[ 148 J 



heat during the winter, it becomes again cooled to its extreme 

 degree foon after the commencement of the following fpring*:; 

 and thus fnfFers as it were an annual tide of temperature, highefl 

 before the beginning of winter, and loweft at the commencement 

 of fpring. 



Hence, by proper obfervations, we have an opportunity of 

 difcovering the proportional increments or decrements of heat 

 communicated to the mafs of earth at certain depths, during the 

 months of any one year, for the purpofe of comparing them with 

 the refpedive increments and decrements of the correfpondent 

 months of any other year : and alfo, from an obfervation of the 

 maximum and minimum of heat in different years, we are 

 enabled to derive the mean temperature of the outward mafs 

 of earth in ixich. years, and thus to make a comparifon between 

 the annual temperatures themfelvcs at different periods. 



At the depth of eighty feet beneath its furface (in covered 

 fituations) the earth is not at all affected, even in many months, 

 by the fuperficial alterations of heat ; and the moft fenfible 

 thermometer is fcarcely competent to difcover any change through- 



* The times of the year at which the maximum and minimum of temperature 

 happen, are different at different depths, earlier at a lefs depth, and later at a greater. 

 Heavy rains about the equinoxes feem to expedite the effeifl : Thus, in the wet 

 autumn of 1787 the temperature began to decreafe in the beginning of Oftober, in 

 fituations, where, in the dry feafon of 1788 it did not take place till the month of 

 December. Count Caflini informed me that the fame obfervation was ufually made 

 in the cave of the Obfervatory at Paris. 



out 



I 



