974 Prof. A. M. Tyndall on the Forces acting 



If the diameters of A and C are not infinite compared with 

 the distance between them, it will still be true that the whole 

 drag is. taken by C at the start and by A at the finish during 

 those stages of the motion when the distance from B to either 

 electrode is still very small. But in the main region between 

 them the direct effect of the drag will be to set the gas in 

 motion; and it is only after this motion has been transformed 

 into static pressures and suctions that the electrodes receive 

 back their shares of the drag momentum. None the less, so 

 long as the motion of the gas is not excessive, that is to say 

 so long as it is viscous, the process will be symmetrical and 

 the electrodes will each end by receiving half the total 

 momentum. In this case also, therefore, the electrodes will 

 be separately in equilibrium when the whole journey is taken 

 into account. 



But now let us consider the effect of a dissymmetry in the 

 form of the electrodes. 



If is of wire gauze and A is continuous, the electrical 

 reactions are unaltered, but the plate now receives nearly 

 the whole momentum from the gas, because gas can flow in 

 through 0. behind the layer B as it moves forward. The 

 electrodes are therefore in this case thrust apart by equal 

 and opposite momenta. If C is continuous and A of wire 

 gauze, the electrodes will be thrust towards one another. 

 Or if the electrodes are of very different sizes, the balance of 

 forces at each separately will be upset. The case of the 

 electric windmill is well known. Also when a fine point 

 discharges against a large plate delicately suspended, the 

 plate moves towards or away from the point according as ii 

 is or is not perforated to allow the wind to pass through it. 

 It appears, therefore, that only when complete symmetry in 

 all the discharge conditions is present can it be said with 

 certainty that each electrode separately will be in equilibrium 

 in respect of the viscous motion of ions between them. 



But when the ions are accelerating under the .force of the 

 field, symmetry is generally lost and the electrodes are no 

 longer separately in equilibrium. Two cases of this type 

 which bear on the present discussion are referred to below — 

 the starting from rest of an ion which later reaches the steadv 

 state of viscous motion, and the projection of ions from 

 one electrode to the other without perceptible drag on the 

 gas._ 



Griven the correctness of Pollock's view of high-speed 

 electrons projected from the cathode with a velocity of 

 l'4x 10 s cm. per sec, Duffield deduces that they must carry 



