We investigated mechanisms involved in inhibition of maize (Zea mays 1.) leaf-elongation growth following addition of nonpenetrating osmolyte to the root medium. l h e elongation rate of the first true leaf remained inhibited for 4 h after addition of polyethylene glycol 6000 (PEC; -0.5 MPa water potential), despite progressive osmotic adjustment in the growing leaf tissues. Thus, inhibition of leaf growth did not appear to be directly related to loss of leaf capacity to maintain osmotic potential gradients. Comparative cell-wall-extension capacities of immature (still expanding) leaf tissues were measured by creep extensiometry using whole plants. Reductions in irreversible (plastic) extension capacity (i.e. wall hardening) were detected minutes and hours after addition of PEC to the roots, by both in vivo and in vitro assay. l h e onset of the wall-hardening response could be detected by in vitro assay only 2 min after addition of PEC. lhus, initiation of wall hardening appeared to precede transcription-regulated responses. l h e inhibition of both leaf growth and wall-extension capacity was reversed by remova1 of PEC after 4 h. Moreover, wall hardening could be induced by other osmolytes (mannitol, NaCI). Thus, the leaf responses did not appear to be related to any specific (toxic) effect of PEC. We conclude that hardening of leaf cell walls is a primary event in the chain of growth regulatory responses to PEC-induced water deficits in maize. l h e signaling processes by which PEC, which is not expected to penetrate root cell walls or membranes, might cause cell-wall hardening in relatively distant leaves was also investigated. Plants with live or killed roots were exposed to PEC. l h e killed roots were presumed to be unable to produce hormonal or electrical signals in response to addition of PEC; however, inhibition of leaf elongation and hardening of leaf cell walls were detected with both live and killed roots. Thus, neither hormonal signaling nor signaling via induced changes in surface electrical potential were necessary, and hydraulic signals appeared to generate the leaf responses.