BY H. B. GUPPY, M.B., SURGEON, R.N. 953 



sand and gravel thinned away, conditions more suitable for tlie 

 growth of coral would be found, and this is the conclusion towards 

 which my soundings pointed. There would thus appear to exist 

 on the outer submarine slope of this barrier-i-eef, in depths of 

 fifteen to twenty fathoms, a belt of detritus dividing into two 

 portions the zone in which the reef-building corals thrive. (I 

 have marked the position of this belt in the section by a cross.) 

 Had my soundings been confined to the upper of these two sub- 

 zones, I should have been justified to a great extent, on reaching 

 the belt of sand and gravel, in concluding that coral did not 

 thrive in depths beyond fifteen fathoms ; but by subsequently 

 extending such, soundings seaward across this band of detritus into 

 the lower or outer sub-zone, I should have exposed the fallacious 

 character of such a conclusion. 



The results of these soundings svipplied me with an explanation 

 of the growth of barrier-reefs in a region of elevation, which I will 

 briefly i-eview in the light of numerous observations I have made 

 in this group on the growth of coral-reefs during the past two years. 



If we imagine an Island, originally formed from the materials 

 ejected from some volcanic vent and bare of coral-reefs, to afford, 

 after the extinction of the subterranean fires, the conditions for 

 growth on its coasts for reef-building corals, a fringing reef of 

 varying width according to the degree of inclination of the sub- 

 marine sloi)e will ultimately invest its shores. In course of time, 

 the detritus of the corals will collect in a band of calcareous sand 

 and gi-avel on the outer slope of the reef, marking the apparent 

 limit of the depths in which the veef-corals are usually stated to 

 thrive. But the vertical and horizontal extension of such a band 

 of detritus will be mainly detei'mined, as my observations on the 

 Choiseul barrier-reef have shewn, by the presence and position of 

 submarine declivities and by the degree of inclination of the slope. 

 In such a zone of sand and gravel corals will not thrive ; but if the 

 submarine slope has a very gradual inclination, as in the case of 

 the barrier-reef of Choiseul Bay, the lower limit of this zone of 

 detritus ma^ lie within the depths in which reef-building corals 

 flourish, and a line of barrier-reef begin lying parallel with the 

 fringing reef, but separated by a deep channel. 



