XL] 
THE FROST FORMATION OF SIBERIA. 
61 
speak correctly, however, the frozen earth begins a little from 
the shore under the sea} For on the coast the bottom often 
consists of hard frozen sand—rock-hard sand,” as the dredgers 
were accustomed to report. The frost formation in Siberia thus 
embraces not only terrestrial but also marine deposits, together 
with pure clear layers of ice, these last being formed in the 
mouths of rivers or small lakes by the ice of the river or lake 
frozen to the bottom being in spring covered with a layer 
of mud sufficiently thick to protect the ice from melting during 
summer. The frozen sea-bottom again appears to have been 
formed by the sand washed down by the rivers having carried 
with it when it sank some adhering water from the warm 
and almost fresh surface strata. At the sea-bottom the sand 
surrounded hj fresh water freezing at 0° C. thus met a stratum 
of salt water whose temperature was two or three degrees under 
0^ in consequence of which the grains of sand froze fast to¬ 
gether. That it may go on thus we had a direct proof when 
in spring we sank from the Vega the bodies of animals to be 
skeletonised by the Crustacea that swarmed at the sea-bottom. 
If the sack, pierced at several places, in which the skeleton was 
sunk was first allowed to fill with the slightly salt water from 
the surface and then sink rapidly to the bottom, it was found to be 
so filled with ice, when it was taken up a day or two afterwards, 
that the Crustacea were prevented from getting at the flesh. 
We had already determined to abandon the convenient cleansing 
process, when I succeeded in finding means to avoid the in¬ 
convenience ; this was attained by drawing the sack, while 
some distance under the surface, violently hither and thither 
of the way, or one of the ways, in which such formations arise, we 
obtain from the known fact that mines with an opening to the air, so far 
south as the middle of Sweden, are filled in a few years with a coherent 
mass of ice if the opening is allowed to remain open. If it is shut the 
ice melts again, but for this decades are required. 
^ Middendorff already states that the bottom of the sea of Okotsk is 
frozen. {Sibirische Reise, Bd. 4, 1, p. 502.) 
