Concrete cutoff walls are usually constructed using a panel-by-panel technique in which primary panels are constructed with space between them and then secondary panels are constructed and inserted in the spaces. A small thickness of residual bentonite cake from the slurry used during excavation usually remains in the construction joints between adjacent primary and secondary panels. The thickness of such bentonite-filled joints should be minimized in terms of the performance of the cutoff wall in controlling seepage. This research experimentally evaluated the effects of a number of design and construction parameters on the thickness of the bentonite cake using data from a case study (the cutoff wall of Karkheh Dam). A physical model test was developed and a number of tests were conducted. The test results showed that parameters such as age of the primary panels, cement content of the slurry, quantity of additives in the slurry, and circulation versus noncirculation of the slurry are responsible for the thickness of the bentonite cake. The results are presented and analyzed.Key words: cutoff wall, plastic concrete, bentonite slurry, bentonite cake, physical model.