26 Barus and Strouhal— Viscosity of Steel. 
here that the torsion imparted was not sufficiently large to pro- 
duce marked permanent set. If we express its intensity by 
2 
oe 
ooo 
Ae 
ose 
| 
(0 4E)SE an and introduce the constants of the apparatus 
g=2, L=380™, 2e=0-082™, 
we find that the moment of the applied torsion couple did not 
exceed 0°5 kg. on centimeter of arm. (Cf. remark on stress 
value, p. 30.) If following Sir William Thomson,* we agree 
that “the molecular friction in elastic solids may properly be 
called viscosity of solids,” then our deduction may be stated, 
the molecular friction in case of steel vs greater in proportion as the 
metal is softer.+ Examples of such relations in divers sub- 
stances are not unknown. Hard steel as regards viscosity and 
*Thomson: l.c., or Thomson and Tait, Natural Philos., I], p. 303, 1883. Our 
present conception of the viscosity of liquids, as well as the hypotheticated 
proportionality of frictional resistance and velocity were introduced by Newton 
(Principia, L. II, Sect. ix, ‘‘resistentia que oritur ex defectu lubricitatis”). 
+ Degrees of thermo-electric hardness are here specially in place. Cf. U.&. 
Geolog. Survey, Bull. 14, p. 65. 
