6o PROTOPLASM 



tions carried out with noii-polarisable brush electrodes it was 

 shown that under these conditions a positive streaming 

 could be seen quite plainly, even if relatively feeble, in pure 

 drops of oil also. Similar experiments performed at earlier 

 dates with platiaum electrodes had partly given the same 

 results, but had more particularly shown that fairly large 

 oil-drops iu soapy glycerine obey distinctly the well-known 

 law that small suspended particles wander towards the 

 positive pole, which they do not do in pure glycerine. 



Influence, of Induction Shocks 



Successfully prepared drops of foam placed in glycerine 

 were frequently exposed to the action of single moderately 

 strong induction shocks, from which, as a general result, the 

 following was fairly uniformly the outcome. The shocks 

 generally produce sudden and frequently most violent con- 

 vulsive movements, which extend in good drops over the 

 whole margin, but may appear more localised in those not 

 so good. In such convulsions the edge pf the drop takes on 

 \a more or less wrinkled appearance, and frequently shrinks 

 \Jack to a not inconsiderable amount. As a rule, it can be 

 plainly observed that the shocks on closing the current 

 have much less effect than those on breaking it. If the 

 effect as a whole is weak, the shocks on closing frequently 

 produce no convulsions at all, while the shocks on breaking 

 still distinctly do so. When the experiments were con- 

 tinued for some time, the definite and invariable result was 

 that the convulsions become feebler, and often cease alto- 

 gether. 



When vigorously streaming drops were exposed to 

 induction shocks, it could frequently be observed that the 

 streaming came to a stop for a moment, or even for as long 

 as a minute, starting again after this interval, and often with 

 increased strength. It is always streaming movements that 

 go on fairly at right angles to the direction of the electric cur- 

 rent, which are thus affected. Occasionally, though seldom, 

 streaming movements which had slowed down were sup- 



