Seed surface area (SSA) may be estimated by WinSeedle Image Analysis System, transformations of seed volume (Hodson method) or by seed index. If fuzzy seed index were related to SSA, SSA could be estimated without delinting seed. Our objective was to compare different methods of estimating SSA using fuzzy, acid- and flame-delinted seed. Seed were produced at two Arkansas locations and included eight diverse genotypes. After ginning, 100-fuzzy seed samples from a total of 10 replicates of field plots were randomly selected and weighed. Samples were flame-delinted, and then weighed, volume displacement determined, and scanned by digital analysis. The same seeds were then acid-delinted, weighed and SSA recalculated in the same manner. Location by genotype and genotype by delinting method interactions were significant for seed index, Hodson SSA and WinSeedle SSA. Volume means were significant for all main effects. Variation among genotypes was relatively similar for each parameter. Lowest correlations were found associated with volume of flame-delinted seed. Otherwise, seed index of fuzzy and acid-delinted seed were highly correlated with seed volume and both SSA measurements. Fuzzy seed index explained more variation in WinSeedle SSA than any other parameter. A regression equation may be used to transform fuzzy seed index to estimated SSA. The strong relationships of fuzzy seed index to delinted seed index, seed volume and the WinSeedle SSA measurements indicate that it can be used as an indicator of SSA when estimating fiber density.