"Experiments were carried out to study the effects of N fertilizer rates and timing of application on the yield and grain quality of a rainfed emmer crop (Triticum dicoccum Shubler) under Mediterranean conditions. The following parameters were analyzed: hulled and net grain yield, hulled index, spikes m(-2), spikelets per spike, kernels m(-2), thousand-kernel weight, biomass, plant height, lodging, grain protein and ash content. In the first experiment, different N rates (30.60 and 90 kg N ha(-1) plus a control not fertilized) were split at three phenological stages (seeding 20%, tillering 40% and stem elongation 40%). In the second experiment, three N doses (30.60 and 90 kg N ha(-1)) were applied to three crop stages (seeding, tillering and stem elongation). In the third experiment, the rate of 90 kg N ha(-1) was distributed in different amounts (90-0-0, 0-90-0, 0-0-90, 45-45-0, 45-0-45, 0-45-45, 30-30-30) at the three mentioned crop stages. Increasing N rates resulted in higher hulled and net grain yield, as well as protein content. Fertilization (from 60 to 90 kg N ha(-1)) applied to tillering maximized hulled and net grain yield. Fertilization (90 kg N ha(-1)) applied to stem elongation gave the highest grain protein content (%) while splitting application (30 kg N ha(-1) each) at three phenological stages maximized protein yield per hectare. Application of half or one-third of 90 kg N ha(-1) to stem elongation improved grain protein content in comparison with applications at sowing, or at both sowing and tillering. The main factor determining higher yields with increasing N rates in this emmer crop was the number of kernels m(-2). None of the yield components accounted for differences in grain yield when timing and splitting application were varied. (C) 2011 Elsevier B.V. All rights reserved."