done on board H.M.S, ' Challenger. 3 617 



above-mentioned cliff had to be skirted, and it was found to consist of a 

 light grey compact doleritic rock. 



From these few observations it may be concluded that the island con- 

 sists of a foundation of older igneous rock ruptured and surmounted by 

 recent volcanoes. That these have been active at no very ancient date is, 

 I think, rendered probable by the perfect preservation of the forms of 

 the cones with their summit craters, and by the fact that the mossy 

 vegetation so luxuriant at their base, and retaining this luxuriance on 

 the certainly older mountain-spurs to an elevation at least equal to that 

 of the top of the cone ascended, has as yet spread up their sides only in 

 straggling isolated patches. 



The evidence afforded by the want of erosion deserves all the more 

 weight when the position of the island is remembered, where, of necessity, 

 the rainfall must be considerable. 



The island of Kerguelen was visited at various points on the north- 

 eastern and south-eastern coasts. The first part touched at was Christ- 

 mas Harbour, already visited by Cook, Ross, and other navigators. It 

 is a deep inlet surrounded by high rocky banks, which are in many 

 places quite precipitous. At the head of the bay there is an extensive 

 sandy beach ; and the ground behind it slopes at a tolerably easy gradient 

 up to the top of the ridge, which is occupied by a large lake. Erom a 

 ship anchored in the harbour an excellent general view of the arrange- 

 ment of the rocks can be obtained ; they are seen to be arranged in 

 apparently perfectly horizontal beds, the separation lines of the different 

 beds being easily traced all round the harbour. Where the sides are 

 not precipitous the summit of the ridge is attained by a series of terraces, 

 and the summit of the ridge is, as might have been expected, almost per- 

 fectly flat. The continuity of the flat-topped surfaces, both of the 

 northern and of the southern ridges, is broken by the two most con- 

 spicuous objects in the landscape, namely, Table Mountain on the north, 

 and one unnamed on the south. This rock-mass does not project above 

 the horizontal hill-top, but rather appears to stand out from it like a 

 huge boulder. The summit of the ridge has been called Mount Havergal ; 

 but it is formed of the ordinary bedded rock, this " neck " of conglomerate 

 not reaching any greater height than that of the contiguous parts of the 

 ridge. These hills belong to a class representatives of which were found 

 again in the south in Greenland Harbour ; and as they resemble each 

 other closely they will be described together. In both places they pro- 

 trude through the horizontal beds of basalt, differing from these and 

 from each other in nature, and without having caused any apparent 

 disturbance in the arrangement of the beds which surround them. 

 The horizontal beds which form the mass of the land are basaltic, and 

 vary from 10 to 20 feet in thickness, being generally compact ; but in 

 ascending the hill beds are met with frequently which contain large 

 amygdaloidal cavities filled with zeolites, principally analcite and heu- 



