AN IMPROVED COLORIMETER 269 



of illumination, the light from which is directed down upon the 

 glass cylinders by a shade reflector. The degree of constancy 

 of the current supplying the lamps is ascertainable by means of 

 a voltmeter. 



The principal difficulty in the construction of the improved 

 instrument has been to obtain proper glass for the cylinders. 



A perfectly white glass tubing of definite and known composi- 

 tion, of proper diameter, with a comparatively thin wall, and free 

 from defects, has been difficult to find. In order that the re- 

 flecting surface and the refraction index of the wall of the cylinder 

 should be identical throughout its length, and to get rid of the 

 linear striations present in the best glass tubing, an attempt 

 was made to have the interior and exterior faces of the cylinders 

 cu^ to a uniform thickness throughout and polished. The prac- 

 tical difficulties in the way of accomplishing the cutting, proved, 

 after much experimentation, to be insuperable, except at a cost 

 that would be prohibitive. The writer therefore was obliged to 

 content himself with obtaining the most satisfactory white 

 standard glass tubing possible. By always using the same make 

 of tubing after it is once settled upon, variations in the readings 

 due to differences "in the absorption spectrum of the glass of the 

 cylinder will be eliminated. 



The rotary cylinder mechanism enables a far greater variety 

 of opaque substances to be successfully determined for color 

 than was possible with the original instrument, and the fitting- 

 out of the binocular with a set of three-way adjustments, enables 

 the instrument to be satisfactorily oriented with respect to the 

 revolving cjdinders. Of course, in the case of liquid substances 

 for which the color is to be determined, or of amorphous powders, 

 the rotary mechanism would not be necessary. 



For the quantitative measurement of color value in flowers, 

 in the study of color inheritance, the writer has found the appa- 

 ratus invaluable, and it seems likely that it will prove useful 

 generally to investigators. In working on color-inheritance in 

 Pelargonium, it is found that clear and definite results in stating 

 the color values of segregates are obtainable with this apparatus. 

 By supplementing these records with Lumiere color photographs 

 a practically perfect series of inheritance records is obtainable. 



THE PLANT WORLD, VOL. 22, NO. 9 



