Hong Zhao

The sampling technique is commonly used in research investigations to more accurately estimate data with greater precision, at a lower cost and in less time. In plant DUS (distinctness, uniformity, and stability) testing, many quantitative characteristic data usually need to be obtained through individual measurements. However, there is currently no scientific method for determining the appropriate sampling size. The minimum number of testing samples for DUS testing was calculated based on the theory of sample size in descriptive studies and was validated through simple random sampling. The results show that the quantitative characteristics for the edible mushroom Shiitake (Lentinula edodes) in DUS testing were uniform. The calculated results show that 10 fruiting bodies for a single measurement were sufficient. Furthermore, the outcomes of the random sampling revealed that the mean of 10 samples did not significantly differ from the mean of all data. When the sample size exceeded 10, Cohen’s kappa statistic suggested that the conclusion of distinctness was very close to the near-perfect agreement. Reducing the number of samples did not affect the uniform assessment. This study suggests that the theory of sample size in descriptive studies could be applied to calculate the minimum sample size in DUS testing, and for Shiitake DUS testing, measuring 10 fruiting bodies was sufficient.

Seasonal (temporal) variations can influence the δ¹³C, δ²H, δ¹⁸O, and δ¹⁵N values and nutrient composition of organic (ORG), green (GRE), and conventional (CON) vegetables with a short growth cycle. Stable isotope ratio mass spectrometry (IRMS) and near-infrared spectroscopy (NIRS) combined with the partial least squares-discriminant analysis (PLS-DA) method were used to investigate seasonal effects on the identification of ORG, GRE, and CON Brassica chinensis L. samples (BCs). The results showed that δ¹⁵N values had significant differences among the three cultivation methods and that δ¹³C, δ²H, and δ¹⁸O values were significantly higher in winter and spring and lower in summer. The NIR spectra were relatively clustered across seasons. Neither IRMS-PLS-DA nor NIRS-PLS-DA could effectively identify all BC cultivation methods due to seasonal effects, while IRMS-NIRS-PLS-DA combined with Norris smoothing and derivative pretreatment had better predictive abilities, with an 89.80% accuracy for ORG and BCs, 88.89% for ORG and GRE BCs, and 75.00% for GRE and CON BCs. The IRMS-NIRS-PLS-DA provided an effective and robust method to identify BC cultivation methods, integrating multi-seasonal differences.

Maximizing rice yield potential has always been the focus of high-yield rice cultivation research. For high-yield rice cultivation and breeding, more research into the link between yield and yield components is essential. In this experiment, 38 rice varieties with different yield types and 185 rice varieties as materials were chosen. The relationships between yield and yield components were studied. The regulation effects of total nitrogen application rate (TNAR) on yield and yield components were observed. The results showed that (1) the grain yield of high-yield varieties was 189.3−195.6%, 76.1−77.7%, and 27.0−28.7% higher than that of super-low-yield, low-yield, and medium-yield varieties, respectively. Compared with rice varieties with other yield types, rice varieties with high-yield type have a higher total number of spikelets. (2) The spikelet number per panicle and total number of spikelets were significantly positively linked with grain yield, but significantly negatively correlated with filled grains and grain weight. (3) With an increase in TNAR (0−340 kg ha⁻¹), the panicles, spikelet number per panicle, and total spikelets of rice varieties with different yield types increased gradually, and the filled grains and grain weight decreased gradually. The higher the TNAR, the more obvious the decrease in filled grains and grain weight. The grain yield of rice varieties with different yield types was the highest under the TNAR at 250 kg ha⁻¹. The main factor contributing to its high yield was the substantial increase in total spikelets. The above results showed that increasing the spikelet number per panicle and total spikelets played a material role in improving rice yield.