done on board H.M.S. e Challenger.' 621 



of the columnar with the bedded rock ; in fact there appeared to have 

 been next to no fusion between the two. 



The corresponding hill on the south side of the harbour is formed entirely 

 of volcanic conglomerate, intersected here and there by dykes, some of 

 which show on the outside the obsidian-like bands produced by rapid cool- 

 ing, which were observed in considerable abundance at Tristan d'Acunha. 



Fossil wood is found on the south side of Christmas Harbour imbedded 

 in the igneous rock. It occurs in stumps and smaller branches. The 

 colour varies from yellowish white to chocolate-brown and black. Its 

 hardness is also very variable. Even in the perfectly white pieces there 

 is still much organic matter remaining. The bark has been transformed 

 into a brown crystalline mass of greasy appearance, which effervesces 

 with acid. The inside of one rather large trunk, the core of which had 

 probably rotted away, was entirely filled up with a mass of igneous rock 

 with elongated cavities filled with crystals. Iron pyrites was occasionally 

 observed. Parts that internally consisted of nothing but trap-rock often 

 presented on the outside the fibrous appearance of the simply silicified 

 wood ; the thickness of this rind, however, w T as insignificant. 



A species of brown coal occurs on the south side of Christmas Harbour 

 between two layers of basalt, and only a few feet above the sea. It is, 

 practically, of no use, being too poor to burn at all alone. 



Near the eastern point of Howe Island much amygdaloid was found, 

 the geodes here consisting almost exclusively of agate. The tops of the 

 hills were thickly strewed with such as, in the lapse of time, had been 

 weathered out of the matrix. Many of these presented a very striking 

 appearance, one of the corners of the cast of the cavity having been 

 neatly planed off, and in some instances even highly polished, in others 

 covered with a natural etching of great beauty. The occurrence of these 

 abraded faces is, I think, a further evidence of the recent prevalence of 

 ice-action over the whole island; and it must be remembered that glaciers 

 actually do reach the sea-level in fiords on the main island not more than 

 twenty miles distant. 



It is worthy of remark that, although amygdaloids are common along 

 the north-eastern side of the island, the nature of the geodes is different 

 in different localities. In Christmas Harbour they are almost exclusively 

 zeolites ; in Cumberland Bay those who have visited it report numerous 

 cavities in the rock filled with quartz crystals, and, indeed, one of the 

 promontories in it is called Crystal Point ; while at Howe Island the 

 silica with which the cavities were filled occurred entirely in compact 

 masses of agate or chalcedony. The cavities were usually quite full, the 

 geodes being solid and forming an accurate cast of the cavity. Where 

 this was not the case the interior presented a finely mammillated surface. 

 I did not observe quartz crystals either here or at Christmas Harbour ; 

 nor did I observe the zeolites, so common in the last-named locality, 

 either at Howe Island or Betsy Cove. 



