Polytechnic Association. 911 



must briefly refer to some elementary principles but tliese I will 

 simply recall to your minds, and hasten to the special department of 

 the subject which at present engages our attention. 



Light, we now fully believe is simply a vibratory or ripple-like 

 motion, and colors differ from each other simply in the number of 

 waves included in a given time or space ; or, in other words, in the 

 lengths of the individual waves which constitute them. 



Thus, a crimson color may be compared to the long, heaving billows 

 of the ocean, while a blue tint would find its parallel in the ripple 

 which the fitful zephyr rouses on some summer's lake. 



Again, we know that when these waves strike obliquely on some 

 medium denser than that through which they have been traveling, 

 but which is nevertheless capable of transmitting them ; their direc- 

 tions will be changed, and that this change will be greater with the 

 little waves than with the great ones. Such an action as this evi- 

 dently leads to a separation between the waves of different lengths, 

 and this separation we call " dispersion," while we give the name 

 " refraction " to the mere act of bending, without regard to its vary- 

 ing degree with reference to the different colors. 



The following diagram will exhibit the case now of most importance 

 to us, in which this action of disperson is developed and utilized. (See 

 Plate I, Fig. 1.) 



Let A, B represent a screen, behind which are located various 

 seurces of light, emitting rays which respectively pass through slits 

 at G, H and I. 



Let C, D be a prism and E, F a screen. If the light-rays pass- 

 ing by G are all of one wave length, then they will be all equally 

 bent by the prism at I and J, and will fall upon the screen in a single 

 line at K. 



Again, suppose that the light passing through the opening H were 

 a mixture of two wave-lengths or colors, one part like that coming 

 through G, and the other of greater wave length, or, as we say, of a 

 lower color. Then that portion which resembles the former will be 

 equally bent, and will fall on the screen in a single line at L, while 

 the other part will be less bent, and will form another single line at M. 



Lastly, if a beam were to come through O, composed of a great 

 number of colors, or waves of various lengths, it would be separated 

 by the prism into as many lines from L to Y ; or if this number were 

 very great the lines would blend with each other, and give a continu- 

 ous band. 



The name spectrum is applied to the resolved, or separated light, 



