To understand the genetic diversity and variation types of the phenotypic traits of Gleditsia sinensis in Guizhou Province, and to lay a theoretical foundation for genetic improvement, germplasm identification, parental selection and cultivar breeding, we selected 70 individuals from 7 wild G. sinensis populations in Guizhou Province as the research objects in this study. The 10 species phenotypic traits of G. sinensis populations were systematically analyzed and comprehensively evaluated by means of ANOVA, principal component analysis, correlation analysis and multitrait comprehensive evaluation. The results showed that: (1) the differences of the phenotypic characters of G. sinensis reached the highly significant level within the population (P < 0.01), and the differences of other phenotypic characters among the populations, except for the number of seeds produced, seed width, seed length plus width and product seed aspect ratio (P < 0.01). (2) The average coefficient of variation of each character in seven populations was 21.16%, and the highest coefficient of variation was found in Kaili (P4) population (24.44%). The variation of pods among populations (29.22%) was higher than that of seeds (11.04%), and the variation mainly came from within population. (3) Correlation analysis showed that there were different degrees of correlation between the traits of G. sinensis seed. Principal component analysis showed that the cumulative contribution rate of the first four principal components (seed size, number of seeds produced by single pod, seed morphological index factor, pod length and seed thickness related factors) was 69.783%, which could basically reflect most information of the phenotypic traits of G. sinensis. According to the comprehensive evaluation of 10 species and fruiting traits of G. sinensis wild population, the comprehensive evaluation of species and fruiting traits of G. sinensis from Huishui County (P7) population was the highest. The results showed that wild G. sinensis in Guizhou Province had abundant phenotypic variation among and within population, and the variation within populations was greater than that between populations, and the variation mainly came from within population.