Sine Galvanometer 377 



proper diameter, tests being made with the arrangement H shown in Figure C. This is 

 not the most precise method by which this work could be done, but it was quite adequate 

 for the purpose. The workmanship was such that the center of the magnetometer and 

 the center of the coil system coincided within a fraction of a millimeter, while the axis of 

 the spool and the plane containing the coil terminals were very nearly perpendicular to 

 the (vertical) axis of the magnetometer. The saddle w^s screwed to the top of another 

 machined brass casting G, provided with two coaxial circles, one fitting over a central 

 circular ring integral with the axle or "center" of the turn-table, and the other fitting 

 over a central circular disc projecting from the bottom of the saddle. The machine work 

 was such that when the turn-table was leveled the magnet hung with its center within 

 a fraction of a millimeter of the center of the magnetometer box. 



The telescope, the scale, and a small electric lamp to illuminate it, together with a 

 shade not shown in Figures A and B, are carried on a vertical brass rod clamped, with 

 vertical sliding adjustment possible, to a horizontal tube which slides in a second tube 

 with a clamp at the end. This tube and a third tube, which carries a counterpoise, 

 are soldered into a ring by which and a screw they are clamped to the support A B, 

 Figure C. 1 



The scale, 10 cm. long and of white pyralin, is divided to thirds, sixths, and thir- 

 tieths of a centimeter. The scale distance can be varied over a range of about 10 cm. 

 from about 34 cm. up. With the telescope used, which is small but has excellent definition 

 and high magnifying power, and either of the steel mirrors, the scale can be read easily 

 to tenths of the smallest divisions. In addition to the adjustments already mentioned, 

 the telescope is provided with fine adjustments in altitude and azimuth, and the scale 

 can be moved laterally a small distance in its support. The scale divisions are so num- 

 bered that the readings increase continuously with clockwise rotation of the mirror, just 

 as the circle readings increase continuously with clockwise rotation of the magnetometer 

 and coil. 



8. The great importance of securing a uniform field, and one whose intensity can 

 be calculated with precision, throughout the region in which the magnet is ever placed 

 for measurements, whether in or out of perfect adjustment, led to the adoption of 

 the type of double coil introduced in 1849 by Helmholtz 2 , and to its being wound from 

 bare copper wire under tension in a single layer in lathe-cut spirals, as suggested by 

 Jones. For the same reason white Carrara marble, already used with satisfactory 

 results for the cores of many coils of precision, notably at the National Physical 

 Laboratory of England, was chosen for the spool. This substance has a permeability 

 differing from unity by a negligible amount. Moreover, after being subjected to a pre- 

 liminary heat treatment its dimensions remain practically unchanged with time and their 

 thermal changes are reversible for small temperature ranges 3 . At the same time its elec- 

 trical resistivity is very high, especially when impregnated with paraffin. 



In order to make it possible to test the insulation of the coils at any time, each coil 

 of the Helmholtz pair was wound in two halves, each consisting of a spiral of the same 

 length and pitch. The two spirals starting from the same plane normal to the axis 180 

 apart. The spirals of one Helmholtz coil were designated as Nos. 1 and 2; those of the 

 other as Nos. 3 and 4. 



This method of winding Jones spirals is due to Ayrton 1 , and has, in addition to the 

 facility it affords for making satisfactory insulation tests, the further advantage, appar- 



1 The telescope and the telescope-and-scale holder were taken froma nother instrument, non-magnetic parts being sub- 

 stituted for magnetic parts in the holder. Tho original design called for the simpler type, with telescope and scale fixed 

 rigidly together and double clamp to the central rod as used in Kohlrauseh's universal magnetometer, which has some 

 advantages. 



2 See Wiedemann's Elektrizitat, vol. Ill, p. 275. 



3 See Souder and Hindert, Scientific Papers, Bureau of Standards, No. 352, Dee. 191. 



4 Jour. Inst. Elec. Eng., vol. 35, p. 18. 



