ON THE CRYSTALLINE STRUCTURE <>K MKTALS. 



diameters. The characteristic of the pits is that they are similar ami similarly 

 oriented over any one grain, hut on pawing from one grain to another, across a 

 boundary, the orientation of the pit is found to have changed. Good examples of 

 this arc seen in tig. 7, where the boundaries are widened out into channels through 

 the presence there of air or gas in the manner described al*>\e. This characteristic 

 if the air-pits is, of course, in complete conformity with the view already stated of 

 the crystalline structure of metals. Additional photographs of air-pits in cadmium 

 are shown in another connection in tigs. 2G, 27, and 28 (Plates 22 and 23). 



Examination of the air-pits in cadmium shows that the forms may be accounted 

 for as sections of hexagonal prisms. It may be concluded that each crystalline 

 element of which the grains of cadmium are built up is a hexagonal prism with 

 plane l>ase.* 



These air-pits are not very readily develojnjd, and the precise conditions which 

 determine their appearance are not easily specified. Many specimens of the metal 

 mav lie cast without obtaining them. That they are not, however, peculiar to 

 i-admiiim is certain, for we have found them also in tin and in sine. Fig. 10 shows 

 them in a glass-cast surface of tin. 



The same photograph illustrates another interesting feature. The surface of the 

 tin is seen to 1*5 covered with a multitude of small dark crystals irregularly disposed. 

 These are evidently inclusions of some foreign matter, which we conjecture to have 

 Ixvn sulphide, because it was olerved that they appeared in large numbers after 

 the metal had, during melting, been directly exposed to a gas flame. The foreign 

 crystals have no definite orientation, and are quite independent of the orientation of 

 tin metal within the grains in which they are emliedded. The crystallisation of the 

 met a I has proceeded around them without check or disturbance, just as in iron 

 containing slag the growth of each crystalline grain ignores the presence of that 

 impurity. It is well known that a slag band is often seen running right through a 

 crystalline grain without affecting the uniformity of orientation of the elements of 

 which the grain is built up. 



Although it must be admitted that a really good development of geometrical 

 bubbles such as those shown in the photographs is exceptional and cannot as yet be 

 reproduced at will, the authors have observed that nearly all bubbles or blow-holes 

 linn id mi surfaces prepared in this way show a distinct tendency to geometrical 

 shapes. A trnlv round bubble, is rarely or never found, and even the larger bubbles 

 ut'teii show a multitude of distinct facets reflecting light at different angles. 



The occurrence of such geometrical pits in surfaces of metals that have never l)een 

 polished or etched may l>e taken as very strong evidence in support of the view that 



* It may In- .i<l<lel in this romuTtioii tli.it liKHKKNs inu.u k> mi the iM-iaienry of nix-sided foinw in 

 the iMilygnii.il Ixmindarios nf the crystalline gtains of cadmium. The form of lioiuidiiry is, however, of 

 little *o! vice in dotorminini: the character of the crysUlliaations within the grain. (Sec his work ' 

 Miiruiicopuche Gefuge <lor Metallu mid l.egiennigeii, Leipzig, lS'J4.) 



