Premature senescence of cotton has been occurring on an increasing scale in China, directly influencing both yield and fiber quality. The most effective way avoiding premature senescence is to obtain stay-green variety in breeding programs. Moreover, plant breeders demand a simple and efficient diagnosis method, and farther understand the genetic basis of premature senescence in cotton. The objective of this paper was to discuss the quick measurement harmless to plant, and to analyze the genetic effects for leaf premature senescence in upland cotton. The SPAD readings of the fourth leaf from the top were measured at the flowering day and also each five days after flowering in nine cotton varieties (lines). The SPAD readings at 35 days after flowering (DAF) were decreasing in premature
senescence cotton varieties (lines). Therefore, the SPAD difference between 35 DAF and the flowering day was calculated as an indicator of the reduction in chlorophyll content; the greater the rate of reduction, the earlier the senescence, and
vice versa. The degree of leaf premature senescence was expressed by the reduction of SPAD or the scale of the green-area in generations (P
1, F
1, P
2, F
2, BC
1, and BC
2) of the 33B×CJ463 cross. Generation mean analyses were conducted to explain the inheritance of leaf premature senescence. The results showed the relative importance of additive effects controlling leaf premature senescence. The estimated minimum number of genes controlling leaf premature senescence was an additive major allele at least with relatively high heritability, suggesting the early selection for late-senescence varieties (lines) was effective in the offspring. The relationships between premature senescence and leaf area per plant were different in the different segregating generations. In a few words, the results of genetic analysis and the relationships between premature senescence and leaf area measured by the reduction of SPAD and the scale of the green-area were relatively consistent.