74 Kubota, Ogino heptanes but some fitting is necessary between eq 1 and the experimental [ 771.Solvent Quality for the Different Classes. For each sample the solvents are listed below in decreasing order of solvent quality: (75%) trans-decalin > cis-decalin > C -c S > S.H. br-Cg > cycloalkanes > S.H. br-C7 > n-alkanes not S.H. br-C9 > tetrahydronaphthalene > highly branched alkanes > toluene > tetrahydrofuran > SnBut4; (63%) cyclopentane > trans-decalin > cis-decalin > methyl c-c6 > cycloalkanes > S.H. br-Cg > S.H. br-C7 > not S.H. br-Cg > n-alkanes > not S.H. br-C7 > highly branched alkanes > toluene > tetrahydronaphthalene > tetrahydrofuran > SnBut,; (33%) cyclopentane > methyl c-C6 > cis-decalin > cycloalkanes > br-Cg S.H. br-C7 > highly branched alkanes > n-alkanes > toluene > tetrahydronaphthalene > tetrahydrofuran > n-c16, n-CI7 > SnBut,. It is to be noted that for two samples, cyclopentane is by far the best solvent. The order of the solvent quality is about the same for the three samples except for cis-and trans-decalin, tetrahydronaphthalene, and for the linear and branched alkanes where correlations of orientations between solvent and polymer are important.This work illustrates the importance of the shape of the polymer and that of the solvent on the free energy of mixing. Large variations in [VI may be found in systems where the usual parameters influencing the solvent quality and chemical and free volume differences are apparently negligible.