412 THE LIFE OF MATTER. 



))l(\s, having- proviousl\' placed at the ))()ttoin of the cavity some frag- 

 ments of a light substance, like cork. The piece of wax is thus between 

 a light body below and a heavy body above. If we wait a few days 

 this order is reversed — the wax has filled the cavity by conforming to 

 it; the cork has passed through the wax and appears upon the surface, 

 while the stones are at the bottom. We have here the celebrated 

 experiment of the flask with the three elements, in which are seen the 

 liquids mercury, oil, and water superposed in the order of tluMr 

 densit}", demonstrated in this case by means of solid bodies. 



Influence of dlfiision. — Ditt'usion, which disseminates li({uids 

 throughout each other, ma}^ also cause solids to pass through other 

 solids. W. Roberts Austen places a little cylinder of lead upon a disk 

 of gold and keeps the whole at the temperature of boiling water. At 

 this temperature both metals are perfectly solid, since gold does not 

 melt up to 1,200 V. nor lead up to SSO*-^. Still, after this contact has 

 been prolonged for a month and a half, analysis shows that the gold 

 has become dili'used to the top of the cylinder of lead. 



Inflnenee of electrolysis. — Electrolysis ofll^'ers another means of trans- 

 portation no less remarkable. By its means we ma}^ force metals, 

 such as sodium or lithium, to pass through glass walls. The experi- 

 ment may be performed in the method indicated by M. Charles Guil- 

 laume. A. glass Inilb containing mercury is placed in a bath of .sodium 

 amalgam and a current is then made to pass from within outward. 

 After some time it can be shown that the metal has penetrated the 

 wall of the bidb and become dissolved in the interior. 



Iiiflaence of riiecJKniical jyressure. — Mechanical pressure is also capa- 

 ble of causing one metal to pass into another. We need not recall the 

 well-known experiment of Cailletet, who, by emplo^nng considerable 

 pressure, caused mercury to sweat through a block of iron. In a 

 more simple manner W. Spring showed that a disk of copper could be 

 welded to a disk of tin by pressing them strongly one against the 

 other. There is formed, up to a certain distance from the surfaces of 

 co'ntact, a veritable alio}'; there is a layer of bronze of a certain thick- 

 ness which unites the two metals, and this could not take place did not 

 the i)ai'ticl(>s ol" both metals mutually intei'peneti'ate. 



4. INTKSTINAI. ACTIVITY OF ALLOYS. 



fStnictare of alloys. — Metallic alloys hiive a reiuarkalde structure 

 which is essentially mobile and which we have now begun to und(M'- 

 stand, thanks to the employnuMit of the microscope. Microscopical 

 examination justifies to a certain degree tlu^ supposition of Coulomb. 

 That illustrious physicist explained th(» physical properties of metals 

 by imagining that they were formed of two kinds of elements — inte- 

 gral particles, to which the metal ow(\s its ])roperties of elasticit>% and 

 a cem(>nt which binds the ])articles and to which it owes its coherence. 



