FORREST SHREVE 
of water. A comparison of the average monthly minimum tem- 
peratures which prevail at the Desert Laboratory and in che bottom 
of the Santa Cruz valley, has already been published by the writer.^ 
The extensive drainage basin of the Santa Cruz, which is bordered by 
high ranges of mountains, receives the cold air flow of such a large 
area that the minimum temperatures in the floor of the valley are 
always much below those of the Desert Laboratory, situated 335 ft. 
above the valley, but only half a mile from its edge. The difference 
between the valley and laboratory minima is greatest on the clear, 
windless nights of spring and autumn, when the valle^^ temperature 
has been as much as 31° F. (30° and 61°) below that of^the laboratory. 
The difference between the monthly mean minima for May is 17.8°, 
that for June 144°, both of these months being dry and their nights 
prevailingly clear. On cloudy nights after heavy rains the minima 
of the two places have approached within 2° (69° and 71°), that of 
the valley being lower. The difference between the monthly mean 
minima for July and August are respectively 7.7° and 8.8°, these months 
being relatively cloudy and damp. 
In Soldier Canon, at an elevation of 5,000 ft. in the Santa Catalina 
Mountains, the temperature of the floor of the cafion has been ob- 
served to be 8° below that of the slope of the cafion 100 ft. above the 
floor. In Bear Canon, at 6,000 ft. elevation in the same range, the 
minimum in the floor of the canon has been observed 7° lower than 
the minimum of the rim of the cafion 1,000 ft. above. Observations 
100 ft. above the floor would probably have revealed an even greater 
difference. The walls of both these canons are clothed with open 
stands of desert and chaparral plants, the latter cafion having extremely 
rocky walls. In Marshall Gulch, at 7,725 ft. elevation in the Santa 
Catalina Mountains, the minimum temperatures are identical in the 
bottom of the gulch and on its rim 275 ft. above. The walls of this 
gulch, or caiion, are heavily covered with forests of pine, spruce and 
fir. The drainage areas of the three mountain canons are, in each 
case, much smaller than that of the Santa Cruz river. 
Our knowledge of the conditions which make cold air drainage 
possible in the desert, leads us to anticipate that it would be less well 
marked, or even absent, at the heavily forested altitudes of the desert 
mountain ranges. The diurnal heating of the soil and other surfaces 
is not so great in the forest, and the nocturnal radiation is retarded 
2 Shreve, Forrest. Cold Air Drainage. The Plant World 15: 110-115, 1912. 
