The concept that a substantial fraction of guest molecules is trapped within the macromolecular network of coals has undergone some modification since the proposal that led to the funding of this project was written. Accordingly, the strategy to be followed has been revised somewhat. It is now predicted that the polycyclic alkanes and long-chain alkyl aromatics to be used in this project as markers for the mobile phase should be extractable from suitable coals by ethylene diamine or a methylcyclohexanone/dimethylformamide mixture. This prediction is being tested. It would follow, if the hypothesis is correct, that a substantial part of liquefaction consists of the liberation of physically trapped material of quite low molecular weight. A further 19 coals have been sealed in pyridine-d/sub 5/ for pulsed free-induction-decay /sup 1/H nmr; this completes the set of 24 coals to be studied by this technique, in an experiment designed to obtain information on the variation of the mobile phase with rank and geological history. A sample from the Illinois No. 6 seam has been extracted with a benzene/ethanol mixture under Soxhlet conditions and at 100/sup 0/C in an autoclave for 3, 24 and 96 hours. A fairly systematic increase in extract yield and in the hexane-soluble portion is observed. The same coal has been extracted under various conditions with other, more polar, solvents, which yielded less or no more hexane-solubles than benzene/ethanol. Indoline has been reported to be a better donor solvent than tetralin. It has now been found that this is certainly true for 3 coals if conversion is defined by THF solubility. Research has been continued in exploring dry catalytic coal hydrogenation as a technique by which to examine the compositions and relationships of the molecular phase and network. It has been confirmed that the impregnated molybdenum catalyst should be present in the fully sulfided form in order to attain high catalytic activity. 22 refs., 3 figs., 17 tabs.