1887. 



on Genesis of the Elements. 



41 



The vacuum most suitable for experiments on tliese earths is one 

 of about the millionth of an atmosj^here. In a vacuum of this degree 

 we find under the action of the induction-spark certain substances 

 phosphorescing or behaving very differently from what they would if 

 similarly treated at a lower vacuum or at the ordinary pressure of the 

 atmosphere. When thus treated, the examination of the spectra of 

 the phosphorescing earths furnishes what I have ventured to call the 

 radiant matter test. 



After a time, on examining the series of yttrium earths in the 

 lowest line of flasks, their phosphorescent spectra are found to have 

 become modified in the relative intensities of some of their lines. 

 Ultimately different portions of the fractionated yttria give the five- 

 spectra approximately shown at the foot of the diagram ; whilst 

 Samaria also appears capable of being split up into two or perhaps 

 three constituents. 



These bodies, it must be clearly understood, are not impurities 

 which may be removed, yttrium or samarium remaining in a pure 

 state after their elimination. On the contrary, the molecule we 

 formerly knew as yttrium has undergone a veritable splitting up into 

 its constituents. 



These constituents I have not as yet formally baptised. For 

 more convenient reference and "discussion I have provisionally 

 ticketed them, as shown in the following Table. 



Table I. 



I will rapidly sketch the most salient features of the rare earths 

 when submitted to the radiant matter test. Some remain unaffected 

 and thus are referred at once to a distinct group. Others have the 

 curious property of preventing the induction-spark passing, and so 

 simulating a non-coi]ducting vacuum, when there is really plenty of 

 residual gas present. The rare earth thoria possesses in the highest 



