150 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



diately vaporized but at greater distances there would be an increas- 

 ing number of solid particles; in fact, in the zodiacal light we have 

 independent evidence of the existence of such material in interplane- 

 tary space. According to this hypothesis, within one radius of the 

 sun's surface the electron is the scattering mechanism but farther 

 out the interplanetary dust cloud does most of the scattering. If 

 the dust cloud moves with the sun, then any details of coronal struc- 

 ture in the outer corona would persist for long periods without 

 changes. For the long streamers which are prominent features of 

 nearly every corona and which unquestionably change slowly with 

 time, the dust cloud explanation is unsatisfactory. When knowledge 

 progresses still further and we learn more about the physical laws 

 that govern matter under solar conditions, then we may be able to 

 find an explanation of the dark lines in the coronal spectrum that is 

 simpler than the assumption of scattering of the sun's light by 

 interplanetary dust. 



Manifestly, for future progress we need more and better spectra of 

 the dark lines in the coronal spectrum taken under clear skies devoid 

 of water vapor in order to be more certain that the Fraunhofer lines 

 are actually coronal in origin and do not come from sunlight scattered 

 in the earth's atmosphere. We need more and better spectra of large 

 dispersion in order to derive more accurate wave lengths of the bright- 

 line spectrum by wliich to find the origin of coroniura and to measure 

 the rotation of the corona. Up to date we have pitifully few spectra 

 of the corona with good definition and adequate dispersion. In the 

 future these spectra must be measured photometrically to as great 

 distances as possible from the sun's surface in at least two directions, 

 at the sun's equator and poles. Needless to say, the photographs 

 should cover as great a range in wave length as possible. More 

 information is urgently needed concerning the degree of polarization 

 of the coronal light to as great distances from the sun as possible. 



In spite of the brilliant work of Lyot in photograpliing the corona 

 outside an eclipse, it is not difficult to predict that coronal observations 

 in the future as in the past will have to depend almost exclusively on 

 the few precious moments of total echpscs which afford an average of 

 1 minute per year for accumulating observations. 



THE FLASH SPECTRUM 



Ever since it was first observed in 1870, the spectrum of the chromo- 

 sphere continues to be one of the most important problems taken up 

 for investigation at each succeeding echpse. In spite of the enormous 

 advances in instrumental equipment and tecluiique, which in other 

 branches of astronomy has appeared to accomplish almost the impos- 

 sible, there are remarkably few good photographs of the flash spectrum. 

 In fact, to check off on one's fingers the really first-class photographs 



