46 REPORT—1863. 
modification of bismuth in zine, and of zinc in the allotropic modification of 
bismuth. 
The above hypothesis explains also why bismuth, when alloyed in the 
liquid state, with traces of lead or tin shows a decrement, but on further 
addition an increment in the conducting-power ; for it may be assumed that 
the metals of Class B, when alloyed in the liquid state with traces of 
another, are also converted into an allotropic modification. This is an import- 
ant deduction ; for it shows that the hypothesis will hold good, not only for 
alloys in a solid, but also for alloys in a liquid state. Passing on to another 
series of alloys belonging to this group—namely, the tin-gold series,—it will 
be seen that the curve representing the conducting-power of these alloys has 
not the typical form of this group. If it be examined, it will be found that 
the causes of the irregularities are chemical combinations. Beginning at the 
tin side of it, we find a slow decrement in the conducting-power to the alloy 
Sn, Au, then a gradual increment to Sn, Au, and from this point a slow de- 
crement to SnAu,. Owing to the brittleness and infusibility of the alloys 
between Sn Au, and that containing 2°7 per cent. tin, no alloy within these 
limits could be pressed or drawn into wire. From the alloy containing 2°7 per 
cent. tin to pure gold, the curve becomes a straight line. 
That the alloys at these turning-points may be regarded as chemical 
combinations, is proved by the following facts (see Plate V.):— 
I. That these points represent alloys of definite chemical composition. 
II. That they represent alloys containing large percentages of each metal, 
Sn, Au containing 60 per cent., Sn, Au 37 per cent., and Sn Au, 13 per cent. 
tin. 
III. That the specific gravity of the alloy Sn, Au is almost equal to that 
calculated, whereas Sn, Au expands, and Sn Au, contracts more than any of 
the other tin-gold alloys experimented with. 
IV. That the percentage decrement in conducting-power of these alloys 
between 0° and 100° does not follow the before-mentioned law. 
VY. That tin and gold dissolve in one another with great readiness; for if 
to melted tin gold be added, it dissolves in the tin immediately, and evolves 
so much heat that it is necessary not to add too much at once for fear of 
losing the contents of the crucible. Copper, on the contrary, placed in melted 
tin, takes a long time to dissolve in it. Assuming that some of the solid gold- 
tin alloys are chemical compounds, we then have examples of solid alloys 
which are solidified solutions of a metal (tin) and a chemical compound 
(Sn, Au) in one another, represented by the part of the curve between pure 
tin and Sn, Au—or solidified solutions of two chemical combinations in one 
another (Sn, Au and Sn, Au), represented by that part of the curve between 
Sn, Au and Sn, Au, and between Sn, Au and Sn Au,. 
After what has been already stated, the third group of alloys will require 
yery little to be said with regard to their chemical nature; it is only 
necessary to point out that most of them are only solidified solutions of the 
allotropic modifications of the metals in one another. The curves repre- 
senting the conducting-power of the different series of alloys all have the 
typical form ; and the conducting-power decreases between 0° and 100° ac- 
cording to the theoretical amount. 
The alloys of copper-silver, however, form an exception ; for some of these 
alloys may be regarded as mechanical mixtures. According to Levol, when 
silver and copper are fused and well stirred together*, if allowed to cool 
* Journ. de Pharm. vol. xvii. p. 111. 
