26 



1 . It must be possible to keep the température constant to 

 within 0.1° C. at least during a longperiod without control. 



2. The température must be quite uniform through- 

 out the thermostat. 



3. No gas may be used for heating. ') 



4. It must be possible to open and shut the ther- 

 mostat quickly without losing much beat. 



5. It must be possible to ventilate the thermostat. *) 



6. It must be possible to rotate an axle running through 

 the thermostat from the outside so as to render it possible 

 in experiments on geotropism to put the plants inside at 

 any desired angle without opening the thermostat, 



7. It must be possible to use the thermostat in expe- 

 riments on phototropism and therefore it should hâve a 

 glass front and a suf&cient depth to ensure a sufficient 

 falling off of the intensity of the light. ^) 



In order to satisfy thèse seven requirements the 

 thermostat had to be différent from the current types 

 used in botanical laboratories, in the first place the method 

 of closing. The ordinary arrangement with a double door 

 always causes a différence of température of several tenths 

 of a degree between the front and the back of the 

 thermostat. Hence I took as base (cf. fig. 8 and plate Ij 

 an inner vessel A, which can be shut hermetically by 

 means of a lid C with nuts, and which stands quite free 

 in a larger vessel B, filled with water, so that the inner 

 vessel is surrounded by it on ail sides. Two stirrers D D 

 placed in the left front-corner and in the right back-corner 

 keep the water moving. They were provided with a straight 

 vertical biade to the full height of the thermostat and 



1) cf. § 8, purity of the air, 



2) cf, § 7, purity of the air, 



3) In the beginning it was my intention to extend this inves- 

 tigation to phototropism too. From lack of time I had to confine 

 myself to geotropism. 



