K I L A U E A. 189 



continuous lava cliff, of the hard metallic kind, like that which is still 

 found on both sides of the sand-hills for several miles. There was no 

 appearance whatever of sand. At present there are three large hills, 

 composed of sand and gravel of a light yellow hue, with little mixture 

 of lava or scoria. The last unite with the lava plain near the sea, 

 which may be observed in some places to flow under them. 



Beds of sand and gravel, similar to those composing the hills, exist 

 for some distance along the sides of the lava streams. From all 

 accounts, the formation of these took place at the time the lava stream 

 joined the ocean, which must have produced a violent sand-storm, the 

 effects of which are rendered evident for a mile on either side of the 

 stream, by the quantity of sand and gravel that is lodged in the pan- 

 danus and other trees. 



From the top of the hill I could perceive no appearance of a shoal 

 having been formed, for the water appeared quite as blue as in mid- 

 ocean. This point I particularly attended to, for it had been reported 

 to me that such a shoal had been formed. The sand-hills appeared to 

 have encroached upon the line of the coast about one hundred feet. 



Through the sand that was near the sea-shore chrysolite was disse- 

 minated in greater abundance than it was met with elsewhere, and of 

 larger size. This mineral is found throughout all the lava formation, 

 in greater or less quantities. To account for the presence of greater 

 quantities of it at this place, it may be supposed that the melted lava, 

 coming in contact with the water, has freed the chrysolite, which the 

 sea has thrown on the shore. 



The width of the lava stream was found to be three-fourths of a 

 mile. The portion of it nearest the sand-hills is in a very confused and 

 rugged state, and there are some large accumulations in mounds, that 

 have been forced up by pressure from above and beneath. It is said to 

 have passed over the ancient village of Nanavalie, and left upon its site 

 and cultivated grounds a deep layer of rock. The natives told us that 

 they had remained till the last moment, hoping the torrent might be 

 stayed or turned aside, and thus save their houses. It however swept 

 on, and they had barely time to remove the few articles they possessed. 

 I was somewhat surprised at the natives making so light of these 

 appalling streams of fire, of which the first notice they have is a few 

 shocks of earthquake, and shortly after a distant fire in the woods. 



I was particularly struck with the difference between the old and 

 recent flows of lava : the old looks the more fresh of the two, and has 

 the smooth dark metallic lustre before observed, without any vitreous 

 crust ; it seems to have flowed over the surface when of the consis- 

 tence of tar. The late flow has a decided vitreous character, with 



