The use of biological assays in research often creates complex Problems of data analysis. Due to the nature of bioassays which respond to the intricate interplay between multiple factors that may either increase or decrease the readout, simple arithmetic or standard linear regression analysis often fails to provide accurate values when extrapolation from a standard curve is employed. Non-linear polynomial analysis could provide a better way for expressing such complex relationships. Using a bioassay for transforming growth factor-beta (TGF-β) employing the mink lung epithelium cell line CCL-64, the authors present a new way to analyze experimental data based on non-linear polynomial statistics to provide accurate quantitative results when other simpler techniques have failed. This technique for data analysis provided accurate consistent results with acceptably small experimental variance. Using this approach, the bioactive concentration of TGF-β in normal rabbit aqueous humor was found to be 2.72 ± 0.61 ng/ml, that is, corresponding to 2. 72 ng of PURE TGF-β2, when referring to a standard curve.