Artificial Light at Night (ALAN) is a fast-spreading threat to organisms, especially in coastal environments, where night lighting is increasing due to constant anthropization. Considering that ALAN affects a large diversity of coastal organisms, finding efficient solutions to limit these effects is of great importance but poorly investigated. The potential benefit of one strategy, in particular, should be studied since its use is growing: part-night lighting (PNL), which consists in switching off the lights for a few hours during nighttime. The aim of this study is to investigate the positive potential of the PNL strategy on the daily rhythm of the oyster Crassostrea gigas, a key species of coastal areas of ecological and commercial interest. Oysters were exposed to a control condition and three different ALAN modalities. A realistic PNL condition is applied, recreating a strategy of city policy in a coastal city boarding an urbanized bay (Lanton, Arcachon Bay, France). The PNL modality consists in switching off ALAN direct sources (5 lx) for 4 h (23−3 h) during which oysters are in darkness. Then, a PNL + skyglow (PNL + S) modality reproduces the previous one mimicking a skyglow (0.1 lx), an indirect ALAN source, during the direct lighting switch off, to get as close as possible to realistic conditions. Finally, the third ALAN condition mimics full-night direct lighting (FNL). Results revealed that PNL reduces some adverse effects of FNL on the behavioral daily rhythm. But, counterintuitively, PNL + S appears more harmful than FNL for some parameters of the behavioral daily rhythm. PNL + S modality is also the only one that affect oysters' clock and melatonin synthesis gene expression, suggesting physiological consequences. Thus, in realistic conditions, the PNL mitigation strategy might not be beneficial in the presence of skyglow, seeing worse for a coastal organism such as the oysters.< Réduire