DISTRIBUTION OF TEMPERATURE ON THE EARTH. 235 
Petersburg. 
In a treatise published in Poggendorff’s Annals (xxiii. p. 54), 
Professor Dove has endeavoured to prove that the form of the 
isothermal lines does not depend entirely upon the direction of 
the wind, and he finds in the above results a new proof for his 
assertion. As the temporary thermal relations depend upon 
temporary currents, so, if the same were the case with the mean 
relation, our results ought to show an agreement between places 
which are isothermally connected. As regards the mean distri- 
bution of temperature, it must be allowed that Iceland is much 
more closely connected with Europe than with North America. 
But as in the variations from the mean it agrees more with North 
America, we see that the cause of this variation cannot also be 
the principal cause for its isothermal position in the general dis- 
tribution of temperature, or in other words, “ the mean direction 
of the wind exerts a smaller influence on the mean temperature 
of a place than the temporary direction does on its modifica- 
tions.” The different distribution of land and sea may make a 
R 2 
