as mean ± standard deviation of triplicate experiments. In order to compare the four organs and to prove that the differences between these organs are not an accident or due to a chance, we utilised the Kruskal-Wallis test. This test rejects the hypothesis that the organs are from identical populations. Moreover, we used ANOVA (at a significant level of p < 0.05) coupled with multiple comparisons of means (Tukey Contrasts) to investigate: the differences between different organ assays and to compare selected M. minima populations. In addition, we completed discriminative analysis, namely PCA, for each organ by the using TPC as a matrix. We evaluated the association between variables by the Pearson correlation method. The levels and the statistical analysis (ANOVA, Kruskal-Wallis test and Tukey test) are shown in Supplementary Tables S1, S2. We used R (version 3.5.1) for all statistical analyses. The utilised packages were: Agricolae, Rcmdr, car, RcmdrMisc, corrplot, tidyverse, hrbrthemes, ggplot2 and RColorBrewer 68-71 .

No abstract

Amplified fragment length polymorphism (AFLP) markers were used to characterize the genetic diversity within and among natural populations and cultivars of Hedysarum coronarium. Twelve populations within Tunisia were evaluated with three AFLP primer combinations. A total of 207 reproducible bands was detected of which 178 (86%) were polymorphic. The great discriminative power of AFLP markers and their ability to represent genetic relationships among Hedysarum plants was demonstrated. Genetic diversity within and among populations was assessed through Principal Component Analysis (PCA) and cluster analysis by using the Neighbor-joining clustering algorithm. AFLP technology has provided evidence of a high degree of intra-and inter-population genetic diversity in H. coronarium. AFLP banding patterns provided molecular markers correlated with the plants' geotropism. In addition, AFLP markers can differentiate wild accessions from cultivars. Moreover, geographical origins did not correspond to population clustering.