b. Excite the above extra-marginal motions of the odometer held near the two plates in succession; and then apply the thumb to the finger in each experiment. Result—Tangential motions are manifested parallel to the diametral motions before displayed. (See figs. 5 and 6.)
We cannot, however, suppose these extra-marginal tangential motions to be the lateral limits of the four great currents, inasmuch as they are obtained by the versed process to that which obtains the central motion; and the question arises, what then are the limits of the central currents?
c. Hold the odometer over the zinc disc at its centre; of course, longitudinal oscillations determined by the proximad current manifest themselves. Then shift its place on the transverse diameter more and more to the left. First Result—For something more than a quarter of the whole diameter, the motion continues longitudinal, proving that the central current has a breadth at least something exceeding half the diameter. Same result on the other side of the centre. Second Result—When the odometer nears the sinistral cardinal point of the zinc disc, its longitudinal proximad motion is replaced by the motion I have called clock-rotation. When it is held near the dextral cardinal point, versed rotation manifests itself.
This second result establishes that the longitudinal proximad current extends laterally to the edges of the disc; but that, when near to them, the force of the co-existing transverse current asserts itself, driving the odometer (on the left) off in a sinistro-proximad diagonal, which ends in the establishment of clock-rotation; on the right driving the central current off, in a dextro-proximad diagonal, resulting in versed rotation. (See fig. 7.)
Parallel and opposite results are obtained by the odometer when these experiments are repeated with the copper disc; and necessarily the clock-rotation appears near the proximad margin of the disc, the versed rotation near the distal edge.
Therefore it is evident that the great longitudinal and transverse currents extend over the whole disc, but not beyond it. Experiment a, section II., and figs. 3 and 4, show that, immediately beyond the cardinal points, single forces are in operation.
Other interesting results follow from trying with the odometer the extra-marginal spaces between the cardinal points.
d. First let the central points between each pair of cardinal points be tried with the zinc disc. Result—(see fig. 9,)—a dextro-proximad current is manifested between the sinistral and distal points, and between the proximal and dextral points; a sinistro-proximad current is manifested between the dextral and distal, and between the proximal and sinistral points—giving the impression that there exist two diagonal forces, comparable to the longitudinal and transverse forces.
Fig. 10 gives the corresponding, but opposite, results obtained upon the copper disc.
It is, however, doubtful whether these currents traverse the whole disc. For if the experiment is made of following each upon the disc, their influence disappears at less than a quarter of the diameter, where the odometer is found to obey on the zinc disc the proximad current, on the copper disc the dextrad current. Or, probably, these currents are the simple expression of the action of two equal forces moving the body operated on by them (at right angles to each other) in the diagonal. These effects thus form a remarkable contrast with the results given in figs. 7 and 8, wherein rotatory movements are manifested; and they seem to show that an essential element in these rotatory movements is, that one of the two currents acting on the odometer must, in the latter case, be of superior force to the other.