NATUHAL rniLOscrnY. 



1C.3 



first and in a 2:reater degroe at tho nojrative pole. An pxamina- 

 tion with a microscope revealed at the broken surface a coni|)lc;t(3 

 molecular change, a crystalline structure haxing taken the phico 

 of the fibrous. He states that, having entered upon some ex- 

 tended experiments in connectioji -with submarine explosions by 

 means of the voltaic current, ho was frequently annoyed by the 

 breaking of one of the wires, and in all cases found tho structure 

 at the broken part crystalline. From these facts he infers that 

 intense currents passed through su!)marine cables must eventually 

 deteriorate them, and counsels their avoidance. The frequent re- 

 versal of the current, in regard to direction, lessens or entirely 

 prevents the molecular change in the wire. — Scientific American. 



REMARKABLE RELATION BETWEEN THE MAGNETISM OF SOME 

 METALS AND THEIR ATOMIC AND SPECIFIC WEIGHTS. BY P. 

 H. VAN DER WEYDE, M. D. 



When we divide the specific gravity of the different metals re- 

 spectively into their atomic weights, we obtain quotients which 

 indicate, not directly, but relatively, the distance of their atoms 

 (upon tlie supposition that the atomic weights indicate really the 

 relative weights of their atoms, which is only probable, but not 

 proved). Comparing those quotients in the subjoined table we 

 iiud the follo\vin<j: remarkable results : — 



Spec. Grav. Atom. Wt. 



Quotient. 

 3.53 remains paramagnetic at white heat. 

 3.59 is only magnetic below bright red do. 



dark 

 600^ F. 

 40 F. 



<( 





1st. The 5 magnetic metals have all quotients below 4. 



2d. The so-called non-magnetic metals have all quotients 

 aoove 4. 



There is, however, one exception to this rule in the case of cop- 

 per, of which the respective specific and atomic weights are 8.8 

 and 31.7, of which the quotient is 3.602; but then it is probable 



