38 Dr. Alder Wright and Mr. C. Thompson. [May 8 r 



present increases with the silver present, giving sensibly the same 

 curve in each series. 



The ahove results obviously lead to the conclusion that silver and 

 zinc form at least two definite compounds, viz., AgZn 5 and Ag 4 Zn 5 : 

 the first of the^e can dissolve lead more freely than can either 

 pure zinc or the second ; so that when the silver and zinc are 

 contained in the lighter alloy in exactly the proportion AgZn 5r 

 the lead dissolved is a maximum. Moreover, it results conversely 

 that lead can dissolve AgZn 5 more freely then either pure zinc or 

 Ag 4 Zn 5 , so that, when circumstances favour the production of the 

 first compound, the zinc dissolved in the heavier alloy is notably 

 increased. It would seem that the compound Ag 4 Zn 5 is so much 

 less soluble than AgZn 5 that lead saturated therewith contains 

 actually less zinc than when saturated with pure zinc ; for the 

 experiments with bismuth-silver-zinc alloys show that the zinc con- 

 tained in the heavier alloys (calculated per unit of bismuth) 

 gradually diminishes to a minimum as the silver present increases, 

 and then regularly increases again, the position of this minimum 

 being sensibly that where the silver and zinc present are in the 

 proportion Ag 4 Zn 5 . This point will be discussed in a future paper. 



When more silver is present relatively to zinc than corresponds 

 with Ag 4 Zn 5 , then this compound is dissolved by lead, and conversely 

 can itself dissolve lead, the more freely the more surplus silver is pre- 

 sent ; so that a more or less rapid rise in each solubility curve is 

 observable when the silver present exceeds that requisite to form 

 Ag 4 Zn 5 with the zinc present. 



The difference between the lighter alloy curves obtained in Series I 

 and II leads to the remarkable conclusion that, when the compound 

 AgZn 5 (containing dissolved lead) is kept fused for some hours, it 

 tends to break up (presumably into free zinc and Ag 4 Zn 5 ), and thereby 

 to throw out of solution more or less of the dissolved lead ; so that 

 after 24 hours' fusion less lead is present in the lighter alloy than 

 after only eight hours' fusion. Obviously, if it were practicable to 

 effect this decomposition absolutely completely, the lighter alloy 

 formed would be simply a mixture of Ag 4 Zn 5 and more or less 

 surplus zinc, saturated with lead; and hence the curve traced out 

 with silver as abscissa and lead as ordinate should rise regularly 

 from the origin up to the point whei-e Ag 4 Zn 5 without surplus zinc is 

 present. 



Owing to the volatility of zinc, we were unable to carry out any 

 experiments with a view to tracing out such a regular curve, by 

 maintaining the alloys in a fused state for lengthened periods of 

 time (several days) ; but we succeeded in effecting the same object 

 by the simple device of eliminating the lead that separated (along 

 with some silver and zinc) on keeping the solution of lead in AgZn 5 



