1886.] New Elements in Gadolinite and Samarskite. 503 



One of the highest of the absorption- spectrum earths is didymia. 

 The spectrum of didymium, as generally met with, is well known, 

 and is given in my paper on " Radiant Matter Spectroscopy : Part 2, 

 Samarium" (par. 135). 



It has long been suspected that didymium is not a simple body, and 

 in June, 1885, Dr. Auer v. Welsbach announced that by a series of 

 many hundred fractional crystallisations he had succeeded in splitting 

 up didymium into two new elements, one giving leek-green salts and 

 the other rose-red salts. The green body he called Praseodymium 

 and the rose-red Neodymium. I have not found that my method of 

 fractionation gives a decomposition similar to this ; probably didy- 

 mium will be found to split up in more than one direction, according 

 to the method adopted ; but by pushing the fractionations at the 

 didymium end of the series to a considerable extent, a change gradu- 

 ally comes over the spectrum. At the lower end the earth gives an 

 absorption-spectrum such as is usually attributed to didymium, but 

 with no trace of some of the bands in the blue end, the one at \443 

 being especially noticeable by its absence. The intermediate earths 

 give the old didymium spectrum, the relative intensities of some 

 of the bands varying according to the position of the] earth in 

 the series, the band 443 becoming visible as the higher end is 

 approached. The highest fractions of all give the band 443 one of 

 the most prominent in the spectrum, being accompanied by other 

 fainter bands which are absent in the lowest didymium spectrum. 



In my note-book, under date 3rd March, 1886, after discussing the 

 absorption-spectrum given by one of my earths rather lower down in 

 the series (fraction —3), and comparing coincidences of the lines with 

 those given by hclmia, erbia, didymia, thulia, and samaria, I remark 

 "the big blue line (X 451*5) is still unclaimed." In ignorance that 

 my friend M. de Boisbaudran was on the same track, and running 

 me somewhat close, I deferred further examination of these fractions 

 till a few months longer work had been performed on them, when I 

 hoped to get fuller evidences of a new absorption-spectrum. This big 

 line in the blue, X 451 '5, now proves to be the characteristic line of 

 dysprosium. This line does not occur in didymium. The next 

 strongest line, X 475, is coincident with a very faint line in the old 

 didymium spectrum, and it also falls within a broad band of samarium. 

 M. de Boisbaudran says that this line is not due either to didymium, 

 erbium, or samarium ; as it follows the same variations of intensity as 

 the other lines of dysprosium he considers it due to the same element. 



The earth ( — 3), which I have already mentioned as giving the 

 broad black band (451*5) of dysprosium, shows this band wider (and 

 therefore presumably stronger) than in the spectrum given by M. de 

 Boisbaudran, without the slightest trace of the band 475, which, 

 according to M. de Boisbaudran, should be wider and almost as dense 



