36 



THE CAKBOHYDRATE ECONOMY OF CACTI. 



for this, the dextrose supplied by the United States Bureau of Standards. 

 The thiosulphate difference between the original standardized copper solu- 

 tion and the value found for the residual copper solution of a determination 

 with sugar represents reduced copper from which can be calculated the 

 TABLE 4. amount of sugar present. With some foresight, the sugar 

 solutions to be determined can be so made up and the 

 amount of copper solution as well as the concentration of 

 the thiosulphate solution used can be so arranged as to fall 

 within the range of the most accurate work. This can be 

 determined by reference to the valuable table of Peters. 



Experiments with a known sugar solution gave the re- 

 sults shown in table 4. 



For very small quantities of sugar a more dilute copper 

 solution can be used, though in this investigation the above 

 copper solution was employed throughout. Thus a solu- 

 tion composed of 17.3 grams copper sulphate, 173 grams sodium citrate, 

 100 grams anhydrous sodium carbonate made up to 1,000 c. c. with water 

 gave good results with small quantities of sugar. 1 



THE ESTIMATION OF THE PENTOSE SUGARS. 



The accurate estimation of pentoses in mixtures with other sugars is 

 associated with many difficulties. It may, therefore, not be amiss to give 

 here briefly my experiences and conclusions regarding the methods of 

 determination, as a great deal of time was spent in casting about for the 

 proper means of estimation of this group of sugars. The method which 

 has found most general application is that of Tollens, or one of the many 

 slight modifications thereof. This method is based upon the property of the 

 pentoses of their partial conversion into furfural by the action of mineral 

 acids. The sugar mixture is distilled with hydrochloric acid (usually 12 

 per cent) and the furfural is collected in the distillate. After neutraliza- 

 tion, the furfural can be determined in a number of ways ; for instance, 

 the gravimetric estimation of the phenylhydrazine or phloroglucine com- 

 pounds; the titration with phenylhydrazine (aniline acetate as indicator) 

 or with Fehling's solution ; or the employment of an excess of phenylhydra- 

 zine and gasometric determination thereof ; finally, the titrimetric method, 

 in which an excess of sodium bisulphite is added to form the aldehyde addi- 

 tion compound and the unchanged bisulphite titrated with potassium per- 

 manganate." Careful examination has shown that all of these methods are 



1 BENEDICT, S. R. A reagent for the detection of sugars. Jour. Biol. Chem., 5, 485- 



487, 1909. 



2 JOLLES, A. Ueber ein neues Verfahren zu quantitativen Bestimmung der Pentosen. 



Zeit. f. Anal. Chem., 45, 196-204, 1906. 



MAQUENNE, L. Les sucres et principaux derives. Page 312, 1900. Paris. 

 TOIXENS, H. In Abderhalden's Biochemische Arbeitsmethoden, II, p. 99, 1910. 

 Dox, A. W., and G. P. PLAISANCE. A comparison of barbituric acid, thiobarbituric 



acid, and malonyguanidine as quantitative precipitants for furfural. Jour. 



Amer. Chem. Soc., 38, 2156-2164, 1916. 



