316 Transactions. — Botany. 



Southern Islands, and occurring in the wliole of them except 

 the Snares, is. according to Bentham (4a, p. 407), closely related 

 to the Andine section Orithropium of Engeron, as is also 

 Celviisia. If, with Bentham and the " Pflanzen-Familien," 

 we consider all the southern species of Ligiisticum to belong 

 to Aciphylla, then this may also be, perhaps, called an 

 " antarctic genus," although most closely related to the 

 northern Ligusticimi. Acayia, Epilobium, Uncinia, and Vero- 

 nica may also well have belonged to this problematical 

 ancient plant-formation. The presence of Stilhocarpa and 

 Bnlbinella present more difficulties. The former, as stated 

 above, is allied to the Chinese araliads, so it either did not 

 belong to the above suggested ancient antarctic plant-forma- 

 tion, or its ancestors in other parts of the Fuegian region have 

 long since died out, these coming, as Wallace suggests, from 

 the north dui'ing the glacial period. As for the presence of the 

 South African genus Bulhinclla, it is one of those remarkable 

 South African relationships such as Peripatns amongst animals. 



The bright colours of the flowers of this formation — blue, 

 red, purple — have already been alluded to. Captain Hutton 

 has shown that such in this case is hardly correlated with 

 the visits of insects (536). It seems to me most probable 

 that these colours originated through the greater amount 

 of sunlight to which the plants would be exposed during the 

 long summer days of a higher latitude.'' On this head Schim- 

 per writes (92, p. 716), " Die in den meisten Eeiseberichte er- 

 wiihnte Farben-intensittit der arctischen Bliithen wirdgewdlm- 

 lich und wohl mit Eecht, als ein Wirkung der andauernden 

 Beleuchtung aufgefasst." 



On the Snares, Ewing Island, and in one or two places 

 on Auckland Island is an Olearia lyallii forest. This is ap- 

 parently just as much in harmony witli its surroundings as 

 is the dominant rata forest ; but I think there can be little 

 doubt, from what has been said in treating of the forest-for- 

 mations of the Auckland Islands, that this Olearia forest was 

 once more extensive, but has been gradually driven into smaller 

 and smaller compass by the encroaching rata forest. In my 

 opinion, the O. lyallii] forest is the ancient forest of the 

 Southern Islands, perhaps, indeed, that of a part of the pro- 

 blematical antarctic continent; but it could not withstand the 

 more vigorous Pliocene invader. The genus Olearia is ex- 

 ceedingly common in New Zealand, there being at least 

 thirty -four species. According to Bentham (4a, p. 405), 



• On the other hand, their luxuriant prrowth docs not point to 

 evolution in a rigorous climate. 



t According to this view O. lyallii and many of the endemic plants 

 are examples of "relict endcmism," and not of ''initial endcmism," to 

 use the term suggested by Shull (94). 



