TRANSACTIONS OF SECTION A. 753 
makers of both countries, magnets were made, and a careful magneto-metric series 
of measurements taken for some months. The French steel showed itself to be far 
superior to the English. The details of the comparison are given in the paper,! 
but the reasons for this marked superiority remain for further investigation. 
9. The Effect of Oxidation on the Magnetic Properties of Manganese Steel. 
By L. T. O’Suua, B.Sc. 
When manganese steel drillings are oxidised they become magnetic, the 
development of magnetic properties being due to removal of manganese by oxida- 
and to the magnetic properties of the oxide of iron (probably magnetic oxide) 
ormed, . 
When the oxidised product is reduced in hydrogen, the iron oxide is converted 
into metallic iron and the manganese remains as manganous oxide (MnO). The 
reduced steel is now powerfully magnetic in virtue of the magnetic properties of 
unalloyed metallic iron. 
During the process of oxidation the proportion of manganese to iron oxidised 
in a given time is only very slightly in excess of the proportion of manganese to 
iron in the original steel. ‘The excess of manganese oxidised is, in all probability, 
due partly to the greater susceptibility of manganese to oxidation, and partly to 
the heterogeneous structure of the steel. 
10. On Testing Iron.2 By J. Swixsurne and W. F. Bourne. 
11. The Compensation of Alternating-Ourrent Voltmeters.” 
By J. SwiveuRye. 
The communication relates to an arrangement for compensating alternating 
voltmeters for changes of frequency. 
It is much more easy to make a current indicator for alternating than direct 
currents, for troubles from hysteresis do not come in, and the slight tremble makes 
the moving part hang freely. If it is attempted, however, to use such an instru- 
ment as a voltmeter, the self-induction makes the reading far too low, and the 
error varies with the frequency. 
To get over this trouble, a voltmeter may have a non-inductive, or nearly non- 
inductive, resistance put in series with its active coil. A coil with an adjustable 
iron core is then put in shunt to the active coil, this shunt coil having a very much 
larger time-constant. The instrument is calibrated with a direct current. An 
alternating current is then put on, and the core of the shunt coil regulated till the 
readings agree with those of the direct current. 
12. Note on a Kinetic Stability of Equilibrium with Electro-magnetic Forces. 
By Professor G. F. Firzcrranp, F.R.S. 
If a perfect conductor move near a magnet there are currents induced in it 
which tend to stop the motion. If the conductor he perfect, the kinetic energy of 
motion will ultimately, if small enough, be all changed into electrokinetic energy 
and the conductor will begin to move in the opposite direction, and when in its 
former position its electrokinetic energy will have been reconverted into kinetic 
energy. For instance, if a perfect conducting shell were placed near three mag- 
netic poles it would be in a state of kinetic equilibrium, if the energy given to it 
by a small disturbance were not great enough to drive it to infinity or into contact 
with the magnet. It is to be remarked that I have assumed the perfect conductor 
to have been brought within a finite distance of the magnet pole without haying 
! Published in the Hlectrician, September 19, 1890. 
2 Electrician, October 1890. 
