XARIJATTYK OF THK EXI-KDITTOX I\ If^ilC. '25 



In places they are undercut and fallen fragments lie on the lagoon platform. The 

 dip of the beds is a few degrees (3° to 4°) to the W.N.W. They consist of alternating 

 sandstone and conglomerate, the latter formed of smooth, rounded pebbles of coral, 

 generally about one inch in diameter, but frequently as much as 4 or 5 inches, and 

 sometimes even G inches to a foot. Slabs and irregular fragments of coral are 

 occasionally present. The smaller coral pebbles present an ojjaque white chalk-like 

 appearance. The cementing substance of these beds, which is more abundant in 

 some layers than others, has a reddish-brown colour and contains, according to an 

 analysis, which Professor Judd kindly had made for me, as much as 25 per cent, of 

 calcium phosphate with some organic colouring matter. It coats round the pebbles 

 as a thin layer formed of many concentric lamellae, and extends from them into the 

 interstices in tubular processes and irregular minute nodules which form a loose, 

 spongy aggregate. On solution in hydrochloric acid a residue consisting of vegetable 

 tissue sometimes remains behind. 



The perplexity which has attended all my endeavours to explain the consolidation 

 of superficial deposits on the atoll is dispelled in this case by the character of the 

 cement, which jDrobably owes its origin to the guano of sea-fowl, and has been 

 deposited by rain-water percolating through loose sand and ]3ebbles, standing above 

 sea-level, and most likely above saturation level. The sandstone of Mateika is also 

 of a reddish colour and may have been cemented in the same manner ; but whether 

 this be so or not, it is clear that in the case of Amatuku consolidation is far from 

 suggesting a negative movement of the strand. It is difficult, however, to account 

 for the alternation of nearly horizontal sand and pebble beds, without assuming that 

 they were laid down under water, and this would involve a subsequent change in the 

 relative level of land and sea. The "phosphates " present in these beds may possess 

 some slight economic value ; at any rate they might be used with advantage by 

 the native population, as a contribution to the manure which they are in the habit 

 of applying to the bananas of the Taro Swamp. 



The le veilings across the islets revealed nothing inconsistent with that negative 

 movement of the strand, of which certain evidence seemed to be afforded by the 

 Heliopora reef; but a doubt still remains in my mind as regards the structure of 

 the pinnacles of the outer glacis ; their base is, I feel sure, formed by the breccia 

 beds, but their upper portion seems, in some cases at least, to consist of coral growths 

 in place. If this should prove to be the case, the course of events during the later 

 history of the atoll would have been more complicated than we have supposed, 

 for we should be able to distinguish four periods, each corresponding to a change of 

 sea-level, as follows : — 



1. Taiisale period .... Sea-level relatively higher. 



2. Breccia beds .... „ lower than 1. 



3. Coral growth of pinnacles . ,, higher than 2. 



4. Existing reef .... ,, lower than 3. 



