a b s t r a c tGrowth and yield formation in rice (Oryza sativa L.) depend on integrated impacts of genotype, environment and management. A rice growth simulation model can provide a systematic and quantitative tool for predicting growth, development and productivity of rice under changing environmental conditions. Existing rice models perform well but are somewhat difficult to use because of the large number of parameters that users must estimate. Experience in modelling wheat suggested that using physiological development time (PDT) as a scaler for phenology and a partitioning index for organ growth could result in fewer parameters while providing good predictability and applicability. RiceGrow was developed using PDT and a partitioning index to quantify relations among rice growth and environmental factors, genotypic parameters and management practices. RiceGrow includes seven sub-models for simulating phenology, morphology and organ formation, photosynthesis and biomass production, dry matter partitioning, yield and quality formation, water relations and nutrient balance. The model was calibrated with three datasets involving various cultivars, sowing dates and N rates at multiple sites. Validation with independent datasets showed the model had good predictability and applicability. The RiceGrow model was compared with the ORYZA2000 model, showing that both provided satisfactory estimates for phenology, shoot biomass and yield. Overall, RiceGrow can be used to predict rice growth and development with varied genotypes, environmental conditions and management practices for multiple uses including scientific understanding, policy formulation and optimizing crop management.

Apple fruits of different cultivars were heat-treated at 38°C for 4 d immediately after harvest, then stored at 0°C for 6 weeks ('Gala') or 4 months ('Golden Delicious' and 'Red Fuji'). Upon removal from storage, fruits were kept for 7 d at 20°C as a shelf-life test. During storage and shelf-life, various ripeness parameters were measured. 'Golden Delicious' and 'Red Fuji' apples were artificially inoculated with Penicillium expansum or Botrytis cinerea before heat treatment, then stored at 20°C for 3 weeks, or 0°C for 6 weeks, and the development of fruit decay was recorded. The respiration rate decreased and the ratio of soluble solids contents (SSC): titratable acidity (TA) increased for all heattreated fruits. For 'Gala' and 'Golden Delicious' apples, heat treatment significantly reduced ethylene evolution, reduced softening during cold storage, and reduced membrane electrolyte leakage after cold storage and shelf-life. Better quality was maintained at the end of the shelf-life for these two cultivars. However, the effects of heat treatment on ethylene production, membrane electrolyte leakage and texture parameters of 'Red Fuji' apples were not significant. The rotting caused by P. expansum or B. cinerea was very severe, especially for apples stored at 20°C. However, decay was completely inhibited in 'Golden Delicious' and 'Red Fuji' apples that had been heat-treated after inoculation. It was concluded that the response to heat treatment varied according to cultivar. Pre-storage heat treatment could delay the ripening of 'Gala' and 'Golden Delicious' apples and maintain storage quality. Although heat treatment gave no beneficial effects on the quality of 'Red Fuji' apples, it could be applied to control disease in this apple cultivar.

Aim To isolate, screen and identify the lactic acid bacteria (LAB) from elephant silage during the process of the fermentation and their effects on the fermentation quality of sweet sorghum silage. Methods and Results The isolated strains were identified based on morphological, physiological and biochemical characteristics as well as 16S rRNA analysis. Seven LAB strains were isolated from elephant grass silage. Three strains (Pediococcus acidilactici (AZZ1), Lactobacillus plantarum subsp. plantarum (AZZ4), L. plantarum subsp. argentoratensis (AZZ7) and one commercial bacteria L. plantarum, ecosyl MTD/1(CB)) were chosen as additives at 6 log colony forming units per gram of fresh sweet sorghum grass in laboratory silos (680 g). Silos for each treatment were opened after 5, 7, 14, 30 and 60 days, respectively. All isolates were characterized as Gram‐positive, catalase‐negative and grow normally in 6·5% NaCl. The strains AZZ1, AZZ2 and AZZ5, were identified as Pediococcus genus while AZZ3, AZZ4, AZZ6 and AZZ7 were Lactobacillus genus. Compared to the control, all the isolates improved the silage quality of sweet sorghum silage, indicated by significantly (P < 0·05) lower pH and ammonia‐nitrogen contents and undesirable micro‐organism counts, and higher lactic acid (LA) contents and ratios of lactic acid/acetic acid. During ensiling, AZZ4 performed better among all of inoculants, indicated by significantly (P < 0·05) decreased on pH and ammonia‐N contents and higher increased on LA contents. Conclusion Strain AZZ4 is recommended as starter culture for sweet sorghum. Significance and Impact of the Study This is the first time to investigate the effects of LAB isolates from elephant grass silage and use them as additives, which is to find out how LAB inoculants improve the fermentation quality of sweet sorghum silage.

Total mixed ration (TMR) silage technology has been practically used to feed ruminants in Tibet. This study was conducted on forage-based TMR to evaluate the effects of supplementing lactic acid bacteria and propionic acid on its fermentation characteristics, aerobic stability and in vitro gas production kinetics and digestibility. Experimental treatments included four variants: (1) TMR supplemented with 10 mL deionised water per kilogram fresh matter (Control); (2) TMR supplemented with 1 × 106 cfu/g Lactobacillus plantarum (L); (3) TMR supplemented with 0.3% propionic acid (P); (4) TMR supplemented with a combination of 1 × 106 cfu/g Lactobacillus plantarum and 0.3% propionic acid (LP). The latter three additives were first dissolved in deionised water and then applied as a water solution (10 mL/kg fresh matter). All treatments were ensiled in laboratory-scale silos for 45 days, and then exposed to air for 12 days to evaluate the aerobic stability of TMR silage. Further, the four experimental treatments were fermented with buffered rumen fluid to measure in vitro gas production and nutrients’ digestibility. The results indicated that all TMR silages possessed good fermentation characteristics with low pH values (&lt;4.18) and ammonia nitrogen (NH3-N) contents (&lt;100 g/kg total nitrogen), and high lactic acid contents (&gt;66 g/kg DM) and Flieg points (&gt;80). The addition of L and LP stimulated a more efficient homofermentation of TMR silage than in the variant without L, as evidenced by higher ratios of lactic:acetic acid. The addition of P had no effect (P &gt; 0.05) on lactic acid production of TMR silage compared with the Control, whereas it decreased NH3-N content (P &lt; 0.05). Under aerobic conditions, L silage showed less aerobic stability compared with the Control silage, whereas P and LP silages were more (P &lt; 0.05) aerobically stable. Compared with the Control, all additives elevated (P &lt; 0.05) the total gas production and in vitro dry matter digestibility of TMR silages. L silage had a higher (P &lt; 0.05) in vitro neutral detergent fibre digestibility than the Control silage. Data obtained from this study suggested that TMR silage based on oat and common vetch can be well conserved with or without additives. Lactic acid bacteria were compatible with propionic acid, and addition of lactic acid bacteria together with propionic acid can improve the fermentation quality, aerobic stability and in vitro dry matter digestibility of TMR silage.

Aims: To examine the characteristics of three isolated Pediococcus acidilactici strains (LTG7, LOG9 and LH9) and evaluate their effects on silage quality, nutritive value and in vitro ruminal digestibility in a variety of forages. Methods and Results: One commercial inoculant Lactobacillus plantarum MTD-1 (G) and three isolated lactic acid bacteria (LAB) strains were measured by morphological, physiological and biochemical tests. All the LAB strains were added to Italian ryegrass (Lolium multiflorum Lam.), tall fescue (Festuca arundinacea Schred.) and oat (Avena sativa L.) for ensiling 30 days in laboratory silos (1 l) respectively. Isolated strains could grow normally at 5-20°C, pH 3Á5-7Á0 and NaCl (3Á0, 6Á5%), and were identified as P. acidilactici by sequencing 16S rDNA. In Italian ryegrass and oat silages, all inoculants obviously (P < 0Á05) increased lactic acid (LA) contents, LAB numbers and in vitro dry matter digestibility (IVDMD), and decreased pH, undesirable micro-organism numbers, butyric acid and ammonia nitrogen (NH 3 -N) contents compared with the corresponding controls. LTG7, LOG9 and G silages in Italian ryegrass and oat had markedly (P < 0Á05) higher LA content and IVDMD, and lower pH and NH 3 -N contents than LH9 silages. In tall fescue silage, LAB inoculants had no obvious (P > 0Á05) effect on fermentation quality, while markedly (P < 0Á05) enhanced IVDMD. Conclusions: Based on our results, strains LTG7 and LOG9 had similar potential with the commercial inoculant G in silage making. Significance and Impact of the Study: Few studies involved inoculation of silage with P. acidilactici in different forage types. Analysis of effects of LAB strains with their physiological and biochemical characteristics help understand how LAB inoculants affect the digestibility.