S. Taber — Growth of Crystals. 535 



In one of these experiments a crystal of potassium sulphate, 

 3"07 millimeters in thickness, was placed between two glass 

 plates, held tightly together by rubber bands. The crystal and 

 plates were then covered by a saturated solution of potassium 

 sulphate and placed in a desiccator. Within a few days volun- 

 teer crystals began to form around the sides of the dish just 

 above the surface of the solution, and a little later crystals 

 appeared on the bottom of the dish. After several weeks the 

 crystal pressed between the glass plates had gone completely 

 into solution, while in the meantime the volunteer crystals had 

 continuously grown in size. 



From these experiments it would appear that when two 

 crystals, similarly placed in a saturated solution, are subjected 

 to unequal pressures the system is in unstable equilibrium. In 

 other words a solution may become supersaturated with respect 

 to a crystal placed under a relatively small pressure, while at 

 the same time it tends to dissolve a crystal that is under a 

 greater pressure. This probably explains why Brulms and 

 Mecklenburg failed to get the results that were obtained by 

 Becker and Day.* 



Experiments with results similar to those described above 

 were performed by placing crystals in warm saturated solutions 

 and allowing the solutions to cool down over night. In these 

 tests supersaturation was induced by cooling, instead of by 

 evaporation. 



In another experiment a saturated solution of copper sul- 

 phate at room temperature (about 20° C.) was caused to flow 

 slowly through a siphon to the bottom of a crystallizing dish, 

 which was kept at a temperature of about 12° C. by surrounding 

 it with running tap water. Another siphon with upturned ends 

 kept the solution in the crystallizing dish at constant elevation. 

 The arrangement of the apparatus is shown in tig. 1. 



Before starting the flow of the solution through the apparatus, 

 a crystal of copper sulphate, held tightly between glass plates 

 surrounded by rubber bands, was placed in the crystallizing 

 dish. The flow of solution was regulated by means of a pinch- 

 cock so that about one liter passed through in twelve hours. 

 At the end of twenty-four hours the growing crystal had 

 pushed the plates apart, overcoming the resistance of the elastic 

 bands, and had increased the distance between the plates from 

 1*21 to 2*63 millimeters. The experiment was discontinued 



* On Dec. 29, 1915, several months after the completion of this work, the 

 writer learned from Dr. J. C. Hostetter that, under the direction of Dr. A. 

 L. Day, he had made experiments similar to those described above and 

 reached the same conclusion. As the publication of Dr. Day's reply to 

 Bruhns and Mecklenburg has been delayed, permission was courteously given 

 to the writer to publish this paper without waiting for the appearance of 

 that reply. 



