184 
DE. A. MATTHIESSEN ON THE SPECIFIC OEAVITT OF ALLOTS. 
Table II. (continued.) 
Lead-Mercury Series (determined by A. Matthiessen). 
Mean of 
Calculated specific 
Katio of volumes, 
Alloy. 
Specific gravity 
found. 
T. 
gravity, from 
volume. 
v+v,. 
V 
Pb„Hff 
11*979 
12-484 
15-9 
15-7 
15-5 
12-008 
1-0024 
Pb Hg 
12-358 
0-9899 
0-9937 
Pb Hi, 
12-815 
12-734 
Lead-Gold Series (determined by A. Matthiessen). 
Pbgo Au 
11-841 
23*3 
11-794 
0-9961 
Pbi„ Au 
12-274 
19*4 
12-171 
0-9916 
Pbg Au 
12-445 
21*6 
12-346 
0-9920 
Pbg Au 
12-737 
21-3 
12-618 
0-9906 
Pb^ Au 
13-306 
22*1 
13103 
0-9840 
Pbg Au 
14-466 
14-3 
14-210 
0-9823 
Pb" Au 
15-603 
14-5 
15-546 
0-9963 
Pb Aug 
17*013 
14-3 
16-832 
0-9894 
From Table II., which gives in the fourth column the ratio of the sum of the volumes 
of the two metals to the volume of the alloy, it appears that the alloys of antimony are 
generally greater in volume than the aggregate of the constituent metals (expand), 
while those of bismuth, silver, gold, and mercury, generally are less (contract) ; and we 
find that the maximum expansion or contraction generally takes place about that point 
when the alloy contains equal volumes of each metal. 
The gold-tin and gold-lead alloys are all very brittle, except those very rich in lead or 
tin : — Sn Au^ to Sng Au are not at all crystalline, and have a glassy fracture ; Sn^ Au 
begins to show a crystalline structure, and has a crystalline fracture ; Sug Au to Sujoo Au 
are exceedingly crystalline ; and Sng Au to Siiig Au all show a fracture like the cleavage 
plane of a crystal. The gold-lead alloys appear all to be crystalline, that is, their sm-- 
face is very much so, but their fracture is glassy. The following alloys expand greatly 
on cooling, so much so, that the liquid metal breaks through the crust, forming large 
or small globules, viz. all those of bismuth-antimony, bismuth-gold, and bismuth-lead, 
which were experimented on ; those of bismuth-tin, from Bigio Sn to Big Sn, the rest of 
the series very slightly; and bismuth-lead, viz. Bko Pb to B^ Pb (Big Pb shghtly), the 
rest apparently not at all. Of the bismuth-cadmium series, Bi Cdg and Bi Cd4 expand 
very slightly, the rest not at all. No concordant results could be obtained with any 
zinc-tin or zinc-cadmium alloys, on account of their very crystalline structure. 
In conclusion, my best thanks are due to Dr. M. Holzmann, Mr. C. Long, and 
Mr. M. Caett for their assistance in carrying out the foregoing determinations. 
