538 S. Taber—Gnowth of Crystals. 



Small crystals of alum colored with chrome alum (not more 

 than 025 millimeter in diameter) were dropped into a warm 

 saturated solution of plain alum, which was then allowed to 

 cool down over night in the laboratory. Next morning, when 

 the crystals were removed from the solution and measured they 

 had an average diameter of 3 centimeters and height of 1*5 

 centimeters. The terraced hollows on the under side ranged 

 from 2*38 to 3*21 millimeters in depth, and, at the top of each 

 of these cavities, the small nucleal crystal, colored with chrome 

 alum, could readily be distinguished ; for it had not dissolved 

 and no white alum had been deposited on its under surface. 

 This means that these small nucleal crystals were elevated 

 through a distance equal to about twelve times their original 

 diameter. 



In another test a small crystal of alum colored with chrome 

 alum was placed in the solution of plain alum and allowed to 

 grow until it had a diameter of 3 millimeters. It was then, 

 removed from the solution, dried, and the under surface given 

 a thin coating of shellac. After the shellac had hardened, the 

 crystal was replaced in a warm saturated solution, and allowed 

 to grow over night as in the previous experiment. At the 

 close of the test it was found that the impervious coating in no 

 way had interfered with the deepening of the hollow, which 

 was similar in every respect to the hollows formed in the 

 crystals that had not been treated in this way. 



Another crystal, with nucleus of chrome alum, was coated 

 with shellac on all faces except the lower face, and then placed 

 in a warm saturated solution as before. Deposition of new 

 material took place first at the base and then gradually it 

 covered all of the shellacked faces. The new faces were 

 parallel to the old, but the new growth was easily broken from 

 the original crystal along the shellacked surfaces, being firmly 

 attached only at the base. The original crystal was elevated 

 through a distance of # 52 millimeter. A crystal, with all of 

 its faces shellacked, did not grow in any direction when replaced 

 in a saturated solution. 



Most of the experiments outlined above were made with the 

 alum crystals resting on an octahedral face, because flattened 

 octahedrons are the commonest forms, and are also the easiest 

 with which to work. When small (not more than 1 to 1*5 

 millimeters in height) these crystals frequently show only 

 octahedral faces, but, as they grow more rapidly near the bot- 

 tom than higher up, the larger ci\ystals assume a terraced and 

 roughly pj'ramidal form due to the alternation of octahedral 

 faces and of octahedral and cube faces as shown in the cross - 

 section, figure 2. In the several experiments made with 

 crystals resting on cube faces, there seemed to be little if any 



