SINTKRING. 59 



cosity is relatively small (anorthite) crystallization begins at a low 

 temperature and proceeds very rapidly, the sintering probably being 

 due to the interweaving of the crystal fibers during their formation. 

 In viscous glasses (albite) sintering also begins at very low tempera- 

 tures the finer the powder and the slower the heating, the earlier the 

 first traces appear. Long-continued heating, even at comparatively 

 low temperatures, yields a perfectly continuous cake (except for the 

 included bubbles) the surface area of which constantly tends toward a 

 minimum. There is no doubt that the sintering of powdered glasses is 

 due to flow in the undercooled liquid and is a phenomenon in viscositv 

 and surface tension. All the feldspar glasses sintered readilv between 

 700 and 900 , depending on the fineness of the powder and the time. 

 Powdered crystalline feldspars do not sinter readily below their 

 melting temperature. Indeed, we were at first inclined to the view 

 that when only pure, dry, stable crystals are present they do not 

 sinter at all, however finely they may be powdered. We observed the 

 phenomenon in natural albite at 1000 , but the crystals were not 

 wholly free from inclusions which may have caused chemical reactions 

 resulting in cementation. Crystalline fluorite also sinters 300 below its 

 melting temperature, but here we were able to establish a decomposi- 

 tion; acid fumes were evolved during the experiment, and the sintered 

 product contained 1 per cent of free lime. Our final experiments with 

 long-continued heating for specific-gravity determinations, however, 

 showed that the purest feldspars which we could prepare, even after 

 they had reached their maximum density, still sinter very slowly. Thus 

 AbiAn 5 powder, which was shown by a determination of its specific 

 gravity to be holocrystalline, formed a compact chalky mass in four 

 hours at a temperature about 1 50 below its melting point ; in three 

 davs the cake was as hard as porcelain. Other feldspars showed the 

 same behavior. It is hardly possible that inhomogeneities sufficient to 

 produce diffusion between portions of different concentration could 

 have existed in these charges. There is considerable indication that 

 some of the crystalline nuclei grow at the expense of others perhaps 

 through exceedingly slow sublimation which may account for it. 



We made repeated attempts to locate some fixed sintering point 

 which should be characteristic of a particular material by means of 

 continuous measurements of the electrical conductivity, but thev 

 all indicated that no such point exists. The conductivity of a 

 dry powder increases enormously after sintering begins and would, 

 therefore, seem to offer a most sensitive test, but the phenomenon is 

 altogether gradual, even with a crystalline feldspar containing only a 

 small percentage of glass. We purpose to extend these observation? 

 to other substances. 



