To elucidate the varietal characteristics of cotton varieties in Yangtze river region， 100 upland cotton of the national regis-tered varieties breeding from year of 2001 to 2023 were collected. Phenotypic identification was conducted for major agronomic traits， yield-related traits and fiber quality traits. It was found that the growth period was gradually shortened， the phenotypic values of yield-related traits were gradually increased， and the fiber quality was steadily improved in the cotton variety breeding in the Yangtze river region. Employing a suite of analytical methodologies， including clustering analysis， correlation analysis， principal component analysis （PCA）， and comprehensive evaluation methods were employed to study 12 key traits. The results delineated a variation coefficient range from 3.06% to 24.37% across traits， alongside a genetic diversity index spanning 0.42 to 2.05. Notably， boll number exhibited the highest variation coefficient， while seed index and boll weight displayed the highest genetic diversity index. The results of cluster analysis showed that the tested varieties were divided into three groups， representing the varieties with poor yield and quality， the varieties with low yield but high fiber quality， and the varieties with high yield but low fiber quality. Correlation analysis indicated a significant positive correlation between the growth period and most yield-related traits， a negative correlation with fiber length and uniformity. PCA revealed that the cumulative contribution rate of the first three principal components at 70.23%， highlighting boll number， fiber strength， and fiber length as major factors contributing to phenotypic variation in cotton. Comprehensive evaluation scores ranged from 0.48 to 2.29， and the top 8 and 2 early maturing germplasm resources with excellent comprehensive performance were selected. The findings contribute valuable insights into the genetic improvement of cotton， offering both theoretical and practical significance.