Daily measurements of mass and phosphorus ( P) were made for Daphnia magna feeding on a high concentration (1 mg C l(-1)) of P-sufficient (molar C: P = 70 - 80, 'P+') or P-deficient (C: P = 1000, ` P-') green algae (Scenedesmus obliquus or Ankistrodesmus falcatus). Animals feeding on P+ algae for 12 d showed a sharp decline in growth during the last juvenile instar (mean +/- SE from 0.57 +/- 0.02 to 0.25 +/- 0.03 d(-1)) and a modest decline in P-content at the first adult instar (from 1.55 +/- 0.01 to 1.46 +/- 0.02% mass). Animals feeding on P- algae showed declines in P- content to as low as 0.84% after 5 d, with the sharpest decline in the first day. Growth of animals feeding on P- algae declined sharply over time and showed a linear relationship with the grazer's P- content. Growth during the first day of exposure to P- algae was relatively high (mean 0.39 +/- 0.03 d(-1)) but approached zero as P- content declined to < 0.90% mass. For animals feeding on P+ algae, the ontogenetic decline in P- content resulted in only a small difference between growth calculated for mass and growth calculated for Daphnia P. In contrast, for animals feeding for 4 d on P- algae, growth calculated for mass was 2 - 5 times higher than growth based on P. Thus, this study shows that short-term growth assays may substantially underestimate the negative impact of P- deficient resources, unless the decline in the grazer's P- content is taken into account. In this situation, measurement of growth in terms of nutrient content ( e. g. grazer's P- content) should provide a better indication of the potential for sustained growth than calculations based on change in mass. This study also supports the ` integration of growth' hypothesis (Sterner & Schwalbach, 2001), in which animals undergoing diel migration between water layers with P-deficient and P-sufficient resources experience a reduced growth penalty.