Japanese knotweed (Fallopia japonica) is an herbaceous invader in North America and Europe with the ability to recruit via several modes (seeds, rhizome, and stem fragments). Within US riparian forests where Japanese knotweed invades, disturbances are common over space and time in the form of flooding and mowing. It remains unknown how disturbance influences the relative recruitment success of sexual and clonal propagules of this riparian invader. Using an experimental approach, we addressed this topic by planting seeds, and rhizome and stem fragments into riparian forest plots with intact Japanese knotweed subcanopies, and into plots with the Japanese knotweed subcanopy removed, simulating disturbance. We monitored canopy openness and propagule fate (shoot emergence and height) over five months. Regardless of treatment, rhizome fragments had a higher establishment probability (85%) than seeds (3%) or stem fragments (16%). Due to a small sample size, we did not analyze the effect of disturbance on seedling recruitment. Disturbance had little effect on the establishment of rhizomes and stems. Disturbance had a significant effect on the size and timing of shoot death from rhizome and stem fragments. During the fall we observed an earlier die-back of shoots from rhizomes in undisturbed treatment plots compared to disturbed treatment plots. Shoot height from rhizome and stem fragments was greater in undisturbed treatments than in disturbed. Our findings are limited to clonal propagules of Japanese knotweed but suggest that disturbance does not increase recruitment or propagule performance.