26 MB. A. MUUEAY ON THE GEOGEAPHIOAL BELATIONS OF 



The genera Phylica, Pelargonium, Mesembryanthemum, Osteosper- 

 mum, and Wahlenbergia are eminently characteristic of Southern 

 extra- tropical Africa ; and I do not find amongst the others any in- 

 dication of an American origin, except a plant referred to Physalis. 

 The I^erns tell the same tale ; of twenty-six species, ten are abso- 

 lutely peculiar ; all the rest are African, although some are also 

 Indian and American." On this sentence, while I implicitly accept 

 its conclusions, I shall only remark : — 1, that Mr. Baker, in his 

 admirable paper on the geographical distribution of Ferns, seems 

 to me to be a little more favourably disposed to America in his 

 estimate of their relationship ; 2, that some of the African spe- 

 cies, as BanJcsia and Protea, may have an Australian significance 

 as well as an African — not that I think that either touches Dr. 

 Hooker's conclusion, but in trying to sum up impartially I do 

 not wish to overlook any point ; and 3, that there is besides what 

 may be called an under layer peculiar to the island itself, and 

 found nowhere else on the face of the globe, sucli as arboreal 

 CompositsB (tree-daisies, as it were) . Dr. Hooker regards these 

 as too abnormal to have their affinities with the plants of neigh- 

 bouring continents made out. I cannot think so if he will lend 

 himself seriously to the work. 



The general result which I draw from the whole flora is, that 

 we have here a compound flora certainly two deep, possibly three 

 deep. We have, in the first place, I believe, a genuine natural 

 Atlantic, that is, European flora ; for in the face of the decided 

 testimony to that effect given by the fauna I cannot accept Dr. 

 Roxburgh's conclusions as to the supposed introduction and natu- 

 ralization of every species having a European habitat. If they 

 can be proved to have been introduced, good and well ; but I 

 object to take the thing to be proved as part of the proof. And, 

 in the next place, I believe we have the traces of an older 

 African flora (why I call it older I will explain when I come 

 to speak of an ancient connexion between Patagonia and South 

 Africa) ; and I believe that both are due to actual continuity, 

 however circuitous or interrupted, with the respective countries 

 the impress of whose floras they bear. 



Before leaving St. Helena, I have just one other argument to 

 adduce in support of its former connexion with the other Atlantic 

 islands, and that is the fact, which has only recently been ascer- 

 tained, or, at any rate, only recently laid down in our maps, that 

 there is a long band of elevated submarine bottom running north 



