General linear models based on physicochemical parameters of monofloral and multifloral bee honey: Part 2.

The objective of the present study is to establish regression models defining the relations between physicochemical parameters of bee honey and the type of honey. 89 samples of bee honey (acacia honey, coriander honey, limetree honey, rapeseed honey, sunflower honey, and multifloral honey) collected from different regions of Bulgaria were used in the study. The following parameters have been defined: water content (WC); content of fructose (F), glucose (G), electrical conductivity (EC), refractive index (RI), and pH. Moderate to strong negative correlation between WC and RI has been found for all studied types of bee honey. In the samples from lime tree, coriander, and multifloral honey, positive but different in strength correlation has been observed - from weak to moderate correlation between pH and EC. The analysis of the samples with acacia honey showed differences in strength and direction of: lime tree and multifloral honey in terms of the correlations between pH and total F + G content; of lime tree, rapeseed, and sunflower honey in the correlations between F+G and RI and between WC and F + G. General Linear Models (1-5) have been compiled based on the connections between the considered physicochemical parameters and the different types of honey, which would allow subsequent estimation of the studied parameters. [ABSTRACT FROM AUTHOR]