164 CARNEGIE INSTITUTION OF WASHINGTON. 



is relatively very great; nothing has to be inserted within the furnace cavity, 

 and the lag is practically nothing; the regulator is often nearly at its best 

 under conditions most unfavorable to other regulators. It has held a small 

 furnace for hours constant to 0.1° at temperatures from 500° to 1400°. 



(19) Temperature distribution in solids during heating or cooling. E. D. Williamson 



and L. H. Adams. Phys. Rev., 14, 99-114 (1919). (Papers on Optical Glass, 

 No. 11.) 



In deciding on the best methods of carrying out various operations in the 

 manufacture of optical glass, we found it necessary to have some idea of the 

 temperature gradients in the pieces during heat treatment. While great 

 precision in absolute magnitudes is generally of minor importance in such 

 cases, the only way to gain insight into the question of the variation of the 

 temperature differences with the shape and dimensions of the blocks and the 

 method of heating is actually to work out numerical cases. 



Equations have been derived for the temperature distribution in soUds of 

 several typical shapes, the solids being heated or cooled according to one of 

 two methods, viz, the surface of the body (1) is continuously heated (or cooled) 

 at a uniform rate, or (2) experiences a sudden change to a higher or lower 

 constant temperature. With these equations a number of calculations have 

 been made and the results of the computations are presented in tabular form 

 and, in certain cases, are also shown graphically. By the use of these tables 

 and graphs it is a comparatively simple matter to determine the temperatures 

 within soHds of a large variety of shapes when, as is commonly the case, they 

 are heated or cooled according to one of the methods mentioned above. 



The equations given are in convenient form for calculation and for showing 

 a number of interesting qualitative relations between the temperature gradi- 

 ents in various solids, and they will probably prove useful in connection with 

 the determination of specific heat and thermal conductivity by dynamic 

 methods. 



While the main interest at the time was in the application to glass manu- 

 facture, the equations are perfectly general, as are also all the quaUtative 

 deductions made. 



(20) The volatilization of iron from optical glass pots by chlorine at high temperatures. 



J. C. Hostetter, H. S. Roberts, and J. B. Ferguson. J. Am. Ceram. Soc, 2, 

 356-372 (1919). (Papers on Optical Glass, No. 12.) 



Of all the ordinary impurities found in optical glass, iron exerts the greatest 

 influence on transmission. The iron-content of the glass arises from pots used 

 as containers during melting as well as from the raw materials. The con- 

 tent of iron in the glass and, therefore, its transmission, would be considerably 

 improved if the iron could be removed from the pot-walls before use. Chlor- 

 ine appeared to be a suitable agent for this purpose, and experiments demon- 

 strated the fact that approximately 80 per cent of the iron could be extracted 

 from the interior of the clay pots and volatilized by the action of chlorine at 

 temperatures easily secured in a pot-arch or glass-melting furnace. Large- 

 scale experiments were carried out at the Bausch and Lomb Optical Company 

 and conditions developed for removing more iron from the bottom of the pot, 

 where the most corrosion takes place, than from the side-walls. Glass melted 

 in these pots showed, in all cases but one, less iron than that made in untreated 

 pots. In the exception noted above, however, more iron was found in the 

 glass made in the treated pot, and it was shown that, although the iron had 

 been volatilized from the pot, more than usual pot corrosion had taken place 

 during melting. The success of the method, then, depends on whether a 

 dense surface can be made in such pots when the iron has been removed, as, 



