66 



NA TURE 



[May i6, 1901 



and then expanding them by blowing. This method gave satis- 

 factory results at once. By it we can produce long tubes and 

 other apparatus like those exhibited to-night, if not at a very 

 quick rate or very low cost, yet with certainty and very much 

 more quickly than before. 



When a tube of silica has been made it can be worked in 

 the flame as easily, though not as inexpensively, as glass. Such 

 a tube can be thickened readily by adding fresh rings of silica ; it 

 can be drawn out to various degrees of fineness and sealed 

 hermetically ; whilst all kinds of joints can be made easily. In 

 one respect silica is easier to work than glass. It never breaks 

 when thrust into the flame, and finished apparatus needs no 

 annealing. 



One precaution must be taken. The eyes must be protected 

 by black spectacles. The glass of which these are made must 



be very dark ; so dark that white hot silica does not look very 

 bright when viewed through it. 



I have spoken of silica as being easy to work. I do not mean 

 you to understand, however, that it is easy to do what you see 

 Mr. Lacell doing to-night. It is not easy to perform any 

 operation of this sort with his wonderful precision, and espe- 

 cially it is not easy to work under the conditions enforced upon 

 him to-night, for he can see nothing of the effects he produces 

 and must adapt his manipulations to my remarks although he can 

 hear the latter only very imperfectly. 



The Properties and Applications of Silica. 

 Vitrified quartz is harder than felspar, but less hard than 

 chalcedony. When cut with a file it breaks like glass. Its 

 conducting power for heat is about equal to that of glass. Mr. 



teller (Comptes rendus, cxxx. 1703) and more recently by Prof 

 Callendar. The former finds its mean coefficient of expansion 

 between o' and 1000° to be O'oooooo7, but from the manner in 

 which his material was prepared I think it is probable that it 

 was not quite pure. Prof. Callendar has, within the last 

 few days, examined the behaviour of a rod of pure vitrified 

 silica prepared by my method. He finds its mean coefficient 

 of expansion to be only O'oooooo59, which is only jV ^s great 

 as that of platinum, and much smaller than that of any other 

 similar substance that has hitherto been studied. He finds also 

 that the expansion of vitrified silica is exceedingly regular up to 

 1000", and that if not heated above 1000' the rod returns very 

 exactly to its original length when cold. Beyond 1000* he 

 found a slight permanent elongation, although the rod was under 

 compression. Prof. Callendar was able to carry his experi- 

 ments up to 1500°, which is very satisfactory, for it shows that 

 vitrified silica remains solid, or practically solid, at this very 

 high temperature. This is an important observation, as less 

 carefully conducted experiments had led us to fear that it became 

 slightly plastic even at as low a temperature as 1000". Above 

 1000° the rate of expansion diminishes rapidly, changing to a 

 contraction at about 1200°.^ On cooling from 1500° to 1200° it 

 expands. 



Fine rods of silica and also quartz fibres are apt to become 

 rather brittle after being heated to redness. But we have not 

 at present detected this defect in the case of thick tubes or rods. 



The transpar.incy of vitrified silica to the ultra-violet rays has 

 been carefully examined by Dr. A. Wynter Blyth, to whom I 

 am greatly indebted. 



The following figure (Fig. 2) illustrates very well the character 

 of the results he has obtained. This figure gives the results of 

 photographing electric sparks taken between electrodes made 

 of an alloy of mercury, tin, zinc and cadmium after passing the 

 light through sheets of quartz, vitrified silica, soda glass and 

 flint glass. The plates of the last three substances were of 

 equal thickness and were carefully prepared for me by Mr. 

 Hilger. 



The results show, as indeed we have found by actual expeti- 



2. Crystal (^>uart^, 3 t 

 5. Do., 4 mm, ... 



4. Vitreous Silic-i 



5. Soda Glass 



6. Flint Gl.-iss ... 



Boys has shown that, even in an atmosphere saturated with 

 moisture, it is a very good insulator. Its density (2 '21)^ is 

 decidedly less than that of quartz (2'66). Its optical pro- 

 perties have not yet been fully studied, but its approximate 

 index of refraction has been determined by Prof S. P. Thompson 

 by means of a small prism cut for the purpose by Mr. Hilger. 

 It is decidedly less than that of quartz. 



The melting point of silica is not known and it is plastic over 

 a considerable range of temperature. When a platinum wire 

 embedded in a thick tube of silica is heated from without by 

 means of an oxy-gas flame, the platinum melts and runs at a 

 temperature at which the silica retains its shape. 



Its rate of expansion has been studied first, by H. Le Cha- 



iiy pupil, Mr. T. Pears, the silica used con- 



NO. 1646, VOL. 64] 



ment, that silica tubes are much more suitable than glass ones 

 for use in studying the spectra of electric discharges. 



The most remarkable property of vitreous silica is its behaviour 

 uniler sudden changes of temperature. We have seen already 

 that tubes of it may be plunged suddenly into an oxy-gas flame 

 without injury, and I have mentioned the fact that apparatus 

 made of silica needs no annealing. But this is not all ; we may 

 drop water on a white hot vitrified silica rod, or plunge white hot 

 silica into cold water, or even, by Prof. Dewar's kind aid, into 

 liquid air without injuring it in any way whatever ; indeed, ex- 

 periments seem to show that the material gains very distinctly 

 in regard to its elasticity when it is thus treated. I need hardly 

 point out how convenient tubes of such a material will be to 



: Fig. 3, shows ; 

 ■X temperature. 



contraction, but coir 



