DEPARTMENT OF BOTANICAL RESEARCH. 81 



air. The other junction is placed in a small thermos-bottle with a 

 mercury thermometer. A portable galvanometer (of sufficient sen- 

 sitiveness to detect changes of temperature of less than 0.1° C.) and 

 a damping-key complete the main parts of the apparatus. The various 

 units are arranged on a board about 7 inches square, which can be 

 screwed to a photographic tripod, and the whole can be easily moved 

 about from plant to plant. The calibrations are made in the laboratory, 

 so that in the field the swings of the galvanometer may be quickly 

 translated into temperatures. The galvanometer is sensitive enough 

 so that it makes no difference how near the temperature of the leaf 

 the second junction may be, it being essential only that its temperature 

 be known. Thus there is no need for ice or warmed water to be carried 

 into the field. The second junction is placed in a thermos-bottle to 

 avoid sudden changes in its temperature while the readings are being 

 made. 



The apparatus has been used to measure the temperature of the 

 leaves of a large number of plants belonging to widely different eco- 

 logical types and growing under very dissimilar conditions. Readings 

 have been taken on both upper and lower surfaces and in all cases 

 the leaves have been in situ on plants Hving in their natural habitats 

 or in a greenhouse. The localities in which work has been done com- 

 prise the desert near Tucson, mountains in the \'icinity of Tucson, 

 and the Santa Lucia Mountains in California. The results have not 

 yet been sufficiently assembled to warrant statements of the con- 

 clusions. The most outstanding fact is the rapidity with which the 

 surface temperature of a leaf growing in the open may fluctuate 

 within a time of 20 to 60 seconds, the change frequently amounting to 

 from 1° to 3° C. Such fluctuations occur also in the air temperature, 

 but are too rapid to be registered on a mercury thermometer. In the 

 greenhouse these fluctuations do not ordinarily occur, either on the 

 plant surfaces or in the air. Changes in atmospheric conditions are 

 without doubt the cause of these fluctuations. If a moderately strong 

 wind is blowing the temperature of the leaf-surface may change as 

 much as 5° in 30 seconds. 



Root-growth in Desert Plants and the Oxygen Supply of the Soil, 

 hy W. A. Cannon. 



In the root-soil aeration there are two features of much importance. 

 One of these relates to the amount of carbon dioxide in the atmosphere 

 of the soil, and the other to the supplj^ of oxygen. In arable lands it 

 appears that at the time of the greatest vegetative acti\dty of soil 

 organisms, of whatever kind, there may be a slight excess of the 

 former and a small decreased amount of the latter. Neither the one 

 nor the other, however, is in sufficient amount to materially affect 

 root-growth. Although no direct quantitative determinations have 



