4 TRAUSTt ElNARSSON 



the dry lava field the lavas weather into sand or dust that is moved by the 

 wind and settles in the depressions. Vegetation is naturally very sparse here, and 

 fossilization would be most unhkely. Only around the springs at the rim of 

 the lava field does the vegetation thrive and here it could be fossiUzed. 



The picture of the Lower Tertiary landscape which I have in mind is that 

 of a very extensive flat lava desert. It is mostly dry and without vegetation, 

 only in occasional depressions is there sufficient moisture to sustain an 

 oasis. Rivers are formed during long periods of relative volcanic quiescence 

 and deposit sand and mud in their lower parts. But when a new intensification 

 of volcanism sets in these rivers may disappear. Thus the vegetation is subject 

 to often repeated changes in the water supply. While an oasis or a river bank 

 is destroyed by a lava flow in one place, a wet depression is formed far away 

 at the front of a new lava flow. The species we would find here after a short 

 time possibly give no true picture of the climate or the vegetation of the 

 country as a whole but may rather reflect which species migrate most rapidly 

 into a new, isolated, wet place. 



Practically every place, where we can study a thick section of the plateau 

 basalts, we find that alternating with sections where a thin sediment separates 

 every two lava banks, there are one or more such parts of the section, con- 

 taining 10-20 lavas, with no sediments at all. These lava successions seem to 

 represent great intensification of the volcanic activity and it seems most 

 likely that during such periods vegetation must have been largely destroyed 

 over an extensive area. 



How large such an area may have been we cannot easily answer at present, 

 because the individual lava groups have not been traced in sufficient detail and 

 over sufficiently large areas. 



In the lower parts of the plateau, traces of frost action are unknown, but 

 in the topmost plateau group such signs are common. In addition there occur 

 moraine-like conglomerates, and sometimes distinct moraines, resting on a 

 striated floor. The conglomerates probably are mainly of fluvial nature, 

 but they often seem to have been reworked by frost. In some cases they also 

 show clear signs of erosion by sandstorms, among them the so-called "drei- 

 kanters". Thus, in this topmost group we have clear signs of a severe climate. 



The age of this group is a very important question, but one not yet fully 

 settled. It may be of Upper Pliocene age, and this is suggested by some 

 approaches to the problem. On the other hand this plateau group is older than 

 a complete peneplanation of the country, a differential uplift, a modelling of 

 the landscape at a base level some 300 m above the present one, and finally a 

 general uplift of 300 m in two or more steps, and a grading of the landscape 

 to the lower base levels. One of the later episodes in this story was the forma- 

 tion of a strandflat at a sea level some 100 m above the present one, and after 

 this strandflat had been formed there flowed lavas that have reverse magnetic 

 polarity which indicates Lower Pleistocene age. 



