It has been recently proposed that Flory-Huggins theory (FH), which fi
nds widespread use in the polymer community, be applied to understand
the partitioning of relatively short alkane chains between an organic
phase and water. Use of this theory, which predicts that the molar vol
umes of the different species play an important role in determining so
lubility, results in a significant increase in the hydrophobic surface
tension estimated from transfer experiments. However, the application
of FH theory to the analysis of alkane solubility has been widely cri
ticized. Here, we derive this theory accounting specifically for the p
ressure of the lattice system, and show that it is appropriate for tra
nsfers from a condensed polymer solvent phase to either a gas phase or
a monomer solvent. The sweeping criticisms of the applicability of FH
theory to partition experiments that have appeared in the recent lite
rature are therefore not valid. A new result is that FH theory is vali
d for treating partition data for solutes of arbitrary shape, as long
as the solvent is chain-like. On the other hand, Monte Carlo simulatio
ns using the chain increment method indicate that Flory-Huggins theory
overestimates molecular size effects by 25% even in the case of ather
mal lattice systems, indicating that the predictions of this theory sh
ould be viewed as first-order estimates of true size effects. The phys
ical origins of Flory-Huggins theory are discussed, and connections ar
e made to Hildebrand's free volume theory and also to Sharp et al.'s i
deal gas derivation of volume effects. The role of molar volume effect
s in transfers of solutes from a gas phase to monomer solution are als
o considered.