BY H. I. JENSEN. 647 



subsidence, we should expect to find the water gradually deepening 

 for a long distance out. For these reasons it is probable that 

 Samoa (and Savaii especially) has been not subject to any 

 oscillations of considerable magnitude in recent time. Minor 

 oscillations (amounting to a few feet) of both positive and negative 

 character may have taken place. Dana, who visited Upolo in 

 1839, gives reasons for believing that a slight depression has 

 taken place at the western end of Upolo, but his evidence is very 

 meagre, and is explicable on other grounds. 



The high and rocky ironbound coast at Papalaulelei suggests 

 a faulted coast. It is, however, quite unnecessary to assume 

 faulting to account for this type. When a lava stream reaches 

 the coral reef it is turned back, and a cool crust forms in front. 

 The surface of the stream gradually rises as new lava wells in 

 underneath. If it should pour over the reef into deep water, the 

 advancing stream will tend to preserve an almost vertical front. 

 It is easy to conceive of the waves working in along the line of 

 the old coral reef, and undermining a basalt flow of this character, 

 producing steep seaward cliffs (see Plate lxi.). 



3. Examination and Description of the Volcanic Flows. 



Most of my time in Samoa was spent on the island of Savaii on 

 the study of the lava flows, and chiefly on that of 1905-06. 



(a) Old Floivs. I had not an opportunity of examining the 

 big "Ma" between Aopo and Asau, but I was informed by several 

 reliable residents of the island that this area supports almost no 

 vegetation, and in many places the lava looks as fresh as if it had 

 been erupted yesterday. In a climate like that of Samoa, the 

 vesicular and scoriaceous basalts would tend to decay quickly, so 

 that such a fresh lava-plain cannot be more than a few hundred 

 years old. It has been suggested by some that it was formed 

 fourteen generations (about 350 years) ago. Mr. W. von Billow, 

 an eminent ethnologist resident on the island, tells me that it was 

 probably formed about the year 1800, basing his calculations on 

 information received from the natives. This estimate appears 

 from the description to be the more likely. The other estimate 



