114 bulletin: museum of comparative zoology. 



form the coral sand beaches, or the greater or less fragments which cover 

 the reef flats. The negro heads occurring on the edge of some of the 

 flats (Plate VIII.) are all composed of beach rock, and remain as monu- 

 ments of the former extension of the beach rock. The reef flats are 

 strewn with the dead corals and shells which once constituted this coarse 

 beach rock. The beach rock where it still stands is eroded, deeply 

 pitted, and honeycombed by the action of the sea, and it resembles lime- 

 stones similarly acted upon in other coral reef localities. The beach rock 

 conglomerate of the Three Isles is however extremely coarse, made up 

 of fragments 'of oyster shells and of all kinds of corals, once cemented 

 together, but disintegrated by the work of the sea. Here and there a 

 small vertical bluff of this conglomerate is left standing, showing its 

 thickness and the height to which it was elevated above high-water 

 mark (Plate XXXIII.). 



I was at first puzzled w r ith the mass of dead corals and of fragments 

 of larger blocks left strewn over the reef flats (both on the inner and 

 outer reefs), in from two to three feet of water at average low tide. The 

 presence of negro heads, remnants of the former elevated coral reef, 

 and of negro heads the remnants of former long stretches of elevated 

 coral beach rock conglomerate, give us a ready explanation that their 

 presence on flats where the sea could not have transported them is due 

 to the disintegration and erosion of the coral reef and of the conglom- 

 erate when they extended over the surface of the reef fiats. 



On Plate XXX. of the Great Barrier Reef (see p. 49), Mr. Kent has 

 given a view of huge masses of consolidated coral rock torn off, as he 

 says, from the outer edge of one of the Capricorn Island group, and 

 hurled far up on the face of the level platform reef. He called these 

 huge masses " Nigger heads." Jukes rightly considers them as indicat- 

 ing a former elevation, and I fail to see that they show " direct evidence 

 of subsidence rather than of elevation," as is stated by Kent, page 50. 



The throwing up of such huge masses of coral rock is well known to 

 occur in all coral reef regions. I have given figures of similar huge masses 

 of rock washed up by hurricanes in the Bahamas. 1 But, as Kent well 

 says, these phenomena " are mostly associated with the cyclonic storms 

 that, during the prevalence of the northwest monsoon, occasionally 

 sweep the reefs with irresistible force, though fortunately over limited 

 areas." The Italics ai-e mine, and attention is called to the " limited 

 areas " as the explanations given by Kent will not explain the occur- 

 rence on these grounds of "nigger heads" on the edge of so many 

 1 Agassiz, Bahamas, Bull. Mus. Comp. Zool., XXVI. No. 1, Plate XXIV. 



