1898.] Funafuti, Rotuma and Fiji. 479 



I had no doubt that it must have consisted of a growing coral 

 bed, of which, Dana remarks 1 : — " The character of a growing coral 

 bed is so strongly marked in its uneven surface, its deep holes 

 and many entangling stems, to the vexation of the surveyor, that 

 in general the danger of mistake is small." 



At Funafuti, eight dredgings were made on the steep slope of 

 the reef between 40 and 140 fathoms, none of which gave any 

 evidence of reef-building corals at these depths. There were a few 

 patches of solitary corals, but even these were rare on the stones 

 brought up. At a depth of 35 — 45 fathoms off the south end 

 I took a haul of the swab and obtained four species of Madrepora, 

 three of which I have identified with M. reticularis, M. angidata, 

 well-known deep living species, and M. scabrosa. No other true 

 corals were obtained, but there were two small pieces of Heliopora, 

 At a greater depth than 30 fathoms, also off the south end, I 

 obtained in two hauls three species each of Pocillopora and 

 Madrepora, and one each of Stylophora and Porites ; there were 

 also several small pieces of Millepora. Off the north of the island, 



1 could not keep in these depths as the slope of the reef was 

 too abrupt, but one long haul, with the shallowest depth 21 

 fathoms, resulted in three species of Madrepora, one each of 

 Heliastrcea and Pocillopora, and also several pieces of Millepora. 

 In all, from depths of over 20 fathoms were obtained 7 species of 

 Madrepora, 4 of Pocillopora, and one each of Heliastrw, Porites 

 and Stylophora, making in all 13 species in four hauls of the 

 swabs. Outside the reef, five hauls of the swabs at depths up to 

 7 fathoms gave in all 11 species, represented by 5 Madrepora, 



2 Pocillopora and one each of Seriatopora, Montipora, Porites 

 (apparently P. arenosa, a very common species on the shoals of the 

 lagoon) and Stylophora. Millepora was, however, very common in 

 all the hauls up to 7 fathoms, and evidently grows extremely 

 luxuriantly at about this depth. 



SECTION III. 



The Food of Corals. 



The conclusion of Murray '\ " that the chief food of the corals 

 consists of the abundant pelagic life of the tropical regions " does 

 not seem to rest upon many direct observations, though the same 

 statement is repeated, or inferred, by nearly all writers on the 

 theory of the formation of these reefs. A further conclusion is 



1 Loc. cit. p. 117. 



2 Coral Reefs. By Chas. Darwin, 1889. App. n. p. 286. 



38—2 



