BRIDGMAN. — ACTION OF MERCURY ON STEEL. 339 



amalgamation spreads more rapidly tlian it does at others. But now 

 the metal is weakened at each of these infected places, and the type of 

 strain is modified as it would be by the presence of a flaw. The 

 strain will be redistributed, the brunt of the strain coming at the point 

 farthest removed from the center, where consequently the pores of 

 metal will be still further distended. That is, at this point the amal- 

 gamation will proceed most rapidly. It is evident that the continua- 

 tion of this process will produce a band of amalgamation travelling out 

 along the radius. When the amalgamation reaches the outside, or ap- 

 proaches sutliciently close, the metal gives way, the crack appearing 

 through the midst of the weakest region, that is, through the center of 

 the amalgamated band. It is evident that when the process of amal- 

 gamation has once started in this way, it will proceed more and more 

 rapidly as the resisting thickness of sound metal becomes less, thus ac- 

 counting for the smallness of the other amalgamated regions. One 

 cylinder was found, however, in which an amalgamated patch had 

 worked its way nearly half way to the outside diametrically opposite 

 the crack. 



Failure to produce amalgamation in the rods subjected to hydro- 

 static pressure is to be explained by the fact that neither can the amal- 

 gamation begin at the surface, because of the thin layer of dirt, nor 

 can the mercury force its way into the steel to begin amalgamation 

 there because the interstices in the metal are closed up by the hydro- 

 static pressure. The same argument of course applies still more to 

 the hollow cylinder submerged in mercury and subjected to pressure 

 on the outside. 



The experiments with the cylinders in which the bottoms were 

 blown off are to be explained in the same way. The amalgamation 

 'Trows most rapidly in the direction in which the distension is greatest, 

 which in this case is diagonally from the corner of the hole. As the 

 amalgamation proceeds it carries the hydrostatic pressure with it. 

 When the region over which this pressure acts has extended so far 

 that the sound metal left can no longer support the stress, it gives way 

 as usual by a clean shear. The fact that mercury was forced through 

 the bottom of a cylinder of soft tool steel (Figure 7), while soft tool 

 steel cylinders of the form of Figure 1 were unbroken by the action 

 of pressure, is probably to be explained by the different strain types in 

 the two cases. In the case of Figure 7 distension in a direction diag- 

 onally from the corner of the hole is great enough to allow amalgama- 

 tion before the pores are closed up by viscous yield. The point has 

 not been worked out in greater detail, however. The disconcerting 

 experiments in which a clean punching was blown out of the bottom of 



