The liquefaction of milled wood (Pinus pinaster) was effected in ethylene glycol acidified with small quantities of H2SO4 as catalyst. The purpose of this paper is to evaluate the influence upon the liquefaction yield of the three operating variables, the maximal temperature (150-280 degreesC), the reaction time at maximal temperature (20-60 min) and the amount of added H2SO4 (0-1.5% on dry wood). The individual effects, as well as the interactions between operating variables, are investigated by using an experimental design methodology. From a Pareto chart, it appears that the most significant effects are clearly those of the maximum temperature and the interaction between acidity and temperature. Such effects can be graphically verified through response surfaces and contour line plots. From a regression analysis, the conversion rate of wood into liquid is simply expressed as a function of the operating variables by a polynomial containing quadratic terms. A statistical model seems particularly appropriate in the case of complex and multi-components, as wood, a kinetic model is nevertheless proposed for the liquefaction of micro-crystalline cellulose. This model accounts for the formation of a carbonaceous solid residue from the liquid product. Such an unwanted phenomenon obviously results in a lower yield in liquid product.