The research that has been presented in this report is a part of an ongoing study on the siltation of tidal harbors. The study deals with the water motion in the harbor entrance, which motion causes the siltation. As yet, too little is known about this complicated time-dependent water motion. The data obtained during the research will be used to calibrate the 3-D numerical model Trisula, so that this model can be used as a tool to predict the water motion in a harbor entrance. With the present knowledge of the transport of cohesive sediments, a better prediction of the siltation of a particular harbor entrance will then be possible. Experiments have been performed in a physical model at the Laboratory of Fluid Mechanics of the Delft University of Technology. In these experiments the influences, on the flow patterns in the harbor entrance and the exchange of heat between harbor and river, of the geometries of the harbor and the harbor entrance, the tidal period and tidal water level changes were examined. Measurements of the time-dependent velocity and temperature fields were made in five model harbors. In the experiments without tidal water level changes three harbors had their length axes perpendicular to the length axis of the river, namely (1) a square harbor of 1 m2, (2) a rectangular harbor of 1 x 2 m2 and (3) a square harbor of 1 m2 with a narrowed entrance of 0.5 m; one harbor, (4), of 1 m2 and an entrance width of 1 m had its length axis at an angle of 45 degrees to the length axis of the flume. In the experiment with tidal water level changes a rectangular harbor, (5), with an entrance width of 1 m and a storage area of 8 m2 had its length axis perpendicular to the length axis of the river. It can be concluded that: details of circulating flows and gyres depend markedly on the geometry of the harbor. the progress of the phenomena after slack water in model harbor (2), that is the development of a new primary gyre, does not seem to depend on the tidal period. As a consequence, the phase difference between the development of the gyre and the accelerating flow in the river increases as the period decreases. the flow pattern in the harbor is highly influenced by the orientation of the harbor entrance. An explanation for this phenomenon is deficient at the moment. in the model harbors, except harbor (3), a quite strong secondary current is present in the gyre. The maximum velocity in the secondary current is on the average 15 per cent of the main flow. This means that a three-dimensional numerical model will be necessary to simulate the flow pattern in the harbor correctly. close to the downstream sidewall, in all model harbors, larger water velocities (20 to 50 per cent larger) were observed near the bottom than higher in the water column. Near the bed high-momentum fluid from the mixing layer between harbor and river appears to be transported into the harbor. tidal water level changes cause an acceleration in the development of the new gyre towards high tide. Towards low tide the development of the new gyre is hindered by the emptying of the basin. in harbors (1), (2) and (4) a large increase in advective exchange takes place around slack water. The influence of turbulence seems to be of secondary importance during this phase of the tide. a narrowed entrance highly reduces the exchange of mass between harbor and river at slackwater. if the flow pattern in the harbor comprises various gyres, in this research harbor (2), the normalized exchange will be less because the secondary,tertiary, etc. gyres do not contribute to the exchange process. wgen the current in the river is around maximum,the flow is quasi-steady for a quite large duration. The exchange then takes place through the mixing layer between river and harbor, that is, it is caused by turbulent motions only. although during slack water a less refined turbulence model is sufficient in a numerical model, the modeling of turbulence is important during the quasi-steady phase of the tide, especially when the geometry of the entrance is more complex (e.g. harbor (4)).