ABSTRACT Grain filling is a critical stage for yield formation and quality development in rice (Oryza sativa L.), particularly in fragrant cultivars where aroma is associated with the accumulation of 2-acetyl-1-pyrroline (2AP). This study evaluated the effects of different nitrogen application rates on grain-filling dynamics, 2AP accumulation, yield, and grain quality in indica fragrant rice. Field experiments were conducted under varying nitrogen levels, and grain-filling parameters were analyzed in relation to 2AP content across different panicle positions. Nitrogen application significantly influenced grain-filling characteristics and 2AP accumulation, with both low and high nitrogen rates promoting higher 2AP levels. Grains located at the bottom of the panicle exhibited higher 2AP content and were positively associated with grain-filling rates, while additional correlations were observed between grain-filling parameters and panicle position. Increased nitrogen application also improved grain yield, processing quality, and nutritional attributes. Overall, a nitrogen rate of 315 kg ha−1 was associated with improved fragrance, yield, and grain quality, likely due to enhanced grain-filling dynamics. However, these findings should be interpreted with caution, as the study did not account for nitrogen application timing, genetic variability among cultivars, or environmental variation across growing conditions.
ABSTRACT This study aimed to develop and select tomato plants tolerant to heat, from F2 populations for industrial processing and fresh consumption, obtained from interspecific crosses between Solanum lycopersicum L. × Solanum habrochaites Knapp & Spooner var. hirsutum (PI 127826). Populations derived from Redenção × PI 127826 (UZIP) and Santa Clara × PI 127826 (UZFS) were evaluated at the seedling, vegetative, and reproductive stages under elevated day/night temperatures (36/28 °C and 40/30 °C). The number of leaves, plant height, and resistance to injuries were assessed, as well as pollen viability and stigma receptivity. In the initial seedling screening, several genotypes outperformed the commercial parents. In the second screening, the cultivars Redenção and Santa Clara confirmed high sensitivity to heat stress, while the wild accession PI 127826 maintained tolerance at all stages. Out of the industrial processing plants, UZIP-40, UZIP-174, and UZIP-245 showed the best performance, with growth and injury resistance comparable to PI 127826. For fresh consumption, UZFS-39, UZFS-85, and UZFS-176 exhibited similar tolerance to the wild parent. These genotypes also maintained high pollen viability and stigma receptivity under heat stress, surpassing the commercial parents. The results indicate that interspecific selection enabled the identification of F2 genotypes capable of maintaining vegetative growth and reproductive functions under thermal stress. These populations represent promising sources for breeding programs designed to improve heat tolerance in tomato, both for industrial processing and fresh consumption.
ABSTRACT Tropical soils face challenges related to degradation, contamination, climatic stressors, and nutrient depletion. Sustainable soil management strategies are required, and modified biochar (MB) has emerged as a relevant material for agriculture, environmental remediation, and carbon (C) sequestration. This bibliometric and functional synthesis reports research on modified and doped biochars applied under tropical conditions between 2020 and 2025. A total of 79 studies from the Web of Science Core Collection (WoS) and Scopus databases were analyzed to elucidate relationships between countries, biomass feedstocks, modification strategies, pyrolysis conditions, and application objectives. An increase in publications was observed after 2022, with China, India, and Brazil leading scientific production. Agricultural and forestry residues predominate as raw materials, while pyrolysis temperatures between 300-500 °C are commonly reported. Metal- and mineral-based modifications are the most frequent strategies, particularly for contaminant immobilization and water remediation, whereas soil fertility applications often rely on unmodified or minimally MB. Across applications, MB frequently outperformed pristine materials in function-specific applications, with individual studies reporting approximately two- to three-fold improvements in sorption capacity or reaction kinetics and contaminant removal efficiencies exceeding 90 %. Agronomic responses were more context-dependent, with some researchers reporting improvements in nutrient availability, nutrient-use efficiency, biomass, or crop yield of varying magnitude depending on the modification strategy and soil conditions. Despite these advances, critical knowledge gaps persist regarding long-term stability.
ABSTRACT Tomatoes (Solanum lycopersicum L.) are one of the most popular vegetables across the globe and their maturity directly affects the taste and market value. Accurate maturity recognition methods ensure that tomatoes are harvested at the optimal time and thus increase yield and quality while reducing post-harvest losses. This paper proposes a lightweight tomato ripeness detection model called GhostNetV2 + Bottleneck Transformer Network + Grouped Shuffle Convolution-You Only Look Once version 8 nano (GBG-YOLOv8n). It is an improved model on YOLOv8n (You Only Look Once version 8 nano) architecture to enhance detection efficiency and accuracy. In the GBG-YOLOv8n model, the original Cross Stage Partial 2-focus (C2f) modules in the YOLOv8n backbone are replaced with GhostNetV2 blocks. In the neck network, both the convolutional layers and the C2f modules are substituted by Grouped Shuffle Convolution (GSConv)-based hybrid convolutions and the Volume-wise Network-GSConv-Cross Stage Partial (VoVGSCSP) cross-stage partial modules. Additionally, the Bottleneck Transformer (BoT) self-attention mechanism is incorporated into the main network. Ablation tests, comparison experiments, and recognition tests were conducted on the tomato dataset. Experiments on the tomato dataset revealed that the optimized GBG-YOLOv8n model achieved 92.6 % mean average precision (mAP)@50 and 76.3 % mAP@50-95, with a 21 % reduction in parameters and floating-point operations per second (FLOPs), while maintaining 74 % of the original model's functionality. The optimized GBG-YOLOv8n model demonstrates a high degree of accuracy and low computational cost for tomato ripeness detection.
ABSTRACT Paspalum rufum Nees is a subtropical species that presents diploid individuals that primarily reproduce sexually and through allogamy due to self-incompatibility, as well as tetraploid races that are self-fertile and reproduce via apomixis. Embryological studies have shown that tetraploid races can occasionally produce offspring through sexual reproduction, making them facultative apomicts. This study aimed to analyze the impact of pollen source on the rate of apomixis in facultative apomictic accessions of P. rufum. Six progenies were obtained through self-pollination and cross-pollination from three facultative apomictic tetraploid (4×) genotypes of P. rufum: Q3756, Q3785, and Q4185. The progeny origin, whether maternal or non-maternal, was determined using molecular markers. Our findings indicate that the pollen source significantly influenced the progeny origin across three P. rufum accessions. In instances of self-pollination by a facultative apomictic plant, most of its descendants originate from aposporous embryo sacs (AES) and are genetically identical to the mother plant, with the percentage of non-maternal plants ranging from 1 % to 6 %. However, when the same genotype undergoes cross-pollination, the percentage of non-maternal offspring can rise to 19 % in the progeny derived from the Q4185 × Q3785 cross. This residual sexuality and the manipulation of pollination conditions could be effective strategies for generating genetic variability in apomictic species for which fully sexual counterparts are not available.
