The Diamond Grace accident at 10:05 July 2 in September 1997 resulted in the oil release of about 1,550kl in the Tokyo Bay. By 21:30 July 4, withdrawal and treatment works of spilled oil were almost finished. This paper describes the process from the survey planning based on the prediction of oil spreading to the survey (observation and sampling of spilled oil, bottom materials, see water and benthos) and analysis of estuary pollution by spilled oil. The survey was carried out as follows; (1)observation of spilled oil and its effect on the benthos, (2)sampling of sea water, bottom materials, Mytilus eduli and spilled oil, (3)oil analysis of these samples. It is revealed from the present study that (1)from the observation in the sea with video camera one month later from the oil spill, (i)no influence of oil on macro benthos was observed, (ii)it is impossible to find out the traces of spilled oil such as slush balls, (2)the spilled oil varies its characteristics such as density and viscosity as the time going, (3)as the oil is dispersed after mixing detergent, the density of each oil partiele in creases and begins to settles down, (4)based on the n-paraffin distribution of spilled oil, bottom materials, and Mytilus eduli, the spilled oil assimilated in bottom materials and Mytilus eduli are estimated at most 1/2 of their original oil already included oil in them.