为了筛选出贵州省较优刺梨品种、能为大规模种植、培育以及刺梨药食商品的加工提供科学依据，本文以贵州省二十二个地区采摘的刺梨果实为原料，测定其可溶性固形物、可滴定酸、可溶性蛋白、维生素C、SOD酶活性、氨基酸含量等基本营养成分，对不同地区的刺梨进行综合品质对比分析。并采用相关性分析、主成分分析、建立综合得分模型以及聚类分析方法对贵州省不同地区刺梨品质进行综合评价。结果表明：不同地区之间的刺梨果实综合品质指标均存在显著性差异（p<0.05），其中维生素C含量差异很大；除SOD活性指标外，其余指标之间存在显著相关性（p<0.05）；仁怀地区的可溶性固形物、总糖、可溶性蛋白和维生素C含量最高，印江地区的还原糖含量和SOD活性最高，望谟地区的氨基酸含量最高，德江地区的各综合品质指标含量均偏高；提取出了三个主成分（特征值>1），累积方差贡献率达到了75.162%；建立综合得分模型F综=0.468F1+0.327F2+0.205F3，得到刺梨品质从优到劣排序为：RH>YJ>DJ>WM>LD>ZA>DZ>PD>HZ>LD>CS>LL>DF>BJ>ZN>KL>FG>AS>MJ>SN>SC>LZ>PZ；在欧式距离为20时，可对不同地区的刺梨综合品质按优劣划分成3类，聚类分析与主成分分析结果基本一致。结合以上分析，可判断出德江、仁怀、印江和望谟这四个地区刺梨果实综合品质较好，且果实品质与海拔存在紧密联系，因此可考虑开发利用来自这些海拔地区的良好的刺梨资源。

To identify superior Rosa roxburghii cultivars in Guizhou and provide a scientific basis for large-scale planting, cultivation, and processing of medicinal Rosa roxburghii products, this study examined Rosa roxburghii fruits from 22 regions of Guizhou by determining their basic nutritional content. Measured values include soluble solids, titratable acids, soluble proteins, vitamin C, and superoxide dismutase (SOD) activity. Overall quality of Rosa roxburghii fruit from different regions was analyzed and compared. Correlation analyses, principal component analyses, and cluster analyses were performed, and a scoring model was constructed to comprehensively evaluate Rosa roxburghii fruit quality. Significant differences were identified in the comprehensive quality indices of Rosa roxburghii fruits from different regions (p<0.05). In particular, vitamin C content varied greatly. Except for SOD activity index, a significant correlation was observed between all other indices (p<0.05). Soluble solid, total sugar, soluble protein, and vitamin C contents were the highest in fruit from the Renhuai region, whereas reducing sugar content and SOD activity were the highest in those from the Yinjiang region. Amino acid content was the highest in fruit grown in the Wangmo region. All comprehensive quality indices were relatively high in fruit from the Dejiang region. Three principal components (eigenvalue>1) were identified, and the cumulative variance contribution rate reached 75.162%. According to the comprehensive scoring model Fsynthesis=0.468F1+0.327F2+0.205F3, the Rosa roxburghii fruit-growing regions are ranked in descending order of overall fruit quality as follows: RH>YJ>DJ>WM>LD>ZA>DZ>PD>HZ>LD>CS>LL>DF>BJ>ZN>KL>FG>AS>MJ>SN>SC>LZ>PZ. When the Euclidean distance is 20, Rosa roxburghii fruits from different regions can be divided into three categories according to their overall quality, and the results of cluster analysis are basically consistent with those of principal component analysis. Rosa roxburghii fruits in Dejiang, Renhuai, Yinjiang, and Wangmo were found to be of better overall quality. Fruit quality is thus closely correlated with altitude. Therefore, good-quality Rosa roxburghii fruits from high-altitude regions should be considered for exploitation and utilization.

贵州省林业局科技项目（黔林科合[2020]2015）；贵州大学培育项目（黔科合平台人才[2017]5788-24）