ON THE RESPIRATORY FUNCTION OF STOMATA. 19 



The chief part of my own work (20) on this subject was 

 carried out in 1893 and the whole of it before Stahl's paper 

 appeared. At that time the only piece of experimental 

 work, among those given above, that was held to be above 

 suspicion or explanation was that done by Boussingault, and 

 on this elementary teaching, at all events in England, has 

 been hitherto based. Feeling the need for direct determina- 

 tions of the intake and output of C0 2 by stomatic and asto- 

 matic leaf areas respectively, and finding all the recognised 

 methods of estimation insufficient for my purpose, I com- 

 menced, keeping specially in view the needs of biological 

 research, to simplify the technique and improve the delicacy 

 of the one which offers the greatest potentiality of accuracy, 

 namely, the volumetric estimation of the CO, by standard 

 solutions of baryta and of an acid. After more than a year's 

 work I have succeeded in designing an apparatus of 

 great chemical delicacy and specially adapted to biological 

 work. 



Thus by its aid the evolution of CO, by a small area of 

 a foliage leaf can be estimated from hour to hour without a 

 break, for any desired time, while for the same area of leaf, the 

 more active absorption of CO, in assimilation can be easily 

 determined for such short periods of time as fifteen minutes, 

 and that at the same time separately for the two surfaces of 

 one and the same leaf area. Further, for the purposes of 

 this assimilation, a current of air containing any desired pro- 

 portion of CO,, however small, can be supplied continuously 

 to the tissue under investigation. Moreover, strictly com- 

 parable experiments can be carried on simultaneously as the 

 apparatus is practically in duplicate throughout. 



The actual estimation of the CO, is accomplished by the 

 well-known method of absorption by baryta solution and 

 titration with hydrochloric acid. The novelty consists in 

 this, that only a very small quantity of baryta solution 

 (under 15 c.c.) is employed in each experiment, and that 

 after the absorption the whole of this is titrated with acid 

 in the tube in which the absorption has taken place. 

 Further, the burettes containing the standard solutions are 

 always in air-tight connection with this absorption chamber, 



