Stratospheric ozone depletion by anthropogenic chlorofluorocarbons has lead to increases in ultraviolet-B radiation (UV-B; 280–320 nm) along the Antarctic Peninsula during the austral spring. We manipulated UV-B levels around plants of Antarctic hair grass (Deschampsia antarctica; Poaceae) and Antarctic pearlwort (Colobanthus quitensis; Caryophyllaceae) for one field season near Palmer Station along the west coast of the Antarctic Peninsula. Treatments involved placing frames over naturally growing plants that either (1) held filters that absorbed most biologically effective radiation (UV-BBE; ‘reduced UV-B’, 22% of ambient UV-BBE levels), (2) held filters that transmitted most UV-BBE (‘near-ambient UV-B’, 87% of ambient UV-BBE levels), or (3) lacked filters (‘ambient UV-B’). Leaves on D. antarctica exposed to near-ambient and ambient UV-B were 16–17% shorter than those exposed to reduced UV-B, and this was associated with shorter epidermal cells at the leaf base and tip. Leaves on C. quitensis exposed to near-ambient and ambient UV-B tended to be shorter (P=0.18) and epidermal cells at the leaf base tended to be smaller than those under reduced UV-B (P<0.10). In order to further explain reductions in leaf length, we examined leaf concentrations of insoluble (cell-wall bound) phenylpropanoids, since it has been proposed that wall-bound phenylpropanoids such as ferulic acid may constrain cell expansion and leaf elongation. In both species, HPLC analysis revealed that ferulic and p-coumaric acid were major components of both insoluble and soluble phenylpropanoids. Although there were no significant differences in concentrations between UV-B treatments, concentrations of insoluble ferulic acid in D. antarctica tended to be higher under ambient and near-ambient UV-B than under reduced UV-B (P=0.17). We also examined bulk-leaf concentrations of soluble (methanol extractable) UV-B-absorbing compounds and found that concentrations were higher in plants exposed to near-ambient and ambient UV-B than in plants exposed to reduced UV-B. We also assessed the UV-B-screening effectiveness of leaves that had developed on plants at the field site with a fiber-optic microprobe. Leaf epidermal transmittance of 300-nm UV-B was 4.0 and 0.6% for D. antarctica and C. quitensis, respectively, which is low compared to grasses and herbaceous dicotyledonous plants found in more temperate climates. While the leaves of Antarctic vascular plants are relatively effective at screening UV-B, levels of UV-B in Antarctica are sufficient to reduce leaf epidermal cell size and leaf elongation in these species, although the mechanisms for these reductions remain unclear.