on the Geography of Plants. 261 



and likewise of August ; p. Ivii., the mean temperature of the 

 year, of summer, and of autumn ; and lastly, p. xxix., the an- 

 nual mean temperature, and the temperature of each of the 

 four quarters of the year, with that of the hottest and coldest 

 months. 



Exclusive of the wish that the author could always have used 

 the same measure of temperature, I must here remark, that 

 although the last of the methods considered appears to be the 

 best, yet neither of them perfectly answers the purpose. In 

 my opinion, the temperature of any place, inasmuch as relates 

 to vegetation, is best shewn by a curve, which exhibits the state 

 of the same during the cycle of vegetation, as given by Wah- 

 lenberg in the Flora Lapponicaj and in his Tentamen de Vegeta' 

 tione et Climate Helvetice. The size of this curve shews, for in- 

 stance, the quantity of heat which the given place enjoys ; and 

 the form of the same shews the manner in which this quantity 

 is distributed during the different parts of the year within the 

 cycle of vegetation. If the curve be very high and narrow, 

 the place has a short but hot summer, as is the case with Eiron- 

 tekis ; if the curve, on the other hand, be low and broad, the 

 summer is long and temperate, as happens generally in the 

 neighbourhood of the sea-coast and in the islands. If the 

 curve have many deviations from a regular form, the tempera- 

 ture is vei-y variable, and so forth. But, as it might be useful 

 to express these proportions by numbers, I would propose the 

 following formula : a number which represents the sum of the 

 quadrate, included within the curve *; then a number for the 

 height, and another for the breadth of the curve, to which must 

 be added one to represent the tortuosities of the same f ; and, 

 lastly, a number for the difference between the temperature of 

 night and day, which is likewise very important, but, in general, 

 from the want of observations, extremely difficult to be obtained. 



• See Wahlenberg's Tentamen, % 83. He proposes this sum as a measure ; 

 but the observation made by himself, ^ 85, viz., that the distribution of 

 heat has an important influence, contradicts it. However, he is not quite 

 satisfied with it. 



t This may be ehewn by a re^lar curve-line drawn upon the curve "of 

 temperature. 



