Relation betiveen the Spectra of Sunspots and Stars .
53 ' " On the .Relation between the Spectra of Sunspots and Stars .
" By Sir Norman Lock yen , K.C.B. , LL. D. , F.R.S. Received June 8 , \#151 ; Read June 16 , 1904 .
As the period throughout which the observations of widened lines have been made at South Kensington now includes two maxima and three minima epochs of solar activity , it has seemed desirable to discuss the results obtained , taking into account the chemical origins of the lines affected in passing from the photosphere to the sunspot nuclei .
This is going on , but in anticipation of its publication , I desire to direct attention to one of the conclusions arrived at in its bearing upon the question of the temperature conditions of the Arcturian and lower type stars , which formed part of the subject of a recent paper.* Since 1894 , when the last discussion of the widened line results was published , !
nearly 10,500 observations of lines in sunspot Spectra have been made at South Kensington .
An analysis of these lines , in respect to their origins , shows that the elements chiefly affected during the period 1892\#151 ; 1903 , inclusive , were Vanadium and Titanium .
The great importance of Vanadium and Titanium in sunspot spectra has also been demonstrated by Father Cortie during his observations in the B\#151 ; D region at S tony hurst , f It was foreshadowed in a previous paper on the chemical classification of the starsS that it seemed probable that , as the result of further work , the " genera " then proposed might have to be split up into " species .
" During the more recent research mentioned above the temperature classification was tested by comparing the relative intensities of the red and ultra-violet ends of the spectra of stars , situated on various horizons of the temperature curve , including Capella and Arcturus , which , according to the original general classification , belong to the same type , viz. , " Arcturian .
" It was found that the spectrum of Capella extended on an average about 70 tenth-metres further into the .
ultra-violet than that of Arcturus , whilst the red portion of the spectrum is certainly stronger in the latter .
That is to say , the general temperature of Arciurus is appreciably lower than that of Capella .
The next step was to see if chemical change accompanied this reduction of temperature , and if so , whether the change was in any way related to the change from the photosphere to the sunspot spectrum .
# 4 Roy .
Soc. Proc. , ' vol. 73 , pp. 227\#151 ; 238 , 1904 .
f 'Roy .
Soc. Proc. , ' yoI .
57 , p. 199 , 1894 .
+ * Monthly Notices ( R.A.S. ) , ' vol. 63 , No. 8 , pp. 479\#151 ; 480 , June , 1903 .
S * Roy .
Soc. Proc. , ' vol. 65 , p. 191 , 1899 .
54 Relation between the Spectra of Sunspots and Stars .
In comparing , for this purpose , the spectra taken with the 6-inch Henry prismatic camera it was noticed that certain lines were relatively intensified in passing from the spectrum of Capella to that of Arcturus .
Similar comparisons of the Fraunhoferic spectrum with the spectra of Capella and Arcturus respectively were next made .
This work led to the following conclusions :\#151 ; ( 1 ) That the line absorptions of Capella and the sun are practically identical ; ( 2 ) that although , speaking generally , the same lines occur in the spectra of the sun and Arcturus , yet in the latter many lines are relatively more intense than in the former .
Moreover , in the great majority of such cases the lines so intensified are probably due to Vanadium and Titanium .
Thus we see that whilst the temperature classification mentioned above certainly places Arcturus on a lower temperature level than Capella and , therefore , the sun , the evidence obtained from a study of the line absorptions of Arcturus and of sunspots indicates very clearly that the temperature of the Arcturian absorbing atmosphere is about the same as that of the sunspot nuclei during the above-mentioned period .
This conclusion justifies the ideas formulated by De la Rue , Stewart , and Loewy that the spots are produced by the downrush of cooler material .
In a recent publication , * which has been received here since the above-mentioned comparisons were completed , Professor Hale suggests that because the lines which are widened in sunspots appear as strong dark lines in Piscian stars , the effect may be produced because sunspots are more numerous in such stars .
From the evidence adduced above it seems a far more probable explanation to suppose that these lines are intensified in sunspots , and strengthened in those stars which have been placed on lower temperature levels than the sun , because the general temperature conditions are similar .
That is to say , the fall of temperature experienced by the metallic vapours in passing from the photosphere to the spot nucleus is of the same order as that to which an absorbing atmosphere is subjected in passing from the temperature conditions of Capella or the sun to that of Arcturus or the lower temperature stars .
* " The Spectra of Stare of Secclii'e Fourth Type " ( ' The Decennial Publications , ' Chicago University , 1903 ) .

