done on board H.M.S. ' Challenger' 615 



be no trace. The shore cliffs consist of a conglomerate, or rather 

 breccia, of generally light- coloured doleritic fragments in a white fel- 

 spathic matrix. These are overlaid, in places, by a bed of old igneous 

 rocks. The peak, which I did not reach, is composed of light grey sub- 

 columnar rock. All along the top of the cliffs, and about 35 feet above 

 the sea, an ancient beach is seen to run ; and the huge caves, at a con- 

 siderable height above high-water mark, afford additional evidence of the 

 rising of the land. The middle island, which I explored, consisted 

 entirely of the above-mentioned breccia, and even at its highest point 

 (150 feet) it was much water-worn. Dykes are by no means so frequent 

 in this as in the other islands. Where they occur penetrating the breccia, 

 the latter appears to be rendered more easily decomposable, the result 

 beiug that the dykes form the axes along which creeks or inlets indent 

 the shore. In the island of Tristan, the gully behind the settlement, in 

 the centre of which the spring rises which supplies the village brook, is 

 formed in a similar way. It is backed by a vertical dyke (if a mass at 

 least sixty yards wide can be called so) of rock similar to that observed 

 on Nightingale, and, like it, it has altered the adjacent rock considerably ; 

 and, probably from the pressure of the sides during its injection, it has 

 a most perfect, almost slaty cleavage, which renders it easily disintegrable, 

 though it does not appear to decompose easily. Dykes of this descrip- 

 tion, but of usually not more than 1 or 2 feet in thickness, traverse the 

 rocks, seen in section in the face of the cliff, both in Tristan and Inac- 

 cessible, in great numbers. Besides these, there are others of a very 

 compact, fine-grained, basaltic rock intersecting the nearly horizontal 

 layers, which form the cliff, in all directions, and showing very marked 

 differences in texture at the sides, where the action on the contiguous 

 rock is apparent. In one of the specimens collected the rock in the 

 interior of the dyke is of the usual light colour, while the edge is marked 

 by a black obsidian-like band of half an inch breadth, as sharply defined 

 from the internal mass as if it had been painted. 



The cliffs, being quite perpendicular and naked, give a good view of a 

 vertical section of the island. At their base the rubbish fallen down 

 from them has accumulated to a heap reaching sometimes quite 100 feet 

 up the face. I had thus the opportunity of observing two sections 

 of what had been valleys or hollows, but which were filled up with 

 finely levigated volcanic material. From the arrangement of the material 

 I should think that it had been erupted by a subaerial volcano and 

 fallen, into a bay, where it had gradually settled. If such has been 

 the case, it furnishes additional evidence of the rising of the land in 

 these parts. 



During the cruise between the Cape of Good Hope and Melbourne 

 three islands were landed on, namely, Marion, Kerguelen, and Heard 

 islands. On the first of these a day was spent. The soft mossy nature 

 of the ground made the walking very heavy, which, combined with the 



