1846.] IVIURCHISON ON THE SCANDINAVIAN DRIFT. S73 



accumulations in question. In numberless tracts however, where 

 the valleys and lakes are of circular or elliptical shapes and of some 

 breadth, the gravel heaps are (as I have already shown) arranged 

 in isolated hillocks, like the Irish " escars." In other places the 

 linear accumulations are so connected by transverse bands composed 

 of similar materials (and the neighbourhood of Hedemora may be 

 specially cited), that it would be difficult to say that any one direc- 

 tion prevailed more than another. An inspection of the map of 

 Forsell will show, in fact, that the outline of the drifted materials 

 has been determined by the physical outlines of the country and the 

 form of the encasing rocks in each tract. 



Some of the wider low expanses of Dalecarlia, subtended by gra- 

 nitic gneiss or low ridges of other crystalline rocks, are singularly 

 exempt from every sort of transported material. Such are the sands 

 and loams, sometimes finely laminated, that are seen in following 

 the Dal Elf through the fertile tracts of Gustafsland, where the 

 overlying erratics are seen only near the higher edges of the arable 

 lands, and between them and the surrounding granitic ridges. 



Intending to describe the palaeozoic and eruptive rocks of this 

 district on another occasion, I will not now advert to the almost in- 

 numerable examples which Dalecarlia affords of promontories of 

 hard rock, rounded off towards the north ; of ridges which by some 

 cause have been broken up in situ and present the external form of 

 volcanic *'cherres" before alluded to ; of osar of various composition, 

 everywhere exhibiting in their imbedded materials proofs of power- 

 ful aqueous action ; and of large surmounting angular blocks, some 

 of which have evidently travelled only very short distances to their 

 present abodes. 



To show however how the general direction of the drifted mate- 

 rials, whether consisting of rounded boulders or of angular erratics, 

 has been from the north or N.N.W. throughout this portion of 

 Sweden, I would here observe, that throughout the tract extending 

 from the sea to near Leksand, all the detritus (with the exception 

 of local detritus of red sandstone near Gefle) consists of gneissose 

 and granitiform rocks, which distribution is coincident with the 

 fact, that the tract to the north and N.N.W. is also entirely com- 

 posed of such rocks. No sooner huMever do we advance as far 

 west as Leksand at the southern extremity of the great Siljan Lake, 

 than porphyries begin to prevail among the transported materials; 

 and furtlier to the N.W. these broken materials are found to in- 

 crease rai)idly as we approach the great masses of porphyry which 

 occupy the higher hills beyond Mora in the parishes of Elf Dal, etc. 

 These rocks are however mixed with fragments of syenitic granite 

 of more recent date, which has in several portions of that tract up- 

 heaved and dislocated those Lower Silurian rocks of which 1 intend 

 to speak on a future occasion. 



The country at the southern end of the great Wenjan Lake is also 

 extremely interesting as showing the north and south direction of the 

 drift, and also the pha^nomenon of a large area of angidar blocks in 

 situ of very different composition from any hitherto alluded to. In 



