232 PROCEEDINGS OP THE AMERICAN ACADEMY. 



The cups were filled with mercury to a height of about 8 mm., and 

 then filled with alcohol to withiu a few millimeters of the top. They 

 usually required cleaning only after several hours' use, when the 

 surface of the mercury consisted of very fine globules, and sharp breaks 

 of the circuit were not made at each stroke of the plunger, as was 

 indicated by the occasional failure of the spark. The length of the 

 spark was kept constant during the observation, and the character of 

 the spark depended much upon the exact adjustment of the height of 

 the mercury cups of the interruptor. While it and the coil were both 

 in action, the height of the cups was adjusted until the sparks came 

 regularly with a peculiar crashing snap, and showed the bluish white 

 thick body that Professor Hertz described. Both the ear and eye 

 were soon so trained that they gave quick and sure information of the 

 character of the spark, but the ear was better than the eye. The ex- 

 act height of the mercury cups was of the utmost importance, a slight 

 difference in height changing the character of the spark greatly. The 

 exact point was best found by concentrating the attention upon the 

 sound, and then slowly raising or lowering the cups, when suddenly 

 there would seem to be a rhythm between the sound of the interruptor 

 and the snap of the spark, which would disappear with further motion 

 up or down. This rhythm would indicate that for every break of 

 contact there was a corresponding pilot spark between the microm- 

 eter balls. 



For measuring the effects produced in the secondary circuit, the 

 bolometer as designed by Paalzon and Rubens * was used with most 

 satisfactory results. The bolometer was constructed according to 

 the description given by them in the paper referred to, and differed 

 only in minor details arising from the circumstances and the materials 

 obtainable. The accompanying diagram is theirs, but the following 

 description applies in all its details only to the instrument constructed 

 for this investigation. 



The bolometer is in reality a double Wheatstone bridge. The four 

 arms of the bridge are the resistances W^, Wg, Wg, W4, of which 

 Wi and W2 are quadrilateral circuits of equal resistance, and Wg and 

 W4 are coils of equal resistance. The quadrilateral A B C D, or Wg, 

 is really a small Wheatstone arrangement, used as one branch of the 

 main bridge. For convenience in description, Wo will be called the 

 " bolometer branch," and the term bridge limited to the Wheatstone 



* Anwendiing des bolometrischen Princips auf electrische Messungen. 

 PoggendorfE, Annalen, XXXVII. 529. 



