110 APPENDIX I. 



A third and last stage of these big coral masses is to be seen in some cases where the undermining by borers 

 has continued until the base has become so small that it is too weak to support the superincumbent mass. 

 They have, therefore, been broken olf from their base by heavy seas, and are noM^ to be found as large free 

 masses in various positions on the beach or some way up the Hurricane Bank. It has not often happened 

 that one has observed unequivocal evidences of the existence of the newer breccia L.2.B. on the ocean side, 

 but here for a short distance along the low breccia cliff on that side it does occur, containing blocks of 

 the older breccia cemented together in the mass, and lying uncoiiformably on the 0.2. E. It is now, 

 ho\yever, being eroded, and worn back to the general cliff-line in common with the older or main breccia sheet. 



Several cocoanut and pandanus trees are lying down the ocean slope of the Hurricane Bank, the drifting 

 back of this farther on to the island having removed the material surrounding their roots. On its inner 

 side a wide area is occupied as a clinker field, made up in some parts of larger and smaller coral blocks, 

 while in parts deep pits and caverns occur between jagged and clinkery coral and breccia, where solution 

 is often found to be in active progress. The width of Funangonga is considerably more than of any othei 

 island on the southern part of the atoll. 



On the lagoon side between the breccia of either end the beach is made up of foraminiferal sand, chiefly 

 Tinoponts haculafus, with a small admixture of the usual beach debris (L.9.B.) rising to the vegetation line 

 or a little beyond, and then gently sloping to near the centre or to where it meets the solution areas or the 

 coarser material. Nearing the centre of the island is similar material, but older and covered by vegetation, 

 the height of which indicates the iiuier part to be the older, while more coral fragments appear on the 

 surface as the clinker field is approached. 



FATATO. 



This island (Plate 8), like the last, is joined at either end to the next one by a long narrow strip of reef 

 which does not equal the normal width of the ocean platform outside the pinnacles together with even a very 

 narrow lagoon platform. Thus, prol)abl v, it has never been, except near each island, the site of a breccia 

 sheet or the usual island, but has remained from the first as a reef not much above low water, Avith little 

 more than loose blocks and dchrif: upon it. The width and character respectively of lioth the ocean and 

 lagoon platforms are remarkably constant around Fatato. The usual pinnacles and rough zone 

 exten<l in a line distant some 30 yards outwards from the low cliffs, ^^'ithin this is the rough zone 

 (0.2. B.), then the worn or smooth zone (O.2.C.), and then 0.2. D. or low cliffs of the breccia sheet. In 

 this O.2.D., near the southern end, patches of L.2.B. or breccia more recent than the usual breccia sheet 

 occur as at Funangonga, and like it enclosing masses of the older breccia inclining at an angle of about 

 20° towards the platform. The differences in dip between it and the older breccia sheet, together with 

 its encircling slabs of the older breccia, assist one in readily recognising its presence. It is here also 

 undergoing erosion, and no deposit now forming in this position corresponds to it, for it more closely 

 resembles the material constituting the Hurricane Bank on the top of the breccia sheet, excepting that 

 a conglomerate sometimes underlies this newer breccia (L.2.B.).* There are several big and high blocks 

 of breccia along this Hurricane Bank, while some of those that appear to be in silu and still united to 

 the breccia sheet are very high and correspond in elevation to some of the high breccia masses that usually 

 appear as outliers, showing the relationship of these and all outliers. The Hurricane Bank is here in 

 places unusually high, but as it slopes inwards to the centre of the island, it is seen to be composed of 

 smaller and much less rugged material than that of the clinker field of the two islands south of this, and 

 solution pits were rarely seen. 



Fatato on its lagoon side closely simulates Funangonga, as at either end the breccia sheet passes across 



* The symbol L.2.B. indicates it to be a lagoon production, but while probably agreeing with the age 

 of the lagoon-formed breccia, this has, no doubt, been formed on the ocean side. 



