along area I in much the same way as was observed radially that is to say, the intensity of induction diminishes in value and suffers retardation in phase. Table IV gives the values of the induction density and the phase-displacements for periodic times of 360, 90, and 45 seconds. In order to compare the diminution of induction density as one proceeds along the axis of the cylinder towards the centre, the maximum average values of induction density over Sections 1 and I have been plotted in figs. 2, 4, 6. The points are marked x and agree fairly well with the corresponding curves. The phase-displacements are shown in Table IV, and are a maximum for the intermediate force. Comparing the maximum average values of the intensity of induction over Sections 2 and II, we see that the corre- sponding differences are not so great and the phase-displacements are less. The maximum average values of induction density over Sections 3 and III are more nearly equal and the phase-displacements are small. Lastly, comparing the maximum average intensities of induction over areas I, II, III, we see that the diminution is not so great as over the areas 1, 2, 3, and the corresponding phase-displace- ments are less.
V. We have seen that with a cylinder of 10 inches axial length, the induced currents do not seriously disturb the approximately uniform distribution of induction density over its section, when rotated in a magnetic field with periodic time 360 seconds. This periodic time corresponds to a frequency of 179 periods per second in a cylinder of axial length O'l cm. In ordinary dynamo-electric machines the frequencies usually met with are very much smaller than 179, and plates O'l cm. thick are used. The inference is, that in cases 'in which as good an approximation to a pure rotating field as has been obtained in these experiments is met with, no serious devia- tion from uniform distribution exists, provided the plates are insulated from one another.
VI. Induction motors for a certain class of work are now supplied with solid iron armatures. Suppose such a motor with an armature 10 inches diameter and 10 inches long, and with one period per revolution, is placed on a supply system having a frequency of 30 periods per second. Fig. 6 refers to a periodic time of 45 seconds, or a frequency of - 022 period per second. To obtain the effects observed at this frequency the armature would have to rotate at frequency 29'978 that is, the effective frequency or "slip," as between the rotating field, which would have to be as uniform as in these experiments, and the armature is 0'073 per cent, of the frequency of supply. To obtain the effects observed at 22'5 second periodic time, the effective frequency would be 0-146 per cent, of the frequency of supply. A slip of 5 per cent, could easily be obtained in practice, and this would correspond to an effective frequency sixty-eight times
