296 Mr. W. B. Morton on the Propagation of 



contain 4*83 per cent, phenol : this concentration is therefore 

 in equilibrium with both ice and solid phenol which had been 

 deposited on the sides of the beaker. The cryohydric mixture 

 therefore contains so little phenol that it may be looked upon 

 as a dilute solution of phenol in water, and its calculated 

 freezing-point, according to van't HofPs rule, would be — 1°'0, 

 in agreement with the observed value — o, 9. Consequently 

 solutions of strength between M and S will deposit phenol 

 on cooling, those between L and S (0 to 4*83 per cent.) ice. 



The diagram is completed by the curve 5lCN which is 

 drawn from Rothmund's observations*, which are indicated 

 by dots ; my own observations (shown by crosses) are in 

 practical agreement with bis and Alexejew's. 



Finally, the curves divide the diagram into regions, with 

 the following meanings : — 



Below LS undercooled solutions of phenol, from which ice 

 crystallizes out, with formation of the saturated solutions LS. 



Below SMNO supersaturated solutions of phenol, from 

 which phenol crystallizes out with formation of the saturated 

 solutions of phenol in water (SM) and water in phenol (NO). 



MONGM, unstable mixtures which separate into the two 

 saturated solutions CM and CN, forming two liquid layers. 



Above LSMCNO homogeneous liquid mixtures. 

 The Davy-Faraday Research Laboratory, Royal Institution, 

 London, October, 1898. 



XXII. On the Propagation of Damped Electrical Oscillations 

 along Parallel Wires. By W. B. Morton, M.A.\ 



IN a paper published in the Philosophical Magazine for 

 ►September 1898 Dr. E. H. Barton has compared the 

 attenuation of electrical waves in their passage along parallel 

 wires, as experimentally determined by him, with the formula 

 given by Mr. Heaviside in his theory of long waves. The 

 results show a large discrepancy between the theory and the 

 experiments, the observed value of the attenuation constant 

 being about twice too large. Dr. Barton discusses several 

 possible causes of error and finds them inadequate, and 

 suggests that the reason of the difference may lie either in 

 (1) the nearness of the wires to one another, or (2) in the 

 damping of the wave-train propagated by the oscillator. To 

 these may be added (3) the consideration that the formulae 

 used were deduced by Mr. Heaviside from the discussion of 

 his " distortionless circuit," in which the matter is simplified 



* L. c. p. 452. 



t Communicated by the Physical Society : read Nov. 11, 1898. 



