The high-power pulsed spallation neutron source of the Japan Proton Accelerator Research Complex (J-PARC) started its operation in May 2008, and it is now used as one of the most powerful facilities in the world for research in the most advanced fields of material and life sciences. 1) However, there are some hindrances to achieving full-power operation at 1MW. The most significant is the cavitation problem. When the short-pulse high-power proton beam is injected into the mercury target, pressure waves are generated by the abrupt thermal expansion of mercury, which may cause cavitation damage to the target vessel wall. Because cavitation damage degrades the structural integrity and seriously decreases the lifetime of the target vessel, 2, 3) technologies to mitigate pressure waves have been crucially needed. Microbubble injection into mercury is one of the prospective technologies to mitigate the pressure wave, and the effect of microbubbles has been investigated experimentally and numerically. 4–7) The injected microbubbles are expected to absorb the thermal expansion of the mercury at the heat source location, and attenuate the pressure wave during the propagation process. 4, 5) Okita et al. showed by numerical simulations that bubbles with radius less than 100 цm are desirable to mitigate pressure waves effectively. 4) To realize these effects, several difficulties must be overcome. First of all, microbubbles must be properly distributed in the area of the peak heat load, which is the source location of strong pressure waves. Since liquid mercury has a higher density than liquid water, microbubbles should be subject to strong buoyancy, and hence, controlling the bubble distribution in flowing mercury may be difficult. The other concern is that the essential cooling of the target vessel by the mercury might be degraded by the injected gas. Since the thermal diffusivity of gases is much lower than that of liquid mercury, accumulated gas layers attached to the surface of the vessel walls could significantly decrease the cooling performance. In this paper, we report the preliminary results of a mercury loop test using a mockup model of the target vessel, performed as one of the studies for a mercury target design with a bubbling system. This experiment was carried out under collaboration with the Oak Ridge National Laboratory (ORNL) using their Target Test Facility (TTF), which is an actual-scale mercury loop constructed for the mercury target development at ORNL. We performed mercury flow experiments to investigate the bubble flow phenomena in the actual target size of J-PARC. We report here the measured bubble size distribution and several interesting phenomena observed in the experiments. One of the difficulties in using liquid mercury comes from its opacity: one cannot see gas bubbles in mercury by optical techniques. In the present experiment, injected microbubbles in contact with the transparent top wall were observed to determine the bubble size distribution in the target vessel. A large void area formed at a downstream flow vane is also reported.