ON GOLD-ALUMINIUM ALLOYS. 225 



showing the X body in pure ivory white, and the network of eutectic somewhat in 

 relief and golden from rubbed gold. When lightly etched, the X body shows in the 

 form of grey blobs in a white and golden network ; it is possible to entirely remove 

 the gold, although at the risk of etching the X body somewhat deeply. The eutectic 

 is somewhat coarse. With 43'3 atoms of aluminium the unetched surface shows only 

 the usual pure white patches of X in a golden network. When lightly etched with 

 aqua regia a power of 50 shows grey or brown blobs in a white network (fig. 24). 

 With a still higher power the striated appearance of the dark blobs of X is seen. 

 There are no traces of purple. A photograph of this is given in order that the 

 relative amounts of X and the eutectic may be noted. We are practically on the 

 horizontal line of freezing points running through G, yet the X body is very far from 

 being the only substance present.* 



Close to 44 atoms of aluminium we pass the point G, and an upper freezing point 

 appears on the branch GH, so that the alloy has three freezing points. The micro- 

 scope now begins to show crystals of purple in the alloy ; at first very little, but as 

 the percentage of aluminium increases, so does the amount of purple, until at H the 

 alloy is pure purple. The photograph of the alloy with 51*2 atoms of aluminium 

 (fig. 26) shows the isolated rows of crystals of the purple AuAl. 2 . But in all the 

 alloys between -G and H there are three substances. This is well seen in the slowly- 

 cooled alloy containing 45 atoms of aluminium, in which the detail is naturally large. 

 A vertical section of this ingot of alloy contains a small amount of purple in large 

 crystals near the top and on one side of the ingot, but most of the section consists ot 

 long-shaped isolated crystals of the X body in a network -which is shown by a power 

 of 20 diameters to be an exquisite eutectic of the E and X bodies. The photograph, 

 before etching, of a horizontal section of this alloy (fig. 27) shows several dark 

 crystals of the purple AuAl 2 , bordered by white X, and surrounding the large 

 patches of white, the eutectic is seen. It is evident that the uppermost fugitive 

 freezing point is due to the solidification of the purple AuAlj, that the well-marked 

 freezing point at the G temperature marks the moment when purple ceases to form 

 and the crystallisation of X begins. At this moment the still liquid part of the alloy 

 has reached the composition given by G. As the alloy continues to cool, the liquid 

 part traverses all the stages of temperature and composition corresponding to the 

 points on the branch GF, crystals of pure X crystallising during this period. Finally, 

 the residual liquid attains the composition of F, and the eutectic network forms at a 

 constant temperature. The facts that the alloys between H and G are the only ones 

 which plainly show three solid constitutents, and that the solid alloy at G is a 



* January 30, 1900. An alloy whose analysis gave 44'1 atoms of aluminium, which therefore must be close 

 to the transition point G, shows after prolonged etching with bromine water a very fine eutectic surrounding 

 isolated spots of the primary crystallisation of X. A photograph of this is given in fig. 25. Figs. 22 and 

 25 are complementary ; both contain the same eutectic, but the large primary crystals are in fig. 22 of 



and in fig. 25 of the X body, dark through etching. 

 VOL. CXCIV. A. 2 G 



