The photographic technique of Leighton (1) has been used to obtain high resolution measurements of photospheric magnetic fields with the following chief results:
1. "Source fluxes", Φ(defined as the average of the absolute values of the total fluxes of positive and negative polarity in a bipolar magnetic region on the day when thisaverage is largest), of bipolar magnetic regions (BMRs) have been measured as a function of the "size", Am, of the associated sunspot activity. [Am is defined to be the area covered by all of the sunspots of a group (umbras plus penumbras of both leading and following parts) on the day when this area is largest.] The source fluxes were found to satisfy Φ = 1.2 Am within about ± 50%, where Φ is in "solar flux units" (1 SFU = 1021 maxwells) and Am is in units of 1018 cm2. Within measurement accuracythere was no imbalance of flux of positive and negative magnetic polarity in the BMRs observed, provided that the photographically-obtained fluxes were supplemented by the fluxes in the umbras of sunspots, as calculated from Mount Wilson observations.
2. Two BMRs were traced for the first few months following their birth. The development of each was semi-quantitatively what one would expect if the random-walk (2) plus solar differential rotation were the only means of flux transport on the solar surface.
3. In quiet regions of the solar surface, magnetic flux is distributed in small bits and fragments, and in the quietest regions, in tiny points less than 3000 km in diameter. Measurements for eight small features yielded fluxes ranging from 0.05 SFU for the very small points to 0.4 SFU for typical network fragments. The corresponding magnetic fieldsranged from 200 gauss for the points to 700 gauss for the network fragments, showing that magnetic fields of a few hundred gauss are not at all uncommon for small magnetic features.
4. Polar faculae curves (3) have been calibrated to givemagnetic flux on the polar caps of the sun as a function of time during the period 1905-1964. The fluxes vary cyclicly with time approximately 90° out of phase withthe variation of the sunspot number for the whole solar disk with time during this same period (provided the sunspot number is given a polarity corresponding to the magnetic polarity of the following sunspots of the relevant hemisphere). The maxima of the polar fluxes vary considerably from cycle to cycle (just as the maxima of the sunspot number vary considerably from cycle to cycle), maximum fluxes ranging from 6 SFU to 21 SFU with 12 SFU being a typical maximum polar flux during the 1905-1964 period.
The principal conclusion is that these measurements are consistent with the hypothesis that all of the magnetism on the sun's surface originates in BMRs, and that it is gradually spread about the surface by the random-walk (2, 4) and differential rotation, eventually causing the cancellation and reversal of the polar fields.
