(1) A preliminary examination of the spectrum of the extreme limb of the Sun, photographed with a large slit spectrograph immediately after third contact at the 1940 eclipse, showed that it is essential to remove the superposed chromospheric spectrum before any significant results with regard to the spectrum just within the limb can be deduced. (2) A method of performing this correction was devised, and the corrected spectrum, arising from less than one second of arc within the limb, at cos ? = 0.05, was found to contain few, if any, emission lines, excepting the hydrogen Balmer series. It is shown that the true change from an emission to an absorption spectrum must have been scanned by the Moon's limb in not more than two seconds of time, possibly in less than one second. To study the transition in detail, as had originally been intended, is therefore a much more difficult problem than hitherto supposed, and cannot be profitably attempted with these spectrograms. (3) Measures of the equivalent widths of 251 lines showed that at cos ? = 0.05 weak lines are as strong, or slightly stronger, than at the centre of the disk. Lines of equivalent width exceeding about 0.13 A. are weakened at the limb, the weakening increasing with equivalent width. (4) An examination of ordinary Fraunhofer line theory failed to reveal any combination of parameters which would explain this result. It is suggested that the clue to this puzzle lies in a roughness of the Sun's surface, probably associated with the solar granulation, which in effect reduces the value of ? for observations near the limb. Quantitative estimates show that an apparent cos ? = 0.05 may be increased to an effective cos ? = 0.20 in this manner, while an apparent cos ? = 0.3 may be increased to 0.39. (5) Accepting a correction to ? or this order, a modified Schuster-Schwarzs-child model is suggested, which within present observational uncertainties fits reasonably well (a) the eclipse observations, (b) Miss Adam's observations of weak lines, (c) Houtgast's observations of strong lines of FeI and CaI. Until more accurate values for the intensity of emission in the continuous spectrum as a function of optical depth are available, and until we have discovered more about the effect of the roughness of the Sun's surface upon centre-limb changes in the spectrum, closer quantitative comparison of theory with observation appears to be premature.
- 1 - natural sciences ; 2 - physics & astronomy ; 3 - astronomy & astrophysics
- 1 - sciences appliquees, technologies et medecines ; 2 - sciences biologiques et medicales ; 3 - sciences medicales ; 4 - techniques d'exploration et de diagnostic (generalites)
- 1 - Physical Sciences ; 2 - Earth and Planetary Sciences ; 3 - Space and Planetary Science
- 1 - Physical Sciences ; 2 - Physics and Astronomy ; 3 - Astronomy and Astrophysics