244 CORAL AND ATOLLS 



If sucli a specimen be watclied, as sand particles are 

 poured over its surface, it will at once be seen tbat there 

 is a very perfect mechanism for ejecting the silt. The 

 delicate tissues, which clothe the rays of the coral, shrink 

 back from the intruding substance ; and a curious wave 

 spreads along the partitions, from the central end towards the 

 periphery, and the foreign body is borne along in the wave. 

 The same process is employed in the ingestion of food ; but in 

 that case the particle is passed on from the periphery to the 

 centre. If an oyster is crushed, and its contents scattered 

 over the surface of a large Fungia, the whole of the soft 

 tissues of the coral shrink together at first, as a result of the 

 stimulus, and then, slowly, successive contractions and ex- 

 pansions sweep the individual particles towards the central 

 portion of the disc, where they disappear from view. Later on, 

 the small bright fragments of oyster shell, which were taken 

 in along with the food, are seen to be migrating towards the 

 periphery, where they are thrown off. The Fungia therefore 

 protects itself by getting rid of the sediment which comes in 

 contact with it ; and the mechanism must need to be a very 

 effective one, for Fungim will live in pools in which vast 

 numbers of particles are moving about, and the skeleton of a 

 dead Fungia always collects large quantities of sand. 



It is probable that all those deep-sea corals which have a 

 flat or wide-mouthed type of growth get rid of the foreign 

 substances that fall within their mouths, after the same 

 fashion as the Fungim; and so they are indifferent to the 

 presence of suspended matter in the water, provided that the 

 deposition of sediment does not proceed at such a rate that 

 it overwhelms them entirely. With the small mouths of 

 the reef-builders this process may be quite impossible, and 

 sediment that settles on a colony cannot be got rid of in this 

 way ; the individual corallites become filled with sediment, the 

 zooid becomes choked, and death ensues. 



The definite knowledge that sedimentation areas are 

 certainly excluded from reef-coral growth, whilst wave-stirred 

 areas are particularly favourable, marks an important stage in 



