August 26, 1897] 



NA TURE 



39: 



In the case of the rings photographed in 1893, the 

 ■evidence that they truly belong to the corona is absolutely 

 conclusive. In the absence of more exact knowledge of 

 the wave-lengths of the radi- 

 ations producing the rings, it 

 is not yet possible to determine 

 if they are represented by dark 

 lines in the Fraunhofer spec- 

 trum, but it can already be 

 stated that, if present at all, 

 they are among the feeble 

 lines. 



A point of some importance 

 is the apparent al)sence of 

 the 1474 K ring from the 

 spectra of the chromosphere 

 and prominences. Its absence 

 from the prominence spectrum 

 was noted by Respighi in the 

 eclipse of 1871. I am not 

 aware of any observation in 

 which the foj-iii of a promin- 

 ence has been observed in 

 1474 light. All these facts 

 seem to indicate that when 

 the 1474 line is observed at 

 the sun's limb without an 

 eclipse, the spectrum of the 

 corona itself is under ex- 

 amination, under the same 

 conditions as those recorded 

 in the eclipse photographs. 



Of the other coronal rings 

 photographed in 1893, those at 

 wave leng^tlis 4217 and 4280 

 are appro.ximately coincident 

 with feeble prominence radia- 

 tions, but since the other 

 coronal rings are not repre- 

 sented in the prominences, the coincidrnccs may be 

 regarded as accidental. 



been photographed as rings in the spectrum of the 

 corona with the prismatic cameras. They have, however, 

 been occasionally recorded as corona lines with slit 



KiG. 21.— Drawing bringing together all the details shown on Kostinsl y's and Hansky's pliotograph 

 (Prominences shown in black.) 



Although H and K are by far the most .intense of the 

 radiations of the prominences, on no occasion h..ve they 



NO. 1452, VOL. 56] 



Fig. 20. — The Corona of 1896, Kostinsky. 



spectroscopes, but it does not seem improbable that in 

 most cases they were produced by prominence light 

 scattered by our atmosphere, as 

 before explained— light of which 

 the prismatic camera takes no 

 account. 



Perhaps the most decided 

 evidence in favour of the ex- 

 istence of H and K in the 

 corona spectrum is that de- 

 pending upon the photographs 

 taken with slit spectroscopes in 

 1886 ; Dr. Schuster states that 

 "the lines end sharply with the 

 corona, and we must conclude, 

 therefore, that in spite of the 

 unfavourable atmospheric con- 

 ditions, there was but little light 

 scattered by our own atmo- 

 sphere in the neighbourhood of 

 the sun." But in spite of this 

 observation. Dr. Schuster has 

 concluded that H and K "do 

 not form part of the normal 

 spectrum of the corona " ; ' and 

 I may add that the prismatic 

 camera strengthens this con- 

 clusion. 



The photographs taken with 

 the prismatic cameras in 1893 

 show a pretty strong ''con- 

 tinuous " spectrum, but it has 

 already been explained that 

 this appearance may have been produced by a very 



Tram., vol. i8o, A p. 341. 



