79 



which might have been more carefully noted and the loss avoided. As 

 the dilution, it' moderate, is of small importance, the object should be to 

 get all or as nearly all as possible of the sugar from the eane. 



The purity of the defecated juices, Table IN', is 1.5 points higher than 

 the dillinsion juices, due to a little destruction of glucose in the clari- 

 fiers. The table shows in addition that there was no inversion in the 

 clarifiers. The scum from the defecators was, as is usual, about as 

 rich in sugar as the juices themselves. These scums were thrown into 

 the ditch, thus entailing a loss of sugar which could and should have 

 been avoided by returning them to the battery. 



For the purpose of comparing the readings of the Brix spindle with 

 the actual total solids obtained by drying and weighing, a number of 

 determinations were made. 



liectangular flat platinum dishes three eighths inch in depth were 

 used and the samples dried for five hours at 100 C. The samples were 

 weighed from a tared flask, about 2 grams being taken in each instance. 

 Duplicates were always made. The use of asbestus as an absorbent 

 agent in drying was also tested. A thin layer of loose asbestus was 

 placed in the bottom of the dish, and the sample dried at and for the 

 same length of time as those samples where the dish alone was used. 



The average of these results in the case of mill juices gives the 

 solids 1.46 per cent, less, and with asbestus as l.CG per cent, less, than 

 the average of the spindle readings. These results arc fully 1 per cent, 

 lower than those recorded by Dr. Crampton at Fort Scott in 1887, but 

 agree closely with results obtained at Douglas, Kans., and at the De- 

 partment this year. In the case of the diffusion juices, the dish alone 

 gave 1.25 percent, solids less, and dish with asbestus 1.40 per cent, less, 

 than the direct readings of the 'spindle. Correcting the percentage of 

 sugar on this basis, in the mill juices it would be .11 per cent, higher, 

 and in the diffusion juices .05 per cent. The purity is also largely in- 

 creased by calculating on the weight of actual solids. 



It may again be remarked here, as was stated in 1887, the use of the 

 Brix spindles standardized for pure solutions of sugar give misleading 

 results, and the solids as determined by direct drying should be relied 

 on. As might be expected, the samples where asbestus was used gave 

 results slightly lower than those without, and it may be stated further 

 that the duplicates with asbestus agreed more closely. The asbestus 

 furnishes a larger drying surface, and less moisture is retained than is 

 the case with a thick film of the sirup. The asbestus should, then, be 

 preferred to the plain dish. 



The average ratio of glucose to. sucrose in the semi-sirups is slightly 

 higher than that in the defecated juices: 



Si mi-sirups 1 



Defecated Juki's 1 



Sucrose. 



3. 86 



