152 BRIDGMAN. 



and is well marked by the striking change of color, from yellow to red. 

 I was not able, however, to make any measurements under pressure, 

 or even to detect the transition, although I made careful search at 20°, 

 130°, and 200°. The explanation is to be found in the gi-eat sluggish- 

 ness of the transition, which is much worse in this respect than Hglo. 

 Gernez ^^ has found at atmospheric pressure that the yellow modifica- 

 tion may be heated to 200° without initiating the transition, and that 

 the red may be cooled to liquid air without starting the reverse transi- 

 tion. He has kept the unstable red modification for two years at room 

 temperature without the transition starting. Furthermore he finds 

 that the transition from one phase to the other is always slow, e^'en 

 when inoculated. The material is evidently unsuitable for experi- 

 ment under high pressures. The only possibility is in finding some 

 catalj^tic agent that will hasten the transition. 



KI is cubic and isomorphous with the ordinary modification of 

 NH4I and the high temperature modifications of XH4Br and NH4CI; 

 it is strange that it showed no new form. KBr and KCl are also 

 isomorphous with KI; judging by analogy with the ammonium salts 

 one would might expect that these two would l^e more likely to show a 

 new transition than the KI. They should be tried. KI is a sub- 

 stance for which Spring ^* announced a large permanent change of 

 volume after exposure to 10000 kgm. at room temperature. I could 

 find no evidence for it. 



Nal is not isomorphous with KI, but it is with Agl. One would 

 expect new forms. 



Lil crystallizes with three molecules of water, the "melting point" 

 at atmospheric pressure being at about 72°, at which it gives up two 

 molecules of water. At 200° this transition point was found displaced 

 to about 10000 kgm. with a change of volume of approximately 0.015 

 cm.^ per gm. It is a matter of considerable difficulty to produce 

 anhydrous Lil, and I did not make the attempt. 



Sbis has been erroneously described as having enantiotropic transi- 

 tion points; it has already been discussed under melting. 



Among other compounds of this class which could be profitably 

 examined is Pbl^: this is known to have a transition at atmospheric 

 pressure, but at a temperature so high as to be beyond the range of 

 this work. 



Summarizing the behavior of the halogen compounds, polymorphism 



33 D. Gernez, C. R., 138, 1695, 1904, and 139, 278 (1904). 



34 W. Spring, Rap. au Cong, international Phys., 1, 402 (1900). 



