306 



JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



further we study it the more we shall be inclined to arrive at a similar 

 conclusion, and to account for the existence of these species in Ireland — 

 and one of the most characteristic of them, the Pipewort, in western 

 Scotland also — by the assumption of their migration across an ancient 

 land surface which once extended across the North Atlantic, via Iceland 

 and Greenland. 



Our study (Ji the West of Ireland flora has led us far afield indeed. 

 Behind these modest, unfamiliar wild-fiowers, which one meets on the 

 Gonnaught roadside, loom problems which take us back through tens 

 of thousands of years, and which involve vast changes in the distribution 

 of land and sea. Indeed, as we have seen, to understand the con- 

 ditions which have determined the presence or absence of species in the 

 district we have been considering, we have to look back to a period 

 almost infinitely remote — back through the Tertiary and Secondary 

 ] >eriods of the geologist to those primeval times w^hen the only vertebrate 

 inhabitants of our globe were the armour-plated fishes of Silurian seas. 



