166 Transactions of tee American Institute. 



iJtliough an incandescent solid or molten globe, sends ns liglit, which 

 before it reaches our planet has passed through sodium vapor ; or in 

 other words, that there must be sodium in the incandescent atmos- 

 phere around the sun. [Applause,] Now what is true of the sodium 

 line is true of the bands of several of the other metals. Iron, for 

 example, when converted into vapor by the heat of the electric lamp, 

 shows us a spectrum consisting of not less than eighty distinct bands ; 

 and each one of these eighty bands coincides absolutely, not only in 

 position but in relative intensity, with the dark bands which cross 

 the solar disc. Then there must be iron in the sun. You can see 

 the force of our argument if you will attempt to calculate upon this 

 doctrine of probabilities the chances of eighty coincidences of this 

 sort. The chances would be as millions of millions to one against such 

 a coincidence happening by mere accident. In like manner, we have 

 been able to prove calcium, magnesium, nickel, chromium, barium, 

 copper, and zinc to exist in the state of vapor around the sun. On 

 the other hand we have equal reason to know that gold, silver, mer- 

 cury, aluminum, cadmium, tin, lead, antimony, arsenic, strontium, 

 lithium, do not exist in the sun's atmosphere, at least to any consid- 

 erable extent. [Applause.] But the bands whose names I have 

 mentioned are far from accounting for all the dark lines of the solar 

 spectrum. These, a? I shall show you, can no more be counted than 

 the sands upon the sea shore. Although we have been able to refer 

 a hundred or more of them to a few well-known metals, the great 

 unnumbered legion remain still unexplained. Thus much, however, 

 seems to be established, that our sun is an immense incandescent ball of 

 molton or solid matter, surrounded by a comparatively dark atmos- 

 phere, which is, however, itself luminous to a less degree, and becomes 

 visible during a total eclipse, when tongues of flame are seen extending 

 beyond the moon's dark disc. The sunlight, in passing tlirough 

 this atmosphere is absorbed, and hence the dark lines. Moreover we 

 have discovered since the last total eclipse that these tongues of flame 

 may be seen even in the full sunlight by carefully directing the 

 spectrum toward the disk of the sun ; and at Cambridge, where Prof. 

 Winlock has observed these bands with great care, it has been estab- 

 lished clearly that three of the bands coincide exactly with the three 

 hydrogen bands to which I previously referred ; and there can be but 

 little doubt that these tongues of flame consist, in great part at least, 

 of burning, or at least incandescent hydrogen. [Applause.] We have 

 also examined the light from some of the brighter stars, and find that 



