1, Ventzke's first modification of Biot's apparatus; 2. Jellet's compensating saccharimeter; and 3. Laurent quartz wedge (Soleil) saccharimeter.

FIGURE 17 (continued.)-Old types of saccharimeters.

4. Soleil-Duboscq saccharimeter (sensitive-tint biquartz plate and double quartz-wedge compensator); 5, Schmidt & Haensch type of Soleil saccharimeter; and 6, Schmidt & Haensch type saccharimeter with Jellet-Cornu polarizer.

FIGURE 17 (Continued).-Modern types of saccharimeters.

1, J. & J. Fric (single wedge); 2, Schmidt & Haensch (Lippich polarizer); and 3. Jobin and Yvon.

FIGURE 17 (Continued).-Modern types of saccharimeters.

4, Bausch & Lomb (American made); 5, Bates type (J. & J. Fric) (adjustable sensitivity type).

saccharimeter capable of being read to 0.01 percent in the hands of chemists at an earlier date.

(c) ADJUSTABLE SENSITIVITY TYPE

In order to build a white-light polariscope with an adjustable sensitivity, the analyzing nicol would have to rotate through a definite angle as the halfshade angle is varied. In such an instrument it is not sufficient that the halfshade angle be merely adjustable, but the far more difficult requirement must be met of having it adjustable without shift of the zero point. This is indispensable for two reasons: first, to permit of making polarizations with the same rapidity as on an ordinary saccharimeter; and second, to eliminate the danger of the observer not properly resetting the analyzing nicol whenever the halfshade angle is changed. The designing of an instrument to meet. these requirements would be comparatively simple if the vibration plane of the analyzing nicol bisected the halfshade angle when the halves of the field were matched. In general, the intensities of the light emerging from the large and small nicols of a Lippich system are unequal. Hence when the analyzer is set for a match, its position is different from what it would have been had the beams been of equal intensity. The angular difference, 8, between the two positions of the analyzer was found by Bates [2] to be

δ tan

(tan)+0.0103 a.

(15)

It is thus evident that if the vibration plane of the analyzer always bisects the halfshade angle, the zero of the instrument will be in error by the angle, 8, and the amount of the error depends on the halfshade angle, a. The zero error, 8, is much smaller and more nearly linear with a than had been suspected. Hence in order to have a negligible change in the zero point as the halfshade angle is varied, it should be necessary to make only a slight modification of the ideal gear ratio in which the analyzing nicol always turns through one-half the rotation of the large nicol of the Lippich polarizer.

The instrument shown in figure 18 was designed at the Bureau of Standards in 1907 [2] to fulfill the theoretical conditions mentioned above, and is one of a number built for the Bureau and the United States Customs Service. It was constructed by the firm of Josef & Jan Fric, Prague, and has been brought to a high degree of perfection. It is double quartz-wedge compensating and has a Lippich polarizing system with the highest grade of nicol prisms. The adjustable sensitivity is attained by a simple mechanism which acts to maintain the analyzer in the proper position to keep the two halves of the field at equal intensities, no matter how the halfshade angle may be varied to suit the pleasure of the observer. In order to accomplish this, the vibration plane of the analyzer is constantly maintained at the angle, 8, to a line bisecting the halfshade angle. A constant match of the field results. Messrs. Fric have succeeded in doing this to such a degree of perfection that there is no observable change in the zero point for a halfshade angle range of 2.5° to 15°. The analyzing nicol and the large nicol of the polarizing system are mounted in bearings and driven by gears from a common horizontal connecting rod. The halfshade angle is varied by turning the milled head which is geared to