144 FLUIDITY AND PLASTICITY If we could always compare the fluidities of liquids at a defini multiple of the molecular limiting volume, it is evident that t. effect of the volume, and therefore of temperature and pressu in so far as they affect the volume, would be eliminated. Such procedure would enormously simplify fluidity relationship As a matter of fact Batschinski has shown that the molecul limiting volume possesses an additive character, the values I.IG too. "50 Fluidity 550 FIG. 55.—Relative volume-fluidity curves after Batschinski using the data c Thorpe and Rodger. 4, Bromine; 5, nitrogen peroxide; 9, isohexane; 11, isoheptane; 13, trimethy. ethylene; 14, isoprene; 15, diallyl; 16, propylchloride; 17, isopropylchloridc 18, isobutylchloride; 19, allylehloride; 20, methylenechloride; 21, ethylenechlc ride; 22, ethylidenechloride; 23, chloroform; 25, perchlorethylene; 68, ethyl benzene; 69, orthoxylene; 70, metaxylene; 71, paraxylene. the atomic constants being H = 4.3, 0 = 8.6, C = 8.8, Cl = 19.2, Br = 24.8, I = 32.0, S = 19.0, a double bond = 3.3 and an iso-grouping = 0.7. For 53 of the substances studied b] Thorpe and Rodger the differences between the observed anc calculated values of the molecular limiting volume do not exceec 2 per cent. The limiting specific volume is approximately 0.301 . of the critical volume which is close to the parameter 6 of var der Waals7 equation. We are now in a position to get a very clear understanding of the law that the fluidity is proportional to the free volume.