344 PROCEEDINGS OP THE AMERICAN ACADEMY. 



occulting screen, the arc on one side the screen, the camera on the other. 

 Rubber corks, placed in the legs of this table, free it from serious vibra- 

 tion even in the neighborhood of high speed shafting. 



The spectrograph employed was made by Bartel of Gottingen. The 

 collimator and objective had each an aperture of 1^ inches. The disper- 

 sion piece was a 2" Rowland grating, 14,438 lines to the inch. 



For the identification of the hydrogen lines which always made their 

 appearance when an atmosphere of hydrogen was employed, this saipe 

 instrument was used visually, a spectrum tube of hydrogen furnishing the 

 comparison spectrum. Ordinarily, for the visual examination of the arc, 

 a small direct-vision instrument was employed. For the examination of 

 the fluorescent light, mentioned at the end of this paper, a Browning 

 single prism instrument was tried with wide slit ; also the direct-vision 

 instrument; but the source was entirely too faint to produce a visible 

 spectrum. The conclusions there stated, viz. that the spectrum of this 

 fluorescent light is not characteristic, rest merely upon the fact that this 

 light presents to the eye the same appearance for all the metals tried, viz. 

 a luminosity resembling fluorescent light in general ; and further, upon 

 the probability that, if this spectrum were concentrated in lines, some of 

 them might be visible in the spectroscopes employed. 



For the purpose of determining the manner in which the rise and fall 

 of the arc occurs when the circuit is made and broken, respectively, it is 

 only necessary to move the eye either to the one side or the other so as 

 to catch the light of the arc a little earlier or a little later than the instant 

 of break. 



Let A represent the arc as seen through an opening in the disk by an 

 eyu looking in a direction parallel to the shaft of the dynamo. Then by 

 moving the head to the right one will see the arc in a later phase ; since 

 the line of vision above and parallel to the shaft will not be intercepted 

 until a later instant. Hence, by moving the arc and head together slowly 



