Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 459 



and a few cryptogams, which now constitute the endemic portion of the present flora. 

 After the removal of the ice-cap the island again became suitable for the support 

 of a more numerous vegetation, which, of course, could only reach it by transport 

 across the ocean. The violent westerly winds, blowing almost continuously through- 

 out the year, effectually barred any immigration from South Africa, which is much 

 the nearest continent, or from Australia, which comes next in point of distance, 

 leaving South America, the most remote of the land-masses, as the only practicable 

 source of supply. 



Schimper admits that the distance is too great to allow the seeds of phanerogams 

 to be transported by the wind alone, although the spores of certain pteridophytes and 

 other cryptogams have probably been brought in that manner. Ocean-currents 

 appear to be out of the question. But in the pelagic birds which exist in such 

 numbers in the Southern Ocean he finds a more probable means of conveyance, and 

 concludes that through the agency of albatroses and other large oceanic birds, driven 

 before exceptionally violent westerly gales, Kerguelen has gradually received the 

 non-endemic portion of its flowering-plants. In support of this view, he points out 

 that most of the Kerguelen plants have seeds specially adapted for such a mode of 

 transport. The " burrs " of Acaena, the barbed utricle of Uncinia, the hooked 

 style of Ranunculus and Azorella, and the awned spikelets of the grasses are all 

 suited for clinging to the feathers of birds ; while the minute seeds of Montia, Tillaea, 

 and Callitriche are very likely to be conveyed in mud adhering to the feet of birds 

 visiting the marshy localities in which they grow. It will be difficult to explain why 

 Kerguelen has such a large proportion of plants having seeds suitable for carriage 

 in this manner, unless the view is adopted that their presence in the island is due to 

 their possession of these adaptations. Schimper closes his remarks on the history 

 of the vegetation of Kerguelen by stating his conviction that since the beginning 

 of the Tertiary period there has been no marked extension of the Antarctic Con- 

 tinent, and that the supposition so commonly entei-tained of the former existence of 

 a vast land-mass uniting it with the other continents is as chimerical as the belief 

 in the fabled Atlantis, or in the hyf othetical Lemuria of some naturalists. 



I have quoted Schimper's views at some length, on account of the clearness and 

 cogency of his arguments, and the evident bearing which they possess as regards the 

 vegetation of the islands to the south of New Zealand. But when the numerous 

 points of diflerence between the two sets of islands are fully considered, and par- 

 ticularly when their very different position with respect to other lands is taken into 

 account — one removed by thousands of miles of barren ocean, the other compara- 

 tively close to a land well stocked with plants and animals — we shall naturally find 

 that a different set of facts will require, in some measure, a different set of ex- 

 planations. 



We have found that the origin of the present flora of the Kerguelen - South 

 Georgia groups of islands can be traced to Fuegia. Now, it is well known that a 

 Fuegian affinity exists in the New Zealand fiora ; and when we come to examine 

 in detail the vegetation of the islands to the south of New Zealand we shall see that 

 this affinity is present in a much more pronounced degree. It is therefore necessary 

 to make a few remarks on the general character of the vegetation of that portion of 

 South America to the south of the Gulf of Penas (S. lat. 47°), which is usually taken 

 to mark the northern limit of the typical Fuegian vegetation. The most recent 



