Nature Vol. 187 No. 4740. September 3, 1960.
In a contribution(1a) to the Institution of Electrical Engineers Convention on Thermonuclear Processes, of which a somewhat fuller account was published later in Nature(1b), I suggested that the electrical discharge theory of solar flares and the associated magnetic storms and aurorae led to the surprising result that the temperature of these discharges must reach values of the order of 100,000,000° K. some where between the Sun and the Earth's orbit. In another account(1c) of this work, I suggested that this need not come as too great a surprise, as temperatures of the order of 1,000,000°K. had been obtained some years earlier in electrical discharges in the laboratory (I. V. Kurchatov, Moacow, 1956), and for long have been known to exist in the solar corona.
Furthermore, a similar increase in temperature is observed when long electrical discharges are propagated down the corresponding, but lesser, density gradients in some stellar atmospheres -- those of the combination spectra stars, such as Z Andromedae and AX Persei, for example. In these, the temperature is initially that leading to the ionization of the metals, hydrogen and helium, that is, probably of the order of 5,000° or 10,000°K. However, by the time the discharge reaches the outer regions of these stellar atmospheres, after periods of the order of 100-200 days, the discharge temperature reaches values of the order of 1,000,000°K. as lines of Fe X and Fe XIV appear in the star's spettra.
Though there was thus some additional theoretical support for the surprising conclusions to which the theory led when applied to these solar phenomena, it is satisfactory to learn that satellite observations made by the U.S. Navy scientists have confirmed(2) that these high temperatures do exist in the disturbances associated with solar flares. As I emphasized many years ago(1d), it is difficult to see how these high temperatures can be built up by any other mechanism, than that of an electrical discharge.
Similar temperatures are reached in galactic electric discharges, and can be measured, as I have suggested, by the proposed 'cosmic gas-velocity thermometer'(1e), of which the basic conception is that these high gas velocities observed in stars and galaxies are always the result of electrical discharges, the velocities of these jets being linked closely with their temperatures(1f), in contradistinction to the assumptions often made in other accounts of these solar jets(3). The only limit to the discharge temperature will be the onset of thermonuclear processes(1g).
On the electrical discharge theory of galactic evolution(1d,g) the main difference between the gas which ultimately forms Population I stars, and that which originally formed Population II stars, is that the former has been subjected to 'thermonuclear temperatures' of the order of 400,000,000°K. for a period of the order of 10-100 million years, during the formation of the spiral arms. It may be, therefore, that it is this difference which has led to the difference in the proportions of the heavier elements observed in these two stellar populations. It is not easy to visualize all the physical characteristics of electrical discharges on such a scale as that envisaged, but at present the value of n, the number of atoms per cm3 of the galactic atmosphere, required to effect the observed chemical change in the time available, appears to be too high. However, it would be interesting to have comments on this suggested explanation of this difference between the two stellar populations.
C. E. R. Bruce
Electrical Research Association; Leatherhead, Surrey.