DERIVATION OF THE FLORA AND FAUNA 323 



It is surprising that, fourteen years later, he vvas converted to wegenerianism 

 in the form modified by Du ToiT, thinking that "this theory would best explain 

 most of the problems in the geographical distribution of the floras of the Southern 

 Hemisphere" (^7.70) — he seems to have forgotten how badly Hawaii fits into 

 the picture. 



GuiLLAUMiN, in his paper on the floristic regions of the Pacific {iiS), devotes 

 a chapter to its geotectonic history. 



Un vaste continent parallele a I'equateur a du relier les regions australiens, canaque, 

 et neozelandaise a I'Amerique du Sud tandis que la region malayo-polynesienne et le 

 domain hawaiien, formant peut-etre un continent, en etaient separes. 



The dissolution and disappearance of this hypothetical continent took place 

 before the end of the Jurassic, so that its direct influence on the distribution of 

 angiosperms must have been slight; Arldt's "Ozeanis" was more useful. The 

 Melanesian extension of Australia-New Zealand, including Fiji, Kermadcc, Tonga 

 etc., persisted longer and is perhaps better founded, but when GuiLLAUMiN thinks 

 that his hypotheses "ne sont pas en contradiction avec les donnees geologiques 

 ni avec ce qu'on salt du relief sousmarin du Pacific", he moves on unsafe ground. 

 Most authors who deny any considerable reduction of the surface of the Pacific 

 admit that Melanesia forms an exception. Thus Guppy (121) who otherwise is a 

 firm believer in oversea migration thinks that "we should rope in Fiji with all 

 the large islands westward and southward as originally continental" because they 

 lie inside the gymnosperm line. All non-coniferous ones are excluded. Germain 

 (/05. 973) goes much farther. Discussing the Hawaiian region he thinks that it 

 may have extended far eastward. The distribution of some animals, for instance 

 the eel, proves that Hawaii vvas "part of an oriental Pacific continent, perhaps 

 also united with the Juan Fernandez and Easter islands", and he suggests that 

 many of the Polynesian elements in Hawaii travelled by way of South America 

 and Juan Fernandez (p. 1009). 



Among botanists Setchell, pointing to the evidence furnished by geology, 

 expresses his opinion on Pacific paleogeography in the followings words. 



I see no necessity of postulating any fundamental changes from the point of view 

 of the permanence of the Pacific Ocean as such, and the purely volcanic origin, prob- 

 ably in Tertiary times, of the islands existing in it, in much the same position as we 

 now find them (279.301). 



Another prominent botanist who has strong claims to the title "Defender of 

 the Oceanic Faith", is Fosberg. He refuses to recognize any Pacific islands at 

 all as continental, even Fiji (pc?. 164 etc.); though the flora is "plutot denature 

 continentale" it is not necessary to count with terrestrial connections to explain 

 its characteristics. Among zoologists, Mayr (779) and Zimmerman (2^8) belong to the 

 same camp, and so do most of them, but it happens that certain animal groups^ 

 particularly the land molluscs, offer serious difficulties and have led otherwise 

 conservative zoogeographers to take refuge in "a mid-Pacific land". So for in- 

 stance MUMFORD (i8j. 247) : 



