PHYSICS. 439 



explanation is not easy without admitting a certain amount of plasticity 

 in ice. {Phil Mag., August, 1884, Y, xviii, 120.) 



Kroucbkoll has-uoticed that perfectly clean platinum readily amal- 

 gamates with mercury, but that aluminum and iron do not. On at- 

 tempting to unite mercury with a scraped surface of aluminum, this 

 metal oxidizes with great facility. If, however, the aluminum be made 

 the negative electrode in acidulated water, this metal becomes strongly 

 charged with hydrogen ; and now if it be dipped into mercury at the 

 bottom of the vessel, it unites with it instantly. On exposure to the air 

 it is tarnished at once. A wire of iron acts in the same way. The au- 

 thor suggests that the action of sodium in facilitating amalgamation is 

 due to its reducing action. {J. Phys., March, 1884, II, iii, lo9.) 



2. Of Liquids. 



E. Wiedemann has calculated, from certain experiments made by Ber- 

 thelot, that the force necessary to separate two adjacent layers of water 

 cannot be less than 55 atmospheres. (J. Phys., December, 1884, II, iii, 

 555.) 



Eeinold and Eiicker have continued their studies on liquid films, and 

 have observed that the disturbing action of the electric current, previ- 

 ously noticed, is due to the transference of liquid by this current in its 

 own direction, thus thinning the film, or the reverse, according to the di- 

 rection. By forming a plane film between two horizontal wires, illumi- 

 nating it by the calcium light and projecting its image on a screen, the 

 authors showed the motions of the bands of color in the direction of the 

 current. {Mature, December, 1884, xxxi, 186.) 



Schiff has determined with care the capillary constant of a large 

 number of liquids at their boiling points. For this purpose he used two 

 capillary glass tubes, one 1.3""" in the bore, the other half as much, con- 

 nected so as to form the legs of a U tube. This tube, after being filled 

 with the experimental liquid, is hung in a wider vessel, at the bottom of 

 which a little of the liquid is kept boiling. From the difference in level 

 of the liquid in these two connected capillary tubes, as measured at a 

 temperature which must be very nearly the boiling point, the surface 

 tension at that temperature is readily deduced. To this surface ten- 

 sion, in milligrams per millimeter, divided by the' relative molecular 

 weight (and multiplied for convenience by 1,000), Schiff assigns the 

 symbol N. He illustrates its significance by pointing out that in the 

 case of a capillary elevation against a vertical wall wetted by the liquid, 

 it represents the number of molecules raised above the free surface per 

 unit length of the wall. The results obtained show that not only is 

 this number N the same for isomeric substances, but that it is often the 

 same for liquids of very different chemical constitution, and that it is 

 possible to replace a certain number of atoms of one kind by a certain 

 number of another kind, without producing in the value of Nan alter- 

 ation which comes within the limit of precise observation. In these 



