12 M. Humboldt a/i Isothermal Lines ^ and the 
gree of latitude, it descends, on the contrary, in North America, 
and Eastern Asia, to the parallel of from 53® to 58®. But the 
direction and the inflexions of this curve of 32“ of temperature 
influences the neighbouring isothermal lines in the same man- 
ner as the inflexions of the magnetic equator modify the lines 
of inclination. To demand what is the mean temperature, or 
what is the magnetic inclination under a particular degree of la- 
titude, is to propose problems equally indeterminate. Though, 
even in high latitudes the magnetic and the isothermal lines are 
not rigorously parallel to the magnetic equator, and to the curve 
of 32° of temperature ; yet it is the distance of any place from 
this curve which determines the mean temperature, as the in- 
clination of the needle depends on the magnetic latitude. 
These considerations are sufficient to prove, that the empiri- 
cal formulae of Mayer require the introduction of a co-efficient, 
which depends upon the longitude, and consequently on the 
direction of the isothermal lines and their nodes with the ter- 
restrial parallels. Mayer had no intention of disengaging the 
results which he obtained from the influence of all disturbing 
causes: He limited himself to the determination of the efiects 
of altitude above the level of the sea, and those of the seasons, 
and the length of the day. He wished to point out the way 
which philosophers ought to pursue in imitating the method of 
astronomers. His Memoir was written at a time when we did 
not know the mean temperature of three points on the globe ; 
and the corrections which I propose after tracing the isothermal 
lines, so far from being incompatible with the method of Mayer, 
are, on the contrary, among the number of those which this 
geometer seems to have indistinctly foreseen. 
Klrwan, in his work on Climates, and in a learned Meteoro- 
logical Memoir, inserted in the eighth volume of the Memoirs 
of the Irish Academy, attempted at first to pursue the method 
proposed by Mayer, but, richer in observations than his prede- 
cessors, he soon perceived, that after long calculations, the 
results agreed ill with observation In order to try a new 
method, he selected, in the vast extent of sea, those places 
Kirwan's Estimate of the Temperature of the Globc^ chap. iii. 
