524 HENRY A. ROWLAND 



With the high dispersion here used the "basic lines" of Lockyer are 

 widely broken up and cease to exist. Indeed, it would be difficult to 

 prove anything except accidental coincidences among the lines of the 

 different elements. Accurate investigation generally reveals some slight 

 difference of wave-length or a common impurity. 



Furthermore, the strength of the lines in the solar spectrum is gen- 

 erally very nearly the same as that in the electric arc, with only a few 

 exceptions, as for instance calcium. The cases mentioned by Lockyer 

 are generally those where he mistakes groups of lines for single lines 

 or even mistakes the character of the line entirely. Altogether there 

 seems to be very little evidence of the breaking up of the elements in 

 the sun as far as my experiments go. 



Even after comparing the solar spectrum with all known elements, 

 there are still many important lines not accounted for. Some of these 

 I have accounted for by silicon and there are probably many more. Of 

 all known substances this is the most difficult to bring out the lines in 

 the visible spectrum although it has a fine ultra-violet one. Possibly 

 iron may account for many more, and all the elements at a higher tem- 

 perature might develope more. Then, again, very rare elements like 

 scandium, vanadium, etc., when they have a strong spectrum, may cause 

 strong solar lines and thus we may look for new and even rare elements 

 to account for very many more. Indeed, I find many lines accounted 

 for by the rare elements in gadolinite, samarskite and fergusonite other 

 than yttrium, erbium, scandium, praeseodymium, neodymium, lantha- 

 num and cerium, which I cannot identify yet and which may be without 

 a name. For this reason, and to discover rare elements, I intend finally 

 to try unknown minerals, as my process gives me an easy method of 

 detecting any new substance or analyzing minerals however many ele- 

 ments they may contain. 



The research is much indebted to the faithful and careful work of 

 Mr. L. E. Jewell who has acted as my assistant for several years. 

 Preliminary publications of results will be made in the ' University 

 Circulars.' 



Among the lastest results I may mention the spectroscopic separation 

 of yttrium into three components, and the actual separation into two. 



