Law of Molecular Attraction. 



489 



of the vapour, direct measurements being not usually avail- 

 able at this point. The same is true of bromo-benzene and 

 iodo-benzene at 30°. The error so caused can be estimated 

 by a method pointed out by Young *, and I give below the 

 probable amount of the error thus introduced. 





Reduced, 

 temp. 



Error. 





Eeduced 

 temp. 



Error. 



Di-isopropyl 



0-55 



0-50 

 0-59 

 0-58 

 0-54 

 0-51 

 0-48 

 0-49 

 0-49 

 0-49 

 0-43 

 0-45 

 0-42 



per cent. 

 1-5 

 09 

 22 

 2-0 

 14 

 10 

 06 

 0-7 

 07 

 0-7 

 Very little 



Nearly correct 



Carbon tetrachloride . . 



Stannic chloride 



Methyl formate 



049 



046 

 0-56 

 0-54 

 0-54 

 051 

 0-52 

 0-51 

 050 

 0-50 

 049 

 0-505 



per cent. 

 0-7 

 0-5 

 1-7 

 14 

 14 

 10 

 11 

 10 

 0-9 

 0-9 

 0-8 

 09 



Di-isobutyl 



Isopentane 



Normal pentane 



Methyl actate 



Normal heptane 



Normal octane 





Ethyl acetate. . 



Methyl propionate . . . 

 Propyl acetate 



ETexametbylene 



Benzene 



Fluor-benzene 



Chloro-benzene 



Bromo-benzene at 30°. 

 Iodo-benzene at 30°... 



Ethyl propionate 



Methyl butyrate 



Methyl isobutyrate . - . 



These errors cause practically the same percentage errors 



in the heats of vaporization and in the constants, making 



them too large. After making these corrections the values 



of the constants at 0° are apparently as much below as above 



the average. Further rather large variations at this point 



are caused bv the necessary uncertainty in obtaining the 



d? 



-^ from the Biot formula at and near its end point. The 



vapour-pressure is here very small, and the -^ sometimes 



only a fraction of a millimetre, and the multiplication of the 

 percentage error is therefore enormous. 



11. Some few misprints which it is not thought necessary 

 to detail were brought to light during the close scrutiny to 

 which the measurements were subjected. 



12. As the critical temperature was approached, the 

 constant of my equation showed a large and persistent 

 decrease for all substances, not accounted for by any facts 

 that were known when I began my work. I devoted much 

 time to a study of these divergences and was led to the 

 discovery of three new facts. 



JFirst. — At and near the critical temperature the values of 

 the vapour-pressure calculated from Young's Biot formulae 

 were never in exact accord with the observed vapour- 

 pressures, the observed vapour-pressures being with one or 

 * ' Stoichiometry,' p. 342. 



