18 



Mr. Carl Barus on the 



13. To facilitate measurement, the greater number of 

 experiments are made at 215°. Polarization was observed ; 

 and for this reason the measurement of temperature-coefficients 

 by removing and readjusting vapour-baths was not tried. The 

 increment of electromotive force is largest immediately after 

 compression (temporary effect), and then decreases to the 

 apparently constant A'alues of the Table. The converse holds 

 for the decrement on removing pressure. At 100° no satis- 

 factory results could be obtained (§ 84). At 310°, although 

 the temporary effects were distinct, the permanent effects are 

 not warranted. 



14. Table III. contains results, given on a plan identical 

 with that of Table I., for the case in which the glass tube 

 (fig. 1) was removed and the steel rod (fig. 4) inserted. In 

 this way the pressure-coefficient of ordinary sperm-oil was 

 found. Table III. also contains results for glass made after 

 the work with oil. 



Table III. — The Insulation and Pressure-Coefficient of Sperm- 

 Oil. j9 = 100atm. 



Pressure-coefficient of glass. 



Series No. 



9. 



R. 



Mean 10 6 X kr. 



10 3 x 

 cR/R. 



vp. 



540 

 573 



1240 

 1140 



553 



10 6 X&r. 



6200 

 6100 

 6500 

 5700 

 4000 



1. 



20°. 



18 x10V 



+5700. 



3300 

 3500 

 8100 

 6500 

 2200 



1. 



100°. 

 2xl0 6 a>. 



+880. 



500 

 550 

 990 

 460 



550 



610 



1150 



550 



910 

 890 

 860 

 840 



1. 



215°. 



300 x 10V 



+330. 



180 

 200 

 230 



550 

 600 

 675 



330 

 330 

 340 



Series No. 



e. 



R. 



Mean 10 6 X k r . 



103 X 

 <5R/R. 



op. 



Glass tube inserted. 



VII. 



100°. 



240xlO R w. 



620. 



VII. 



215°. 



500 x 10 3 o>. 



171. 



347 

 373 



90 



94 



186 



570 



580 



570 



560 



1000 



10 6 x£: 



600 

 640 



157 

 170 

 186 



The pressure-coefficient of sperm- oil is therefore decidedly 

 larger, and the specific resistance of this oil is usually smaller, 

 than that of glass (cf. § 19, Tables IX. & X.) ; but the break- 

 down of resistance with rise of temperature is more marked 

 in the latter case. Nevertheless the data of Table III. cast 

 grave doubt on the trustworthiness of the data of Table I. ; 

 and hence a search must be made for more highly insulating 



