290 On the Construction of the Photophone. 



circular mouth of the mirror cone, and the general appearance 



of the mirror and cylinder 

 will be that presented in 

 figure 2. 



With an angular aperture 

 less than 90°, the depth of 

 the mirror from back to 

 front must be greater than 

 the length of the cylinder ; 

 and the mirror, however 

 prolonged, could not bring 

 more rays to the surface of 

 the cylinder except they 

 underwent more than one 

 reflexion. 



If the angular aperture 

 should be greater than 90°, 

 the diameter of the cone 

 that will reflect the effective rays will be less than that of the 

 90° cone, and hence cannot gather as much light. 



One advantage possessed by such a mirror cone of the 

 form specified above over any other form, parabolic or other- 

 wise, is that all the rays meet the sensitive surface of the 

 cylinder at normal incidence, and the loss by reflexion will 

 be therefore a minimum. 



(7) In preparing to repeat the Photophone experiments, the 

 author has constructed sundry cells in a manner somewhat 

 differing from that adopted by Professor Bell. 



Finding it laborious to cut and fix the alternate disks of 

 mica and brass, he constructed a cell by winding brass wires 

 spirally round a glass tube so that the successive convolu- 

 tions did not quite touch. Selenium was afterwards applied in 

 the interstices, and alternate convolutions were connected 

 metallically, the wires being cut and then soldered with alter- 

 nate junctions. Afterwards two parallel wires, wound side by 

 side as in a double-threaded screw, were employed. One of 

 these cells was found by the author on Oct. 19th to have as 

 small a resistance (in the dark) as 240 ohms. On account, 

 however, of the wire not adhering firmly to the glass, and 

 from other causes, the arrangement, though far more easily 

 constructed than the built-up cell, was not satisfactory. 

 Taking a hint from Mr. Shelford Bid well, who has recently 

 published a communication on the Photophone in i Nature,' 

 the author has constructed cylinders of slate grooved with a 

 fine double-threaded screw, in which the parallel wires are 

 laid. These cells prove much more satisfactory. Experi- 



