ABOUT TERRESTRIAL TEMPERATURE. 91 
always the predominating one, as, from the mobility of its parts, might be anti- 
cipated. 
48. I shall not, I hope, be thought to push the application of the formula too 
far, if I suggest an explanation of the excess of 2°°8 Fahr., which it gives for the 
parallel of 75° as in Table III. I might urge that our knowledge of the isother- 
mal lines in that extreme latitude is so limited, as to leave a wide margin of 
uncertainty in the determination of the mean temperature. This is no doubt 
quite true. But I attribute to another and a manifest physical cause this ap- 
parent anomaly. The amount of the third term of our Formula (par. 33), de- 
pends on the measure of the land, and it is of course here subtractive. But in 
the 75th parallel, though I have estimated the amount of land on the parallel at 
only 28°6 per cent. of the entire circumference (see Table II., and Plate III. 
fig. 1), or a diminution of no less than 20 per cent. from lat. 70°, we know how 
far this is from representing truthfully or practically the physical condition of 
those Arctic wastes. For all the year, except a very few weeks, it is not too 
much to say, that three-fourths of the ocean of lat. 75° is encased in solid ice, 
having the same physical characters superficially (and therefore affecting the 
climate) in the same manner as the adjacent snow-clad land. We must there- 
fore practically regard the land as forming far more than its due share of the 
earth’s surface in those regions, since the water has lost its convective and 
equalizing qualities in relation to temperature. Now, we shall find that if we 
increase the coefficient of the third or land term of the formula from 0°29 to 0°38 
(which seems a very moderate expression of the conditions referred to), the 
excess of calculation in Table ILI. for lat. 75° wholly disappears.* I believe, 
indeed, that it ought to be still farther corrected in the same direction, which 
would justify us in giving to the pole of a water sphere a temperature slightly 
higher than + 12.°5, and thus diminishing the negative errors of the formula 
for lat. 60°—65°, while in lat. 70° the balance would be restored by the sensible 
effect of the frozen sea. With these suggestions, however, I leave my formula as 
it at present stands to await farther researches. 
* Among other tests to which I have put my hypothesis, I have calculated what ought to be 
the magnitude of the coefficient L’ in the third term of the formula, so that it might in each 
parallel represent the numbers of Dove (Table I.), supposing the other constants of the formula to 
be exact. 
PITLOCHRY, September 1858. 

VOL. XXII. PART TI. 2K 

