Avian incubation dramatically reduces the abundance and diversity of microbial assemblages on eggshells, and this effect has been hypothesized as an adaptive explanation for partial incubation, the bouts of incubation that some birds perform during the egg-laying period. However, the mechanisms for these antimicrobial effects are largely unknown. In this study, we hypothesized that microbial inhibition is partly achieved through removal of liquid water, which generally enhances microbial growth, from eggshells, and experimentally tested this hypothesis in two ways. First, we placed the first- and second-laid eggs of tree swallow (Tachycineta bicolor) clutches in unincubated holding nests with either ambient or increased water on eggshells. Second, we added water to eggshells in naturally partially incubated nests. We compared microbial growth on shells during a 5-day experimental period and found that, as predicted, both unincubated groups had higher microbial growth than naturally partially incubated controls, and that only in the absence of incubation did wetted eggs have higher microbial growth than unwetted eggs. Thus, we have shown that water increases microbial growth on eggshells and that incubation nullifies these effects, suggesting that removal of water from egg surfaces is one proximate mechanism for the antimicrobial effects of incubation.