April A,, 1889] 



NATURE 



541 



dysprosium. The absorption spectrum of dysprosium 

 contains the two bands X 753 and X 451-5, the residual 

 holmium having a spectrum consisting of the remaining 

 two bands, X 640 and 536. 



Phosphorescence Spectra. 



I will now deal with phosphorescence spectra. Not a 

 few chemists and physicists, conspicuous among whom is 

 Ed. Becquerel, have carefully studied the phenomena of 

 phosphorescence. Phosphorescence may be excited by 

 elevation of temperature, by mechanical action, by elec- 

 tricity, and by exposure to the rays of the sun, and the 

 light thus given off, for example in the case of fluor-spar, 

 has been examined by means of the spectroscope. In 

 my own spectroscopic research I have dealt with the 

 phosphorescence occasioned by the impact of the mole- 

 cules of radiant matter upon certain phosphorescent 

 bodies, or what I have ventured to call molecular 

 bombardment. 



It is not necessary for me to describe the mode of 

 procedure further than to say that the substance under 

 examination is placed in a very high vacuum— a vacuum 

 which varies in degree in the case of certain earths. In 

 such a vacuum, when submitted to the action of the 

 induction current, substances phosphoresce very differ- 

 ently from what they do when treated similarly at the 

 ordinary pressure of the atmosphere. Under such cir- 

 cumstances the spectroscopic examination of matter 

 affords what I have called the radiant matter test. The 

 number of substances which are thus phosphorescent is 

 very considerable. Glass of different kinds, according 

 to its composition, phosphoresces with various colours. 

 Phenakite (glucinium silicate) phosphoresces blue ; spodu- 

 mene (aluminium and lithium silicate) gives off a rich 

 golden-yellow light ; whilst the emerald phosphoresces 

 crimson, and the diamond, being exceptionally sensitive 

 and brilliant, throws off a bright greenish white light. 



The ruby, one of the minerals I examined earliest in 

 this manner, glows with a rich brilliant red tone, quite 

 independent, as regards its depth and intensity, of the 

 colour of the stone as seen by daylight ; the pale, almost 

 colourless specimens, and the highly-prized variety of the 

 true " pigeon's blood," all phosphoresce with substantially 

 the same colour. 



This method of observing the constitution of the rare 

 earths, duly aided by delicate and prolonged chemical 

 processes, has permitted us to push our investigations 

 further than had previously seemed practicable. It 

 enables us to determine whether we have reached the 

 end of our investigations— a consummation which had 

 hitherto been vainly sought. It has enabled us to prove 

 that yttrium, samarium, &c., are not simple, homogeneous 

 bodies. But what of the constituents into which they 

 have been thus resolved } Suppose we refine them down 

 until each displays merely one spectral band— what then } 

 Is each one of such bodies, barely differentiated from its 

 neighbours chemically or physically, entitled to rank as 

 an element ? If so, as I pointed out in the address which 

 I had the honour to deliver before you in March last, we 

 shall have to deal with further perplexing questions, 

 arising in part from the relation of such elements to the 

 periodic system. In a discussion of the elements, not as 

 yet published, Dr. Wundt maintains that their possible 

 number cannot exceed seventy-nine. But I myself see no 

 definite and sufficient reason for limitation to this number. 

 If these bodies are not elementary, possessing as they do 

 the properties commonly regarded as characteristic of an 

 element, we must be prepared to show why not ? 



Whatever rank may ultimately be assigned to these 

 substances, they must, for convenience sake, have names 

 as soon as our knowledge of their properties is in a suf- j 

 ficiently advanced state to allow of their removal from the 

 suspense account. 



The Yttriu.m Group. 



Yttrium— the old yttrium— proves now to be not a 

 simple element, but a highly complex substance. I have- 

 come to the conclusion that it may be split up certainly into 

 five and probably into six constituents. If we take these 

 constituents in the order of their approximate basicity — 

 the chemical analogue of refrangibility— the lowest of 

 these constituents gives a deep blue band, Ga ; then fol- 

 lows a strong citron band, G8, which increases in sharp- 

 ness until it may be called a line ; then a red band, GC : 

 then a deeper red band, Gi; ; and lastly a close pair of 

 greenish blue lines, Gi3. Following these are frequently 

 seen Ge, Gy, and G^, the yellow, green, and red com- 

 ponents of samarium. 



A possible explanation of the existence and nature of 

 the new bodies into which " old yttrium " has been split 

 up, and of parallel cases which will doubtless be found 

 on closer examination, is this. Our notions of a chemical 

 element must be enlarged ; hitherto the elemental mole- 

 cule has been regarded as an aggregate of two or more 

 atoms, and no account has been taken of the manner in 

 which these atoms have been agglomerated. The struc- 

 ture of a chemical element is certainly more complicated 

 than has hitherto been supposed. We may reasonably 

 suspect that between the molecules we are accustomed to 

 deal with in chemical reactions, and the conponent or 

 ultimate atoms, there intervene sub-molecules, sub-aggre- 

 gates of atoms, or meta-elements, differing from each other 

 according to the position they occupy in the very complex 

 structure known as " old yttrium." 



The arguments in favour of the different theories are 

 as yet not unequally balanced. But the assumption of 

 compound molecules will perhaps account for the facts, 

 and thus legitimate itself as a good working hypothesis, 

 whilst it does not seem so bold an alternative as the 

 assumption of eight or nine new elements. 



I have just mentioned that the earth heretofore called 

 yttria, and supposed to be simple, has been split up into a 

 number of simpler bodies. Now these constituents of the 

 "old yttria are not itnpurities in yttria, any more than 

 praseodymium and neodymium are impurities in didy- 

 mium. They proceed from a real splitting up of the 

 yttrium molecule into its components, and when this 

 process is completed the "old yttria" has disappeared. 

 If these newly-discovered components on further examina- 

 tion should be found worthy to take the rank of elements, 

 I think, as first discoverer, I am entitled, by the custom 

 prevailing among men of science, to name them. For 

 the present, and until their investigation is more ad- 

 vanced, I designate them by provisional symbols. One 

 of the most distinct characteristics of " old yttria " is its 

 very definite spark spectrum. To which of its components 

 this spark spectrum belongs I am not yet able to say. It 

 is possible the particular component to which the spark 

 spectrum is due yields no phosphorescent spectrum. It 

 is also possible that the spark spectrum, like " old yttria," 

 may prove to be compound, and then the well-known 

 lines it contains will have to be shared between two or 

 more of the newly-discovered bodies. 



I wish emphatically to re-state that at present no single 

 component of old yttria can lawfully lay claim to what 

 may be called the paternal name ; and it seems to me 

 that in the present state of the question no one is entitled 

 to call one of the new bodies " yttria," and to characterize 

 the remainder as impurities. 



Interference of Phosphorescence Spectra. 



A recent discovery of some beautiful spectra given by 

 the rare earths when their pure oxides are highly calcined, 

 shows the remarkable changes produced in the spectra of 

 these earths when two or more are observed in combina- 

 tion. It has likewise opened to me a wide field of in- 



