652 Professor T. E. Thorpe [March 4, 



The fact, discovered by Amagat, that water under great pressure is 

 more expansible by heat than at ordinary pressure, may also be equally 

 well explained on this hypothesis. Increasing temperature, as we 

 have seen, works in two directions on the volume of water — but as 

 yet nothing is exactly known of the effect of pressure upon the volume- 

 change per degree of temperature of an aggregate consisting solely 

 of one kind of water-molecules ; but the probability is that such an 

 aggregate of molecules would behave like a gas. The anomaly 

 found by Amagat gradually disappears as the pressure is increased. 

 This finds its explanation in the fact that with gi*adually increasing 

 pressure the number of ice-molecules becomes less. Amagat also found 

 that the anomaly became less marked as the temperature was increased; 

 this also is explained by the circumstance that as the temperature 

 increases the number of the ice-molecules diminishes. 



The same hypothesis explains the fact that under pressure the 

 temperature of the point of maximum density becomes lower, and 

 it also affords a reason for the circumstance that the freezing point of 

 water becomes lowered by pressure. 



It has been observed that water at low temperatures becomes 

 colder when subjected to pressure, which may be explained by the fact 

 that in order to convert ice-molecules into molecules of the second 

 kind, heat is required, which can only be furnished by the compressed 

 liquid. 



As regards the influence of pressure on viscosity, we have only to 

 assume, as analogy indicates, that the greater the number of ice- 

 molecules in solution the more viscous becomes the liquid. If we 

 add soluble matter to water, its viscosity increases. Sea water is more 

 viscous than pure water, and the greater the amount of salt in solu- 

 tion the greater becomes the viscosity. If by pressure we diminish, 

 for any particular temperature, the number of ice-molecules in solu- 

 tion, it must follow that we diminish the viscosity, which is what is 

 observed. 



Now, in the light of Professor Eontgen's explanation, the behaviour 

 of water is no longer " anomalous." Its normal properties are exactly 

 similar to those of any other liquid. The so-called anomalies are 

 simply due to the circumstance that the " solid " form of water is 

 specifically lighter than the liquid form. The peculiar form of the 

 curve showing the relation between viscosity and temperature in the 

 case of water at low temperatures, arises from the progressive and 

 rapid increase of the number of the ice-molecules. In this special 

 particular water is not peculiar. Studies on surface energy, on vapour 

 pressures and densities, and on optical characters, have shown that 

 this hypothesis of molecular complexes is well founded, and it is 

 remarkable that many liquids, especially hydroxyl combinations, in 

 which there is reason to assume the existence of such complexes, also 

 exhibit curves of viscosity very similar in character to that shown by 

 water. 



The mathematical expression of the relation of the viscosity of 



