ii2 THE NEW WORLD OF SCIENCE 



Crystallization results from super-saturation, just as in any 

 solution, and when the glass is coolecl too rapidly crystals may 

 also be formed even to the extent of devitrification. 



At one time each of two manufacturers had much trouble 

 with cloudy glass. It has been shown that this may be due to 

 chlorides or sulphates in the potassium carbonate, or in the 

 case of medium and dense flints, to excessive arsenic. Some 

 observers think the material to be present in colloidal form, 

 for when glass containing selenium, copper, gold, etc., is cooled 

 slowly a high color results, but if cooled quickly, the glass is 

 often clear. Potassium carbonate with o.i per cent sulphur 

 trioxide gave good results. When 0.4 per cent, was reached 

 the pot glass was milky at the edges, and when 0.75 per cent, 

 was present, the entire mass was spoiled. The skilful use of 

 high temperatures is said to be a good remedy for this lack 

 of chemical homogeneity. 



Physical homogeneity is just as important, for strains cause 

 deformity of optical surfaces, give astigmatism, and may even 

 lead to cracking of lenses. It is out of the question to produce 

 high-grade optical parts with glass not free from strain or 

 internal stress hence the necessity of fine annealing. The 

 softening point of glass is the temperature at which it flows 

 under its own weight, while the temperature at which it yields 

 slowly under loads approaching in magnitude its crushing 

 strength is its practical annealing temperature. Heretofore 

 annealing has been done by cooling so slowly that there is no 

 large temperature difference between the surface and the center 

 of a pot of glass. This requires expert manipulation and 

 extreme regulation of the temperature fall, as well as exact 

 pyrometric control. Drs. Adams and Williamson of the Geo- 

 physical Laboratory arrived at the conclusion that if the pot 

 were held long enough at the high temperature to allow internal 

 stress to be removed by the molecular movement of the glass, 

 yet with the temperature below that at which the glass would 

 flow, annealing could then take place much more rapidly. 

 While under this treatment strains would again be set up, they 



