268 HERBERT F. ROBERTS 



Figure 4 A. is a transverse sectional drawing of the drive mecha- 

 nism, reduced about one-third. Spindles A and A x are rotated 

 by friction discs B and B', making contact at K and / on annular 

 C, which, in turn is operated by a motor drive (M, in fig. 1), 

 connected with the annular gear C, by a round belt at V. By 

 turning a thumb-screw F (Si, in fig. 2), spindle A may be 

 shifted so that contact can be effected at I between the friction 

 discs B and B', attached to spindles A and A', thereby releas- 

 ing contact at K, and effecting a reversal in direction of rotation 

 on the part of spindle .4. The shifting of spindle A is accom- 

 plished by the thumb-screw F forcing the wedge G against the 

 collar H, which is held in place on spindle A by means of a set- 

 screw. The annular C, is carried on spindle D, which is held in 

 place by collar E. Figure 4B shows a lateral longitudinal section 

 of the drive mechanism, the parts being lettered to correspond 

 with figure 4. 



A further improvement upon the mode of use of the original 

 tintometer, lies in the employment of artificial light of constant 

 quality and value instead of ordinary daylight. It is plain that 

 in the employment of daylight as a means of illumination sev- 

 eral sources of error are presented. The relative intensity of 

 the different spectrum rays varies according as the sun is over- 

 head or near the horizon. The available rays in solar illumina- 

 tion vary also according to the absorption spectrum of the sur- 

 rounding surfaces. It is manifest that a color determination 

 with the tintometer will not be the same for the same object 

 when a red brick wall opposite the laboratory is replaced by a 

 green field or a mass of white buildings. A laboratory window 

 admitting only a north light from a clear sky would be reason- 

 ably satisfactory, except for the difference in the absorption spec- 

 trum due to the apparent position of the sun with respect to the 

 earth's horizon from morning to evening and during the solar 

 year. Added to these variations, which have to do with local 

 disturbances in the relative spectral content of the available 

 solar rays, there is the added practical difficulty of using the 

 tintometer at all during dark gloomy weather. 



It has therefore been arranged to use the instrument in a ven- 

 tilated dark room, using two 25-watt tungsten lamps as the source 



