15 



lose them gradually when these get overgrown, so we establish 

 the connection between reefs and their lagoons, and again 

 between the brackish lagoons — Sir Emerson Tennant calls 

 " Gobbs'* — and the marsh, losing it in maana or dry 

 grass land. But as we proceed up the country we can at a 

 glance trace by the vegetation, &c., the very gradual change 

 of maana into patina, and as we follow coral to maana so- 

 we proceed from maana to patina, and the link is complete, 

 although the latter part is graduated to an almost inappre- 

 ciable extent. Thus we see the improbability of any fossils 

 being found far inland, owing to their rapid decomposition, 

 as the elevation of the ground enables vegetation to encroach 

 on the hollow they were originally preserved in. 



Again, in the case of the mountains, we have no diffi- 

 culty in mentally stepping from one hill to another, untii 

 we reach the sea, and stride down with equal steps beneath 

 its waves, and nothing is wanting in this self-evident gra- 

 dation of elevation. We find also, that as a rule the hills rise 

 at an uniform rate one over the other, from the rock on the 

 beach to Pedro tallagalla ; of course the further inland we 

 go, the more the wear of ages has escarped the rocks and 

 scooped out the valleys, a natural effect of long-continued 

 elevation. 



Wherever the receding sea leaves an elevated coast line, 

 the cocoanut and a few other trees at once establish them- 

 selves, to the exclusion of all others. Now the cocoanut is 

 peculiar and worthy of notice in the way it roots itself, not 

 deeply r for the wear of the breakers allows no deep deposit 

 over the rock, as the beach rises above their influence, but 

 cramped as they are to a few feet of soil, one 

 tree's growth so permeates the surrounding earthy as to des- 



