Abstract:

To establish the comprehensive evaluation system of pepper seedlings on heat tolerance， 30 pepper germplasms were randomly divided into 2 groups， which were used for heat resistance predictive model construction （containing 20 materials） and model validation （10 materials）， respectively. The high temperature environment was set in artificial climate chamber. The 10 morphological and physiological indexes of pepper seedlings were determined， and the heat tolerance of pepper seedlings were evaluated by principal component， subordinate function， cluster analysis and stepwise regression analysis. The results showed that among the 10 indexes， 8 indexes were significantly affected by high temperature， which could be transformed into 3 independent comprehensive indexes and the accumulative contribution rate was 76.474%. The 20 materials were divided into 4 categories： heat-resistance， medium heat-resistance， medium heat-sensitivity and heat-sensitivity by systematic clustering method based on the comprehensive evaluation value （D value）. The prediction equation was constructed by stepwise regression analysis， and the heat tolerance of the test materials was predicted by this equation. The predicted values were significantly correlated with the D value， indicating that this equation was reliable for predicting the heat resistance of pepper seedlings， and could be used to evaluate heat-tolerance of pepper at seedling stage. By comparing the heat injury index （HII） identification method and the comprehensive evaluation method to identify the heat resistance of pepper seedlings， it was found that the identification results of the 2 methods were generally similar. The HII method was insufficient to distinguish the heat-tolerance of moderately tolerant materials， but because of its intuition and convenience， it could be used as a reliable index for heat-tolerance identification， and used for preliminary evaluation of heat-tolerance.

Key words: pepper, heat tolerance at seedling stage, comprehensive evaluation system, prediction equation, heat injury index

摘要：

为建立辣椒苗期耐热性综合评价体系，以30份辣椒种质为试验材料，随机分成2组，分别用于耐热性预测模型构建（20份）和模型验证（10份）。利用人工气候室模拟高温环境，测定10个与高温相关的形态和生理指标，应用主成分分析、隶属函数、聚类分析、逐步回归分析法综合评价辣椒材料的耐热性。结果表明，在10个指标中，有8个单项指标受高温影响显著，其可转化为3个相互独立的综合指标，累计贡献率达76.474%；基于综合评价值（D值）的系统聚类将20份材料分为抗、中抗、中感和敏感4大类；通过逐步回归分析构建了预测方程，并以此方程对验证组10份材料的耐热性进行预测，结果与D值极显著相关，表明此方程对辣椒苗期耐热性的预测具有较好的可靠性，可用于辣椒苗期耐热性评价。通过比较热害指数法和综合评价法鉴定辣椒苗期耐热性，发现2种方法的鉴定结果总体相似，热害指数法对于中等抗性材料的鉴定区分度不够，但其更直观、便捷，可作为耐热性鉴定的可靠指标和进行耐热性初步评价。

关键词: 辣椒, 苗期耐热性, 综合评价体系, 预测方程, 热害指数