RECENT ADVANCES IN SCIENCE 33 



by the compressibility curve. The descending portion must 

 be ascribed to a change in the nature of the water itself, for 

 the comparatively small amount of urethane present can 

 scarcely be effective. Now it is known that the compressi- 

 bility of water diminishes as the temperature rises, a fact 

 which has been attributed to the breaking down of the trihy- 

 drol molecules into simpler forms. It is reasonable, therefore, 

 to suppose that urethane — -and in fact any solute whatsoever 

 — possesses the property of dissociating the trihydrol, and 

 consequently diminishing the compressibility as an initial 

 effect. The observation that the position of the well-known 

 minimum volume of water can be shifted towards lower tem- 

 peratures as the pressure increases is likewise due to the 

 changing molecular complexity of water. The decomposition 

 of the trihydrol also affords an explanation of the generalisa- 

 tion made several years ago by Tammann, namely, that the 

 volume of a solution varies with the temperature in the 

 same sense as the volume of the pure solvent under a higher 

 pressure. 



The ascending portion of the compressibility curve in the 

 case of urethane is easily attributable to the fact that urethane 

 itself is more compressible than dihydrol, and consequently 

 when the concentration of the urethane becomes sufficiently 

 great the compressibility of the mixture as a whole increases. 



As regards the solution volume curve, Richards points out 

 that its form " is precisely what one would expect if water 

 contained an appreciable amount of trihydrol in dilute solutions 

 and an appreciable amount of monhydrol in concentrated solu- 

 tions with dihydrol present throughout." This is the conclu- 

 sion arrived at by Bousfield and Lowry in 19 10. 



Next, in connection with the surface tension curve it is 

 found that at first the falling off is very marked, and that 

 later, in solutions containing 40 per cent, or more of urethane 

 — from which presumably the trihydrol has been eliminated 

 — the surface tension decreases much less rapidly but more 

 regularly. Here, again, the two markedly different branches, 

 of the curve must be ascribed to the changing constitution of 

 water. When the region, 40 per cent. — 50 per cent, urethane 

 is reached; — -at which point the solvent has become relatively 

 simple in constitution — Richards finds that the surface tension 

 7 and the compressibility /8 are related to one another by 



