626 Mr. Shelf ord Bidioell [March 11, 



of selenium. The effect produced by light is increased by lengthen- 

 ing this arrangement. We might go on increasing the wires with 

 advantage until they were 10 or 12 feet long or more, every addi- 

 tional inch of length producing an increase of " sensitiveness " as it 

 is called. But a cell of this length would be cumbersome and un- 

 wieldy to use, and in fact could hardly be lifted without being 

 destroyed. Dr. Werner Siemens therefore adopted the device 

 (among others) of coiling up the wires so as to form a double s^nral, 

 and thus made a convenient and portable cell of great sensitiveness. 

 But it is very difficult indeed, as I know by experience, to produce 

 these double wire spirals of any considerable size without the two 

 wires touching one another at some point. After many attempts I 

 succeeded in producing spiral cells about f inch in diameter, but I 

 found it impossible to exceed this size, and as it was not large enough 

 I adopted a simple and very effective variation of Siemens's method, 

 which I believe has not been previously suggested. A copper wire 

 is wound after the fashion of a flat screw around a narrow slip of 

 mica, the threads of the screw being about y^^ inch apart. Beside 

 this wire, and at a distance of ^^ inch from it, a second wire is 

 wound in exactly the same manner, each of its turns coming midway 

 between two consecutive turns of the first. Care is taken that the 

 two wires do not touch each other at any point. Over one surface of 

 the mica a film of melted selenium is spread, and after being worked 

 smooth and uniform it is crystallised. By this means the two wires 

 are connected with each other through half their entire length, by 

 a series of very narrow strips of crystalline selenium. 



I have here a tiny selenium cell which has been constructed in 

 this manner. Each wire makes about six turns, and the area of the 

 selenium upon its surface is about half that of a threepenny piece, its 

 thickness not much exceeding that of a sheet of ordinary note-paper. 

 Its resistance, though it is very high relatively to that of good con- 

 ductors, is, compared with anything of the kind that I had ever seen 

 before, remarkably low, and its sensitiveness to light is great. When 

 a batswing gas-flame is held at a distance of three inches from it, its 

 resistance is less than one- third of that which it measures in the dark. 

 Larger and more carefully constructed cells, of course, show better 

 results. No form, however, that I have tried (and I have made 

 several dozens) has in my hands been superior to that of a double 

 flat screw. 



When cells are to be made of any considerable size, the labour of 

 winding on the wires with sufficient regularity is very great. By 

 the help of a lathe this difficulty may be reduced to a minimum. The 

 method of proceeding is this : — A cylinder is turned of hard wood, of 

 length and diameter slightly greater than that of the proposed cell. 

 This cylinder is cut longitudinally into two equal parts, and between 

 the two semi-cylinders thus formed a slip of mica is placed sandwich- 

 like. The ends are secured with screws, and the whole is smoothed 

 down in the lathe. When the edges of the mica are quite flush with 



