590 Holsts Studies in Glacial Geology. 
the sand and mud was washed away, leaving the stone and gravel 
deposits of the rivers in the shape of ridges. Hummel was the first 
one to recognize the fact that the existence of an inland ice must be 
pre-supposed as the indispensable agent in forming such ridges; ` 
but he regarded them as formed beneath the ice in tunnels exca- 
vated by percolating waters. Finally, in 1876, Holst published his 
new theory, which in 1878 was also used by Warren Upham in his 
report of the geology of New Hampshire. Dr. Holst’s theory 
stands now without a rival. 
The following short extract of Dr. Holst’s paper may be sufficient 
to explain his views. He says :— 
“ For a correct interpretation of the origin of gravel-osar it is of 
main importance to answer the question how it was possible for 
running water to deposit its silt in such shape as that of an ordinary 
gravel-ose.” There can be no further dispute that these deposits are 
old river-beds. How, then, have they happened to assume the 
form of elevated ridges, rising above the adjacent country? ” 
The water at the surface of the melting glacier gathers in the val- 
leys of the ice-sheet, whence it extends its coastward course in rivers 
whose beds are cut down in the ice-sheet. The ice along these rivers _ 
melts faster than that farther off, and in melting it gradually releases 
its contents of moraine matter. This matter will then follow the 
water, although at a far slower rate, down to the said valleys, and is 
‘finally—at least partially—swept along into the rivers, where the 
gravel-grains will be worn and their angles will be more or less 
rounded off, according to the swiftness of the current. This same 
factor will also regulate the assorting of the material and determine 
the place where each individual boulder, pebble, or grain shall be 
dropped. Layer upon layer will thus accumulate in the river- 
bed, and, when finally the entire glacier has melted away, the accu- 
mulated silt of the former river-bed will present itself as a ridge 
elevated above the surrounding ground ; it is an ose. 
“Tosumup: ` 
“ Osar are formed in running water. 
“ No running water could lift all this matter to the considerable 
altitudes where we often find it. 
1 Geol. of N. H. Vol. 3., pages 14-176. 
1 Swedish äs (pronounced ose), plur. dsar—osar—not “ osars,”’ aS it 18 
written by some authors. 
- 
