SOME COXTEMPOK.IHy .IIH.IM r.S l.\ I'lDSICS- If 471 



abrupt breaks; of the priinar\ rirruit prodiux-tl h\ a mechaniial inter- 

 rupter at the rate of a thousand per second. Kach of these volta>;e- . 

 impulses exeites a spark between the doiiblet-e\ linders, accompanied 

 b\ a hJKhly-iiainfXMl oscillation which radiates what the authors 

 desirilie as "a very short \va\e-train with from tiO lo SO', of the 

 energy concentrateil in the first half-wavelength." This hi^h dampinj,' 

 is deplorable, as the waves arc inconvenient to measure and must Iw 

 regardetl as mixtures of sine-waves of difTerent frequencies. The 

 gap Ix-tween the cylinders is of the order 0.01 0.02 mm.; it changes 

 rapidh' and irresularK- as the opposing surfaces are eaten away by 



Fig. 2 — Photograph of the Oscillator I'scd by Nichols and Tear. (Physical Review) 



the sparks (tungsten was chosen by Nichols and Tear instead of 

 platinum in the hope, justified by the event, of diminishing this 

 trouble). 



The rays issue through a mica window in the fnjnt of the containing- 

 cylinder and are formed into a plane-parallel beam by an enormous 

 double-convex paraffin lens (these objects are shown in the photo- 

 graph. Fig. 2). Paraboloidal mirrors can be and have been used 

 insteatl of the lens. In the sketch of F'ig. 3, Li represents the lens; 

 the plane-parallel beam proceeds to the mirror A and thence to the 

 mirror B, which is really the pair of mirrors on the left-hand end of 

 the apparatus of which Fig. 4 is a photograph. In this apparatus, 

 the "Boltzmann interferometer," the upper mirror slides backward 

 and forward (left to right and right to left, in the picture) along the 

 guides, controlletl by the screw; it remains always parallel to the 

 lower and stationary- mirror. Half of the plane-parallel beam falls 

 upon each mirror, and the two reflected halves travel side by side 



