IVM) 



tJkler (Mid I) roirn i ikj-I'!!, c!roli/sis and 



deposit as liquid globules which eventually fell automatically 

 into a small glass spoon kept vertically below the cathode. A 

 sectional view of a cell is shown to scale in figure 1. Tins 

 diagram also indicates the additional convenient accessories 

 employed in the production of trees. It was found that trees 

 were invariably formed at about 0° C, hence the statement 

 that the figure is intended to suggest ice in a porcelain develop- 

 ing tray with the crystallizing dish supported on the four feet 

 of an inverted brass stool, affords an adequate explanation. 



Fig. 1. 



The current used was about 0'28 ampere giving a mean cur- 

 rent density of 0*007 and 6 amperes per square centimeter at 

 the anode and cathode respectively. These means are, of 

 course, extremely rough since the distribution of electricity 

 was far from uniform, especially at the cathode tip. 



Since the melting point of gallium is about 30° C it may 

 occur to the reader to look upon the scheme of cooling the 

 electrolyte as a perfectly obvious procedure for the production 

 of trees. The necessary condition for the formation of these 

 solid structures is, however, not as simple as might be expected 

 at first thought. With three cells in series and at room tem- 

 perature (23° C.) it sometimes happened that one or two of the 

 cathodes would continue to present liquid globules while the 

 remaining cells or cell respectively would give the solid phase. 

 Stroking a globule with a fragment of solid gallium often 

 started the growth of a tree, but not invariably. The deposi- 

 tion of a tree at room temperature seems to depend upon a 

 number of factors, such as the alkalinity of the electrolyte, the 

 curvature of the surface of the liquid globule, etc. These 

 idiosyncrasies are on a par with the predisposition to pro- 



