84 PROFESSOR FORBES’ INQUIRIES 
parallel. In an intermediate latitude, which seems to be about that of the Medi- 
terranean Sea, these opposite influences are neutralized. These effects may be 
traced both on the common map of isothermal lines, and in M. Dove’s maps of 
the “Thermic Anomaly.” In the former, the curvature of the isothermal lines is, 
in tropical countries, usually concave towards the equator over the continents,— 
that of Africa, for example,—but it is convex towards the equator, or the iso- 
thermals bend to the south in passing over the continental portions of higher 
latitudes, as in Siberia. In the maps of the ‘“ Thermic Anomaly,” the effect of 
land is still more apparent, being denoted by ovals of relative heat in the con- 
tinents of the warmer latitudes, while ovals of relative cold occupy the vast 
northerly areas of Asia and America. 
31. Sixthly, From an inspection of the above-mentioned charts, it is pretty 
evident that about the 40th or 45th degree of north latitude is, on an average, that 
where the distribution of land and water is a matter of indifference as regards 
temperature. We have seen in the last paragraph that the climatic influence of 
land increases in both directions from the neutral parallel, as the climate assumes 
more of the tropical or arctic character. As a first approximation, we will 
assume that the term depending on the land is effected by the co-efficient cos 2X, 
which makes it zero at 45° of latitude, and gives it a positive value below that 
latitude and a negative value above. Of course, any odd power of cos 2A 
would answer to the same conditions. Hence the Land term of the formula will, 
in terms of this and two preceding paragraphs, be thus expressed :— 
C. L. cos 2a 
where C is a constant expressing the excess of equatoreal temperature on a sphere 
all land above that of one all water; and L is the fraction of land compared to 
the circumference of the parallel. It will also be more in accordance with physical 
principles if we take for L its average value over a certain space north and south 
of the parallel under consideration. 
32. Collecting the terms of the formula, we have 
T,=A+B cos "0+C. L. cos 2a 
which contains four unknown quantities, A, B,C and m. My next procedure was to 
obtain these constants by elimination between four equations furnished by Dovr’s 
temperature of the parallels (19). These being projected in a diagram, and 
an interpolating curve drawnamongst them, the ordinates of that curve* were 
taken for the latitudes 0°, 30°, 50°, 70° north, as expressing best the course of 
climate observed in one hemisphere. The equations were (on Fahrenheit’s scale) 
* It may be satisfactory to add, that the results obtained by using M. Dovz’s numbers without 
any modification lead to an almost identical result when the same latitudes are employed. 
