376 ROBERTS EUTYRO-REFRACTOMETER ZEISS BUTYRO-REFRACTOMETER: THE CONVERSION OF SCALE- READINGS TO REFRACTIVE INDICES. BY C. C. ROBERTS, M.A., A.I.C. (Read at the Meeting, November 1, 1916.) IN the book of directions sent out with the Zeiss butyro-refractometer there is a table showing the refractive index for sodium light corresponding to every tenth degree of the scale of the instrument. For 0" of the scale the refractive index is 1.4220, and for 100" it is 1-4895. In a paper by Leach and Lythgoe (ANALYST, 1905, 30,176) i t is pointed out that the change in the refractive index corresponding to a rise of 10" in the scale- reading is less between 90" and 100" than between 0" and 10".If, however, the increase of the refractive index per 1" of scale-reading decreases at a uniform rate, it will be possible to express the relation between the refractive index and scale-rertding by the formula 1000[n],,= 1422 +ax - bx2, where [?&ID is the refractive index for sodium light and xis the scale-reading; a and b are constants.Using the values of [nIn for 0", 50", and 100" on the scale, I find that 1000[n],= 1422 + 0.817~- 0-00142~~. Using these values for a and b, I find that the results calculated by the formula agree perfectly with those in the table supplied with the instrument a t 10, 30, 40, 60, 80, and 90, and that a t 20 and 70 there is a difference of one unit in the fourth place of decimals; from this I think that the formula may be safely w t d for con- verting intermediate scale-readings into the corresponding refractive indices. The following will perhaps be found more convenient for calculation : [%ID= 1.4220 +0*00142~ DISCUSSIOX.Mr. E. R. BOLTON said that by the use of this formula a chart could be construc- ted from which the figures could be read off without calculation. I t could be constructed to read to the fourth decimal place, which probably would not be given very accurately by the slide-rule, while the butyro-refractometer almost gave the fift,h place.While referring to this subject he should like to put in a plea for uniformity of temperature in the recording of refractive indices. He had made it a practice not to record a refractive index at any other temperature than 40' C. ; but one found figures obtained at 15", 20°, 25') 40°, or 45'. Mr. ROBERTS said that his object in working out the formula in the first instance had been to provide a ready means of converting readings when a chart had been mis- laid-as had happened in his own case.