Leonard Hawhes — Rock Stream in East Iceland. 99 



part of the material composing the " Hraun" comes from Skumhottur, 

 but I am not satisfied that "the liparite is for the greater part 

 intrusive here" (2, p. 276). The acid series is estimated to have 

 a maximum thickness of 1,800 feet, and though the actual junction 

 was covered with snow (June month) the basalts did not show any 

 signs of disturbance commensurate with that to be expected if the 

 main part of the series were intrusive. The acid rocks here are 

 similar to those which the author has described from the neighbouring 

 Seydisfjord (4, pp. 391-2), consisting of tuffs, breccias, obsidian, and 

 liparite. One breccia underlying the main mass contains large 

 fragments of liparite, pitchstone, and obsidian, and the farmer at 

 StakkahlvS has some petrified tree trunks which were obtained here. 

 Thus with proof of the extrusive nature of part of the series, and in 

 the absence of indications of intrusion with the exception of some 

 dykes, it seems probable that the main mass was extruded in 

 Tertiary times, as is the case with so many occurrences of acid rocks 

 in East Iceland (5, p. 468). 



The chief objection to the hypothesis of a lava origin is the nature 

 of the "Hraun" itself. Examination revealed the fact that the 

 rock stream is not, as has been stated, exclusively composed of 

 spherulitic liparite, but contains pitchstone, obsidian, pumice, and 

 basalt. A beautiful mahogany obsidian is especially noteworthy, 

 and all the rocks mentioned are to be seen in situ in the 

 Skumhottur mountains. In the gorge of the Hraun a the "Hraun" 

 is seen to be fragmental and composite to the base and to contain 

 much gravelly material. 1 I saw no sign of a liparite dyke in the 

 gorge, or any evidences of fumarolic action in situ as reported by 

 Thoroddsen. 



Perhaps the most interesting part of the rock stream, which is 

 about one and a quarter miles in length, is its termination in the 

 main fjord valley. Near the fjord the blocks become smaller until 

 the final fan of .fine fragmental material is reached, resting on the 

 floor of the Bardarstadadal, north of the FjarSara. These deposits 

 are seen in PL VI, Fig. 1. South of the river, at and to the west 

 of Sseverendi, are a number of conical mounds commonly composed 

 of liparite fragments, with obsidian and basalt less frequently. One 

 of the largest mounds was elliptical in plan, 100 feet by 80 feet, 

 with a maximum height of 23 feet, and contained a block of liparite 

 7 feet in diameter. Proceeding westwards the mounds become 

 fewer and further between until about half a mile from Saeverendi 

 the last one occurs close to the southern wall of the valley. Many 

 of the mounds are exclusively formed of one type of rock, some of 

 obsidian and others of liparite, and they furnish the clue to the 

 problem of the origin of the "Hraun". Owing to its jointing 



1 In connexion with the lava hypothesis as advanced by Thoroddsen it is of 

 interest to note that the block surface of some lavas in the Cordilleras of 

 South America was taken by Humboldt and de Boussingault as evidence 

 of eruption in fragmentary form, but Scrope considered this an " improbable 

 hypothesis", pointing out that the fragmentary nature of lava streams is 

 a property restricted to their upper and under surfaces, the main mass being 

 compact (6, p. 70). 



