218 GEOLOGICAL DISTRIBUTION. 



not more than fifteen or twenty being common to Europe. Evi- 

 dently the oceanic abysses, with their deep layers of cold water, 

 constitute an almost insuperable barrier to the free migration of the 

 animals belonging to this class. It is a little remarkable that 

 no larger jiroportion (about fifty per cent.) of species should be 

 common to the European and American " Boreal " sub-regions. 

 The influence of land-barriers in shaping distribution is still more 

 marked than that of the sea. This is best seen in the case of the 

 Mediterranean and Red Sea moUuscan faunas, where, of eight hun- 

 dred and eighteen species dredged by M 'Andrew in the Gulf of 

 Suez, only three were found to be identical with forms occurring in 

 the Mediterranean ! For a long time it was supposed by naturalists 

 that not a single moUuscan species occurring on the west coast of 

 the Isthmus of Panama reappeared on the Atlantic side, and, if it 

 is now known that this supposition was not absolutely in accord- 

 ance with the facts, it must be admitted that the number of recog- 

 nised transgressional forms, thirty-five out of five hundred to six 

 hundred,'* is very insignificant. With the facts here stated before 

 us, it cannot be doubted that the broad dis2)ersion of animal life 

 in past periods of the earth's history was not only conditioned 

 by favourable climatic circumstances, but, in a marked degree, by 

 the absence of barriers to a free migration. That the amount of 

 land-surface permanently exposed during the Paleozoic era was 

 insignificant, in comparison with that exposed at the present day, 

 is strongly indicated by the vast extent covered by the more 

 ancient marine deposits, from which it would appear that the 

 seas were at that period practically continuous throughout their 

 broadest expanse. But this general supposition is based upon the 

 hypothesis of the permanence of continents and oceanic basins, 

 which, as has already been remarked, has much in its favour, but 

 is still far from being in the nature of a demonstration. Grant- 

 ing the probability of continental upheavals from the oceanic 

 abysses, then, naturally, must all conjectures regarding the rela- 

 tionship between land- and water-surfaces, and the existence of 

 interposing barriers, be valueless. It may be hastily concluded 

 that the circumstance of very broad distribution is, in itself, strong 

 evidence tending to show that such continental elevations did not 

 take place, and that there was a true permanency in the position 

 occu2:)ied by the sea. This need not necessarily have been the 



