CHAPTER V. 



TRISTAN DA CUNHA, INACCESSIBLE ISLAND, 

 NIGHTINGALE ISLAND. 



Settlement of the Island. Geological Structure. Vegetation. Teirpera- 

 ture of Fresh Water. Pliylica arborea. Rigorous Climate. Con- 

 dition of the Settlers. Inaccessible Island. Rock-hopper Penguins. 

 Tussock Grass. Penguin Rookeries. Peculiar Land Birds. Noddies 

 and other Sea Birds. Southern Skuas. Wild Swine. Change of 

 Habits of Penguins. Nightingale Island. Vast Penguin Rookery. 

 Seal Caves. Rocks Worn by the Feet of the Penguins. Mollymauks 

 and their Nests. Derivation of Seamen's Names for Southern 

 Animals. Dogs run Wild in a Penguin Rookery. Migrations of 

 Penguins and Seals. Insects, etc., of the Group. Flowering Seasons. 

 Sea Beans. Relations of the Flora. 



Tristan da Cunha, Oct. 15th, 1873.— The .ship arrived at 

 Tristan da Cunha on October 15th. The island of Tristan da 

 Cunha is one of a group composed of three, the other two 

 being called Nightingale and Inaccessible Islands. Besides 

 these, another small island, Gough Island, lies about 200 miles 

 to the south and somewhat to the east of Tristan da Cunha, 

 and from its vegetation would seem to be naturally included 

 in the group. 



Tristan dti Cunha itself lies in Lat. 37° 2' 48" S., Long. 12° 

 18' 20" W., distant westward from the Cape of Good Hope, 

 1,550 miles, and about one-third farther from Cape Horn, lying 

 nearly on a line drawn between the two Capes ; it lies 1,320 

 miles south of St. Helena. The island is about 16 square 

 miles in area,* it is nearly circular in form, its highest point is 

 8,326 feet above sea level. 



The latest information concerning the inhabitants of the 

 island, e.xtant at the tinie of our visit, is to be found in the 

 "Cruise of H. M.S. ' Galatea,'" p. 28 (London, Allen & Co., 

 1869). In this account reference is made to the various men- 



* I regret exceedingly, that owing to ignorance of the nature of a 

 German geographical square mile, I concluded that Grisebach had, in his 

 " Veg. der Erde," made an error in describing the area of Tristan as two 

 geographical square miles, and that I stated this in "Journ. Linn. Soc." 

 Bot., Vol. XIV., p. 328. 



93 



