1884.] 



On Magnetic Polarity and Neutrality. 



being 25 per cent, for iron, and 33 per cent, for steel of its propor- 

 tionate previous force. 



The most conclusive experiments, however, will be seen in Nos. 8, 

 9, 10, and 11. No. 8 being a brass tube coated with an exceedingly 

 thin transparent coat of iron (I was unable to measure its thickness), 

 this thin coating of iron was easily raised to its saturation by a feeble 

 battery, from which point no increase of battery power had the 

 slightest effect, giving always o, 95. The extraordinary effect of 

 thinness was seen on taking off the inducing influence, no perceptible 

 movement of the needle on the balance occurred, indicating that its 

 retaining power was the same as its capacity or 100 per cent, of 

 retention; vibrations and hammering which reduce a solid bar at 

 once to perfect neutrality had no effect. I have, however, marked it 

 down as o, 94 as a limit of experimental error ; if we assume only 

 80 per cent, of its previous force it is still sufficiently remarkable. 

 At No. 9 where the brass was coated to a measurable thickness of 



millim., we already see a better return to neutrality, having now 

 only 50 per cent, of proportionate remaining magnetism ; at No. 10, 

 where we have 1 millim. thickness, we have improved our neutrality by 

 having only 20 per cent., and at No. 11 we have, by means of an 

 increased thickness of 1 centim., almost completely allowed the 

 balancing waves of opposite polarity perfect formation, the remaining 

 magnetism now being only 3 per cent, of its proportional power ; and 

 if we even neglect the proportional power, we see that the 1 centim. 

 bar has far less remaining magnetism than that of 1 or millim., 

 whilst its magnetic capacity is far higher. 



No. 4 shows that while bundles of wires have a higher remaining 

 magnetism than solid (due to the want of homogeneity allowing 

 perfect formation of the opposing waves), still from their increased 

 surface exposed to the inducing effect they give a higher magnetic 

 effect, the differential effect (as that employed in temporary electro- 

 magnetic and induction coils) is greater, being here for a solid bar 

 960-29=931, and for the bundles of iron wire 1,268-142=1,126 

 useful effect.* 



The effect of thickness even upon finely divided iron such as filings 

 is shown in No. 5, where the remaining magnetism is only 9 per cent, 

 against 50 per cent, as shown in No. 1, and we have precisely similar 



* If we require a continuous magnetic effect, as in the field magnets of dynamo 

 machines or small constant electro-magnets for extremely feeble electromotive force, 

 solid cores or bundles of wire of large diameter should be employed, but as the time 

 of charge and discharge increases with the diameter, it would be unsuitable for 

 electro-magnets requiring rapid charges, such as those employed for telegraph 

 relays, large electro-magnets requiring several seconds to charge them to saturation, 

 while extremely small electro -magnets may be saturated in the part of a 



second. 



