state of the Atmosphere for the Science of Meteorology. 43 



culars are to be found, chapter vii. p. 104 et seq. But let us 

 proceed to those propositions. 



I. The average temperature which prevails at any certain 

 place is not that which is generated there by the action of the 

 sun, &c, and which would depend simply on the latitude and 

 the elevation of the ground, but is remarkably changed by the 

 influences of other regions, particularly by the action of the 

 winds. 



II. That average temperature, such as it is obtained any- 

 where from observations during a series of years, for the dif- 

 ferent months or days of the year, will by no means always 

 prevail at those places for the determined month or day of 

 each single year. On the contrary, observations give gene- 

 rally great variations ; and it is precisely the magnitude of 

 these variations which it is of the utmost importance to learn. 



III. What we asserted regarding the temperature in pro- 

 position II. applies equally to all meteorological indications; 

 it is of great importance to become acquainted with those va- 

 riations of the barometer, and of the force and direction of 

 the wind. 



IV. The most efficient means for prognosticating the 

 weather are, the employment of the electric telegraphs and of 

 self-registering instruments, because they facilitate and make 

 possible a tabular union of the variations mentioned in II. 

 and III. 



In the following remarks these propositions are more fully 

 developed. 



Art. I. No one can call in question our first proposition, 

 we shall not therefore demonstrate it; w r e will only observe, 

 that it is the winds that modify this temperature; we will en- 

 deavour to ascertain how much is to be ascribed to each wind. 



Let © represent for longer, for shorter spaces of time, the 

 mean theoretical temperature of a place, and let MT and mt 

 represent the mean temperature deduced from long series of ob- 

 servations, OT and ot the observed mean temperature for that 

 longer or shorter space of time at the same place in a given 

 year ; then we need only annex some distinctive symbols to 

 these three signs in order to show of which space of time we 

 are speaking. Therefore, when we speak of years or seasons, 

 we shall annex the first letters of the words year, y; winter, xv; 

 spring, sp. ; summer, su. ; autumn, a, at the foot of the great 

 letters; and when we speak of months, we shall annex the 

 first letter of the name of the month after the small letters. 

 Thus ©z/ will signify the mean temperature of the year, 0/*the 

 mean temperature of February, which ought to prevail at a 

 certain place, if that place did not receive warmth from and 



