220 PHYSIOLOGY OF NUTRITION. 



crystallising glass K is supported by an arrangement of wire, which 

 is attached to the under side of the wooden, or better metallic, 

 cover closing the cylinder. The temperature of the air within the 

 cylinder is indicated by the thermometer T. We may experi- 

 ment with twigs of Salix fragilis, which even in the dark trans- 

 pire very vigorously, since their stomata do not close in absence of 

 light. A very leafy twig is cut off, put into a glass containing 

 water, the surface of which is covered with a layer of oil to prevent 

 evaporation from it. The glass is now placed in the large porce- 

 lain dish, after this, and also the crystallising glass If, has been 

 supplied with Calcium chloride. If at the commencement of the 

 experiment, and at the end of every four hours, we weigh the glass 

 and willow twig, we obtain information as to the rate of transpira- 

 tion of the twig in air very free from aqueous vapour. We now 

 replace the Calcium chloride in P and K by water, and let the 

 twig transpire for four hours in air which is now well supplied 

 with aqueous vapour. Having determined the relatively slight 

 loss by transpiration under these circumstances, the twig is once 

 more exposed to air poor in aqueous vapour. It is best to place 

 the apparatus in a room with a north aspect, because in such a 

 room the variations of temperature in the course of a day are only 

 slight. 



In investigating the influence of temperature, the twigs are first 

 kept at a low temperature, P and K having been charged with 

 Calcium chloride. We then warm the air in the apparatus, and 

 expose the object to the higher temperature. A willow twig gave 

 off (always in air poor in aqueous vapour) in every four hours 

 at 21 C. 5-2 gr. of water, and at 32 C. 8'5 gr. of water. 



It is clear that in the last experiment the rate of transpiration 

 must be accelerated, not only by the high temperature of the air, 

 but at the same time from the fact that the object absorbs a rela- 

 tively large quantity of fluid from the warm water. To exclude 

 this latter factor we may employ the following apparatus (see Fig. 

 85), which I also used to demonstrate the dependence of the rate 

 of transpiration on the amount of water present. On the ring of 

 the stand St rests a shallow porcelain dish, P, perforated in the 

 middle. The bell-glass 6r, provided a thermometer, 2 1 , should be 

 rather large (30 cm. high and 15 cm. in diameter). Through the 

 cork K of the water-containing U-tube, Z7, pass the thermometer 

 T and the base of a shoot (e.g. Syringa), the leaves of the shoot 

 being covered by the bell-glass G. The shoot is fixed in the hole 



