226 



They advocate for the purpose of securing greater uniformity and exactness among 

 analysts the application to all polarizations of a correction formula " JV + 0.00031 tN, 

 where N is the observed reading on the Ventzke scale and t is the difference 

 between the temperature of observation and that at which the polarimeter was stand- 

 ardized." In answer to my criticisms of such a correction when applied to raw cane 

 sugars Messrs. Watts and Tempany reply as follows: 



While not disputing the accuracy of the statement concerning the effect of tempera- 

 ture on levulose, we would point out that the process of determining the polariscopic 

 test of a sugar js purely arbitrary and conventional. We take it that the polariscopic 

 test of any sample of sugar is the rotation produced t>y it when tested in such a way 

 that a sample of chemically pure sucrose tested under precisely similar conditions would 

 give a reading of 100. The 100 point of the Ventzke, or any other sugar scale, is based 

 on the rotation of a standard weight of sucrose, dissolved in a standard volume of water, 

 at a standard temperature. If at any oth-T temperature this weight of pure sucrose 

 will not give a rotation of 100 on the scale, the scale has been altered; consequently, 

 allowance must be made for this alteration in the scale when polarizing commercial 

 sugars under these conditions. 



The above criticism of my previous article is, however, not a valid one. We could 

 say with equal justice: Consequently, allowance must be made for this alteration in the 

 scale when polarizing molasses or honey or condensed milk or glucose or any other sub- 

 stance which is polarized upon a saccharimeter. The only scientific conclusion which 

 could be drawn is allowance must therefore be made for this alteration in the scale 

 when polarizing pure sucrose; to include commercial sugars and other substances is 

 too sweeping and unwarranted a generalization. It is true that the 100 point of the 

 sugar scale of a saccharimeter is based upon the rotation of a standard weight of c. p. 

 sucrose under certain standard conditions; this sucrose, however, is a means of stand- 

 ardization and nothing more. A definite weight of milk sugar can be made to read 

 100 upon any saccharimeter and this weight is used for the estimation of milk sugar in 

 milk products. To apply a correction formula for sucrose in such cases would of course 

 be an absurdity. 



Quartz may also be used for standardization, and is so used, the 100 point of the 

 French sugar scale being based upon the rotation of a plate of quartz 1 mm thick. It 

 might be said, following the same line of argument as that of Messrs. Watts and Tem- 

 pany, that because a standard plate of quartz always polarizes 100 irrespective of 

 temperature upon a quartz-wedge saccharimeter, the scale has not been altered and 

 consequently no allowance at all should be taken of temperature in the work of polari- 

 zation, a conclusion of course perfectly true as regards quartz but not of other sub- 

 stances. Similarly the conclusions worked out for chemically pure sucrose -for a given 

 type of saccharimeter are true for chemically pure sucrose but for nothing else, neither 

 for mixtures of sucrose with other substances nor for products which contain no sucrose. 



The International Commission for Uniform Method of Sugar Analysis in 1900 decided 

 that it was permissible, as in tropical countries, to adjust saccharimeters to a higher 

 standard temperature than 20 C. This adjustment may be made by changing the 

 quartz wedges of the instrument, by increasing the normal weight of sugar, by increas- 

 ing the length of the observation tube, or in other ways. When only local comparisons 

 are involved it is advisable and advantageous to make such an adjustment; there is 

 a serious objection, however, against having several separate standards for universal 

 work, since comparisons are no longer possible upon a large class of low-grade sac- 

 charine products. Two saccharimeters, for example, one standardized for the rotation 

 of sucrose at 20 and one standardized for the rotation of sucrose at 30, will give, of 

 course, identical results for pure sucrose, but not for a raw cane sugar, nor for a cane 

 molasses, nor for a large class of other products. Having adjusted our saccharimeter 

 to any desired standard temperature, this standard temperature must be rigidly 

 adhered to if identical observations are to be always obtained between different 



