126 Professor Merian on Ground-Ice 
therefore be supposed, that water which is near the point of con- 
gelation, has a tendency to remain above, and that, consequently, 
the ice should begin to form at the surface. But although this 
is the case with regard to standing water, it is different with 
respect to streams ; in them the ground-ice is really formed up- 
on the soil which constitutes their bed. Any person may easily 
convince himself of the fact, in a pretty severe winter, by exa- 
mining what takes place in a river which carries ice, and which 
is shallow enough to let him see the bottom. In the winter of 
1823, the canal of St Alban, which conducts the waters of the 
Birse through the town of Basle, carried a considerable quantity 
of ground ice. The clearness of the water was such that ob- 
jects were distinctly visible at a depth of three feet. The bed 
of the canal in this place is covered with rolled pebbles. Where- 
ever a projection was perceived at the bottom of the water, in 
the deep places as well as those less so, there was seen a bundle 
of bits of ice which had been formed there, and which at a dis- 
tance presented the appearance of cottony tufts. In several 
places nearly the whole bottom was covered in this manner, and 
the fasciculi were detached from time to time, and came up to 
the surface of the current, which is very rapid. The fasciculi 
which were taken up by the hand from the bottom, presented ex- 
actly the same appearance as those which rose to the surface of 
themselves, and which were floating about in great quantities. 
They were composed, like the latter, of small rounded and agglo- 
merated plates of ice ; so that no doubt could exist with regard 
to the origin and mode of formation of the floating ice. The 
uniform and peculiar arrangement of the ice which appeared at 
the bottom of the water, entirely excludes the supposition that it 
might have been precipitated from the surface. 
The following explanation of this phenomenon is what M. 
Merian considers the most natural. If it be true, that, in win- 
ter, running water is first cooled at the surface, it is also true, that 
its constant agitation, especially when aided by a wind blowing 
in a direction contrary to that of the current, continually mixes 
the water of the surface and that of the bottom, notwithstand- 
ing the inconsiderable difference of their specific gravities. The 
temperature of the bottom and that of the surface, even in pretty 
deep rivers, does not present any remarkable difference, while 
