CREATION OF THE LEEWARD SHORELINE 423 



lenticular terrace perhaps 50 meters wide and three-fourths of a meter 

 high. Doctor Anderson furthermore states that in the winter of 1906- 

 1907 the river ran along the broad upper terrace which I have just de- 

 scribed from his plate. 



Leeward Shoreline 



The chattering vibrations of the mountain due to friction of the erup- 

 tive mass ascending through the conduit of the volcano caused the coastal 

 plain or apron from Eichmond village northward for more than 5.5 kilo- 

 meters to be shaken off into the depths of the sea. This apron seems to 

 have been at least 100 meters wide in places — at any rate, the negro vil- 

 lages of Wallibu and Morne Eonde were located on it. At the time of 

 my first visit, in May, 1902, the whole coastline was abrupt except where 

 the larger streams discharged into the sea, and even at these points the 

 under water slope seemed to be steep. The shoreline at the mouth of 

 the Wallibu was almost straight, and was scarcely 50 meters beyond the 

 bluff line of the Eichmond and Wallibu estates. The river mouth was 

 embayed 100 to 120 meters at this time, between benches of new and old 

 ash whose tops were 20 to 22 meters above the sea. In March, 1903, the 

 Wallibu had pushed the shoreline of its mouth nearly out to the points of 

 these benches, while the benches themselves had lost some of their sea- 

 ward extent through the action of waves and currents. The shoreline 

 •was still but slightly convex, and continued to be at right angles to the 

 line of direction of the lower valley. In June, 1908, the shoreline was 

 strongly convex and apparently 100 meters farther out than in 1903, 

 while the northern bench had lost about half its 1902 extent through ero- 

 sion by river and sea. 8 It was evident, furthermore, that much of the 

 material brought down by the Wallibu had been spread along the coast 

 northward by the marine currents. The history of the whole coast from 

 Eichmond village (just south of the mouth of the Wallibu) northward to 

 Larikai point has been the same — the new or augmented headlands had 

 been cut back by the waves, the deltas of the streams (Wallibu, Wallibu 

 Dry, Eozeau, Morne Eonde, Trespe, and Larikai rivers) had been pushed 



* This statement is directly contrary to one made by Doctor Anderson in his supple- 

 mentary report : "The fan of the Wallibu in 1902 extended beyond the coastline, and 

 was very steep. Gushes of hot mud came down it, and tended continually to build it 

 up. In 1907 it scarcely extended beyond the coastline, and both have receded consider- 

 ably" (Philosophical Transactions of the Royal Society of London, series A, vol. 208, p. 

 281), but a careful study of my own and other photographs of the mouth of the Wallibu, 

 together with my field notes, confirms me in holding to the statements made in this 

 paper. No accurate survey of the region has been made, as far as I know, since the 

 eruptions began. 



