280 



On a Method of Studying Transpiration. 



Now 



We will suppose that in an experiment on the effect of illumination we 

 find the transpiration-rate in the light (relative humidity 70 per cent.) to 

 be 120 ; while the rate in the dark (humidity 80 per cent.) is 75. We must 

 multiply 75 by 35/25. The product 105 is the transpiration in the dark 

 room (humidity 80 per cent.) reduced to humidity 70 per cent., and there- 

 fore now comparable with transpiration in the light, i.e. 120. Thus 

 Transpiration in light _ 120 _ 114 

 Transpiration in dark 105 100 ' 



Sir Joseph Larmor has been good enough to point out to me that it is 

 possible to get a rough idea of the temperature of the leaf at full saturation, 

 i.e. of the leaf temperature which in fig. 9 produces the amount of transpira- 

 tion (or what may be perhaps called distillation) equal to AF. The oblique 

 line, or curve of transpiration, cuts the horizontal at 105, i.e. at 5 per cent, 

 above saturation. The figure shows that, in supersaturated air, i.e. 5 per 

 cent, above saturation, transpiration is nil. The hypothetical degree of 

 supersaturation should be a measure of the transpiration AF at the satura- 

 tion point, and therefore of the internal temperature which can distil off 

 water in saturated air. Assuming* the temperature of the air to be 16° C, 

 the vapour pressure would be 13"51. If we add 5 per cent, to this we get 

 14 - 2, which is the vapour pressure corresponding to 16 - 8° or O^ C. above 

 the temperature of the air. There seems no improbability in leaf-respiration 

 producing, under the conditions of the experiment, a temperature of roughly 

 1° C. above that of the atmosphere. In my earlier experiments I concluded 

 that the transpiration curve DEFG cut the horizontal at 107. This would 

 have given a temperature 1 - 1° C. above that of the air, instead of - 8° C. 



It should be noted that the distance AF, i.e. the amount of transpiration in 

 saturated air, will depend on the general temperature, since respiration is 

 greatly influenced by temperature. We have some evidence on this point, 

 but the experiment needs careful repetition. 



It is remarkable that, as far as I know, the method here used for plotting 

 the relation between transpiration and relative humidity has not been 

 employed. If Le Clercf had treated his results in this way, he might, 

 perhaps, have obtained a result like mine. 



I cannot conclude without expressing my indebtedness to Miss D. F. Pertz 



for much kind help in the laboratory. 



* The figure is a diagram not taken from any one experiment, 

 t Le Clerc, ' Ann. Sci. Nat.,' 1883, vol. 16. 



