THE BIOLOGY OF THE FUNAFUTI ATOLL AND TJEEF FORiNLVTIOX. 149 



(B.) Experiments on the amount of carbon dioxide in the water of the lagoon antl 

 the ocean. 



The method of dealing with this question was, for the want of better appliances, a 

 someM^hat rough and ready one. Yet, since the experiments were carried out on the 

 same lines, the results, for purposes of comparison, may be considered accurate. A 

 saturated solution of lime-water was added to the water to be treated in the propor- 

 tions of one of the former to three of the later. The lime was obtained by burning 

 coral fragments in the furnace of the boiler of the drill machinery. 



The amount discovered in the water of the ocean is large, and its presence there, 

 especially at the greater distance from islands, cannot be attributed to decaying coral, 

 and the difference in the amount of carbon dioxide in the ocean and lagoon respec- 

 tively is so slight as to make it improbable that more solution of the reef-rock is 

 taking place inside than outside the reef In an enormous lagoon like that of Funafuti, 

 where, especially at high tide, communication between ocean and lagoon is so 

 extensive, one would hardly expect much difference in the carbon dioxide contents. 

 But the fact that so much of this gas is present in the water of the ocean suggests 

 that if it is a factor in the solution of the reef, then it must act on the ocean face as 

 well as in the lagoon. 



(C) Experiments on the evolution of gases by coral. These were carried out on 

 the suggestion of Professor T. W. Edgeworth Davjd, who had read Mr. Gardiner's 

 remarks on the subject in his paper on the coral reefs of Funafuti, Botuma and Fiji.'"" 



The determination of the evolution of gases by organisms of so low a development 

 as corals is in itself a difficult task and one hardly to be attempted in places 

 where it is a matter of impossibility to keep everything under the necessary control. 

 So that results obtained in the field by makeshift apparatus can hardly be looked upon 

 as conclusive in a matter of such great importance as the possible evolution by an 

 animal organism of gases other than carbon dioxide. 



This is what was done. An attempt was made to collect the gases, as they were 

 given off, in bottles fixed over inverted funnels. The latter were fastened to boards, 

 which, by means of legs, could be placed over coral. From the start the possibility of 

 obtaining gas was considered very small. A coral, PociUopora ven-ucosa, with its 

 polyps fully extended, was watched in glass jars upwards of an hour at a time, during 

 which not more than one or two of the minutest bubbles were seen to rise to the 

 surface, and they might easily have been air bubbles which had adhered to the coral 

 during its removal from the sea to the bottles. Further, the gases as they were evolved 

 might be immediately dissolved by the surrounding water and hence not discoverable 

 by the above method. The only sure one would be an analysis of the water before 

 and after the corals were placed in it. 



Then a,rose the difficulty of finding growing coral which was not closely associated 

 * 'Cambridge Phil. Soc. Proc.,' vol. 9 (1895-98), p. 483. 



