In this study, we investigated variation for tolerance to water deficit in 47 lines of Medicago truncatula Gaertn., M. ciliaris (L.) All., and M. polymorpha L. collected from different ecogeographic regions in Tunisia. Plants were cultivated in the greenhouse under 100 and 30% of field capacity and they were harvested at the flowering stage. Nineteen morphophysiological parameters were measured for each species. Our results showed that variation of parameters were explained by the effects of species, line within species, treatment, species × treatment interaction, and treatment × line within species interaction. Drought treatment explained the most variance for the measured parameters. Of the 19 traits, 12 showed significant differences for tolerance to water deficit between M. truncatula, M. ciliaris, and M. polymorpha. M. ciliaris was the latest‐flowering and it exhibits the highest biomass under both treatments. High broad‐sense heritabilities (H2) were noted for most of parameters under control treatment and drought stress. Positive correlations were found between transpiration rate (E) and aerial and root fresh weights, and between photosynthetic rate (A) and stomatal conductance (gs), and between relative water content and number of nodules. Studied lines formed five groups based on drought response indices (DRI). A first group is formed by 23 most tolerant lines, a second group and a fourth group are constituted by 17 moderately affected lines, and a third group and a fifth group have 8 sensitive lines. Tolerant lines of three species can be good candidates in future breeding programs.

Medicago truncatula is an omni-Mediterranean species grown as an annual forage legume. In addition to its small genome size and simple genetics, M. truncatula harbors several attributes which make it an attractive model legume. In this study, we investigated morphological and photosynthetic responses to water deficit in parental genotypes of M. truncatula recombinant inbred lines. Ten parental lines were cultivated under three water regimes (100% of field capacity (FC), 50% FC and 30% FC) and were harvested at flowering time and at the end of their lifecycle. Results from ANOVA showed that variability of measured parameters was explained by the effects of line, treatment and their interaction with treatment factor recorded the highest values. Out of the 27 traits, 14 were influenced by the line x treatment interaction. High to moderate broad-sense heritability (H²) were observed for most of the traits under control treatment and drought stress. Most of the correlations between measured traits were positive under the three water regimes. The flowering time was positively correlated with aerial and root growth rate. The tolerance of lines to water deficit seems to depend, in particular, on their ability to maintain higher photosynthetic activity. In 30% FC, principal component analysis clustered lines into two groups. The Jemalong A17 line was the least affected for most phenotypic parameters. Hence, all populations of recombinant inbred lines derived from crosses between Jemalong A17 and the remaining lines were useful for the identification of the genetic determinants for tolerance to water deficit in M. truncatula.