COLD AIR DRAINAGE 



113 



is 0.5°F. (0.2°C), occurring on the night after a fall of 0.98 inch 

 (2.5 cm.) of rain. 



On several occasions readings of the minimum for the same 

 night were made in the mountains on ridges and in the valleys 

 1000 feet below. These were taken in the humid mid-summer 

 months, and the readings differed by from 4.5° to 8°F. (2.5° to 

 4.4°C), which is to say they were not quite so great as the differ- 

 ences between the two laboratories. On an evening in late Sep- 

 tember a comparative set of temperature readings was taken on 

 the floor of a canon at 5000 feet elevation (1525 m.) and on the 

 side of the canon 100 feet (31 m.) above the floor. The first of 

 the readings was taken five minutes before the sun set in the canon. 



In returning to the floor of the canon after making each of the 

 readings on the side, it was possible to notice an abrupt change 

 of temperature at a particular level, which level became higher as 

 the evening advanced. The floor of the canon was filled with a 

 veritable stream of cooled air with a very definite surface. At 

 4.30 the next morning the sky had become heavily overcast and 

 the difference had fallen to 5°F. (2.7°C.) . The greatest difference 

 of temperature due to cold air drainage was, accordingly, nearly 

 the same as the average apartness between the two laboratories 

 during September. The evidence afforded by this series of read- 

 ings and by the first autumnal effects of frost on vegetation in 

 narrow canons, shows that the stream of cold air is always a shal- 

 low one, probably never exceeding 60 feet (18 m.) in depth. 



In order to determine a true temperature gradient for the moun- 

 tains from the data in hand it is necessary to compare the three 

 stations situated on ridges with the Desert Laboratory, and to 

 compare separately the 8000 feet canon station with the Accli- 

 matisation Laboratory. The accompanying diagram (fig. 1) 



THE PLANT WORLD, VOL. 15, NO. 5, MAY, 1912 



