26 



RECORD AND DISCUSSION OF TEMPERATURES. 



amplitude is a maximum) about three hours after noon, the whole mean range 

 (obtained graphically) being four hours during the year. In the months of Novem 

 ber, December, and January, there are two minima, one at about 6 A. M. (whether 

 this represents the primary or secondary minimum cannot well be decided from 

 these observations alone), and the other at about 9 P. M. During the remaining 

 months of the year, there is but one minimum during twenty-four hours, which 

 occurs at 1 A. M. 



At about 1\ in the morning, and a little before 8 in the afternoon, the tempera 

 ture equals the mean temperature of the day, excepting the months of November, 

 December, and January. The greatest deviations from these hours are +14 hours. 

 For the remaining three months, the means are reached at 9i in the morning, and 

 also between 4 and 10 in the afternoon; the latter (as well as another hour at 

 about 4 A. M.) being very irregular. 



For three months of the year, the diurnal variation is exhibited graphically. 



The diagram shows the maximum diurnal variation in the month of April, the 

 secondary minimum variation in the month of July, and the diurnal variation in 

 the month of November when nearest to its absolute minimum. + indicates a 



&amp;lt; ID &amp;gt; temperature than the mean of the day. The month of November ex- 

 1 lower j 



hibits considerable anomalies. In July the sun never set, and in November he 

 never rose above the horizon. 



In the following diagram, I have exhibited the annual march of the diurnal 

 amplitude for each month. 



