Il6 TRISTAN DA CUNHA GROUP. 



in Tristan da Cunha ; we found them also, and besides an 

 lulus was very common, and several spiders. 



From what the Germans told him, Von Willemoes Suhm 

 concluded that there were two butterflies, a Va?iessa and an 

 Argytifiis, in the island ; if so, these may no doubt be attracted 

 by the scarlet blossom of the Pelargonium^ so abundant in the 

 island, and fertilize it, and act as a stimulus to the preservation 

 of its colour, and to some extent account for this. 



Otherwise one must regard this case as an instance of the 

 survival in an island, where it is now without function, of a 

 flower the bright colour of which was developed originally in 

 the progenitors of the plant on a continent amongst numerous 

 insects. 



Though some of the plants in the Tristan da Cunha group 

 appear to flower all the year round, others have their regular 

 blooming season. This is the case with the Pelargonium and 

 the Tea plant. The Pelargonium blossoms, according to the 

 Germans, in the middle of summer. Large numbers of the 

 plants come into blossom at the same time, so that the beach 

 is thickly strewn with the coloured petals fallen from the cliffs. 



The Tea plant was nowhere found in blossom in October, 

 though it was abundant. The Phylica trees were all in the same 

 stage of development, bearing fully formed but green fruit. 



The existence of the Cape Horn current sweeping up to the 

 islands, may account for the presence of many South American 

 plants in them. The part of the Brazilian current which turns 

 from the coast of South America, and runs across to the 

 Tristan group, brings with it many seeds to the islands, but 

 these, being tropical, do not germinate. The seeds are cast 

 upon the beach at Tristan, and are familiarly known amongst 

 the islanders as sea beans, from a belief that they grow at the 

 bottom of the neighbouring sea. 



Two of these seeds were shown to me ; one of them was a 

 bean of a tropical American tree, the other was the seed of a 

 Gui/andina* also tropical, which seed, singularly enough, is 

 also cast up sometimes at Bermuda, and is there called a sea 

 bean, and worn on watch chains as a curiosity, and I believe 

 as an antidote to drowning. 



Sir Joseph Hooker, in his lately published account of the 

 Botany of Kerguelen's Land,t writes: "The flora of Tristan 

 da Cunha, Nightingale and Inaccessible Islands, is essentially 

 Fuegian, with an admixture of Cape genera, but with none of 

 those characteristic of Kerguelen's Island. Of Cnpe types it 



* See page 15. 



f Transit of Venus Expedition, Botany. " Obsei vations on the Botany 

 of Kerguelen's Land," p. 8. By Sir J. D. Hooker, P.R.S. 



