1. Above- and belowground biomass allocation is an essential plant functional trait that reflects plant survival strategies and affects belowground biomass estimation in grasslands. However, due to the difficulty in distinguishing fine and living roots, field-based studies show large uncertainties in estimating the biomass allocation. In addition, how plant density regulates biomass allocation across various species remains poorly addressed. 2. To address this knowledge gap, we conducted greenhouse manipulation experiments to explore patterns of above- and belowground biomass allocation and its regulation by planting density in six common grass species of different functional types (i.e., C3 vs C4; annuals vs perennials) in China’s temperate grasslands. The six species were Chenopodium glaucum, Cleistogenes squarrosa, Leymus chinensis, Medicago sativa, Setaria viridis, and Stipa grandis. 3. We found that the mean values of root to shoot ratio (R/S) ranged from 0.04 to 0.92 across the six species, with much lower R/S values in annuals than in perennials (C. glaucum and S. viridis vs C. squarrosa, L. chinensis, M. sativa and S. grandis) and in C4 plants than in C3 plants (C. squarrosa vs L. chinensis, M. sativa and S. grandis). For C. squarrosa, increasing the planting density decreased the shoot biomass fraction (SMF) but increased the root biomass fraction (RMF) and R/S. In contrast, for the other five species, planting density had nonsignificant effects on the SMF, RMF, and R/S. In addition, the planting density significantly affected the allometric relationships between above- and belowground biomass. 4. Synthesis. Our results suggest that R/S values obtained from field investigations are severely overestimated and that the pattern and density regulation of the biomass allocation vary across species of different functional types. Our findings provide important insights into approximating difficult-to-measure belowground biomass with easier-to-measure aboveground biomass in grassland ecosystems, and also provide important theoretical foundations for grass-based livestock husbandry practices.

Using traditional photosynthesis-intercellular CO2 concentration (A-Ci) response (TACiR) curves to obtain the maximum rates of ribulose-1,5-bisphosphate carboxylase oxygenase carboxylation (Vcmax) and electron transport (Jmax) is time-consuming and labor-intensive. Instead, the rapid A-Ci response (RACiR) technique provides a potential way with high efficiency. However, multiple parameter settings of RACiR technique for different plant life forms remain unclear. Here, we used Li-Cor 6800 to test the applicability and optimum parameter settings of RACiR curves for evergreens and herbs. We set 11 groups of [CO2], i.e., R1 (400-1500 ppm), R2 (400-200-800 ppm), R3 (420-20-620 ppm), R4 (420-20-820 ppm), R5 (420-20-1020 ppm), R6 (420-20-1220 ppm), R7 (420-20-1520 ppm), R8 (420-20-1820 ppm), R9 (450-50-650 ppm), R10 (650-50 ppm) and R11 (650-50-650 ppm), and made contrasts between TACiR and RACiR curves. We found that Vcmax and Jmax calculated by TACiR and RACiR overall showed no significant differences across 11 [CO2] gradients (P<0.05). The efficiency and accuracy of R2, R3, R4, R9 and R10 showed higher superiority than others. Moreover, the accuracy of manual empty chamber correction method was higher than the automatic method. In conclusion, the RACiR technique could be generally used to obtain photosynthetic parameters with higher efficiency than traditional methods for various life forms.