PROFESSOR W. C. ROBERTS-AUSTEN ON THE DIFFUSION OF METALS. 397 
Table D.—Tube 4. Platinum in Lead. 
1 
1 
Number 
of section 
of diffu¬ 
sion tube. 
|_ 
2 
Weight in grammes 
of metal from each 
section— 
3 
Per¬ 
centage of 
platinum 
in each 
section. 
4 
Corrected 
for 
density to 
equal 
volumes. 
5 
Divided 
by 
00048556. 
6 
Theoretical numbers 
for— 
7 
h 
2 Wia 
(by inter¬ 
polation). 
lead- 
platinum 
alloy. 
A 
platinum. 
2 yar 0 ' 18 - 
h 
—t==019. 
2 y/H 
i 
317 
0-2795 
8-817 
9'509 
1958 
1927 
2023 
0-183 
2 
3-85 
0-3435 
8-922 
9-613 
1980 
1817 
1895 
0-201 
3 
3-49 
0-2690 
7-707 
8-235 
1696 
1614 
1664 
0-196 
4 
4-47 
0-2761 
6-177 
6-517 
1344 
1352 
1368 
0-175 
5 
2-62 
0-1280 
4-885 
5-093 
1049 
1067 
1054 
0-194 
6 
412 
0-1471 
3-571 
3-682 
758 
793 
760 
0-191 
7 
362 
0-0891 
2-462 
2-514 
517 
556 
513 
0-189 
8 
3-40 
0-0491 
1"444 
1-462 
301 
368 
324 
0-195 
9 
3-73 
0-0342 
•917 
0-925 
190 
230 
193 
0191 
10 
3-57 
0-0169 
•473 
0-475 
102 
137 
108 
0191 
11 
3-26 
0-0099 
•304 
0-305 
62 
83 
60 
0-189 
12 
3-28 
0-0074 
•226 
0-226 
46 
57 
39 
0-186 
Sum 
48-556 
Mean . 
0-190 
h = l - 27 centims. cold = l - 303 centims. at 492°, therefore ht — 11'76. 
t = 696 days, therefore Jc = 169 sq. centims. per diem. 
In figs. 3, 4, the diffusions which are given in Tables B and C are plotted in thick 
lines, with distance and concentration as coordinates. The curves in dotted lines 
give theoretical distributions for two values of which nearly agree with the 
h 
experimental results, and between which the true value of 
2 \/ let 
appears to lie. 
In these curves the horizontal length of the figure, which represents the height of 
the diffusion-tube, is divided by the sectional lines into as many equal parts as there 
are sections of the tube. The concentration in each section was marked by a hori¬ 
zontal pencil line, and the continuous curve shown in fig. 2, and in figs. 3, 4, was 
drawn through these lines in such a way that the area included in each section still 
represented the average concentration in it. This method was adopted to avoid the 
slight error which would have been introduced if the average concentration had been 
plotted at the mid-point of each section. This mode of plotting, fig. 2, in which the 
area a — a , b = h ', and c = c, renders it impossible to specify the points through 
which the curve has been drawn, but it may be remembered that the average position 
of the curved line in each section represents the result of one analysis. 
The dotted lines are plotted from figures calculated from Stefan’s tables for two 
