548 S. Tdber — Growth of Crystals. 



lower temperature than the reservoir by surrounding it with 

 running tap water having an average temperature of about 

 12° C. The chamber was first filled with fragments of a 

 broken battery cell together with a few small crystals of copper 

 sulphate to induce crystallization, and then the interstices 

 were filled up with clean, fine quartz sand. Upon admitting 

 the solution, care was taken to remove, as far as possible, all 

 air bubbles from the apparatus. In the beginning the flow of 

 the solution was regulated so that about two liters passed 

 through in twenty-four hours, but toward the close of the 

 experiment, as the passageways became choked with crystals, 

 the flow became slower and slower. After starting the experi- 

 ment, it was kept in continuous operation until the chamber 

 was ruptured, which occurred at the end of 9 days in one test 

 and 7 days in another. 



After one of these tests, the mass filling the ruptured 

 chamber was immediately broken to pieces and carefully 

 examined. The outer portion of the mass was thoroughly 

 cemented together so that the surface in contact with the glass 

 was almost smooth, but near the center some of the spaces 

 were only partly filled by deposition. The crystals of copper 

 sulphate had sharp boundaries along their contact with the 

 quartz grains and cell fragments. No fibrous or columnar 

 crystals could be found, and, when freshly broken pieces were 

 examined under a bifocal microscope, magnifying to 78 

 diameters, no copper sulphate crystals were observed in the 

 pores of the cell fragments except occasionally where there 

 were unusually large openings. 



After completing the other test, the ruptured glass chamber, 

 with its contents, was set aside and allowed to stand undisturbed 

 for several weeks. The solution remaining in the chamber 

 at the close of the experiment gradually reached the surface 

 by capillary attraction, and evaporated where exposed to the 

 air along the fractures in the glass. The evaporating solutions 

 built up crystals in the fractures, and the glass fragments were 

 pushed farther apart, but none of these crystals showed a 

 columnar or fibrous structure. 



During these investigations another observation was made 

 that throws some additional light on the manner of crystal 

 growth under varying conditions. After a test in which the 

 flow of solution through the glass crystallizing chamber had 

 been nearly stopped by crystal growth, the apparatus was dis- 

 connected without rupture having occurred, and, with the tub- 

 ing still full of solution, was placed in a beaker, as shown in 

 fig. 5, where it was allowed to stand undisturbed for three 

 months during the summer. The pressure of the solution 

 standing in the longer tube caused the solution to flow very 



