THE PRODUCTS OF PHOTOSYNTHESIS 193 



not be distinguished.^'''^ Davis and Sawyer have shown that a mixture 

 of 0.01 gram of arabinose plus 25 grams of sucrose yield 20 per cent 

 more furfural than without the sucrose, while 0.02 gram arabinose plus 

 0.25 gram sucrose (the proportion usually present in the plant extract 

 which they were examining) was 15 per cent high for pentose alone. 

 The presence of hexose sugars in great excess is therefore apt to cause 

 serious error in the determination of pentoses by means of the furfural 

 method. 



Another fact which must be taken into consideration in this method 

 is that strong hydrochloric acid has a destructive effect on furfural so 

 that low results are obtained. The passage of a slow stream of steam 

 through the distillation mixture is sufficient to carry over the furfural and 

 thus avoid the destruction thereof by the strong acid according to Pervier 

 and Gortner.^^ 



A variety of methods have been employed for the determination of 

 furfural. The reagents which have been used include phenylhydrazine, 

 barbituric acid, sodium bisulfite, Fehling's solution and phloroglucin. The 

 latter in which the insoluble compound formed with furfural is weighed 

 has found most general use. On the basis of the difference in solu- 

 bility of the furfuralphloroglucide and the methylfurfuralphloroglu- 

 cide in 96 per cent alcohol at 60° the amounts of pentoses and methylpen- 

 toses can be differentiated. A critical discussion of the details of these 

 methods is given in the book of van der Haar and also in the papers by 

 Pervier and Gortner. The latter have also devised a method in which potas- 

 sium bromate is used which has some advantages over the phloroglucine 



method. 



In general the method of pentose determination by means of furfural 

 formation is not altogether satisfactory, as it is apparently very difficult 

 for different workers to obtain reproducible results. 



An alternative method of determining the pentoses in plant material 

 is based upon the fermentation of the hexose sugars and the determina- 

 tion of the remaining unfermentable sugars.^^ Here, however, great care 

 must be exercised to obtain strains of yeast which do not affect the 

 pentoses and by special tests to determine that the residue is completely 

 fermented, as galactose ferments very slowly under certain conditions, 

 and that the unfermentable residue is actually pentose and not some other 

 unfermentable sugar. If this has been established the pentoses then can 

 be determined by means of copper methods. 



'^"Warnier, Rec. trav. chim., 17, 377 (1898). Brauns, Fhann. JVeekblad. 46, 

 326 (1909). Krober, Rimbach and Tollens, Zeit. angczv. Chem., 15, 508 (1902). 

 Kluyver, Dissertation Delft., 1914. Davis and Sawyer, Jour. Agri. Sci., 6, 406 



"Pervier and Gortner, Ind. and Eng. Chcm., 15, 1167, 1255 (1923); 16, 97 

 (1924). Extensive bibliography. 



"Davis and Sawyer, Jour. Agri. Sci.. 6, 406 (1915). Spoehr, Pub. No. 287, 

 Carnegie Inst. Washington, p. 36 (1919). 



