66 



THE COMPRESSIBILITIES OF THE MORE IMPORTANT 



Upon comparing the individual data concerning the compressibilities 

 and atomic volumes, it is seen that in many cases these are more or less 

 parallel properties, as has already been shown concerning the alkali metals. 

 This parallelism, however, seems to hold, even approximately, only be- 

 tween similar elements. The comparison of magnesium, calcium, zinc, 

 cadmium, and mercury gives the compressibilities and the atomic volumes 

 in parallel sequence, as in the case of the alkali metals. Tin and lead bear 

 a similar relation to each other; but with the less compressible elements 

 possessing very high boiling-points, this parallelism seems to cease, and on 

 comparing elements of different character, the relation completely fails. 



Often the coefficients of expansion are seen likewise to be parallel with 

 the values of the compressibilities, especially in those cases where the 

 atomic volumes are nearly identical. Evidently, however, the relationship 

 between these quantities is not that demanded by the rule of Dupre : 



For example, if sodium is taken as 0.000015, potassium should be, accord- 

 ing to Dupre's rule, 0.000021, but it is in reality 0.000031 ; the coefficients 

 of expansion being, respectively, 0.000072 and 0.000083. Again, accord- 

 ing to Dupre's rule, sulphur should be ten times as compressible as sele- 

 nium ; but they are, as a matter of fact, almost alike. It is not necessary, 

 however, to go as far afield as this to discover the uncertainty of Dupre's 

 rule. The single case of water, having at 1 and 100 the coefficients of 

 expansion of 0.000006 and -f- 0.00077 respectively, and the compressi- 

 bilities of 0.000051 and 0.000048 respectively, would prove that it is not 

 infallible, although it takes cognizance of causes which undoubtedly affect 

 compressibility. No surprise need be excited by this lack of parallelism, 



