458 SUBANTARCTIC ISLANDS OP NEW ZEALAND, [Systematic Botany. 



Trie vascular plants of the Kerguelen - South Georgia group of islands thus 

 number thirty-seven. From the point of view of their geographical distribution 

 they can be classed in the following manner : — 



1. Six endemic species, showing a percentage of 16'2. Two of these {Pringlea 

 antiscorbutica and Lyallia kerguelensis) constitute monotypic genera, the first with 

 no near allies, the second related to the Andine Pycnophyllum. The remaining four 

 are more or less closely allied to South American (Fuegian) species. 



2. Twenty-seven species also found in Fuegia, the percentage to the entire flora 

 being 70"9. Six of these are absolutely confined to the two localities. 



3. Twenty species found in the New Zealand group of islands, a percentage 

 of 54-1. Of these, three are found nowhere else. 



These figures show very clearly that the chief affinity of the florula is with Fuegia, 

 nearly three-quarters of the whole number of species being found in that country. 

 Assuming, therefore, that the majority of the vascular plants of the Kerguelen- South 

 Georgia Islands have been derived from South America, the question at once arises 

 as to how their transport was affected. Kerguelen Island is situated more than 

 4,000 miles from Cape Horn, a distance so great that most botanical geographers 

 have refused to entertain the idea of seeds being carried by the prevailing westerly 

 winds or floated by ocean-currents. This has led to the supposition that land once 

 existed between Fuegia and Kerguelen, either directly connecting the two localities, 

 or in the shape of detached islands placed at comparatively small distances apart. 

 This view, first advocated by Sir J. D. Hooker in the " Flora Antarctica," has been 

 wholly or partially accepted by most writers on the subject, apparently more on 

 account of the absence of any other hypothesis offering a reasonable explanation of 

 the facts than from any arguments which can be deduced from the present contour 

 of the ocean-bed, the geological history of the islands, and the precise nature of 

 their inhabitants. A few years ago, however, the question was attacked by the 

 late A. F. W. Schimper, the talented author of " Plant Geography upon a Physio- 

 logical Basis," and, although he did not live to publish the results of his inquiry, 

 they have lately g,ppeared, under the careful editorship of Dr. Schenck, in one of 

 the memoirs descriptive of the work of the German Deep-sea Expedition in the 

 " Valdivia." As Schimper's memoir has a close application to the islands to the 

 south of New Zealand, it is advisable to give a brief sketch of the conclusions he 

 arrived at. 



He points out, first of all, that the present flora (and also the fauna) consists 

 of two distinct elements — first, an endemic element of considerable antiquity, com- 

 prising a fair number of mosses and other cryptogams, but with respect to flower- 

 ing-plants limited to the isolated Pringlea, and possibly Lyallia and Colobanihus 

 kerguelensis ; secondly, the remainder of the flora, consisting of species almost 

 wholly found in Fuegia, and presenting all the appearance of recent immigrants. 

 He then proceeds to show that the configuration of Kerguelen and its physiography 

 amply prove that in comparatively recent times, and after it had assumed its 

 present limits, it has been subjected to glaciation on such an extensive scale that 

 the island was probably covered by an ice-cap almost as complete as that which now 

 enshrouds the Antarctic Continent ; and that Heard Island, to the south of Ker- 

 guelen, is now just emerging from this glacial period. Unable to endure such a severe 

 climate, the original flora perished, with the exception of two or three flowering-plants 



