MEGAPOD1US MOUNDS. 147 



rather uneven ground. It seemed to be composed 

 of earth and fragments of coral, of which large quan- 

 tities were lying- on the beach on one side of it. Its 

 lower portions spread over the roots of several old 

 trees, which were now growing through it. Near 

 the top of the hill, 200 or 300 yards from the beach, 

 and 60 or 80 feet above it, was another mound, 

 smaller, but apparently newer and more perfect. 

 This was a pretty regular cone, on a circular base, 

 the circumference of which was seventy -seven feet. 

 It was about eight feet high, the slope of its side 

 measuring about fourteen. The top was flattened, 

 and about three feet in diameter. This had like- 

 wise begun to encroach on one or two neighbouring 

 trees, of which it had buried the stem to the height 

 of one or two feet. Externally, it seemed to be 

 composed of loose earth and fragments of stone, 

 mixed with pieces of stick. Many of the stones 

 were as large as the first. We began digging into 

 it about two-thirds of the way up on one side, but 

 found the interior so firmly compacted together 

 with sticks and stones and roots of trees, that it was 

 no easy task, and after working for two or three 

 hours, were obliged to get two men to assist us. 

 With their help, we at last cleared off three or four 

 feet of ear tli and got into the interior, where we 

 dug down, and presently came upon a broken egg. 

 This was an old one, the yolk being partly con- 

 solidated and rotten. We then turned up several 

 spots full of fragments of egg-shells, and then 



l2 



