Empirical records have proven global climate change to be an indisputable fact, with an important contribution from the increase in atmospheric carbon dioxide (CO₂) concentration. According to the projections of the Intergovernmental Panel on Climate Change (IPCC) in 2007, global CO₂ concentration is expected to double in the middle of the 21st century compared to its pre-industrial level. As the main substrate for plant photosynthesis, elevated CO₂ concentration will directly influence the growth and development of all terrestrial higher plants, especially those grown as crops. Rice (Oryza sativa L.) is one of the most important crops in the world and is the primary staple food in Asia, as well as China. Many studies have indicated that increasing CO₂ concentration generally increases the grain yield of rice, but it is unclear whether this CO₂ fertilization effect varies with alteration in the source-sink relationship of plants. In order to answer this question, we designed an experiment with treatments of elevated CO₂ concentration and source-sink manipulation of the hybrid rice Shanyou 63 by using a rice Free Air CO₂ Enrichment (FACE) facility at Jiangdu (119°42'0"E, 32°35'5"N), Yangzhou, China, in 2011. Rice plants were grown under two levels of CO₂ concentration (ambient and ambient + 200 μmol/mol) from transplanting until maturity. Source-sink manipulation was achieved through cutting off the whole flag leaf (LC, leaf cutting) or half of the spikelets at heading (SC, spikelets cutting; remove every other primary branch of a panicle). The results showed that under the CK (control, no leaf or spikelet cutting) condition, elevated CO₂ concentration increased grain yield by 32% (P < 0.05), which was mainly due to the increase in spikelet number per square meter (+26%, P < 0.05) and was partly due to the non-significant increase in fertility. On average, CO₂ elevation increased grain yield by 55% (P < 0.01) for LC-crops, with the increase being much larger than that of CK-crops. This higher response was mainly attributed to the dramatic increase in fertile grain percentage (+28%, P < 0.05), filled grain percentage (+23%, P < 0.05), and average grain weight (+19%, P < 0.05). By contrast, for SC-crops, the yield response to the high CO₂ level (+25%, P= 0.07) was much lower than that of CK-crops, which was related to the trends of down-regulation in fertility. Similarly, elevated CO₂ concentration increased the final aboveground biomass by 39%, 43%, and 28%, for CK-, LC-, and SC-crops, respectively, with the effect being significant for the former two. Compared with CK, LC- and SC-treatment at heading significantly decreased grain yield by 40% and 45%, respectively. The former was mainly due to the great decrease in grain fertility, while the latter was linked with the reduction in total spikelet number by half. Compared to CK, LC-treatment at heading significantly decreased the leaf, stem and sheath, panicle, and above-ground biomass at maturity by 29% (P < 0.05), 32% (P < 0.01), 28% (P < 0.01), and 29% (P < 0.01), respectively; SC-treatment reduced the biomass of the corresponding plant parts by 15% (P= 0.24), 33% (P < 0.05), 47% (P < 0.01), and 17% (P < 0.05), respectively. These results indicated that elevated atmospheric CO₂ concentration greatly increased the productivity of hybrid rice. Furthermore, this fertilization effect was enhanced by decreasing the source-sink ratio (i.e., leaf cutting at heading), but was reduced by increasing the source-sink ratio (i.e., spikelet cutting at heading).