555 



same, we discover, that the distribution qf heat between 

 summer and winter takes place according to fixed propor- 

 tions ; that is to say, that the variations do not exceed cer- 

 tain limits : but these oscillations, which I have recently ex- 

 amined in a particular paper (M6m. de la Socie'te' d'Arcueil, 

 torn. iii,y are still considerable enough, to have a sensible 

 influence on the culture of the plants necessary to man. If 

 we chose to characterize a vine-climate, it is not enough to 

 say, that the mean temperature of the year must be above 

 8*7° or 9° ; we must add, that in order to have wine worth 

 drinking, the winter must not be below -f 1°, and the sum- 

 mer not below 18*5° or 19°. Now on the New Continent, 

 in the United States, the winters are lower than 0°, in 

 those places, where the mean heat of the year does not ex- 

 ceed 9°. On the isothermal line of 9°, we often see the 

 thermometer fall to — 18° in the range of transatlantic cli- 

 mates. 



It results from these considerations, that, to give an exact 

 idea of the climate of a place, we must determine, not only 

 the mean of the year, of the seasons, and of the months, but 

 also the variations which take place at different hours of the 

 day and night during a time of some extent. When we have 

 a great number of observations to calculate from, 1 believe 

 we ought to determine the annual mean, and the mean of 

 each month, by taking the mean of fifteen years (conse- 

 quently by the help of 10950 partial observations) : and then 

 choose the month, the temperature of which is the nearest 

 to that of the mean of fifteen years, to give as an example of 

 the progress of the thermometer at different hours of the day 

 and night. This method of presenting the observations, as 

 they have been successively made in the space of a month, 

 appears to me preferable to adopting the rule of taking from 

 10950 observations the mean of each day. To characterize 

 a climate, we must not deprive it of what I would venture 

 to call it's physiognomy, it's particular features, the sudden 

 transitions from one temperature to another, the variations 



