

 influx of a deep-sea fauna, or that elevation might develop laminarian 

 or littoral conditions, so that the variations of the fauna which are 

 possible, without seriously affecting the distribution of land and water 

 on the earth's surface, are more numerous than might at first sight 

 appear. Similarly, when a land emerges from the sea, the life which 

 covered that area is displaced, and the rising land divides the marine 

 fauna, so as to cause the group to be pushed apart as the island 

 enlarges into a continent, until by following the littoral or laminarian 

 zones which offered the conditions of least resistance to their move- 

 ment, it has happened that the fauna, which was originally temperate, 

 has in part become arctic ; in another part has become tropical ; while 

 temperate provinces are found in temperate latitudes. If a land sur- 

 face becomes enlarged, the life diffuses itself, so that the species are 

 few relatively to the area ; but if the land is undergoing depression, 

 then the fauna and flora become rich, because the area of land 

 decreases without any corresponding decrease in the variety of life. 

 The whole doctrine of the identification of strata by means of organic 

 remains, rests upon the assumption that the strata identified were 

 deposited in the same natural history province. If strata are defined 

 to be of the same age from their fossils, it can only be because the 

 old life province can be traced continuously from one point to the 

 other. But great practical difficulties result from the imperfection 

 of our knowledge of the past distribution of life ; for though 95 per- 

 cent, of the shells of the Ked Crag are still living, a much smaller 

 percentage of British shells will be found in the fauna of Portugal ; 

 so that contemporary fauna*, even when separated by moderate dis- 

 tances, may have less in common than faunas of different geological 

 ages which are superimposed on each other. 



Identification of Strata by Fossils. As taught by William Smith, 

 the identification of strata by fossils was a matter of observation. He 

 used fossils as an aid in tracing the strata through England. The 

 results, as displayed in his books and maps, were so remarkable, that 

 the principle, which in his hands was sound, came to be regarded as 

 a universal law, though no conditions were imposed on its application. 

 The British strata were taken as types, and British fossils found in dis- 

 tant places spread William Smith's names for the beds over Europe. 

 And before long the great divisions of geological time, thus marked out 

 conveniently, came to be recognised over distant parts of the globe, and 

 their ages were determined by their fossils. This was not a scientific use 

 of fossils, for it ignored the origin and distribution of species. It led to 

 the theo-geological conception of successive creations and extinctions 

 of life on the earth, which marked off the range of species exactly. 

 But when it came to be seen, by detailed examination of strata, that 

 there was no such homogeneous group of life in a period of time, but 

 that the life changed all through a formation, approximating to the 

 older life in the bottom beds, and to newer life in the top beds, we 



