Strains Mechanically Applied, 



Table II. — Electrical Effect of Motional Annealing. 

 Glass-hard Steel, 2/9 = '081 centim. 



159 



Twist 



^0. 



T. 



''t. 



^^^. 



Twist 

 No. 



r. 



''t. 



^'t. 



0. 







39550 



+ 60 



5. 



-•102 



39560 



+60 



1. 



-•099 



39380 



-120 



6. 



+ •102 



39530 



+30 



2. 



+•098 



39350 



-150 



7. 



-■102 



39500 



+tJ 



3. 



-•102 



39570 



+ 70 



8. 



+ •102 



39510 



+ 10 



4. 



+•102 



39430 



- 70 











6. The results of Table I. are in excellent accord with my 

 earlier data ; and the oscillatory march of the viscous incre- 

 ment can be represented in the same way. If the tangents 

 be constructed at the same time-point, in each of the curves 

 of Table I., there appears to be some similarity between the 

 march of these results (A(0 — 0')/t) and the corresponding- 

 march of rt in Table II. But interpreted by the data of § 2, 

 this similarity is only qualitative in kind. In other words, 

 whereas the increment of viscosity due to successive alterna- 

 tions of twist is decidedly greater than one half of the incre- 

 ment of viscosity due to annealing glass-hard steel at 100°, it 

 appears that the electrical effect in the first instance (motional 

 anneaUng) is practically negligible in comparison with the 

 electrical efPect of thermal anneaKng. In Table II. the total 

 interval of variation of A?'< is about one half per cent, of n ; on 

 the other hand, the variation of n due to annealing at 100° is 

 from 10 per cent, upwards. It follows that in glass-hard steel 

 there are two distinct ways in which viscosity may be 

 appreciably increased, a result corroborating § § 3,4. Again, 

 if the possibilities of viscous motion are to be fully given, it 

 is essential to postulate groups of atoms, as well as the some- 

 what less definite groups of molecules, both varying in de- 

 grees of stability from point to point of the sohd mass. 

 Hence slight positional change of the elements of the atomic 

 configurations, or of the molecular configurations, due either 

 to stress not exceeding a critical value or to mere secular 

 subsidence, must in general involve an augmentation of the 

 viscous quahty. 



7. Having arrived at this result I desire to inquire some- 

 what more in detail into the viscous relations of the motional 



