Viscous Traction of Lead and Tin Alloys. 199 



arises whether the introduction of the idea of a coefficient 

 can be really helpful in the case of these alloys and such like 

 substances. It must, however, be considered that, as in the 

 case of permeability in magnetism, a coefficient, though a 

 function of the force applied, may yet render assistance in 

 unravelling the intricacies of complicated phenomena. 



Particular mention must be made of the behaviour of those 

 alloys containing more than 10 per cent, of tin. It may be 

 at once pointed out that the curves for the alloys 98 Pb and 

 90 Pb and for the pure metal Pb are quite comparable in 

 general contour. With increasing load there is the same 

 approximation to the quasi-liquid state, the viscosity falls to 

 the same limiting value, but for the alloys 85 Pb and 17 Pb 

 much the same mobility is observable, no matter what load 

 is operating. 



As mentioned already, it must be remembered that the 

 application of the term " coefficient of viscosity " to these 

 cases may be somewhat misleading. It is a moot point 

 whether such a term can be applied fairly with its usual 

 understood connotation, viz. that it is a constant. The re- 

 ciprocal expression "coefficient of mobility" might perhaps 

 be better as not presenting an idea already stereotyped by 

 usage. The chief difficulty is the determination of the 

 limiting stress, before which no displacement other than that 

 within the elastic limit occurs. That there is such a limit 

 even for liquids appears certain. Consequently the method 

 used in this paper, of assuming coefficients of viscosity to be 

 comparable when determined with the same stressing force, 

 is to be regarded only as an approximation. This will be 

 quite evident when fig. 3 (p. 200) is considered. Here are 

 plotted rates of extension against loads, and it is clear that 

 comparable rates of flow can be determined only when the 

 initial rate is neglected ; in other words, when the intercept 

 of th^ tangent on the load axis should be subtracted from the 

 total load. Up to the present the data at hand are not 

 sufficient for this correction to be made in every case. 



General ( 'onclusions. 



1. When a wire composed of pure lead or tin or alloys of 

 these metals is stretched beyond its elastic limit (in these 

 cases this limit is exceeded by a very small load) the rate of 

 extension proceeds quite steadily, and the material undergoes 

 a viscous flow. From this rate of flow a coefficient of viscous 

 traction may be deduced for any given load. 



