Precision Method of uniting Optical Glass. 285 



lenses were found to be strained when examined by polarized 

 light. A few spots which bad developed in the usual way 

 from minute dust particles had blown out into flat bubbles 

 in the flint glass, indicating that they contained gas under 

 great pressure, as we had already inferred. 



It appears, therefore, that the flint and crown glasses 

 commonly used in object-glasses differ too widely in an- 

 nealing temperatures for successful union. Owing to the 

 large number of factors which limit the nature of the classes 

 to be employed, this is one of the most difficult cases to which 

 the method has been applied. 



Polarimeter Tubes. — A polarimeter tube having windows 

 permanently fixed in position by this method was pre- 

 pared without difficulty. The tube was of soda-lime glass 

 and the windows of plate glass ; maximum temperature 

 470° C. The joint was perfect, and in spite of the fact 

 that different glasses were employed, the windows showed 

 little strain — the amount was found not sufficient to inter- 

 fere with measurements of rotation in any way (see figs. 9 

 and 10, PI. V.). For many purposes this form of tube 

 should be very convenient, especially in respect of ease 

 of cleaning and freedom from corrosion. As soon as 

 possible we intend to apply this construction to jacketed 

 tubes ; they would somewhat resemble the ordinary pattern 

 of glass Liebig condenser, and should, with careful heating, 

 withstand very high temperatures. The nature of the hot 

 circulating liquid would be determined only by the tempe- 

 rature required, and not by the powers of resistance of the 

 cement, as is the case at present. 



Union of Fused Silica. — An experiment with component 

 parts of this material showed that there was no essential 

 difference between its behaviour and that of glass; the tem- 

 perature required for union was naturally found to be much 

 higher. The upper limit of temperature to which silica glass 

 apparatus may be heated is determined by its power of 

 devitrifying ; this takes place at about 1200°-1300° 0.*, and 

 is accompanied by great decrease in mechanical strength. 

 It is obvious that a very small amount of devitrification 

 would be sufficient to spoil silica glass for optical work> 

 especially as it takes place more readily where there is access 

 to the air, i. e. at the surfaces. The highest temperature 

 which we can use is therefore somewhat below 1200° C; 

 fortunately a preliminary experiment has showm that a still 

 lower temperature will suffice. 



* Blackie. Trans. Faraday Soc. vii. p. 158 (1911) ; Orookes, Proc. 

 Eoy. Soc. A. lxxxvi. p. 406- (1912). 



