Comparisons of mineral and non-mineral prediction methods for bitter pit in ‘Honeycrisp’ apples

Abstract

Bitter pit is a physiological disorder that can develop during storage and result in major economic losses of susceptible apple cultivars such as ‘Honeycrisp’. Prediction methods to determine the risk of disorder development have focused primarily on mineral composition of the fruit, but there has also been interest in non-mineral based methods. In this study, we have compared non-mineral methods with mineral analyses for bitter pit prediction. Fruit of ‘Honeycrisp’ were harvested three weeks before anticipated commercial harvest and at commercial harvest. Mineral contents in peel from the calyx-end were measured in fruit at both harvest times. In year 1, fruit were kept at 20 °C (passive method), dipped in 2000 mg L−1 2-chloroethylphosphonic acid (ethylene method) or 0.8 M MgCl2 (Mg method). Treated fruit were then kept at 20 °C for three weeks. At commercial harvest, fruit were stored at 0.5 or 3 °C with and without 1 week of conditioning at 10 °C. The Mg method showed toxicity on the fruit, which was hard to distinguish from bitter pit. Therefore, in year 2, only the passive and ethylene methods were used, and fruit from the commercial harvest were stored at 3 °C with or without conditioning. In year 3, only the passive method was used because of its simplicity and lack of registration of ethephon for this purpose. Fruit from the commercial harvest of 38 orchards in three growing regions were stored at 3 °C after conditioning. Fruit were stored for two months in year 1 and four months in years 2 and 3, and assessed after 4 d at 20 °C. Multivariate analysis shows that the passive and ethylene methods for fruit harvested three weeks before the anticipated harvest had higher or similar correlations with the actual bitter pit after cold storage than those from minerals. Although actual bitter pit incidence was higher than that predicted from the passive method, the method results in predictions that are similar to mineral-based predictions, and also is more straightforward for commercial application.