356 REPORTS ON THE STATE OF SCIENCE.—1912. 
quite possible that in these experiments traces of grease or of oxide 
hindered perfect contact, as negative results were equally obtained 
when all the filings were of copper or of silver respectively, whereas 
- Spring’s experiments with carefully polished and cleaned surfaces 
showed that union was sufficiently intimate to tear the metal when an 
attempt was made to separate it, showing that molecular interpenetra- 
tion must have occurred to some extent. Spezia supports his views 
by reference to the native copper of Keweenaw Point, Lake Superior, 
which frequently contains inclusions of native silver ®* which have 
not diffused into the surrounding copper in the course of geological 
ages. This author also points out the importance of distinguishing 
between uniform (hydrostatic) pressure and differential pressure causing 
flow. The latter may be expected to have the greater influence. Many 
experiments *? show that great hydrostatic pressure, even when long 
continued, is without influence on the properties of metals. 
It was an important contention of Spring that only those reactions 
were favoured by the application of pressure which proceeded with 
diminution of volume. This principle, so important in relation to 
systems containing a gaseous phase, has received less consideration in 
respect to solid systems. 
The investigation of Roberts-Austen, recorded in the Bakerian 
lecture for 1896, has become classical, owing to the well-designed 
character of the experiments and to the accuracy of the quantitative 
method employed. The case selected for special study was that of 
the diffusion of gold into lead. Discs of an alloy of lead with 5 per 
cent. of gold were accurately faced and pressed into contact with 
cylinders of pure lead. The specimens were then kept at constant 
temperature for several weeks, and the diffusion was measured by 
cutting the lead cylinder into dises and estimating the proportion of 
gold in each disc by the usual method of dry assay. The gold alloy 
being placed at the base of the cylinder, diffusion had to take place 
upward against gravity. The following results were obtained after 
heating at 165° for thirty days :— 
| 
| 
No. of Section 
| | | 
Weight of Section Au é os H 
from Base in Grams | per cent. | Diffusivity | 
/ | 
| | 2 ——N 
1 | 0-64 0-039 | | 
2 | 2-33 0-030 0-005 
3 | 2-02 | | 
0-015 
Further experiments at different temperatures gave the following 
yalues for the diffusivity (expressed in sq. cm. per day) °*:— 
MOOS el ey cel od | ere eC ch ht ree Ae. Jie 0700002 
G52 eat. Bas ONS he Me tay Rees tees he U0 00b 
BOOSTS eee! as a oe ae Be OOS 
O51) TREN RRA: GRR RADY Wet Benen 'h 430080 
550° (in*liquid lead)» is [0! siece tenet. tow 5 bon dD 
> 61 G. T. Jackson, Amer. Jour. Sci., 1845 [i.], 49, 81; Compt. rend., 1845, 20, 593. 
62 See also O. Faust and G. Tammann, Zeitsch. Physikal. Chem., 1910, '75, 108 
6 Phil. Trans., 1896, 187A, 383. 
64 A, Fick, Pogg. Ann., 1855 [ii.J, 94, 59. 
