I04 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. 



pied either by land or by sea of no great depth. If the difficulty 

 about the depth of the intervening ocean is overcome, there is no 

 improbability in the suggestion that at some period of geological 

 history an important continent, having connexions with South 

 America, South Afrieaj and New Zealand, may have occupied the 

 Antarctic area. I have already referred to the fact that many 

 biologists regard the present distribution of terrestrial life as 

 evidence of original dispersion from an arctic centre. But unless 

 we are acquainted with the distribution in past times of various 

 groups of animals and plants, there is always a liability to regard 

 the stage on the road of migration from which the present repre- 

 sentatives of any group diverged as the original centre of distribu- 

 tion. Unless we can trace the actual line of migration (and that 

 we may probably never succeed in doing), how are we to tell whether 

 placental Mammals, for instance, appeared first at an arctic centre 

 and diverged thence to Africa, Asia, and America, or whether the 

 original stock came from a southern continent, for instance, South 

 America, and after travelling to the northern hemisphere and mi- 

 grating into Asia or Europe, ramified thence again into the Oriental 

 and Ethiopian regions ? During the period of migration and evo- 

 lution great changes would take place in the country whence the 

 type originally sprang; and as each fresh and improved branch ap- 

 peared, it would spread forward to new regions and backward to the 

 country of its ancestral stock, where it might either exterminate in 

 the struggle for existence those descendants of its own ancestors 

 who had not progressed in structure, or live on beside the more 

 favoured races that had progressed sufficiently to hold their place. 



That some families of living animals may have originated in the 

 southern hemisphere is shown by such examples as the Amphishce- 

 nidce, Aglossa^ and Characinidce, and especially by the Galaxiidce 

 and Haplocliitonidce. 



In this connexion there is one series of palseontological facts of 

 some interest. At particular geological horizons, apparently through- 

 out the world, there is a sudden appearance of terrestrial animals 

 and plants belonging to orders or even subclasses not rei)resented in 

 older strata by any probable ancestors. Amongst the most remark- 

 able instances is the sudden appearance in the Upper Cretaceous 

 epoch, almost or quite simultaneously, of xicanthopterygian fishes 

 and Dicotyledonous Angiospermous plants *, now the dominant and 



* I have beard lately of Dicoiyledonous Angiospernis in Lower Cretaceous 

 beds in America, but the age of the beds appears to have been determined by 

 the Icssil plants, an unsatisfactory method which has often led to error. 



