456 Dr. J. Kerr on a new Relation between 



action is contrary to the actions of the five preceding liquids. 

 The material employed latterly was the best olive-oil to be had 

 from the apothecaries. 



The oil was let into the cell through filtering-papers, which 

 probably gave up some of their finest fibres to the thick and ad- 

 hesive liquid ; and one or two of these fibres, though very fine 

 and short, were enough to mask the principal effect. From 

 very small fibres I could never cleanse the liquid perfectly till I 

 applied a simple device for removing them by electric action, after 

 the cell had been closed and filled. Two fine wires, properly in- 

 sulated from each other near their ends by fine (vaccine) tubes 

 of glass, were let down as one piece through the central boring 

 into the cell, one wire being connected with the prime conduc- 

 tor of the slowly worked machine, and the other wire uninsulated : 

 the fine fibres were at once attracted, and arranged themselves 

 as a chain or woof between the bare ends of the two wires, which 

 were then withdrawn as one piece without being discharged. It 

 was found unsafe to apply this piece a second time. When a 

 second application was required, a new piece had to be made of 

 clean materials. 



The oil was in this way cleared of all visible specks of solid 

 matter ; but it did not yet work well when examined under elec- 

 tric action in the polariscope. Sparkling points appeared in all 

 parts of the electric field, darting hither and thither irregularly. 

 These were fine air-bubbles, produced either by the preceding 

 operations upon the liquid, or by occasional spark-discharges 

 through the cell : they generally rose out of the way more or less 

 completely in a few minutes. Accordingly it often happened in 

 these experiments, that a plate of oil which gave no trace of the 

 principal effect at first, acted very well after it had been left un- 

 disturbed for some minutes, giving the effect at last with per- 

 fect distinctness and regularity. But if the liquid was disturbed 

 anew by accidental discharge through the cell, things were at 

 once in much the same condition as at first. 



T may mention, finally, that- in the course of the experiments, 

 whenever the electric force reached any considerable intensity^ 

 distinct movements were generally observed in the oil. They 

 bore no resemblance to the current-motions which are observed 

 under similar conditions in alcohol and other imperfectly con- 

 ducting liquids ; they rather presented the appearance of pulsa- 

 tions, quick and irregular heavings of the liquid. Whatever 

 the nature of these movements, and whatever their relation to 

 the electric force, I see no reason to think that they had any 

 influence on the principal effect. 



All the experiments on carbon disulphide were repeated with 

 olive-oil, the first (29) being most frequently repeated and espe- 



