206 MK. JUKES-JJJtOWNE AND VB.OF. IIARIIISON 



Dr. II. 13. Guppy, in his excellent and suggestive paper on the 

 Coral Kecfs of the Solomon Islands*, has offered an explanation of 

 the formation of outer reefs which can be ap^jlied to the Barbadian 

 reefs, and explains the relation of the submarine wall to the surface 

 reef. The features of the Solomon-Island reefs, according to his 

 description, resemble those of the Barbados reefs. Their submarine 

 profiles often show two terraces, first a ledge from the edge of the 

 shore-reef to a depth of 4 or 5 fathoms, then a steep declivity to 

 20 fathoms, below which is a broad j)lateau gradually sloping to 

 25 fathoms, and ending in a steeper slope to deep water. Dr. 

 Guppy's soundings showed that at the base of the first declivity 

 there was always coral-sand and debris, which prevented the growth 

 of living coral, but that farther out on the plateau beyond this sandy 

 zone living coral again occurred. He points out that where a shore 

 has such an outline the conditions favour the formation of a harrier- 

 r«c/,because, if the outer part of the lower plateau or terrace lies within 

 the depth in which reef-building corals can flourish, they will grow 

 freely there and will be forced to grow upward by the contiguity of 

 deep water on one side and sand on the other ; in this way an 

 outer barrier-like reef may be initiated, separated from the shore- 

 reefs by a sand-covered channel. 



Applying this explanation to Barbados, we find that where no 

 part of the inner coral-shelf rises to the surface, the lower plateau 

 is continuous, and only small patches of sand occur on it ; but 

 where reefs like the Cobbler's E-eef exist, the action of the breakers 

 leads to the formation of a debris-slope below them, and yields a 

 deposit of coral-sand which prevents the growth of coral ; the 

 current impinging on the reef probably helps to make this sand- 

 bank a narrow one by driving the sand along the slope ; outside 

 the sandbank the water is still shallow enough for the growth of 

 reef-corals, and consequently they have grown upward in the form 

 of a ridge or barrier, the outer edge of which is limited by the 

 depth of the water. 



One reason for the vigorous growth of coral at this eastern end 

 of the island is probably that the polypes find a superabundant 

 supply of food in the microzoa brought by the Gulf-stream, which 

 here first impinges on the island. 



The supposition that such a shelf or plateau extending to 25 or 

 30 fathoms did exist along this shore before the growth of the 

 coral is confirmed by the contour of the sea-floor beyond the ter- 

 mination of the outer reefs. The lower line in fig. 3 (p. 203) is a 

 profile running S.W. from the shore near South Point. The sound- 

 ings in this direction from 20 to 30 fathoms all prove sand or sand 

 and coral fragments, and the conditions do not appear to be favour- 

 able for coral-growth ; so that we may infer that the profile 

 represents the original contour slightly masked by sandy deposits. 

 It is, however, difficult to understand wherefore the two lines of 

 reef should end off so suddenly, though this abrupt termination 



'^ Prou. Eoy. Soc. Edinb. 1885-8G, p. 857. 



