given, as the author does not wish the paper to show a greater accuracy than the experiments are entitled to.
The way in which the magnetism is crowded to the surface at the higher speeds can be well seen in Table II. With a periodic time of 22'5 seconds the value of maximum average induction density near the surface is 10,770, for a maximum average over the whole cylinder of 6850. With a periodic time of 360 seconds a maximum average over the whole cylinder of 6970 is obtained with a maximum average near the surface of 7170. The values of maximum induction density at the surface are 17,000 and 7000 in the above examples respectively.
II. Alternating Magnetic Force. It is interesting to compare the results just given with those obtained from a cylinder 12 inches (30'5 cms.) diameter when under the influence of an alternating mag- netic force directed along the longitudinal axis. This cylinder is of mild steel and has probably nearly the same magnetic and electric properties as the cylinder in fig. 1 at the same temperature. Holes are drilled in the 12-inch cylinder in a plane at right angles to the longi- tudinal axis in much the same manner as in the rotating cylinder. Taking the areas as bounded by the centre lines of the drilled holes, the data in Table III have been obtained, and the curves in figs. 3, 5 have been plotted therefrom.* In fig. 3 the periodic time was 10'3 minutes, corresponding to a frequency of 150 periods per second with a cylinder O'l cm. diameter. In fig. 5, the periodic time was 2'6 minutes, corre- sponding to a frequency of 600 periods per second with a cylinder O'l cm. diameter. We see by Table I that figs. 2 and 4 can refer to a cylinder of diameter and length O'l cm., when rotating at frequencies 179 and 7 17 t periods per second. Comparing figs. 4 and 5, we see that there is a similarity, but with the alternating magnetic force the shielding effect is more severe than with the rotating field.
This is perhaps better seen by referring to fig. 7, where the points marked x are taken from fig. 5, and should be compared with curve 90. It will be seen that with the alternating magnetic force, the severity is about as great as at 1434 periods per second (curve 45) with the rotating force. Similarly, by comparing figs. 2 and 3, and the results obtained therefrom in fig. 7, namely, curve 360 and the points marked <8>, Ave see that for the same diameter of cylinder (O'l cm.) the effects are much more marked with the alternating force. The points lie fairly well on the curve numbered 180. The same remarks refer to the maximum average over the whole section.
It should be noted that the area of No. 1 coil in these experiments, which is the same as the innermost coil in the case of the alternating
- See Wilson, ' Roy. Soc. Proc.,' vol. 68, p. 218. The figures in italics in the table
at the end of this paper have been re-calculated, owing to an error in estimating the areas.
