40 Dr. Alder Wright and Mr. C. Thompson. [May 8, 



the proportions representing those due to any given point on the 

 earlier portion of curve No. 1, fig. 5, and kept in a state of fusion 

 for eight hours or more, just the same effect was produced; i.e., a 

 small quantity of heavy alloy subsided, leaving a lighter alloy, the 

 composition of which was not altered materially by cutting off the 

 separated heavy alloy and fusing again for eight hours more. 



Uniting together all the observations thus made, we obtained the 

 following series of figures, representing the limiting compositions of 

 the normal lighter alloys formed, i.e., the compositions below which 

 no further reduction in lead percentage could be obtained by keeping 

 in a fused state for several hours longer : 



Series III. Limiting Composition of Lighter Alloys. 



Silver. Lead. Zinc. 



11-51 2-37 86-12 



17-27 2-76 79-97 



24-49 2-65 72-86 



28-11 2-68 69-21 



37-54 3-14 59-32 



46-23 3-56 50-21 



52-78 3-60 43-62 



61-12 6-40 32-48 



64-58 10-00 25-42 



66-25 10-84 22-91 



On plotting these figures it is obvious that they give a curve 

 (No. 3, fig. 5, dotted line) sensibly identical as regards its latter 

 part with the corresponding portions of the curves obtained from 

 Series I and II, the differences iu no case being greater than the 

 amounts that may reasonably be ascribed to unavoidable differences 

 in the average temperature and in the small temperature fluctuations 

 in the different experiments. But the first part of the curve is 

 wholly different, the rise to a maximum at AgZn 5 and subsequent 

 fall being wholly eliminated, and a regular ascent being visible 

 instead, precisely as should be the case were the compound AgZn 5 

 wholly decomposed in each case, so that no excess of lead could be 

 taken into solution over and above that due to the normal solubility 

 in the particulai' mixture of silver and zinc (or of Ag 4 Zn 5 and surplus 

 zinc or silver) present. In similar fa.sh.ion, we found that if the 

 bottom portions of the compound ingots obtained after eight hours' 

 fusion, and containing unduly high percentages of zinc, were cut off, 

 and fused again separately for another period of eight hours, a small 

 quantity of lighter alloy rose to the top, whilst the heavier alloy 

 formed after the second fusion gave, on analysis, percentages of zinc 

 and silver yielding a perfectly regular curve, instead of the abnormal 



