A novel tubular module was used to investigate the direct osmosis concentration process in the case of tomato juice. This module, was constructed, according to given specifications, by PCI UK and consisted of an external stainless steel shroud accommodating, internally, a set of two identical RO membrane tubes having no support lengthwise and properly sealed at their ends. The process performance was measured in terms of water permeation flux and its response to changes of the process parameters was experimentally assessed and established. The process parameters which were investigated in the course of this study were : the kind of osmotic medium, the viscosity of osmotic medium, the osmotic medium concentration, the juice temperature, the juice flow rate, the juice concentration and the membrane thickness. Sodium chloride brine was found to be the best osmotic medium, among the six which were tried, and this was due to its very low viscosity. The above parameter appears to be of paramount importance regarding the effectiveness of an osmotic medium. Higher osmotic medium concentrations yielded to higher osmotic permeation rates. Increasing the juice temperature was found to markedly increase the permeation flux. However, only a slight enhancement of flux was observed by increasing the juice flow rate. Moreover, higher juice concentrations up to approximately 12 degrees Brix led to a lowering of the osmotic flux. Finally, as far as the membrane thickness was concerned, a strong trend was revealed for exponential increase of permeation by shifting towards lower membrane thicknesses. This trend however needs to be further investigated as an inadequate number of experimental points were obtained due to lack of additional membranes.