EEEF OF BABELTHDAP. 245 



reef is naturally much narrower; starting from Roll, it takes 

 at most twenty minutes to go across the ridge of the island to 

 the east coast. The narrowest part is a little farther to the 

 south; according to my measurements the island here is at 

 most 3,700 feet * wide. In this place, too, Priedrichsen gives the 

 island a breadth of some miles. Gloss below Aibukit to the 

 south, the island suddenly widens, without, however, anywhere 

 reaching the considerable width of fourteen to fifteen miles at- 

 tributed to it by Fredrichsen ; I am convinced that even at the 

 widest part it is at most from seven to eight miles across. 



The reef which runs round the whole of the narrow island 

 exhibits the following remarkable peculiarities of structure. 

 At the northern point and on the west it is at a considerable 

 distance from the land ; at the latitude of my house (7° 38' N.) 

 it was from four to five miles from a small hill on which I had 

 set up my theodolite. Between the inner reefs lying close to 

 the shore and the exterior reef is a channel of from five to six 

 thousand feet in average width, and from thirty to forty-five 

 fathoms deep. This communicates with the ocean by three 

 channels, which certainly do not lie opposite to large rivers 

 or even brooks. The one marked on Eriedrichsen's map as 

 ' Kavasak passage ' is very narrow, and certainly not navigable 

 for ships. The second, called by Friedrichsen ' Woodin's 

 passage,' is placed by him much too far south ; its position is 

 more accurately given on the accompanying map. I passed 

 through this canal with Captain Woodin himself in the 'Lady 

 Leigh ' in 1861. The third channel is the widest; it is almost 

 due west from the highest point of Babelthuap, which is de- 

 signated in Friedrichsen's map as 'Eoyoss Aremolongui.' The 

 inner lagoon channel runs almost parallel to the coast ; it is not 

 very wide, and the depth varies between thirty and forty-five 

 fathoms, and into it debouch, almost all at a right angle, a 

 number of channels which intersect the surface of the shore reef 

 in all directions; one of the last led us as,much as 1,200 paces 

 inland at high water, and we there anchored close to the per- 

 pendicular bank of living coral which formed the shore of the 

 canal. 



* German feet, about 3,810 English feet. 



