and Electrical Resistance of Iron. 223 



The oxygen-absorbing coil was, however, quite distinct from 

 this. It consisted of two bare iron wires, connected in 

 parallel, wound side by side on a cylindrical asbestos frame, 

 and surrounded by a loosely-fitting asbestos mantle. Before 

 the experiments they were heated by an electric current to a 

 temperature at which they would readily absorb oxygen. 



The coverings of the ring-magnet formed a mass of asbestos 

 about 6 cm. in diameter ; a small hole through them enabled 

 one to see what sort of temperature had been reached in 

 the middle of the ring. 



The coil for absorbing oxygen was then made fast to the 

 ring-magnet, and the leading wires brought through a cork 

 (previously boiled in linseed oil), which was then inserted into 

 the glass jar and covered with melted sealing-wax (fig. 5). 

 Finally, the terminal wires were soldered to those leading 

 away to the rest of the apparatus. 



The following particulars relate to the last two ring-magnets 

 constructed : — 



Core of ring specimen A. 



Description of iron Charcoal 



Iron. 



Number of turns of core 4 



Density of iron (after experiments) .... 7*775 



Total sectional area of iron "131 



Mean diameter of ring 2-35 



Number of turns of primary 28 



„ per cm. of magnetic circuit .... 3-792 



Field due to 1 amp. in primary 4 - 765 



Mean sectional area enclosed by primary. '72 



Number of turns of secondary 15 



Resistance of iron at 18° "0486 



B. 



Swedish 

 Transformer Iron. 

 5 turns 

 7-461 



T43 cm.' 

 2-23 cm. 

 29 turns. 

 4T43 turus per cm. 

 5-206 C.G.S. units. 

 •83 cm. a 

 17 turns. 

 •0477 ohm. 



Electrical Arrangements. — Reduction of Results to Absolute 

 Measure. — The various electrical circuits communicating with 

 the ring-magnet are indicated in fig. 6, which needs little 

 explanation, except perhaps with regard to the arrangement 

 for standardizing the ballistic galvanometer. 



For this purpose an auxiliary ring-magnet of known dimen- 

 sions and having a non-magnetic core was used. Its secondary, 

 in series with that of the iron ring-magnet, formed, in circuit 

 w T ith the galvanometer and a coil of 100 ohms, the secondary 

 circuit, as it remained throughout the experiments. On re- 

 versing a known current in the primary of the auxiliary coil, 

 the resulting throw of the ballistic galvanometer was a measure 

 of its sensitiveness to induction alterations in that particular 

 secondary circuit, whether taking place in the core of the 

 auxiliary coil, or in that of the iron ring-magnet. Such 



