TEMPERATURE 71 



obtained by averaging twenty-four hour-temperatures recorded by the ther- 

 mograph. The method employed by Meyen*, of deriving the mean 

 directly from the maximum and minimum for the day, is not accurate ; from 

 a large number of computations, the error is always more than two degrees. 

 On the other hand, the mean obtained by averaging the maximum and 

 minimum for the day and night has been found to deviate less than i de- 

 gree from the mean proper. This fact greatly increases the value of maxi- 

 mum-minimum instruments if they are read daily at 6:00 a.m. and 6:00 p.m. 



103. Temperature curves. The kinds and combinations of temperature 

 curves are almost without number. The simple curves of most interest 

 are those for a series of stations or habitats, based upon the level of three 

 feet, or the surface, or the daily mean. The curves for each station represent- 

 ing the different heights and depths and the season curve of the daily means 

 for a habitat are also of much importance. One of the most illuminating 

 combinations is that which groups together the various level curves for a 

 series of habitats. Other valuable combinations are obtained by grouping 

 the curves of daily means of dift'erent habitats for the season, or the var- 

 ious station curves. 



104. Plant temperatures. The direct effects of temperature as seen in 

 nutrition and growth can be ascertained only by determining the tempera- 

 ture of plant tissues. The temperatures of the air and of the soil surface 

 have an important effect upon humidity, and water-content, and through 

 them upon the plant, but heat. can influence assimilation, for example, only in 

 so far as it is absorbed by the assimilating tissue. The temperatures of 

 the leaf, as the most active nutritive organ of the plant, are especially im- 

 portant. While it is a well-known fact that internal temperatures follow 

 those of the air and soil closely, though with varying rapidity of response, 

 this holds less for leaves than for stems and roots. Owing to the very 

 obvious difficulties, practically nothing has yet been done -in this important 

 field. A few preliminary results have been obtained at Minnehaha, which 

 serve to show the need for such readings. Gravel slide rosettes in an air 

 temperature of 24 C. and a surface temperature of 40 C. gave the follow- 

 ing surface readings: Parmelia, 40, Eriogonum, 38.6, Arctostaphylus, 

 35, Thlaspi, 31.8, and Senecio, 31. The leaf of Eriogonum Havum, 

 which is smooth above and densely hairy below, indicated a temperature of 

 31.8 when rolled closely about the thermometer bulb with the smooth sur- 

 face out, and 28 when the hairy surface was outside. The surface read- 



^ Meyen, F. J. F. Grundriss der Pflanzengeographie, 12. 1836. 



