96 



KANSAS Academy of science. 



modification of Fehling's solution. Upon defecating the juice by means of basic 

 lead acetate, and examining the solution with the saccharimeter, it was found that it 

 possessed only slight optical activity, and was sometimes dextro-rotatory, sometimes 

 lievo-rotatory. It was obvious that there was present some substance other than 

 cane sugar, which affected the polarized ray, and nearly balanced the rotation pro- 

 duced by the cane sugar. The cane sugar was therefore determined by the chemical 

 method; i.e.,hj inversion and titration with the copper liquor, correction being 

 made for the reducing sugars originally present. In some cases the cane sugar was 

 also determined by Clerget's method. In this, the polarization of the juice before 

 and after inversion gives the necessary data for calculating the percentage of cane 

 sugar. The results obtained by these methods on the four samples analyzed are 



given below: 



ANALYSES OF WATERMELON JUICE. 



September 17.. 

 September 18.. 

 September 23.. 

 September 26.. 



CANE SUGAR. 



ClergeCs 

 melhod. 



3.43 

 3.31 

 3.32 



Method 

 by inver- 

 sion. 



2.39 

 3.01 

 3.11 

 3.12 



Total 

 reducing 

 sugars. 



4.03 

 4.18 

 3.46 

 3.11 



Specific 

 gravity. 



1.033 

 1.032 

 1.031 

 1.030 



The optical results indicating the presence of levulose, it seemed desirable to de- 

 termine the amount of this sugar present. This was accomplished by destroying 

 the levulose by Sieben's method as described by Wiechmann in the Journal of 

 Analytical Chemistry, vol. IV, p. 257. 



A considerable amount of the juice obtained September 26th had been preserved 

 by adding five per cent, of lead solution to it. On October 30th it was analyzed 

 again as before, and in addition a determination of levulose was made. For this 

 purpose some of the juice was freed from lead by adding sodium sulphate, and di- 

 luted so as to contain about 2+ per cent, of sugar. To 100 cc. of this, 60 cc. of six 

 times normal hydrochloric acid were added, and the whole heated on the boiling- 

 water bath for three hours. By this means levulose is said to be entirely decom- 

 posed while dextrose remains unaltered. After neutralizing the acid and filtering, 

 a determination of the reducing sugar left will give the dextrose originally present 

 in the juice, plus that resulting from the inversion of the cane sugar. Having de- 

 termined the total reducing sugars in the original juice and in the inverted juice, 

 it will be seen that we possess the necessary data for calculating the amount of lev- 

 ulose and of dextrose present in the original juice. The result was as follows: 



Cane sugar, by Clerget's method, 3.30 per cent.; chemical method, 3.27 per cent.; 

 total reducing sugars, 3.14 per cent.; glucose or dextrose, .69 per cent.; levulose, 

 2.45 per cent. 



The large percentage of levulose present is a very interesting fact. The author 

 hopes to study the subject more fully another season, and to confirm his assumption 

 that the laevo-rotatory sugar found is levulose. The sugar has at least these proper- 

 ties of levulose, viz.: A very sweet taste; a strong lasvo-rotation; a reducing action 

 on copper solution; and is destroyed by the hydrolytic action of hydrochloric acid. 



