1909.] OK A THEORY OF ATOLL FORMATION. 671 



The Coral Island Question. 



Dr. F. Wood Jones, F.Z.S., exhibited lantern-slides, models, 

 and specimens to illnstrate the formation of coral structures. The 

 following is an abstract of his remarks : — 



The full communication upon which the demonstration was 

 based will be reproduced in its complete form elsewhere. The 

 present note consists merely of a brief resume of the problems 

 requiring solution, a criticism of those theories already advanced, 

 and a proposal of some new suggestions arising out of the study 

 of the atoll of Cocos Keeling. 



(i.) The problems demanding exjplanation. 



A theory that is to be satisfactory must not be limited in its 

 application to any one form of coral structure, but must account 

 for the origin of all those forms of reef and island that are built 

 up of coral. It must take notice not only of the larger land 

 masses, and the more obvious geogiuphical structures, but must 

 embrace the actual growth tendencies of the coral colonies them- 

 selves, for it is merely by an aggregation of such colonies that 

 these structures are made up. The submerged coral bank, the 

 barrier reef, the fringing reef, and the atoll must all receive an 

 adequate explanation ; and this explanation must be compatible 

 with the actual processes that may be observed to take place in 

 the individual colonies of a reef. 



In the case of the atoll — the most highly developed of all the 

 coral structures — ^the theory must satisfactorily account for the 

 presence, and development, of all the several parts that enter into 

 its composition. Finally, no explanation must be considered as 

 adequate that carries us only to the stage of the developed atoll, 

 for it must also agree with, and account for, the tendencies of its 

 known after-history. 



The problems connected with the develo]3ment of the atoll are 

 to be correctly gauged only from a proper appreciation of the 

 whole of its structure. The explaining of the origin of an 

 extensive coral reef situated upon a large ocean plateau, the 

 mere raised rim of which constitutes dry land, is a problem 

 different from that involved in accounting for a ring of islands, 

 when these islands are supposed to be the summit of an abruptly 

 rising oceanic peak. 



The question of the actual contour of the elevation of the 

 ocean floor upon which the atoll itself takes origin, is a very 

 important one ; and, in the case of the Cocos Keelings, the 

 soundings of the cable routes have accurately determined the true 

 proportions of this island basis. In this case, the bank that rises 

 from the bed of the ocean is by no means a steep one, and it is 

 only the drop from the reef edge into comparatively trivial depths 

 that may be termed at all abrupt. For the rest, the basis consists 

 of a gradually shelving slope of Globigerina and Radiolarian ooze 



