THE DISCOVERY OF NEW ELEMENTS. 239 



nations fluorine, hitherto almost unknown. Spectrum analysis has 

 brought to light the existence of an entire series of elements with char- 

 acteristic spectra, and it has been possible to demonstrate the i)rc'sence 

 of one of these, helium, in the sun, before it was known that it also 

 entered into the composition of our globe. The conclusions drawn by 

 D. Mendelejeff from the law of periodicity also led to the discovery of 

 several elements whose existence was theoretically indicated before the 

 chemist had isolated them. 



I will again recur to the results obtained by Mendelejeff, but I will 

 now mention scandium discovered in 1879 by Mlson in euxonite, gado- 

 linite, and yttrotitanite. This metal, of whose oxide only a few grams 

 exist, and which, perhaps, no one except its discoverer has ever had in 

 his hands, has a considerable scientific importance, because its atomic 

 weight of 44, determined by Nilson, is precisely that indicated by 

 Mendelejeff for ekabor, an element whose existence had been foretold 

 by the law of periodicity. 



As early as 1794 Gadolin had separated from the gadolinite of 

 Ytterby an earth which he called ytter earth, and which later was 

 recognized as formed of erbine, terbiue, and of ytter properly so called. 

 These earths were again found in a great number of rare minerals, but 

 the oxides derived from these minerals appeared to be different in 

 nature and behavior, seeming rather to be mixtures very difficult to 

 break up into their constituent parts, their various elements giving no 

 very characteristic reactions. It was necessary to have recourse to 

 spectrum analysis and to the determination of their atomic weights, 

 trying to isolate them by repeated fractionation, either by the action 

 of the sulphates of potash or ammonium or by the partial decomposition 

 of the nitrates under the action of heat. We will not enter here into 

 the details of these operations which are special researches in the strict 

 sense of the word, and whose results are not yet clearly established as 

 to certain points. The major part of the work has, however, been 

 accomplished in the last quarter century; and not only has it given us 

 more precise ideas concerning scandium and yttrium, but it has also 

 revealed the existence of numerous other rare elements whose decom- 

 position does not appear impossible, among which we will cite : Erbium,^ 

 holmium,^ thulium,^ dysprosium,'' terbium,'' gadolinum,'' samarium,* 

 decipium,^ ytterbium." Lucium,'° recently announced by P. Barriere, 

 has been since contested. 



1 Cleve, Comptes rendus, 91-381. 



2Id., ibid., 89et91. 



3 Id., ibid. 



''Lecoq de Boisbaudran, ibid., 102. 



^Delafontaine, Ann. de cbitu. et de phys., 14. 



*^Lecoq de Boisbaudrau, Comptes rendus, 102. 



Ud., ibid., 89. 



^Delafontaine, Comptes rendus, 87. 



^Marignac, Comptes rendus, 87. 



it-W. Crookes, Cbem. Zeitung, 1896. 



