254 O. N. Rood—Nature and Duration of 
to the eye of the observer placed ten or twenty inches behind 
it, an appearance like that in fig. 8. It is evident that by in- 
creasing the rate of rotation of the disc, or by diminishing the 
angle between its open sectors the termination of the outer 
series of illuminated sectors, may be made to coincide with the 
beginning of the interior series, when the included angle will 
give a measure of the total duration of the phenomenon. In 
practice the interior sector was cut in a second and smaller disc, 
which was placed on the same axis with the first, a portion 0 
this latter being removed so as to allow of any angular adjust- 
ment within 90°. The larger disc was graduated on its circum- 
ference. In some of the experiments this arrangement 0 
movable sectors was repeated for each quadrant, but as a gene- 
ral thing it was found safer to confine the operation to a single 
quadrant. Besides this, the exterior portion of the outer sector 
was provided with a prolongation, consisting of 
small slit, which was very useful in furnishing 
information regarding the number of the dls- 
charges in a given case, as with sectors 2$° in 
width the images frequently overlapped. Most 
of the dise observations were made with a train 
3. 
Form and duration of the discharge of the larger jar, with brass 
balls as electrodes, 
In this set of experiments the length of the simple induction 
4, spark, blunt brass points being used as elec: 
trodes, was 48 millimeters. 
Striking distance 1 millimeter.—The form 
others. Fig. 4 roughly represents the ap- 
pearance with the revolving mirror. _ The 
* The actual duration of sparks called in this paper instantaneous has already 
been given . 
