E. L. Troxell — Plaster of Paris. 



205 



water, for the gum retards its penetration. As has been shown, 

 the additional water detracts from the strength of the plaster 

 and, as in the other cases cited, causes some of the weakening 

 effect which may be ascribed to the gum. 



A test of the strength of twenty -four briquets made with 

 gum arabic in the original mixture gave the curves (fig. 3). 

 It begins showing the result of a one ninety-sixth solution, 



Fig. 4. 



soo 



400 



I 300 



2 200 



100 



2. A 6 S 10 12 



Parts of Gum Arabic in 48 of water. 



Fig. 4. The effect of soaking the hardened plaster in various solutions of 

 gum arabic. A, B, G, D, each represents 24 briquets, while E is based on the 

 tested strength of 12. 



which gave the very high average for four briquets of 509 

 pounds per square inch. The average for four briquets made 

 with a 1/48 solution is over a hundred pounds less. The 

 lowest point in the curve is where a 1/8 solution was used, 

 and the strength is less than half that given by the 1/96 

 solution. There is a slight rise to the end of the curve rep- 

 resenting the average strength of briquets made with a one- 

 fourth of gum. 



There is an irregularity in this curve simulated by one (same 

 fig.) based on the tests of eighteen briquets made earlier. This 

 is probably a coincidence since it is unlikely that there should 



