884 



SCIENCE 



[N. S. Vol. XXXIII. No. 858 



the surrounding physical and geological 

 conditions. 



The reefs themselves have been very 

 largely — in some instances, predominantly 

 — made up of lime-secreting organisms 

 other than the so-called reef-building 

 corals, such as calcareous algffi, foraminif- 

 era and corals other than true reef build- 

 ei-s, many of which have a wide depth 

 range. 



The characteristic features of coral-reefs 

 — the central shallow lagoon and the sur- 

 rounding rim of living coral with deep 

 water outside — are mainly to be explained 

 by biological, chemical and mechanical ac- 

 tivities continuously in operation at the 

 present time, there being vigorous growth 

 of all lime-secreting organisms wherever 

 the conditions of life are most favorable, 

 and less vigorous growth and even death 

 of these organisms where the conditions 

 are unfavorable. A detailed study of the 

 favorable and unfavorable conditions for 

 different species in an existing atoll 

 seemed to Agassiz a great desideratum at 

 the present time and I am delighted to 

 learn that this is now being undertaken by 

 American naturalists under the auspices of 

 the Carnegie Institute. 



In small atolls, where the surrounding 

 reef is very extensive relatively to the en- 

 closed lagoon, the lagoon tends to become 

 filled up by the accumulation of coral 

 sand, the deposition of carbonate of lime by 

 the living organisms of the atoll being in 

 excess of that removed in solution and by 

 mechanical means ; where the atoll is large, 

 and the encircling reef is — relatively to 

 the size of the lagoon — small, then the lime 

 removed from the lagoon by solution and 

 currents is greater than that deposited by 

 living organisms; hence the lagoon be- 

 comes deeper and wider. The lagoon of 

 Diego Garcia appeared to have increased 

 considerably in area in this way between 

 1837 and 1885. 



It is undoubtedly true that many coral- 

 reef regions have been recently elevated. 

 The circular atoll and barrier reef can not 

 be accepted as evidence of subsidence; the 

 characteristic features of coral reefs would 

 be very similar in a stationary, in a slowly 

 sinking or slowly rising area, although 

 each would show secondary modifications. 

 It matters not whether the change of sea- 

 level be due to crustal movement, to attrac- 

 tion of elevated continental land, or to 

 the accumulation or the melting of polar 

 ice-masses. 



When coral plantations rise from a sub- 

 merged bank, the corals and other lime- 

 secreting organisms situated towards the 

 seaward edge would from the first have 

 the advantage ; they would hence reach the 

 surface, before the central portions, where 

 the corals would be in a position more or 

 less unfavorable for vigorous growth. A 

 shallow lagoon would thus be formed, 

 which might subsequently be cleared by 

 solution, and mechanical action of many 

 of its living coral plantations. 



The coral atoll, on reaching the surface 

 would, he admitted, in very many cases 

 advance seawards on a talus of its own 

 debris, expanding like a fairy ring, and it 

 seemed to him more than probable that the 

 boring at Funafuti atoll was driven down 

 into such a talus, with an underlying Ter- 

 tiary base. 



The red earth which is found on coral 

 islands and supplies the food for plant 

 life, is chiefly derived from the disintegra- 

 tion and decomposition of floating pumice, 

 which is frequently thrown up by the 

 waves on the reefs. 



These results of Agassiz depend on a far 

 greater number of original observations, in 

 widely scattered areas, than have been 

 made by all the other authorities on coral 

 reefs put together. 



When we attempt to survey the life- 

 work of Alexander Agassiz, we are aston- 



