51 229 



gas samples from Grafarbakki is probably due to the fact that these samples con- 

 tained a little oxygen. Wliile the gas was in the collecting bottle the small amount 

 of sulphuretted hydrogen became oxidized and consequently disappeared. 



According to the analysis, the spring gases at Grafarbakki contain a small 

 percentage of carbonic acid gas, and it is not inconceivable that the presence of 

 this gas and the larger volume of sulphuretted hydrogen in the spring gas are to 

 some extent altributable to similar causes. 



Around the hot springs at Grafarbakki a slightly larger deposit is to be seen 

 on the stones than at Laugards or Reykir. At Reykir the deposit is infinitesimal 



That there is a greater deposit at Grafarbakki is, I think, partly due to the 

 carbon dioxide contained in the gas; 1 can see no other reason why the deposits 

 should be larger here than at the other two places mentioned. 



As far as the carbon dioxide contained in the gas and the silicious sediments 

 around the springs are concerned, the "livers" at Grafarbakki more resemble the 

 third type of hot springs, i. e. hot springs with considerable silicious sediments, 

 silicious sinters. Of the hot springs I investigated in the summer of 1906, the 

 "hvers" at Hveravellir are the most important representatives of this class. The 

 beautiful, cream-coloured silicious sinters here form two eminences of considerable 

 height and of great extent. 



The spring gases given off by these springs are characterized by a great 

 amount of carbon dioxide, though they contain very little sulphuretted hydrogen. 

 As before mentioned, the carbon dioxide contained in the spring gases probably 

 tends to the production of siHcious sediments. The gas sample Reykjafoss No. 1 

 is of a similar composition. This spring must therefore be considered amongst the 

 springs of this third type, in spite of its very small deposits. 



The spring gas from Hveravellir contains more oxygen than that from Reykja- 

 foss No. 1, and perhaps this has some influence on the solution of the silicious 

 compounds in the boiling spring water, although there are probably other more 

 decisive factors. 



In this connection it is worthy of mention that the alkaline springs at Reykja- 

 foss seem to be very changeable. A good example of this is "Litli Geysir," 

 which was previously a ver yactive geyser, but which has now been inactive for a 

 long time. 



A similar example is to be seen in the great basin from which I collected 

 gas samples in the summer of 1904. It was then a very energetic hot spring, but 

 the basin is now filled with clear, tranquil water, and no gas is evolved. 



The changes in the hot springs at this place seem to have been closely con- 

 nected with the earthquakes which have taken place, no doubt on account of this 

 group of hot springs being situated just on the outskirts of the district most sub- 

 ject to earthquakes. ' In such districts comparatively large displacements must take 



' In connection with the influence of earthquakes on the hot springs of Iceland, see Ji. Thoroddsen's 

 paper in Geogr. Tidsskr. 15, 109, 18i)'.l-l'J00. 



30" 



