performed in individual sensory booths with controlled humidity and temperature.
Sensory analysis. Descriptive sensory analyses were carried out according to Alonso-
Fernández et al. . Thirteen trained panellists were selected for turnip greens and
turnip tops sensory evaluation in accordance with . Fifteen attributes were
considered according to ISO norms : aroma intensity, leaf color, leaf brightness,
stalk and leaf firmness, resistance to cutting, moistness and fibrosity in mouth,
sharpness, sticks to palate, bitter, acid sweet and salty taste, aftertaste persistence, and
abnormal aroma. All descriptors were quantified using 10-cm no structured intensity
scales , except abnormal aroma which were evaluated on two-point scales.
Reference values for each attribute are shown in Table 2. In all cases a rating of 1 was
considered ‘slight’ and a rating of 10 as ‘high’.
Statistical analyses. A combined analysis of variance across environments was
performed for each sensory trait. Analysis were made independently for each of the two
plant organs evaluated (leaves and shoots). Varieties were considered as fixed effects
and environments were considered as random factors. Comparison of means among
varieties was made by Fisher’s protected least significant difference (LSD) at P=0.05
. Simple correlation coefficients (p < 0.05) among sensory traits were made in order
to determine which traits better explain the sensory attributes of turnip tops and turnip
greens. Total and individual glucosinolate content and total and individual phenolic
compound content were quantified in the same set of varieties. Part of these results was
published by Francisco et al. . Therefore, simple correlations (p < 0.05) between
these secondary metabolites and sensory characteristics were made in order to establish
the relationships between them. All statistical analyses were made using SAS .