May 27, 1920] 



NATURE 



395 



specialised form of the appliance, the sacchari- 

 meter, is practically indispensable in the sugar 

 factory. 



Mention must also be made of the simple polar- 

 iscope in its application to the glass industry, 

 where it is employed for detecting strains in glass- 

 ware due to faulty annealing. Not infrequently 

 glass articles, imperfectly annealed, are destroyed 

 on the cutting-wheel after a good deal of time 



-Twyman's apparatus for the determination of annealing temperatures. A, B, F, G, optical 

 system ; C, electrical furnace ; D, pyrometer ; £, temperature recorder. 



has been spent on their partial decoration. Use 

 of the polariscope to detect strains is not new ; 

 makers of optical glass have, naturally, long 

 availed themselves of it ; but as regards ordinary 

 glassware the method has been brought more 

 prominently under notice as a result of war con- 

 ditions, and the " strain viewer " is now becoming 

 more generally known in glass works. The prin- 

 ciple involved is merely that of the well-known 

 transmission of polarised light through crossed 

 Nicol prisms when 

 crystalline or semi- 

 crystalline material is 

 placed between them. 

 Well-annealed glass 

 leaves the field of the 

 instrument practically 

 uniformly dark; 



strained glass pro- 

 duces patches or bands 

 of light, the intensity 

 and colour of which 

 give some idea of the 

 amount of strain. 



In this connection it 

 may also be men- 

 tioned that certain 

 other faults in glass 

 can readily be de- 

 tected by means of 



X-rays. This discovery has proved very useful in 

 making the best qualities of optical glass, by 

 preventing the use of material in which "air- 

 blows " had formed. 



Twyman's apparatus may also be noted here. 

 It is used for determining the annealing tempera- 

 ture of glass (Fig. i). The method evolved for 

 this purpose is applicable also to metal objects, so 

 far, at least, as the removal of stress is concerned. 

 Trouble during the machining of metals such as 

 manganese-bronze, owing to distortion through 

 NO. 2639, VOL. 105] 



stress, might often be obviated by proper anneal- 

 ing of the articles. 



Among recent developments, perhaps the most 

 notable is the fact that the spectroscope, in one 

 adaptation or another, is beginning to take a 

 definite place as an adjunct to industry. This 

 follows upon the progress which has been made 

 in fitting the instrument to quantitative work. In 

 fact, it is the spectrometer, rather than the spec- 

 troscope proper, which is prov- 

 ing its value to the manufac- 

 turer. Hartley's work on 

 quantitative spectrum analysis, 

 dating from the eighties of last 

 century, may be regarded as 

 the pioneer investigation. He 

 showed that the ratios of the 

 intensities of lines in the spec- 

 trum of an element do not re- 

 main constant whilst the quan- 

 tity of that element is de- 

 creased, and he introduced the 

 term "persistency " to indicate 

 whether a particular line appears at a definite 

 concentration of the substance emitting it — 

 e.g. I per cent., o-i per cent., and so on, 

 of the total material under examination. This 

 work of Hartley's was followed by that of 

 Pollock and Leonard in Dublin, and of 

 Gramont in France — to mention only three names 

 out of many. Meanwhile, the earlier torms of 

 spectroscope have given rise to the more perfect 

 " constant deviation " wave-length spectrometer 



Fig. a. — Print from negative taken on spectrograph with wave-length scale. Top strip, sample of reputed "pure 

 tin." Middle stripj sample of commercial tin. Bottom strip, short exposure of copper spectrum. The 

 presence of coppjr in the csmmercial tin is shown by the presence in the corresponding spectrum strip of 

 the two copper lines at 3247 and 3974. A trace of copper is present in the "pure" tin. 



and the quartz spectrograph, with the result that 

 it is now practicable even to carry out quantitative 

 analyses of metals bv means of their spark spectra 



Gramont uses two types of sparking apparatus 

 (see Comptes rendus, 1918, clxvi., 95). In one of 

 these the substance under examination is contained 

 in a crater formed in one pole of the apparatus; 

 in the second type the substance is fused in a 

 platinum vessel, a spark being passed from a 

 thin rod into the fused material. 



