THE GEOLOGY OF FUNAFUTI. 71 



Unfortunately we did not obtain specimens of the highest points of the breccia at 

 Motusanapa, but we have no reason to suspect that it is in any way different from 

 the breccia which we have examined in thin sections under the microscope from other 

 locahties in the atoll. 



The breccia sheet has apparently at one time covered the reef proper around 

 perhaps the whole of the rim of the reef platform, though- now there are wide 

 stretches of this from which it has been almost completely denuded. Nearly 

 all the islets of the atoll have a foundation of this material, or at least a 

 rampart of it, along their shore line facing the ocean. In some cases there is 

 an outer line of extremely dense breccia marking the most advanced position ocean- 

 wards occupied by that material at a time when it formed the foundation for 

 islets which have since been driven inwards over the reef platform towards the shore 

 of the lagoon. Such an outer line is well seen between the islets of Pava and 

 Mulitefala at the north end of the atoll (see Plates 12 and 13, and see Section 1 

 of Plate 13, which shows how the ancient islet of Fualifeke has been driven lagoon- 

 wards from its former position), and also in the south-western and southern part of 

 the atoll from Fuagea past Tutanga and Avalau to Nukusavalivali. (See Plates 3, 4, 

 and 5.) It is to be noted that all the high breccia outliers, or breccia negro-heads, are 

 situated between this line and the present breccia rampart. As regards the 

 subdivisions of the breccia sheet, 0.2. A. represents the portions of it exposed in the 

 central parts of the islands, where at low tide it is subjected during the heavy inter- 

 mittent tropical rains to the corrosive action of humous acid. The surface is very uneven, 

 bristling with sharp points. It is typically developed at the main island and Tutanga. 



O.2.B. represents a portion of this breccia sheet forming the third zone inwards 

 from the edge of the reef platform at low tide (see Plate 17 and Section by 

 Mr. G. H. Halligan on line C-D). Most of the space between the outer line 

 above referred to and the breccia rampart belongs to this zone. It is even more 

 jagged and more bristling with sharp points than O.2.A. A zone of this rock is in 

 some cases found on the lagoon platform as well as on the ocean platform. The 

 corrosion is apparently due partly to rain-water, partly to sea-water. 



The reason why the jagged points are not broken off by the force of the waves is 



that this zone is at present out of range of the loose shingle at the base of the 



Hurricane Bank. In addition to the high outliers already referred to, there are 



in places within the zone small outliers and " negro-heads," mostly just covered at 



high tide. The lower figure on Plate C shows some of these "negro-heads" at 



Fatato Islet, on the lagoon platform of the reef. We agree with Professor Agassiz's 



general opinion that such " negro-heads " are of the nature of outliers, and do not 



represent once loose blocks of reef limestone cast up upon the reef platform by storms 



and subsequently cemented down. Our observation at Funafuti led us to the 



conclusion that these "negro-heads" indicate downward movement of the shore line.* 



* It should be stated, hoH^ever, that Messrs. Halligan and Finckh consider that these blocks are not 

 in situ, and they do not therefore look upon them as evidence of such a movement of the shore line. 



