428 ALEXANDER AGASSIZ 



we dropped anchor, and we arranged to coal at La Pey- 

 rouse Bay, on the south side of the island, where the swell 

 interferes less with coaling alongside another steamer. 

 By 9 o'clock we were anchored there and at work in full 

 blast. While we were coaling we went on shore to col- 

 lect plants and insects and to examine some of the plat- 

 forms on which the natives of former days used to erect 

 their stone images — a more desolate-looking place I 

 don't know except the great lava flows of the Sandwich 

 Islands. But this island is perhaps worse for its weak 

 attempts to grow grass on which the owners try to raise 

 sheep and cattle. The walking about all day was most 

 tiresome ; we were glad, indeed, to get on board again 

 where you could put your foot down without fear of 

 twisting your ankle in a hole between two pieces of rock. 



To-day we started off early to see the quarries where 

 the stone images are made. We got some nice little 

 ponies, most sure-footed beasts, and a great improve- 

 ment over walking. The stone images are found in all 

 stages of finish in the interior slopes of a volcano at the 

 east end of the island, Rana Koroka, as well as at those 

 of the outer face. They are cut out of a comparatively 

 soft volcanic ash which hardens with exposure. They 

 cut all round them and when free stand them up on 

 end to finish them, and then carry or slide them by main 

 force to the platforms which line the seacoast of the 

 island. 



One of the finest of all the platforms is near the cra- 

 ter, but the images are no longer on their base. They 

 are all flat, lying at the back of the platform. They must 

 have been knocked down by an earthquake or some 

 cataclysm, as it looks as if the whole work on the island 

 was suddenly put a stop to. 



