S. Taker — Growth of Crystals. 537 



are lifted by growing crystals are always the same as the 

 amount by which their hollows are deepened. The results 

 obtained in the experiment described below may be taken as 

 typical. 



A crystal of alum having a thickness of 2 , 18 millimeters was 

 placed in a warm saturated solution and covered with a small 

 glass plate. The solution was allowed to cool slowly in the 

 crystallizing dish, which was left over night on a table in the 

 laboratory. After about fifteen hours the crystal was taken 

 from the solution, dried and remeasnred. It was found to 

 have increased 0*62 millimeter in total thickness, raising the 

 glass plate through that distance, the cavity on the under side 

 was deepened by 054 millimeter, and a shallow hollow, hav- 

 ing a depth of 0*06 millimeter, was formed on the upper sur- 

 face of the crystal. The increase in the depth of the cavities 

 and the increase in the height of the crystal, therefore, differ 

 by only 0*02 millimeter, which is well within the limits of 

 error in making the measurements. All measurements were 

 made with a screw micrometer graduated to 0'01 millimeter or 

 with a spherometer similarly graduated, and the figures given 

 above are the average of several independent determinations. 



The hollows found on the under side of crystals are familiar 

 to all who have grown them in glass vessels, for they are com- 

 monly formed on uncovered as well as covered and weighted 

 crystals. Bruhns and Mecklenburg argue that the presence of 

 these cavities does not indicate a lifting of the crystal by 

 growth from below, and in support of this conclusion cite the 

 experiment performed by Kopp in 1855, which they repeated 

 with similar results.* They advanced the hypothesis that the 

 formation of the narrow rim which surrounds the cavity and 

 on which the crystal rests is due to adsorption, stating that the 

 height of this rim does not increase with the growth of the crys- 

 tal, and that it moves outward because of solution on the 

 inner side, while it is continuously being built up anew on the 

 outer edge. They further state that the height of the rim 

 " about 0"12 millimeter " remains approximately constant dur- 

 ing its outward growth, and that this is possibly connected 

 with the thickness of the layer of solution between the crystal 

 and its support, f It is clear that these statements do not agree 

 with the results obtained in the experiments that the writer 

 made with weighted crystals, but in order that there might be 

 no uncertainty in this matter the following tests were con- 

 ducted with crystals in free contact with the solution except, 

 of course, on their under side, where they rested on the bottom 

 of the crystallizing dish. 



* Loc. cit,, p. 101 and footnote 15, p. 113. 

 % Ibid., pp. 101-105. 



Am. Jour. Sci. — Fourth Series, Vol. XLI, No. 246.— June, 1916. 

 37 



