232 PROCEEDINGS OP THE AMERICAN ACADEMY 



These strips were not measured very accurately, but the following 

 were carefully measured : — 



Narrow. f Double. Triple. 



Red 18 paces 23 paces 34 paces. 



Yellow 29 " 43 « 56 " 



Green 34 " 51 " 67 " 



"We see from the above table that when the strips are three times 

 as wide, the distance at which they can be seen is just about double. 

 My conclusions therefore are as follows : — 



1. Herbert Spencer's instance cannot be accepted as an evidence of 



heterogeneity. 



2. Distance is an important element in our perception of color. 



Every one may be said to be color-blind in reference to the stars. 



3. The three factors, distance, intensity, and amount of surface from 



which the illumination proceeds, should be considered in the loca- 

 tion of colored signal-lights. 



4. Spectrum analysis of very faint stars does not aid us in determining 



their color, on account of the faintness of the lines. 



No. XXI. — SIMPLE APPARATUS FOR ILLUSTRATING 

 PERIODIC MOTION. 



Bt John Trowbridge. 



Presented Dec. 10, 1879. 



There are many ways of showing Lissajous' experiments, and on 

 many accounts the tuning-fork method is the best ; but the apparatus 

 is expensive and cannot be readily obtained. The following forms of 

 apparatus can be made by any one at a trifling expense. 



No. 1. — Graphical Method. 



The apparatus represented in Fig. 1 draws the curves before an 

 audience on the screen, and does not require a vertical lantern. A is 



