Birge—An Unregarded Factor in Lake Temperatures. 999 
lake filled with a very viscous fluid, like glycerine, viscosity 
would be far more effective than thermal resistance in opposing 
mixture. 
The amount of the viscosity of water at temperatures be¬ 
tween 0° and 30° may be seen from the following table. 
Table of viscosity, from Landolt and Bornstein. 
Temperature. 
Dynes per sq.. cm. 
Relative viscosity 
0°. 
0.01778 
2.23 
5°..... 
0.01510 
1.89 
10°. 
0.01303 
1.03 
15°..... 
0.01134 
1.42 
20°. 
0.01002 
1.20 
25°. 
0.00891 
1.12 
30°. 
0.00798 
1.00 
From this it appears that water has sufficient viscosity to 
offer some resistance to the action of the wind which attempts 
to move the particles on each other and thus to mix them. It 
appears also that the viscosity is very small, but that it is 
greater at low temperatures and that it increases at a rate 
which rises as the temperature falls. 
Viscosity offers a hindrance to mixture which cannot be 
stated in terms of ergs. It has been impossible to ■ find a 
quantitative relation between thermal resistance and viscosity 
so as to ascertain exactly how much the increase of the latter 
at low temperature would affect the influence of the wind. Yet 
it seems clear that it does not have a great influence. The 
present question is one of the relative influence of viscosity 
at different temperatures, and viscosity plainly increases far 
more slowly at low temperatures than thermal resistance dimin¬ 
ishes. 
If we attempt to mix in a unit of time a column of water, 
the area of whose base is 1 sq. cm. and whose height is 1 m., 
viscosity will offer a resistance to be overcome. From the 
table given above it appears that this resistance is about twice 
as great at 4° as at 30°. The thermal resistance to mixture 
for a temperature difference of 1° decreases 37.5 times if the 
average temperature falls from 30°' to 4°.- Between 10° and 
4° the thermal resistance decreases over ten times, while the 
