DEPARTMENT OF BOTANICAL RESEARCH. 77 



gradients. Rainfall data for the summer season have been secured 

 at seven stations, at vertical intervals of 1,000 feet, for the third 

 year, and they are found to form a far more satisfactory basis for 

 determining the relation of rainfall and altitude than do data from 

 the scattered Arizona stations of the Weather Service. The winter- 

 temperature data for the same stations are confined, perforce, to the 

 absolute minimum for the interval between the last visit of autumn 

 and the first of spring. These data have, for a second time, exhibited 

 a lower minimum at 6,000 feet elevation, in the open chaparral zone, 

 than at 7,000 feet in a pure stand of yellow pine, or at 7,700 feet in 

 the spruce and fir forest surrounding the montane garden. The 

 difference of minimum between 6,000 feet and 7,700 feet is 6° F. 

 Such a reversal of the temperature gradient is not to be related to 

 cold-air drainage, for the stations at 6,000 and 7,000 feet are on 

 ridges, >vhile that at 7,900 feet is in a deep canyon which receives the 

 cold-air flow from the main ridge of the range. The explanation of 

 the phenomenon is probably to be sought in the greater nocturnal 

 radiation of the soil in the unforested altitudes as compared with the 

 heavily timbered elevations, and suitable experimental tests of this 

 possibility will be sought during the coming winter. 



The Water-relations of Plants, by B. E. Livingston, 



The systematic study of the fundamental water-relations of plants, 

 in charge of Professor B. E. Livingston, has continued through the 

 year, as have also Professor Livingston's climatological studies in 

 relation to plant distribution in the United States. As has been pre- 

 viously announced, these climatological studies are planned as a part 

 of a collaboration on environmental conditions and phyto-geography 

 between Professor Livingston and Dr. Shreve. The manuscript and 

 charts of this work are now nearing completion, having been greatly 

 delayed by unforeseen complications in the climatological calcu- 

 lations, etc., the work thus consuming more time than was expected. 



The investigation of the water-relations has definitely reached 

 the stage where quantitative knowledge of some of the subterranean 

 environmental conditions is necessary, and the past two years have 

 been largely devoted to the elaboration of physical methods which 

 may be of value in this connection. The growth of plants depends, 

 as far as water-relations are concerned, upon the amount of water 

 available within the plant body. This, in turn, is a function of two 

 variables: (1) the possible rate of water-supply to the water-con- 

 suming tissues and (2) the rate of water consumption, mainly loss by 

 transpiration. 



The first part of these investigations involved studies of trans- 

 piration and the conditions controlling it. In this connection a 

 fairly satisfactory set of quantitative methods has been devised 



