Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 469 



arctic) types in the New Zealand flora is best explained on the supposition that, 

 under more favourable climatic conditions than at present exist, they once occupied 

 the now barren Antarctic Continent, and have spread northwards to New Zealand 

 and Tasmania. And even if the great depth of the ocean to the northwards of 

 Antarctica should seem to preclude the possibility of a continuous land bridge, it 

 is legitimate to speculate on the probability of the Antarctic Continent having ex- 

 tended considerably to the northwards of its present limits, and of New Zealand 

 having stretched southwards over the plateau on which the southern islands now 

 stand. On this supposition we provide a practicable route for the passage of the 

 so-called " antarctic types " northwards, and we supply a sufficiently large land- 

 area for the gradual development of the endemic portion of the flora of the southern 

 islands. 



But, granting the previous southwards extension of New Zealand, there are 

 good reasons for believing that it occurred far back in geological time. The large 

 percentage of endemic plants in the flora of the islands, and the very distinct character 

 of many of them, unmistakably point to a lengthened period of isolation. Again, 

 each of the separate islands possesses its peculiar species, a fact just as distinctly 

 indicating a prolonged isolation of the individual islands. In other words, the peculi- 

 arities of the endemism of the islands go far to prove that they have not only been 

 isolated from New Zealand for a long period, but that their separate existence has also 

 been of lengthened duration. I do not mean by this to imply that the islands may not 

 have been of larger size during comparatively recent times. On the contrary, the 

 abundant signs of denudation everywhere visible around their shores show that their 

 area has been greatly reduced by the stormy waves of the Southern Ocean. Sir James 

 Hector has calculated, for instance, that the Auckland Islands had a diameter of not 

 less than fifty miles in early Tertiary times. 



Another argument against any recent land connection can be drawn from the 

 composition of the New Zealand element in the flora of the islands. Over 68 per 

 cent, of the total flora consists of species at present living in New Zealand. But 

 when these are examined in detail it is impossible not to feel surprised at the absence 

 of many well-known New Zealand plants, to all appearance quite as well adapted 

 for existence on the islands as any that now live there. If the New Zealand element 

 had travelled across a continuous land bridge, the absence of these plants is inex- 

 plicable ; but if, on the other hand, the immigration was trans-oceanic, then only 

 those species would cross that possessed adaptations favouring their dispersal in that 

 manner. 



One important fact has not yet been referred to : it is the influence on the 

 character of the flora which must have been exercised by vicissitudes of climate. 

 Palaeontological facts prove that the Antarctic Continent once supported a numerous 

 phanerogamic flora, the existence of which, under its present climate, has become 

 an utter impossibility. There is little doubt that the Kerguelen - South Georgia 

 groups of islands have recently passed through a climatic period so severe as to 

 warrant the belief that they were covered by an ice-cap almost as complete as that 

 which now enshrouds the Antarctic Continent. In the case of South Georgia, the 

 whole of the phanerogamic flora perished ; while in Kerguelen only a few species 

 survived, now constituting the endemic portion of the present flora. Unfortunately, 

 previous visitors to the southern inlands do not seem to have specially attended to 



