10 RESPIRATION. [CH. I 



of C0 2 which is not the case in apparatus in which water 

 takes a part. We have not found the Bonnier-Mangin 

 apparatus so easy to work with as Timiriazeff's eudio- 

 meter, but this is doubtless due in great part to our not 

 having made with it any continuous series of observations. 

 It is claimed by Aubert that the determination of the 

 oxygen, nitrogen and C0 2 in a sample of gas can be 

 finished in a quarter of an hour. 

 (6) Intramolecular respiration. 



To demonstrate the fact, the following simple form of 

 experiment may be tried. 



Soak 6 peas in water for 12 hours, when the seed- 

 coats can easily be removed without injury to the embryo ; 

 the removal of the testa is necessary to avoid introducing 

 air with the peas, the object of the experiment being to 

 show that C0 2 is produced in the absence of free oxygen. 

 Fill a test-tube with mercury and invert it in a mercury 

 trough, which should stand in a strong wooden tray. This 

 precaution is advisable in all experiments involving the 

 use of mercury, so that if any accident occurs the mercury 

 may not escape and get into the cracks of the floor. 



Pass the peeled peas one at a time under the rim of 

 the test-tube so that they float up into the mercury, and 

 occupy the upper end of the test-tube. On the following 

 day it will be found that the test-tube is half full of gas, 

 and the peas are therefore clearly visible, instead of being 

 partly hidden by mercury. 



A few drops of water are now passed in under the test- 

 tube rim with a bent pipette, and a fragment of caustic 

 potash added from below, in this way a strong solution 

 of KHO is supplied, by which the C(X is absorbed. 



