BY H. I. JENSEN. 655 



lava was thrust in under the cinder and scoria masses first 

 ejected, and the latter were floated and pushed along on the 

 molten matter beneath and behind. It was not till September, 

 1905, that any actual "live" lava was seen by Mr. Williams, 

 who kept a teen eye on the progress of events. Pere Mennel 

 recognised a distinct lava flow on September 3rd, moving towards 

 the north-west; and a second towards the south on October 14th \ 

 but these were only short flows. 



(c) The Floiv of 1902. — This lava flow, which I also visited, is 

 situated about 10 miles in a direct line from the more recent 

 crater. It belongs to the Aa type, consisting of fragments of all 

 sizes, from cinders the size of peas to blocks many yards in 

 diameter, piled in wild confusion. The lava is vesicular and 

 scoriaceous. The small patches of Aa in the 1906 flows occur 

 where the lava has overwhelmed a very humid scrub or jungle ,- 

 and hence has cooled quickly frbm beneath. The flow of 1902 is 

 situated a couple of miles north-east of Mauga Afi, and some four 

 miles south-south-east of the town of Aopo. Heuce, it is called 

 the Mauga Mu of Aopo. The crater, which is very indefinite and 

 of the pit type, and has built no cone, is 4500 feet above sea-level 

 on the northern slope of the main range. The flow is short, and 

 bifurcates some distance down the side of the mountain. Fog 

 and rain prevented me from making a thorough examination of 

 this locality, and from getting photographs; however, it appears 

 to me that the fragmentary nature of the material met with is 

 only partly due to the causes which usually produce the Aa. 

 Probably the volcano of 1902 was formed by the opening of a 

 fissure in the side of the Mauga Afi. The lava has risen slowly, 

 and partially cooled in the vent, after which, increased pressure 

 has blown to fragments the cooling plug. This process was 

 probably repeated frequently, and served to pile up angular 

 fragments and scoria round the fissure. A subsequent small 

 exudation of lava has carried these masses with it down the 

 mountain side. The stream is less than half-a-mile wide, and 

 only a couple of miles long. Sulphurous vapours are still emitted 

 at the crater. 



