PHYSICS. 275 



reaches its greatest distinctness in between 36 and 108 minutes ; at 700 ° in 

 between 4 and 9 minutes, though after this the coalescence of ferrite continues. 



The degradation of the ferrite network seems due primarily to the belated 

 coalescence of additional ferrite into masses which conceal that network, 

 though the balling-up of the network itself through surface tension probably 

 contributes. 



The internal ferrite and cementite are slower to coalesce into visible 

 masses in cooling down, and are more quickly reabsorbed by the neighboring 

 austenite when the temperature again rises through the transformation 

 range, than the network, a fact which, at least in the case of cementite, may 

 be related to the pseudomorphism of the internal masses and the probable 

 idiomorphism of the network. 



But at constant temperature within the transformation range, after the 

 cementite network has reached its maximum prominence, the internal ce- 

 mentite coalesces into masses which, probably through some optical illusion, 

 are much greater than our theories explain, and even conceal the network. 

 Still later, the internal cementite gradually disappears, by a process akin to 

 sublimation, by progressive resolution in the surrounding austenite and re- 

 precipitation. This migration of the cementite usually again makes the net- 

 work very prominent by thickening it up, but it sometimes forms large, 

 shapeless cementite masses. A like process may perhaps go on in the case 

 of ferrite also. 



Experiments to determine the upper boundary of the transformation 

 range seem nearly complete. 



In addition to the foregoing, which refers to the iron-cementite diagram, 

 considerable work on the iron graphite diagram has been done. Some of 

 the flake-graphite of cast iron has been proven to be formed before solidifi- 

 cation has progressed far. While this may be primary graphite, it may be 

 formed by the very rapid resolution of solid cementite in the molten iron 

 and its reprecipitation as idiomorphic graphite. 



Experiments to determine the lower boundary of the transformation 

 range of this diagram have been begun. 



Nichols, E. L., Cornell University, Ithaca, New York. Grant No. 777, 

 allotted December 15, 191 1. Systematic study of the properties of mat- 

 ter through a wide range of temperatures. (For previous reports see 

 Year Books Nos. 4-10.) $3>oco 



I. Fluorescence and phosphorescence: 



During the year a large amount of time has been given to the preparation 

 of the report recently published by the Carnegie Institution of Washington 

 under the title "Studies in Luminescence," by Edward L. Nichols and Ernest 

 Merritt (publication 152, pp. vm + 22 5- I9 12 )- 



In the meantime we have extended our observations on the fluorescence 

 and absorption of the uranyl salts with reference to the more detailed deter- 



