ON THE CRYSTALLINE STRUCTURE OF METALS. 
299 
crystals had grown up to the line of the weld and there ended quite abruptly. It 
must not, however, be supposed that this weld line was mechanically weak; it 
proved on trial to be quite as difficult to cut or tear the metal along the weld as in 
any other direction. The weld therefore behaved as a true inter-crystalline boundary, 
differing only in the absence of eutectic, and therefore forming a barrier to crystalline 
growth. Fig. 37 shows the appearance of such a weld in section after annealing 
and etching, at a magnification of 30 diameters. The line AB is the weld. As these 
experiments were made on commercial lead, we were prepared to find that, as a 
mere matter of probability, some eutectic would have occasionally found its way into 
the welding surfaces, but this seems to have happened only very rarely. We 
examined some forty specimens, and only in two instances did we see a slight amount 
of crystalline growth crossing the line of the weld. We think that we are justified 
in attributing these rare exceptions to the accidental presence of impurity. 
We then went a step further. If we have in a welding surface an inter-crystalline 
junction which acts as a barrier to crystalline growth owing to the absence of eutectic, 
then if a suitable eutectic be supplied, growth should occur there as elsewhere. 
Our first experiment was to interpose a thin but continuous layer of lead-bismuth 
eutectic between the lead discs in welding ; the specimen was then annealed for 
several days at 200° C.—well above the melting-point of the eutectic—but' on 
examination it was found that the layer of eutectic had persisted as such, and 
allowed no growth to cross it. But in this case the film of eutectic introduced at the 
weld was continuous, and the conditions were therefore analogous to those which hold 
at the boundaries of unstrained crystals, where, as we have pointed out, growth does 
not occur. To make the experiment conclusive it was necessary to have a discon¬ 
tinuous film of eutectic at tine weld. We accordingly tried another experiment, 
introducing only a few small flakes of the same alloy, and after annealing we found 
that crystalline growths had crossed the line of the weld in many places. This 
experiment was repeated many times, various impurities being used, such as the 
lead-tin-cadmium-bismuth eutectic, pure tin, cadmium, bismuth, and mercury. All 
these gave the same result, showing considerable growth across the weld after 
prolonged annealing at 200 C., but the amount of growth observed varied very much. 
Fig. 38 shows a typical example of crystals that have grown across the weld ; the 
line of the weld is still faintly indicated by a discontinuous line, CD, probably repre¬ 
senting an included impurity of a noil-metallic character, around which the crystals 
have grown much as they grow around the slag in wrought-iron. 
In order to remove all doubt as to the action of the impurities which were intro¬ 
duced, and particularly to obviate the possible contention that their action was 
either purely mechanical or else of the nature of that of the “ dirt ” more or less 
requisite in many chemical actions, certain further experiments on welds in lead 
were made. In these, the matter introduced at the weld was— 
2 Q 2 
