Continental relative humidity (RH) is a key-climate parameter. However, there is a lack of quantitative RH 15 proxies suitable for climate model-data comparisons. Recently, a combination of climate chamber and natural transect calibrations laid the groundwork for examining the robustness of the triple oxygen isotope composition (d 18 O, d 17 O) of phytoliths as a new proxy for past changes in RH. However, it was recommended that besides RH, additional factors that may impact d 18 O and d 17 O of plant water and phytoliths be examined. Here, the effects of leaf anatomy, leaf development stage and day/night alternations are addressed from the 20 growth of the grass species F. arundinacea in climate chambers. Plant water and phytoliths are analyzed in d 18 O and d 17 O. Silicification patterns are examined using light and scanning electron observation of phytoliths. The isotope data show the increasing contribution of evaporated epidermal water to the bulk leaf water, from sheath to proximal and apical leaf blade. However, despite this isotope heterogeneity, d 18 O and d 17 O of the bulk leaf water can be predicted by the Craig and Gordon model, in the given experimental conditions (high 25 RH). Regarding phytoliths, their forming water (mainly epidermal) is, as expected, more impacted by evaporation than the bulk leaf water. This discrepancy increases from sheath to proximal and apical blade and can be explained by the steepening of the radial concentration gradient of evaporated water along the leaf. However, we show that because most of silica polymerizes in epidermal long cells of the apical blade of the leaves, the d 18 O and d 17 O of bulk grass phytoliths should not be impacted by the diversity in grass anatomy. 30 The data additionally show that most of silica polymerizes at the end of the leaf elongation stage and at the transition towards leaf senescence. Thus, climate conditions at that time should be considered when interpreting d 18 O and d 17 O of phytoliths from the natural environment. At least, no light/dark effect was detected on the d 18 O and d 17 O signature of plant water and phytoliths of F. arundinacea. However, when day/night alternations are characterized by significant changes in RH, the lowest RH conditions favoring 35 evaporation and silica polymerization should be considered when calibrating the phytolith proxy. This study contributes to the identification of the parameters driving the d 18 O and d 17 O of bulk grass phytoliths. It additionally brings elements to further understand and model the d 18 O and d 17 O of grass leaf water, which influences the isotope signal of several processes at the soil/plant/atmosphere interface. 40