934 Scientific Intelligence, 



meter was deflected, the direction of deflection changed with the 

 reversal of the primary induction. No satisfactory explanation 

 of this deflection is given. Mr. Trouton, who has worked with 

 Professor Fitzgerald upon the Hertz effect, states that the phe- 

 nomenon observed by MM. Sarasin and de la Rive was noted by 

 Hertz, u It seems as if a vibrator did not send out a line spec- 

 trum," so to speak, but sends out a band spectrum, the center 

 of which is the brightest. The period then, of a vibrator is that 

 belonging to the center of this band. "It is obviously of import- 

 ance lor the central period of the resonator employed to coincide 

 with the velocity of the disturbance, for this is presumably the 

 period given by theory. This is practically done when arrang- 

 ing their relative sizes, so as to obtain greatest intensity." Mr. 

 Trouton does not agree with Cornu in his view of the important 

 bearing of the experiments of MM. Sarasin and de la Rive 

 upon Hertz's work. — Nature, Jan. 30, 1890, p. 295. j. t. 



8. Note on the Absolute Viscosity of Solids, Liquids and 

 Gases ; by C. Barus. (Communicated.) — In case of gases and 

 of mobile liquids, viscosity is expressible in gjcs units with 

 facility, and data are available in great number. This is not true 

 for solids, nor even for viscous liquids. Suitably integrating the 

 elementary transpiration equation, I obtained two methods, by 

 which the viscosity (rj) of glycerine can be measured in a few 

 minutes, and the viscosity of substances of a pitchy consistency 

 (marine glue) in a few months. Thus I found for glycerine 

 7/=5 ; for marine glue, t/=200X10 8 ; for paraffine, 7£>2X10". 

 Even ifiarine glue or hard pitch is therefore at least 20 billion times 

 as viscous as water at the same mean atmospheric temperature. 



For solids I devised two new methods, one of which is compar- 

 ative and somewhat crude, the other direct, dynamic and sharp. 

 I found for hard steel, during the first hour after twisting just 

 within the elastic limits, at mean atmospheric temperature, 

 77=10" to 6X10 17 ; for soft steel, t/=6X10" to 6X10 18 , under the 

 same conditions. Limits must here be given because solid 

 viscosity, even apart from the material operated on, is essentially 

 a function of time, strain and temperature, of a kind which can- 

 not well be tersely specified. Both viscosity and its rate of 

 increase with time under the same given conditions, are magni- 

 tudes which increase together. 



With these data in hand I submit the following g/cs scale of 

 the solidity of the three states of aggregation : I. Gases and 

 vapors, 77=10" 5 to ?; II. Andrew's critical state, ?]■=? to ?; 

 III. Liquids, ?]— ? to 10 2 ; IV. Viscous liquids, 7/=10 a to 10"; 

 V. Solids, ?/= 10" to 10 20 . Since for ether at 30°, 77=9 X 10"* and 

 for oxygen at 0°, 77=2 X10" 4 , it is not improbable that the critical 

 «tate may be definable by a narrow limit of viscosity. Data are 

 wanting. 



The scale proposed indicates the positively astounding range 

 of variation of the chief variable of our material environment, — 

 bearing in mind that throughout the whole of this great interval 

 viscosity nowhere transcends the scope of the senses. 



