METABOLIC PROCESSES IN THE PLANT. 279 



,'- c.c., dips into a vessel containing mercury. The porcelain dish, 

 />, the bottom of which is perforated, contains the research material, 

 resting on moist filter paper or moist glass wool. It is supported 

 on a glass tripod. The shallow glass, g, contains potash solution 

 to absorb the Carbon dioxide evolved. The whole arrangement, 

 supported on a nickel-plated stand, is placed when in use in a 

 glass case, the front and back of which are parallel to one another, 

 and which is filled with water. We are thus enabled to con- 

 veniently regulate the temperature, and keep it constant. 



At the commencement of an experiment we draw the mercury 

 a little way up the tube r by suction at r l9 close the well-greased 

 stop-cock, and then wait for about half an hour before proceeding 

 to read the position of the mercury in r. The readings must 

 always be made by means of a telescope. We naturally determine 

 also the length of the column of mercury, together with the tem- 

 perature and barometric pressure. It is necessary to know the 

 capacity of the bell-glass, of the tube r, and of the tube r t as far 

 as the stop-cock. From this volume is to be deducted, however, 

 the aggregate volume of the objects in the apparatus (dish con- 

 taining potash solution, glass tripod, porcelain dish, moist filter 

 paper, research material). The volume of these we ascertain in 

 part by immersion in water contained in a graduated vessel. 

 Great stress is to be laid on accurate readings of temperature and 

 barometric pressure. The reduction of the volumes can easily be 

 effected by employing the formula given in 104. 



The most important source of error in the method lies in the 

 fact that at the commencement of the experiments we cannot at 

 once make the readings, but have to wait for about half an hour 

 till the temperature is in equilibrium. But during this half -hour 

 the plants are already taking up Oxygen, and to make allowance 

 for this, we correct the actually found volume of Oxygen absorbed 

 by calculating the quantity corresponding with half an hour, and 

 adding this to it. This correction is naturally a near approxi- 

 mation only when the research material during the period of 

 observation does not undergo any important change in its rate 

 of respiration. 



The potash solution employed for absorbing the Carbon dioxide 

 is accurately weighed. It must be nearly concentrated and per- 

 fectly clear. At the end of each experiment we pour the potash 

 solution into a small flask, dilute with water, precipitate the 

 Carbonic acid with Ba C1 2 , and filter. To wash the precipitate on 



