September 17, 1903J 



NATURE 



481 



The metals chiefly referred to are copper, silver, lead, tin, 

 manganese, chromium, iron, cobalt, nickel, palladium, gold, 

 and iridium. 'Several of these, such as silver and gold, 

 have lately been distilled in vacuo by Krafft. 



Banded Flame Spectra. 



Well-defined groups of elements yield banded flame 

 spectra which have a similar constitution ; thus magnesium, 

 zinc, and cadmium yield bands composed of fine lines, de- 

 graded towards the violet, while fluted band spectra of 

 beryllium, aluminium, and indium were found to be de- 

 graded towards the red. Thallium also yields a fluted 

 spectrum ; gallium gives a line spectrum ; lanthanum gives 

 bands degraded towards the red ; palladium gives bands 

 in the nature of flutings composed of fine lines ; germanium 

 gave very faint indications of bands ; rhodium and iridium 

 both lines and bands. It became manifest that elements 

 belonging to the same group in the periodic system of 

 classification exhibited banded spectra which are similarly 

 constituted, and hence similarly constituted molecules of the 

 elements have similar modes of vibration, whether at the 

 lower temperature of the flame or at the higher temperature 

 of the arc or spark. Banded spectra are thus shown to be 

 connected with the periodic haw. 



A great advantage is to be derived from an investigation 

 of banded spectra from a theoretical point of view, as well 

 as from the application of this method to the analysis of 

 terrestrial matter. While the spectra are easily obtained, 

 they can be applied in a very simple manner to the chemical 

 analysis of minute quantities of material, and may readily 

 be made quantitative. 



M. Armand de Gramont has described a method of obtain- 

 ing spectra of metals and metalloids by means of a spark, 

 and has given the analysis of eighty-six mineral species. 

 The novelty and importance of his work lies in the method 

 of obtaining spectra of such constituent substances as 

 chlorine, bromine and iodine, sulphur, selenium and 

 tellurium ; also phosphorus and carbon when in a state of 

 combination, as sulphates, phosphates, carbonates, &c. 



There is a possibility of utilising this method for the 

 quantitative determination of carbon, sulphur, and phos- 

 phorus in iron and steel during the process of manufacture. 



Definition of an Element. 

 In a discussion on the question of the elementary character 

 of argon in 1895 it was pointed out by me that argon gave 

 a distinct spark spectrum by the action of condensed sparks, 

 and therefore, on this evidence alone, it must be regarded 

 as an element. The fact that it gave two spectra under 

 different conditions was not opposed to, nor did it invalidate, 

 this evidence, because such an element as nitrogen not only 

 emits two spark spectra, but the two spectra can be readily 

 photographed simultaneously from the same spark dis- 

 charge. 



It was proposed by M. de Gramont at the International 

 Congress in Paris in 1900, and agreed, that no new sub- 

 stance should be described as an element until its spark 

 spectrum had been measured and shown to be different from 

 that of every other known form of matter. 



This appears to me to have been one of the most im- 

 portant transactions of the Congress. The first application 

 of this rule has resulted in fhe recognition of radium as a 

 new element : it is characterised by a special spark spectrum 

 of fifteen lines which have been fullv studied and measured 

 by Demar^ay. It shows no lines of' any other element. 



Another application of this rule has recently been made 

 by Exner and Haschek with preparations of the oxide of 

 an element obtai.ned by Demar^ay, and named europium. 

 It exhibits 1193 spark lines and 257 arc lines. 



I have already mentioned that one feature strikingly 

 shown in the spectra of chemically related elements was the 

 wider separation of the lines in pairs, triplets, or other 

 groups ; was in some way related to the atomic mass, since 

 the separation was greater in those elements the atomic 

 weights of which were greater. Kayser and Runge, and 

 also Rydberg, have shown that in the series of alkali metals 

 the differences between the oscillation frequencies of the 

 lines are very nearly proportional to the squares of the 

 atomic weights. Runge and Precht have recently shown 

 that in every group of elements that are chemically related 

 the atomic weight is proportional to some power of the 



NO. 1768. VOL. 68] 



distance separating the two lines of the pairs of which the 

 spectrum is constituted. In other words, if the logarithms 

 of the atomic weight and distance between the lines be 

 taken as coordinates the corresponding points of a group 

 of elements which are chemically related will lie on a 

 straight line. Applying this law to the determination of 

 the atomic weight of radium they find that the strongest 

 lines of the new element are exactly analogous to the 

 strongest barium lines, and to those of the closely related 

 elements magnesium, calcium, and strontium. The 

 intervals between the two lines of each pair in the principal 

 series, and in the first and second subordinate series, if 

 measured on the scale of oscillation frequencies, are equal 

 for each element, and the same law holds good for the 

 spectrum of radium. From this the value 257 8 was found 

 for the atomic mass of the element. This does not quite 

 accord with the number obtained by Madame Curie, who 

 found it to be 225. It will be interesting to see which 

 number will eventually be proved to be the more correct. 



It is now many years since I first pointed out that the 

 absolute wave-lengths of the lines of emission spectra of. 

 the elements are physical constants of quite as great im- 

 portance in theoretical chemistry as the atomic weights;, 

 in the light of recent discoveries this statement may be saii; 

 to be now fully justified. 



Radio-active Elements. 

 From the study of rays of measurable wave-lengths we ■ 

 have lately sailed under the guidance of M. Henri Becquerel 

 into another region where it is doubtful whether all the rays ■ 

 conform to the undulatory theory. In fact some of the rays 

 are believed to be charged particles of matter, charged, that 

 is to say, with electricity. Beyond doubt they are possessed 

 of very extraordinary properties, inasmuch as they are able 

 to penetrate the clothing, celluloid, gutta percha, glass, and 

 various metals. They are, moreover, endowed with a no 

 less remarkable physiological action, producing blisters and 

 ulcerations in the flesh which are difficult to heal. It is 

 an established fact that such effects have been caused by 

 only a few centigrams of a radium compound contained im 

 a glass tube enclosed in a thin metallic box carried in the- 

 pocket. 



From this we can quite understand that there is no- 

 exaggeration in the statement attributed to the discoverer. 

 Prof. Curie, by Mr. W. J. Hanmer, of the American Insti- 

 tute of Electrical Engineers, that he would not care to- 

 trust himself in a room with a kilogram of pure radium, 

 because it would doubtless destroy his eyesight, burn all 

 the skin off his body, and probably kill him. 



It remains for me to express regret that without an undue 

 extension of the time devoted to this Address it would have 

 been scarcely possible to afford adequate treatment to the 

 absorption spectra of inorganic compounds, particularly 

 those of the rare earths, and such also as afford evidence 

 of the chemical constitution of saline solutions ; or of 

 organic compounds closely related to coloured substances 

 and dyes, the investigation of which leads to the elucida- 

 tion of the origin of colour, and serves to indicate the nature 

 of the chemical reactions by which coloured substances may 

 be evolved from those which are colourless. 



Chemistry is popularly known as a science of far-reaching 

 importance to specific arts, industries, and manufactures;: 

 but it occupies a peculiar position in this respect, that it 

 is at one and the same time an abstract science, and one- 

 with an ever-increasing number of practical applications. 

 To draw a line between the two and say where the one ends 

 and the other begins is impossible, because the theoretical 

 problem of to-day may reappear upon the morrow as the 

 foundation of a valuable invention. 



SECTION C. 



GEOLOGY. 



Opening Address by Prof. W. W. Watts, M.A., M.Sc, 

 President of the Section. 

 There are two circumstances which invest the fact of my 

 presidency of the Section this year with peculiar pleasure 

 to myself. The first public lecture I ever gave was in the 

 Town Hall at Birkdale in 1882, and the first of the fifteen 

 meetings of the British Association which I have attended 

 was that held in Southport in 1883. 



