FLUIDITY AND THE CHEMICAL COMPOSITION 125
But it is not clear that this is unavoidable, for in the case of
benzene we note that the compound differs from hexane by eight
hydrogen atoms, and since we found in the allyl compounds that
the absence of a pair of hydrogen atoms is compensated for to the
extent of 114.4 we obtain for the calculated value, 59.2 X 6 —
95.7 X 6 + 114.4 X 4 = 238.6. The observed value is 311.9;
hence the association is 1.30, which is somewhat larger than the
value obtained by Traube of 1.18. It has usually been believed
that the more compact and symmetrical the molecule was, the
lower would be the temperature required to give it a certain
fluidity. In disregarding constitutive influences entirely for the
time being, as we have done here, we suppose that benzene would
require the same temperature as a straight chain hydrocarbon
containing four "double bonds." If Traube's value of the
association or some other value less than 1.30 is correct, we will
be compelled to assign a positive value to the ring grouping in
order to increase the calculated value. In other words, the
evidence at hand indicates that the effect of the ring grouping
is not to make the compound less viscous but more so. This is
so contrary to earlier belief and to the probabilities of the case
that it seems preferable to await further data before assigning
any value to the ring grouping.
Having been unable to detect the effect of constitutive
influences upon fluidity with the data at hand in the halogen,
oxygen, or ring compounds, we have left remaining one positive
evidence in the value which we have found it necessary to give
to the "iso-grouping." This effect is not large but it is fairly
uniform and quite outside of the observational error. We cannot
believe that normal hexane and heptane are sufficiently associated
to account for the higher temperature above their isomers
required to give them a fluidity of 200. If then an iso-grouping
affects fluidity it is probable that there are other constitutive
influences, but the solution of this problem evidently requires
more data, particularly among the higher homologues. In this
connection the reader should have regard for the relation of
fluidity to volume, to be discussed later.
Before closing the chapter on fluidity and chemical composition
and constitution, we may add that constants calculated for a
fluidity of 300 give an association which is invariably a little