There have been numerous attempts to create and/or develop harvestable tidal areas where juveniles of the natural Manila clam Ruditapes philippinarum may be abundantly stocked. To prevent erosion and/or promote the colonization of natural juveniles through improving the physical stability of bottom sediment, it is important to understand the mechanisms that regulate dispersal and recolonization of benthic juveniles. The physical transport of bottom sediment is known to have a substantial effect on the spatial distribution patterns of infaunal bivalves in intertidal soft-bottom habitats. During the summers of 2004 and 2005, we conducted field experiments to identify the physical transport mechanisms for newly-settled postlarval Manila clams until growth to a size of 1 mm shell length at the Banzu tidal flat of Tokyo Bay in Japan. We used sediment traps to collect natural and released hatchery-reared juveniles, in parallel to acquiring measurements of seawater flow and observations of the spatial distribution of newly-settled natural juveniles. There was a sharp increase in the number of clams that were collected in the traps placed on the soft-bottom surface when τ_w (wave shear stress) exceeded 0.3 N m^-2. Because the τ_c (advection shear stress) was far lower than τ_w (τ_c/τ_w =1/60–1/130), the initiation of juvenile transport appeared to depend primarily on wave-generated oscillatory flow. The number of trapped juveniles regressed linearly to the weight of the sediment that was simultaneously collected in the trap (R² = 0.99, 0.81), which indicated that clams were transported in a similar way to that of sediment grains, despite juveniles and sand particles exhibiting different physical properties (size and specific gravity). Hence bedload transport may have resulted from the biologically induced adhesion of juvenile clams to sediment grains and/or their burrowing behavior. In the release-recovery experiment of marked juveniles, a larger number of clams were recovered from traps that had been placed downstream of the water current from the release point. A denser distribution of the natural Manila clam population settled in mid-July 2004, and subsequently moved several hundreds of meters inshore within a one month period. The concurrent monitoring of bottom flow during a total 4 wk period in the summer of 2004 indicated that τ_w frequently exceeded the incipient threshold of bedload transport (assumed to be 0.3 N m^-2). Consequently, the bedload movement of Manila clam juveniles in the study area was expected to be initiated at a wave shear stress that was greater than the incipient threshold, and in a downstream direction of the advection current. Because juvenile clams in the summer population appeared to be frequently subjected to hydrodynamic stress, which forces juveniles to move and halt incidentally in the early benthic stages, physical transport is likely to contribute to the changing pattern of juvenile distribution at the Banzu tidal flat of Tokyo Bay in Japan.