PHYSICS — HOWE. 233 



The formation of the central "pipe" is due to the cooling and hence con- 

 traction of the different layers of the mass seoliotachically, i. e., at different 

 rates inter se. In the early stages of solidification the outside of the mass, 

 especially if it is cast in a cold iron mold, cools much faster than the deeper 

 seated solid layers. The early excess of contraction of the outside, caused by 

 this excess of cooling, is resisted by the lagging interior, with the result that 

 the outer layers are virtually stretched beyond their normal dimensions; so 

 that, when solidification is complete, the interior, which in the latter part of 

 the cooling has to cool through a greater range of temperature and hence 

 has to contract more than the outside, no longer suffices to fill that outside 

 completely, and this deficit of volume of the interior is represented by a 

 central cavity, overlying the region in which the last of the solidification 

 occurs. This same excess of contraction of the earth's crust in its early 

 stages should later throw that crust into great compression, which may be 

 an important element in volcanic and earthquake phenomena. 



Blowholes themselves tend in effect to expand the volume of the interior 

 as a whole, without changing its outer dimensions, and thus to lessen the 

 deficit or pipe. 



In case of steel ingots this pipe may reach very deep into the axis and, 

 because it is hard to weld up, may compel the manufacturer to discard as 

 much as a third of the ingot in order to get sound unpiped metal. To avoid 

 this some makers of steel of a composition favorable to welding have pur- 

 posely allowed blowholes to form rather abundantly, so as to prevent the 

 formation of a pipe, and, relying on the ease with which such steel welds, 

 have tried to get flawless metal by welding these blowholes up in the process 

 of rolling the ingot out into its final form, such as that of a boiler-plate. 



This procedure is of great economic importance, in that it enables the 

 steel-maker to avoid the serious discarding which would be necessary in case 

 his ingots were free from blowholes and hence deeply piped. But many 

 intelligent metallurgists have condemned this practice on the ground that the 

 closing of blowholes was impossible, because the gas which they contain must 

 remain ever present during the rolling, even though somewhat compressed. 

 The present investigation seeks to learn by two lines of inquiry whether 

 the gas of the blowholes is qualitatively absorbable, and whether the sides of 

 the blowholes themselves are qualitatively weldable, under the conditions of 

 actual manufacture. Both lines proceed by comparing the metal in slabs 

 cut from the original ingot without rolling with metal cut from a boiler- 

 plate into which that same ingot was rolled, and cut in such a way as to 

 separate and distinguish those parts of the metal in the plate which had orig- 

 inally been porous when in the ingot from those which had originally been 

 compact. 



The first line showed that the enormous differences in density which ex- 

 isted between the porous and the compact parts of the ingot were practically 

 completely obliterated in rolling the metal down into a boiler-plate. In one 

 case the initial difference of 16 per cent in density was completely removed; 



