520 



PROFESSOR KNOTT ON SOME RELATIONS BETWEEN 



effects, altogether different from those for the other and thinner wires. The magnetic- 

 lagging begins negative, changing to positive for the twists ±45° and ±90° when, the 



Twist. 



Max. Area 



7T. 



Current 

 corresponding. 



Current for Zero 

 Area. 



45° 

 90 

 100 



21 

 15 



8 



16 

 •54 

 •2 



1 

 1-3 



•48 



45 



90 



100 



110 



135 



21 

 30 

 25 

 26 

 none 



10 



•8 



•5 



•4 



none 



? 

 3-4 

 1-83 

 1-1.4 



none 



current has reached a value of about 3 amperes. For the twist ±135°, however, it 

 continues negative throughout up to the highest current used. It is possible that this 

 discrepancy is to be explained as due to the greater strains which the thick wire suffers 

 when twisted. There were indeed many irregularities in the manner in which the thick 

 wire went through its cyclic changes. 



The cyclic graphs, instead of having the smooth sweeps so characteristic of the graphs 

 for the thinner wires, had in most cases very irregular broken forms. One of these 

 is shown in fig. 6, PI. IV. Again, while the twisting was being applied, the wire gave 

 audible clicks, proving that the straining was accompanied by cracking and wrenching. 

 Then the effect of tapping the wire was, in many cases, very peculiar, the first effect 

 being sometimes of an opposite character to the final effect. A single tap occasionally 

 did more in changing the magnetic moment than a whole twisting to the opposite 

 extreme. These facts show that the wire was really too thick for such experiments, or 

 at any rate was not annealed to a sufficient uniformity. 



20. Effect of Tapping. — In Table VI. certain of the numbers are marked with an 

 asterisk. In these cases the wire was vigorously tapped all along its length at the 

 different stages. In all instances the effect of tapping is to increase the negative lag. 

 The range, however, is not greatly changed. The following is given as a specimen of 

 the effect of tapping upon a complete cycle. The twisting was through ±90° ; the wire 

 was of diameter 0'87 mm. ; and the current was 1*98 amperes. 





Not Tapped. 





Tapped. 





Twist. 











Intensity. 



Lag. 



Intensity. 



Lag. 



-90 



+ 242 



+ 240 



+ 2 



+ 237 



+ 244 



hr 

 — 1 



-45 



+ 198 



+ 194 



+ 4 



+ 170 



+ 210 



- 40 







- 4 



- 40 



+ 36 



-108 



+ 112 



-220 



+ 45 



-190 



-200 



+ 10 



-208 



-168 



- 40 



+ 90 



-241 







-242 







\n this particular case, a distinct positive lagging is changed by tapping into a pro- 



