ON GOLD-ALUMINIUM ALLOYS. 221 



are a rough brilliant silvery white. Moreover they are rounded, and have no 

 u|>|>":ir;iiirr <>t' Living Ix-cii sqU6M0d in!., iL- intrrsti<vs >f rryst!ils uhv.-uh Conned. 

 Si. that it is liiird ti xiv which of the two ingredients crystallised first ; but the higher 

 the power one uses the more disposed one is to say that the white spots were the 

 primary crystals. With a power of 1000 or 1500 the ground shows a detail in it of 

 small spots like minute grains of wheat, some of which are white. Probably this 

 alloy represents the eutectic as fairly as that with 21*6 atoms. 



With 23 '1 atoms (fig. 10) the blobs of white are very uniformly distributed in a 

 dark field, filling about one-third of it. They are a brilliant silvery pitted white. 

 At 2 3 '6 atoms the white blol have increased in area, and form a beautiful object 

 when examined with a low power. At 25'9 they fill two-thirds of the field (fig. 11), 

 and are a remarkable example of rectangular crystallisation. At 26*5 atoms the 

 white blobs have grown into elongated polygons, only separated by a broken network 

 of dark lines ; in other words, the D body is nearly pure, the slowly -cooled 26*1 atom 

 alloy (fig. 12) gives a good idea of these.* 



This slowly-cooled alloy shows between the masses of white a network of golden- 

 brown, which broadens at the angles. These broader patches of mother-substance 

 under a j>ower of 700 diameters prove to be full of minute polygons, bounded by 

 bright, and often white, lines. This we presume is the structure of the eutectic 

 corresponding to the state C. This mother-substance was formed after surfusion. 

 The pattern is found on every patch of the mother-substance, although the boundary 

 lines of the polygons are often incomplete (fig. 13). The section at 27 '2 atoms 

 (fig. 14) is very remarkable; it consists, like the preceding one, of polygons separated 

 by lines which are now fine and broken ; but the polygons are in groups, some of 

 the groups being of a greyish-white rough substance, while other groups are a very 

 smooth brilliant ivory-white. With oblique illumination the rough polygons show 

 the effect of changing from bright to dark, as the stage is rotated through a right 

 angle. Examination with a high power, however, forces one to the conclusion that 

 the rough and smooth patches are of essentially the same material. The photograph 

 of this section, with a magnification of 50 diameters and normal illumination 

 (fig. 15), shows how very misleading a photographic reproduction may be. It does 

 not give the impression of being the homogeneous substance that it really is. Even 

 with a high power one cannot say to a certainty whether the interstices between 

 the polygons are filled with the same eutectic as that found in the 26 '1 atom alloy 

 or with a new one : this point, if settled, would determine the formula of the 

 Dbody. 



At 28 '3 atoms we have a complete change ; with oblique illumination, as in figure 

 (fig. 16), one MM scanty groups of primary crystals of a new substance on a uniform 

 ground. Some of the new crystals, examined with a power of 1000, show traces of 



* This photograph was taken with oblique illumination, ami the white and dark patches are of the same 

 material differently orientated. 





