TEMPERATURE ON THE ELECTRIC CONDUCTING-POWER OF ALLOTS. 199 
it, on being rapidly cooled, was 2*0683 at 12°*8, showing again an increment. On the 
fourth morning, at 10°*5, it was 2*0275, and after being kept for 5 hours at 100° and 
rapidly cooled, it was 2*0837 at 11°*7. There must be, therefore, with some of the 
bismuth alloys, some disturbing cause, which may act either in the one direction or the 
other, for on investigating the Bi Sn 8 series the opposite effect is produced. This dis- 
turbing cause may be so great that, as in the case of Bi 4 Pb, it appears as if the conduct- 
ing-power increases with an increase of temperature. Other alloys of bismuth and lead, 
rich in bismuth, give the same results. As yet, we have not had time to investigate 
thoroughly this curious property of the bismuth alloys ; we hope, however, to be able 
shortly to do so, as well as explain the reason of these remarkable exceptions to the law, 
that the conducting-power of alloys decreases with an increase of temperature. 
3. Respecting the parts the metals take in the conducting-power of their alloys, we 
are at present unable to give any definite data ; we did hope at one time to have deduced 
them with the help of the results in this memoir. It is scarcely necessary to point out 
that in many cases the composition of the alloy may be deduced from its conducting- 
power in the same manner as it may be from its specific gravity ; for as 
Po : Pc : : X 100 o : X 100 o , ( 1 ) 
then if Po and X 100 o be determined, Pc being known (=29*307), X' 100O can be calculated, 
and from it the relative amounts of the component metals for 
, a?c + (100— x)d 
Aoo°— 100 
where # represents the volumes per cent, of the one metal, ( 100 —#) those of the other, 
and c and d their conducting-power at 100 °. 
Thus the observed conducting-power of the gold-silver alloy at 100° is 14*05, and its 
percentage decrement 6*49, 
,, 14*05x29*307 fl0 iC 
* 100 °= 6^9 =63*45, 
therefore 
co AC . 7 l*56*a? + 55*90*(100 — ar) 
bd*40— 100 
755=15*66#, 
48*20=#. 
The amount of silver in the alloy was 47*92 volumes per cent. Again, the platinum- 
silver alloy, containing 19*65 volumes per cent, platinum, conducts at 100° 6*49, and 
loses in conducting-power between 0° and 100° 3*10 per cent. ; calculating in the same 
manner the percentage amount of silver, we find it equal to 82*67 instead of 80*35. 
The values deduced for the percentage amounts only agree in a few cases well with 
those found by analysis, as slight errors in the determinations materially affect them ; 
for instance, if the conducting-power of the gold-silver alloy were equal to 14*20 at 100 ° 
* Observed conducting-power of silver and gold at 100° (Philosophical Transactions, 1862, p. 24). 
2 E 
MDCCCLXIV. 
