90 C. Barus — Isothermals, Tsopiestics and 



rate in the lapse of time, is the chief feature. The case, how- 

 ever, is much more complex, for the restitution of volume is 

 greatest in the axis of cylinder where the flow is a maximum, 

 and it is accompanied by a series of distortions of the kind 

 given in figure 2. Here G shows the shape of the cylinder 

 or button immediately after cutting it off of the column H, 

 surrounded by the steel transpiration tube. After 24 hours, 



when both were left free from pres- 



T f^5) ^ sure, G passed into the form /, and 



*• |\\jl .ZTinto K. The emergence of the 



T k&. cone begins visibly, soon after cut- 



1-4 ting. After long waiting, a cavity 



sometimes dimples the top of the 



button as shown at L. I have also 



Figure 2.-Sectional diagrams, obtained cup-shaped deformations, 

 showing the elastico-viscous clefor- ,., ,, mi -i ■ i _ > 



mations of exuded cylinders of the jlke M. ihe height of the project- 



charge C : Fig. i, external pressure ing cone bears no obvious relation 

 tein £ zero - to the length of the cylinder. Thus 



on flat buttons '6 cm in diameter and less than •4 cm high, the 

 apices of the conoids will often be at a distance of '3 cm and 

 .|cm f rom the geometric base. 



In general, therefore, the originally plane right sections of 

 the transpiring column tend to become j3lane after stress ceases ; 

 or plane right sections of the axially stressed column, tend to 

 bulge out conoidally, symmetrically around the axis, and in a 

 direction opposite to that of the stress, when stress ceases. 

 Thus the experiment points out in a beautiful way how much 

 residual viscosity resembles a slowly reacting elasticity.* Com- 

 plete restitution of form cannot occur because of the dissipa- 

 tion of energy. 



7. Viscosity and Pressure. — Inasmuch as marine glue is a 

 nonconductor and highly viscous, much time must be allowed 

 before temperature and pressure can be assumed to have pene- 

 trated the mass uniformly. Again whenever the oil of the 

 compressor accidentally reaches the transpiration tube, the slip 

 error is enormously increased and the results are worthless. A 

 single charge will not therefore outlast many experiments. 

 These are the chief reasons why much time has to be spent on 

 the work and why it is difficult to coordinate the results. 

 Another annoyance is the unavoidable lack of homogeneity of 

 the charge, and the possibility of a reservoir correction. It 

 will therefore be expedient to briefly indicate the method by 

 which trustworthy results were eventually reached. 



* A suggestive example of the gradual passage of true elasticity into true 

 viscosity is given by the phenomena observed on stretching a string of (dry) 

 vulcanized india rubber deposited from solution. If the string be suddenly re- 

 leased, elasticity and viscosity appear as the two extreme phases of contraction, 

 terminating a continuous series of intermediate phases. 



