12 STUDIES ON FKUIT KESPIRATION. 



The method of double titration employed by F. W. Morse 1 and 

 by Brown and Escombe 2 was used, in which phenolphthalein and 

 methyl orange are employed successively. It was found convenient 

 to rinse the soda solution into a precipitating jar of about 650 cc 

 capacity. If much carbon dioxid had collected the volume of the 

 solution to be titrated was increased considerably by adding water, 

 and it was found necessary to stir rapidly, and to have the tip of 

 the burette well down in the solution; these precautions tend to 

 avoid loss of carbon dioxid due to the presence of local excess of acid. 3 

 Normal hydrochloric acid was added until the solution was colorless. 

 Methyl orange was then added and the titration finished. The 

 sodium hydroxid solution used contained a small known amount of 

 carbonate whose value expressed in terms of normal hydrochloric 

 acid was subtracted from the results of each titration. Ktister 4 

 has critically reviewed the volumetric methods for the determina- 

 tion of carbon dioxid. He found that this method of double titra- 

 tion was faulty on account of the fact that sodium bicarbonate is 

 faintly alkaline to phenolphthalein, and low results for carbon dioxid 

 may consequently be obtained. If the titration with hydrochloric 

 acid is carried to the point where the solution is colorless to reflected 

 light while still containing a trace of red color when compared with 

 a control, Kiister found that accurate results were obtained, but 

 notes that the end point is empirical. He states also that methyl 

 orange is strongly colored by carbon dioxid and the titration with 

 this indicator should be carried to a normal tint, which is denned 

 as that of an aqueous solution of the indicator, of the same concen- 

 tration as the one used in the titration, saturated with carbon dioxid. 

 The writer was not familiar with the work of Kiister until the study 

 was well under way, and it was then considered best not to change 

 the method. No control was used in judging when the phenol- 

 phthalein pink disappeared, and the end point was thus essentially 

 the empirical one described by Kiister. The error caused by the 

 uncertainty of the end point can hardly amount to more than 1 

 per cent of the amounts of carbon dioxid found. Very accurate 

 results were obtained by Kiister by using the Winkler 5 method in 

 which solution of barium chlorid is added in excess. The titration 

 is then made without filtering, using phenolphthalein. This there- 

 fore appears to be the best volumetric method of determining carbon 

 dioxid. 



1 Loc. cit. 



2 Loc. cit.; Royal Society, London, Philos. Trans. 1900, (B) 193 : 289. 



3 Certain minor refinements used by Brown and Escombe in tbeir work on leaves, in which very small 

 amounts of carbon dioxid were estimated (using sodium hydroxid prepared from sodium, using dilute 

 acid in titrating, and taking measures against contamination from the atmosphere of the solutions to be 

 titrated during their transfer from the absorption apparatus), were omitted, as blank determ i nations were 

 found to give constant results irrespective of slight changes in technique. 



4 Zts. anorg. Chem., 1896, 13 : 127. 



6 CI. Winkler, Massanalyse; see also Bockmann, Untersuchungsmethoden,pp.408,411,and413; through 

 Kiister, loc. cit. 



