PHYSICS. 337 



not occur if it were practically possible to have the axis of the wire 

 pass accurately tlirough the center of mass of the vibrator. Another 

 error arises from the fact that in a wire recently suspended the tor- 

 sional vibration perio<l is always slightly greater than when the 

 wire has been long used and frequently oscillated (Phil. Mag., Novem- 

 ber, 1S8G, V, XXII, 414; Nature, July, 1886, xxxiv, 283.) 



Barns and Stroubal have continued in the laboratory of the U. S. 

 Geological Survey their researches upon the effect of temper upon the 

 structure of glass and steel, and have published: "A note on tbe 

 structure of tempered steel," "Strain effect of sudden cooling exhibited 

 by glass and steel" (two papers), "Note on the hydro-electric effect of 

 temper in case of steel," and on " The viscosity of steel and its rela- 

 tions to temi)er." (Am. J. Sci., May, June, 1886, III, xxi, 38C, 439; 

 September, October, December, 1886, III, xxxii, 181, 276, 444.) 



Lehmaun has observed certain remarkable spontaneous changes of 

 form in solid crystalline bodies, produced evidently by their interior 

 forces. These changes were observed with the microscope and in the 

 following substances : Quinohydrodicarbouic ether, protocatechic acid 

 and ammonium chloride. The hist of these substances, for example 

 dissolved by the aid of heat, in aniline thickened with a little resin 

 crystallized in leaflets which are parallelograms of 44°. But on grad 

 ually lowering the temperature, these crystals change into others hav 

 ing a pale green color and angles of 60°. This change may commence 

 at different points in the same crystal even, and so twist them, the 

 movement at the ends of the crystal having force enough to dis- 

 place the entire mass. The other bodies mentioned show the same 

 property. (Wied. Ann., xxv, 173; J. Phys., November, 1886, II, v, 479.) 



Fonque and Levy have experimented to determine the velocity with 

 which vibrations are propagated through the ground. In their pre- 

 liminary experiments the vibration was i)roduced by the fall of a steam- 

 hammer of 100 tons, at theCreusot works. The transmitted vibrations 

 were observed b^^ means of a mercury surface; their arrival, as well as 

 the instant of fall, electrically transmitted, being registered by hand on 

 a revolving cylinder. At 1,200 meters distance the jar produced by the 

 blow ceases to be perceptible directly to the senses but is distinctly 

 seen in the mercury. In the permian sandstone of Creusot, a velocity 

 of 1,200 meters per second was recorded, in a direction parallel to the 

 strata and 1,050 meters at right angles to their direction. The dura- 

 tion of the disturbance w^as nearly a second. At Meudon, where the 

 terrace is formed of a thick layer of Fontainebleau sand, the propaga- 

 tion of the vibration is much less rai)id, 320 to 360 meters only per 

 second. But the duration of the disturbance was much longer, being 

 about 5 seconds at 500 meters and 3^ seconds at 250 meters. Subse- 

 quently the authors employed an apparatus constructed by Breguet, in 

 which a sensitive gelatino bromide plate was made to turn about a beam 

 of light reflected from the mercury bath. A shutter opens automati- 

 H. Mis. 600 22 



