ELECTKODYNAMIC QUALITIES OE METALS. 
697 
184. But, lastly, I found between the quasi-elastic part of the magnetization pro- 
duced by alternately applying and removing the magnetizing force, and the initial 
reverse magnetization produced by the application of a reverse magnetizing force, a 
very surprising difference. The former, as stated above, is greater when the wire is 
pulled than when unpulled : the latter is less in the wire when pulled than when free 
from pull, but not by so great a difference; and the whole magnetizational effect of 
reversing the current suddenly must therefore be greater when the wire is pulled than 
when unpulled, and is found to be so. 
185. The following series of experiments, I XXXI., performed in November and 
December 1874, all on one and the same piece of steel wire, first confirmed the conclu- 
sions inferred (§182) from the preliminary investigation of the previous March, then 
reproduced with more regularity the immediate experimental results of that preliminary 
investigation, and lastly discovered the very remarkable phenomena described in § 184. 
186. The general order of procedure followed was this. The image on the scale of 
the ballistic galvanometer was watched by one observer, while a second stood by to 
make or break circuit of magnetizing current, or put on and off weights, on word of 
command from the first. When the image is seen to be steady on the scale, the number 
at which it stands is read and recorded as “ z ” (zero). The order “ make ” or “ break ” 
or “ on ” or “ off ” is given by the first observer and executed suddenly by the second. 
The first observer reads and records the greatest or least number on the scale reached by 
the image in consequence of the electromagnetic impulse produced by the operation. 
Finally, the excess (positive or negative) of this reading above the immediately preceding 
zero is written down and marked “ M ” or “ B,” or “ On ” or “Off,” as the case may be. 
187. The connexions chanced to be so arranged that, with the direction of current 
invariably used in the “ M’s ” of Series I XXIX., the effect of M was to throw the 
image to the left, or in the direction of decreasing numbers : thus until Series XXX. a 
negative number always shows increase of magnetization in the direction of that 
produced by M. The numbers actually written down by the observers during the 
experiments are shown in the following extracts from their day-book, for a few of the 
series, chosen as examples to precede the abridged Tables of results given below for all 
the Series I XXX. 
