136 HEAT OF LOWEST STRATUM OF AIR. 



evening, and then from these three or four obser- 

 vations respectively to deduce the mean.*" 



If the object be only to obtain from the daily 

 means that of the month or of the year, you can 

 get a good approximation to these, from the 

 temperatures daily observed at any two hours of 

 the same name, especially at ten or at four o'clock, 

 morning and evening. 



By means of comparisons made during several 

 years at various places in the temperate zone, it 

 has been found that the thermometer generally 

 stands at its lowest in the middle of January, and 

 that from that time it mounts, slowly at first, 

 more quickly in April and May, then more slowly 

 again, till by the end of July it reaches its highest 

 mark. Then begins its descent, which being 

 slow at first, is hastened during September and 

 October, and lasts, becoming slower again, to 

 January. 



According to Kaemtz, the 14th of January is, on 

 the average, the coldest, and the 26th of July the 

 warmest day in the year, while the mean tempera- 

 ture of the year falls about the 24th of April, and 

 the 21st of October. 



The mean temperature of the surface of the 



* This process would probably yield even more satis- 

 factory results, if a thermometer was employed, the move- 

 ments of which were retarded by clothing it with a non- 

 conducting envelope. 



