510 



Dr. E. J. Mills on Electro striction. [Dec. 13, 



(3) The electrostrictive effect brings vividly to the mind the ordinary 

 phenomenon of contraction during cooling — differing, perhaps, but little 

 from that phenomenon, except in the circumstance that it is brought 

 about without necessary change of temperature. The electrolyte is, in a 

 manner, a melted metal. This view did not, of course, admit of experi- 

 mental illustration in the case of the metals to which reference has been 

 made. It was, however, conceived that substances other than metallic 

 should yield the result in question, and a specimen of hard paraffin was 

 selected for trial. This was heated to a point not far from that of 

 fusion, and thermometer 454 was dipped into it several successive times. 

 The following table shows the constriction obtained. 



No. of dip. Total ascent. Temp. 



1 -256 49 



2 -444 30 



3 -633 — 



4 -447 30 



5 -369 30 



6 -321 30 



These results were confirmed by a second set of experiments. In each 

 set, the coating of paraffin was found to he vertically fissured before the 

 conclusion of the observations ; this accounts for the gradual diminution 

 of the effect. 



(4) Electrostricting metal appears to be in a partially unstable condition. 

 Thus a decided fall is observable in a coated thermometer after the 

 deposit has had its surface disturbed by sand-papering or fifing; this 

 was noticed both with silver and copper. On the other hand, the con- 

 strictive effect is increased by heating the deposit to 100°, or by keeping 

 in ice. Thermometer C, for example, having had its zero raised electro- 

 strictively to 3°*52, showed an elevation to 5 o, 00 after heating for ten 

 minutes to 100°, afterwards falling permanently to 4°*86. Thermometer 

 50 had its zero raised in a similar manner to 1°*87 ; this was increased to 

 2°-96 by a temperature of 100°. In another experiment with the same 

 thermometer, a like increase from 0°-65 to o, 79 was recorded ; the latter 

 number rose spontaneously to 0°*90 in the course of six months. Again, 

 two similar thermometers (103 and 104) were electrostricted at about 

 10° C, and the zero was found to rise by heating, as before. The zero of 

 103 then fell gradually, during nine months, from 4°-73 to 4°-37 ; that 

 of 104 fell during two months from 2°-01 to l o, 80, and then rose during 

 seven months to 2°*14. Thermometer 101, made exactly on the same 

 pattern, but left uncoated, exhibited the usual continuous ascent of zero 

 during the same period, viz. from o, 07 to o, 31 in nine months. All 

 three thermometers showed identical and normal readings at 100°. Eacts 

 like these prove that the nature of the deposit requires further elucida- 

 tion. In any case the best evidence of electrostrictive effect is gained, 



