PHYSICS. 377 



used in registering- tlie current for domesticligLting, there are two cells 

 througli which a shunted portion of the current Hows, in one several 

 times that in the other. In these cells are plates of coi)per immersed 

 in sulphate of copper solution. These are weighed every month, and 

 from the increase in weight the quantity of the current which has passed 

 through the cell is given. This, multi[»lied by the shuut, gives the 

 total current in the house for the time. In the second and more deli- 

 cate form two copi^er plates are suspended in an electrolytic cell con- 

 taining copper sulphate from the arms of a balance. The apparatus is 

 placed in a shunted circuit, say, of xoVo' ^J the action of the current 

 copper is dissolved off one plate and deposited on the other. The heavier 

 one falls to a certain point, then automatically reverses the current. 

 The other side now becomes heavier and goes down, and the current is 

 again reversed. The beam thus oscillates, and its oscillations are reg- 

 istered on a dial. By properly adjusting its parts each tip may be 

 made to correspond with a definite quantity of current. By combining 

 the delicacy of the mirror method with a delicately constructed weber- 

 meter, Edison has been able to measure in one minute a current so 

 slight thr.t it would deposit only ten milligrams of copper in the course 

 of a century. {Cat. Gen. Off., Paris Exh., 1G2; Nature, July, 1881, xxiv, 

 291.) 



Bra'ckett has described a new form of galvanometer for powerful cur- 

 rents, based on the tangent-galvanometer principle. Two rings of cop. 

 per or brass are turned so that the one passes within the other. They 

 are then both cut on one side, the smaller placed within the larger, one 

 of the ends of each united firmly by a metal plate, and pieces of vulcanite 

 put between the rings to make them concentric. The other ends of the 

 rings and the united ends are attached to three binding-screws. The 

 iuslrument may be used as an ordinary-tangent galvanometer with 

 either of the rings, or, by combining, them it may act differentially on 

 the needle, owing to the different distances of the rings. The instru- 

 ment works well in practice. {Am. J. Sci., May, 1881, III, xxi, 395.) 



4. Electric sparJc and light. 



Deprez has applied to the induction coils made by Carpentier a new 

 form of interrupter, designed by himself. From his study of the action 

 of the coil he concluded (1) that the current should be broken as soon as the 

 maximum magnetism is attained in the core; and (2) that it should be 

 re-established as soon as possible thereafter. {J. Phys., August, 1881, 

 X, 3G0.) 



Bottomley has described some curious experiments with vacuum tubes. 

 The tubes are exhausted very completely and sealed up without elec- 

 trodes. If one end of a long tube like this be api)lied to tiie prime 

 conductor of an ordinary frictional machine, the other end being held 

 in the hand, the tube becomes charged as a double Leyden jar, the end 

 next to the machine being positive without and negative within; while 



