Esmark on the Geohgical History of the Earth. 117 



tumuli, that had been all opened. Urburhill, which reaches 

 out to the sea, lay upon the right hand, and, on advancing far- 

 ther over the plain, you see the dwelling-house of Howkelie, at 

 a little distance to the right, in a valley which stretches up into 

 the hill. At the upper end this sandy plain was bounded by a 

 glacier^dike or rampart, which extended across the whole valley. 

 As this glacier dike is remarkable, and, so far as I know, the 

 only one of its kind lying close to the level of the sea, in a 

 district* where you find only a few heaps of perpetual snow 

 in hollows of the mountains, where it slopes to the north-east, 

 at the height of from two to three thousand Rhenish feet 

 above the sea, I must be a little more particular in describing 

 it. Its length across the valley, from mountain to mountain, 

 is 2250 feet, its perpendicular height above the plain 100. At 

 one of the ends where it approaches to the mountain, it is bro- 

 ken through, so that there the highest part of its brink is not 

 above twelve feet higher than the plain. This opening or 

 breach is not above 200 feet broad. The dike itself consists of 

 coarse gravel and sand, mixed with a great number of immense 

 blocks of gneiss, which is the prevailing kind of rock in the 

 mountain. We find this gravel and sand not only heaped up 

 across'^the valley, but pushed up in great quantity on the oppo- 

 site side of the dike, to the length of 1400 feet towards ihe 

 mountain. 



The whole bottom of the valley is covered with a lake, which 

 is called Howkelie Water, of the same breadth as the length of 

 the dike, and extending about ten thousand feet up the valley. 

 The people in the neighbourhood say that it is one hundred fa- 

 thoms deep. As the surface of the lake is only ten feet higher 

 than the plain on the other side of the dike, and as this, there- 

 fore, where it is lowest, is two feet higher than the surface of 

 the lake, there can be no run from it on this quarter. The wa- 

 ter has its outgate at the other end, by a fall of a few feet, into 

 a similar lake, and from this again, by a little fall at Vasbotten, 

 it passes into a larger lake, called Ejewater, from which it 

 soon after runs into Lyseford. These three lakes lie in a semi- 

 circle. 



From this description it will be easy to see that this dike 



* By observations I made, Stavanger Church lies in Lat. 58' 57' 56". 



