Abstract:

Objective: To establish a prediction method for soluble solid content (SSC) of peach. Methods: Hujingmilu
peaches were investigated for the relationships of SSC with peel color value, single fruit weight, firmness with skin, and
firmness without skin based on multiple regression analysis, linear regression and quadratic polynomial regression. Results
showed that 1) stable multiple regression equations between SSC and each of the other quality indexes could hardly be
established based on the original data; 2) stable regression equations could neither be established using respectively a
value (a*), hue angle (h) and a*/b* (b* is b value) as the independent variable based on the grouped data. There was a very
significant linear relationship between single fruit weight and SSC; however, the quadratic polynomial model established
provided a poor prediction. Poor prediction of SSC as a function of firmness with skin and firmness without skin respectively
was also observed by linear regression. Yet the quadratic polynomial regression equations SSC = − 0.128 2 × (firmness with
skin)2 + 1.403 5 × (firmness with skin) + 11.418 0, and SSC = − 0.481 8 × (firmness without skin)2 + 1.975 0 × (firmness
without skin) + 13.290 0 had higher coefficient of determination (R2) and indicated very significant regression relationship.
Conclusion: It is feasible to predict the maturity of peach by analyzing the relationships of SSC with firmness with skin and
firmness without skin based on quadratic polynomial regression.

Key words: peach, soluble solid content, regression analysis