THE CHEMICAL EXAMINATION OF THE MATERIALS FROM FUNAFUTI. 367 



in character ; its lower portions would have their cavities permanently filled with sea- 

 water, while, in the upper part of the reef, the sea-water flows to and fro with the 

 rise and fall of the tide. During the boring operations, sea-water was constantly 

 conducted down the bore-hole in order to carry away the fine particles produced by 

 the grinding of the diamond -crown. The cores thus saturated with sea-water would, 

 of course, retain some of the soluble salts on being brought to the surface and dried. 



It is evident, however, that no useful purpose would be served by determining the 

 exact amount of these soluble salts which were retained by different specimens. 

 Various accidents, such as the draining away of water from them, the exposure to rain 

 or similar conditions, would deprive the exact proportion in which these soluble salts 

 were present of all significance. It was determined therefore to neglect altogether 

 these soluble constituents, and at the outset all the samples were thoroughly washed 

 with distilled water before being submitted to analysis. 



The Funafuti cores, as brought home, all exhibited a more or less pronounced 

 brownish-yellow tint on their exposed surfaces. On being cut open, however, the 

 cores were found to be creamy white in colour, except between the depths of 687 and 

 748 feet, where the rock exhibited an earthy texture and a chalk-like whiteness, as 

 it also did at a few points below. The yellow pieces of core when placed in dilute 

 acid had the coloured film removed from them at once, and it was evident that this 

 external colour was accidental — due to the deposition of a slight film of iron oxide, 

 derived from the boring tools, rods, and lining tubes. The analyses were therefore 

 performed on samples taken from the interior of the cores, where free from this iron- 

 staining. 



In some of the samples brought from Professor Sollas' boring, from a small depth 

 below the surface, there were naturally exposed surfaces of the solid cores which showed 

 a much darker brown tint, due to a film deposited upon them. It was found quite 

 impossible to separate this film for analysis, but various tests showed that it contained 

 iron, though manganese could not be detected in it. I think this separation of iron- 

 oxide must have been due to the action of vegetation, which had extracted ferru- 

 ginous salts from the soil and deposited the oxide on the decay of the plant-tissues. 

 But it is remarkable that where, as we shall see in the sequel, there is so small an 

 amount of mineral matter available, with the exception of calcium carbonate, plants 

 should be able to find the iron compounds and take them up into their tissues. 



4. Proportion of Organic Matter in the Samples. 



The proportion of organic matter retained in the calcareous skeletons of various 

 organisms is liable to constant variation on the death of the organism, and is very 

 difficult of determination. In the case of corals, we meet with special difficulties, for 

 one part of a mass of coral may be dead and undergoing decay, while other portions 

 of the same mass may be living. This is well shown in some of the specimens sent 



