NARUATTAT. OF THK F.XPKDITION IX IsOfi. 1.'^ 



The lagoon mound is not shown in Darwin's diagram, which represents the surface 

 of the islet as continuously sloping from the outer ridge to the lagoon, as it is 

 expressly said to do in tlie descriptive letter-press. Darwin was fully acquainted, 

 however, with both the existence of the mound and its origin, for he states a little 

 further on (p. 16) that " the little waves of the lagoon heap wp sand and fragments 

 of thinly branching corals on the inner side of the islets on the leeward side of the 

 atoll .... but the land thus added is very low." 



On Funafuti, the distinction between the ocean ridge and the lagoon mound is 

 sometimes very sharply marked ; on the broader parts of Funafuti islet they are 

 not only obviously independent features, but also are widely sej^arated from each other 

 by an intervening plain, the black and rugged surface of which lies somewhat below 

 high-tide level, so that at full " springs " sea-water oozes up through it and gives 

 rise to scattered pools. This plain may be called the " central flat " (J, fig. 4) ; it is 

 the exposed surface of a sheet or series of sheets of consolidated coral breccia, resem- 

 bling the stratified coral rock and calcareous sandstone of the outer and inner 

 platforms. 



As the islet grows narrower to north and south the central flat becomes corre- 

 spondingly reduced in breadth and the outer ridge and lagoon mound approach 

 one another till they meet, and then pass into each other, till at length scarce 

 any sign of a depression can be seen between them. 



The structure of the central flat is best displayed in the interior of Funafuti 

 islet a few yards north of the site of the third bore-hole ; here it had been eaten 

 away by the solvent sea-water, leaving steep-sided gullies some 3 or 4 feet deep, 

 on the sides of which sheets of coral breccia were plainly visible. During a falling 

 tide, when a foot of water covered the floor of the northernmost one of these 

 gullies, I have seen little whirlpools, like those made when the plug is removed 

 from the bottom of a bath, swirling over holes in the rock, and thus affording 

 visible evidence of the free communication which exists between the middle of 

 the island and the sea. In other cases where the holes were numerous enough to 

 form a sieve no whirlpools were formed. A more striking evidence of the ready 

 access which is open to the sea is to be found, as will appear later, in the Mangrove 

 Swamp. 



Glancing backwards over the observations we have already made, it will be seen 

 that on the seaw^ard side of the islets a sheet of hard limestone exists and extends 

 from the nullipore rim up to the ocean ridge, beneath which it disappears, but on 

 crossing this ridge a similar rock is again met with on the central flat, and though 

 for a second time lost when we trace it up to the inner mound, it, or a rock very like 

 it, reappears to form the cliffs and marginal floor of the lagoon. This suggests at 

 once that the hard rock is continuous from lagoon to sea, nor is this supposition 

 unsupported by evidence ; for at the eastern end of the little islet of Pava, one of the 

 northernmost islets of the atoll, the rock of the ocean glacis can be seen extending 



