The plant material used in this work consists of three Moroccan alfalfa cultivars: the Marocaine cultivar originating from the Tata region, the Ifrane cultivar originating from the Guelmime region, both of which are located in the arid and semi-arid regions of the Anti-Atlas. The Demnate cultivar originates from the High-Atlas mountains, in addition to an introduced variety, Siriver. The seeds were surface disinfected with 6% sodium hypochlorite solution for 5 minutes and thoroughly washed 7 to 8 times with distilled water to remove traces of sodium hypochlorite solution. The study was conducted in a completely randomised design with three replicates of each treatment and40 seeds per Petri dish were used to test the germination of the four lucerne varieties. The seeds were placed in Petri dishes containing two layers of filter paper soaked with 0 (distilled water), 50, 100, 150, 200, 250, and 300 mmol∙L⁻¹ NaCl. Germination was carried out in an incubator in constant darkness at 25˚C.

The seeds were used in a factorial design combining two factors in a randomized complete block design. The two factors were variety with 4 levels and salinity with 7 levels. Three replications with 40 seeds per box were used for the germination test with three replications for each treatment and for each cultivar. Germination was monitored daily for 8 days, counting germinated seeds every 24 hours; radicle extension is the event that ends germination and marks the start of seedling growth [17] . Seeds are considered germinated when the radicle has extended at least 2 mm beyond the seed coat [18] .

Several germination parameters were calculated to characterize salt tolerance, including the germination rate (GR), final germination percentage (FGP), mean germination time (MGT) [19] , and germination index (GI) [20] .

The germination rate was calculated by using the following formula:

GR = ∑ n i s × 100

The final germination percentage formula is:

FGP = n f s × 100

The mean germination time is a measure of the rate and time-spread of germination (lower values indicating faster germination) [19] . It was estimated as:

MGT = ∑ n i t i n i

The germination index (GI) described by Maguire [20] is a measure of seedling vigor and includes not only germination but also emergence characteristics. High values obtained using this expression mean higher seedling vigour. The value of GI was higher when the seeds germinated earlier. GI was calculated by using the following formula:

GI = [ n 1 1 + n 2 2 ] + ⋯ + [ n x x ]

where, n_i represents the number of newly germinated seeds on day i; n_f represents the number of germinated seeds at the end of the experiment; S represents total number of seeds in an experimental unit (germinated/viable/non-germinated); t_i is the number of days from the start of the experiment to day i; n is the number of germinated seeds at each day after sowing and 1, 2, … and x is the corresponding day of germination.

The kinetics of germination is expressed by the number of seeds germinated at 1, 2, 3, ..., 7 days after sowing. It is a parameter that makes it possible to better understand the ecological significance of the germination behavior of the studied cultivars, as well as the series of events that begin with the stage of water absorption by the seed and end with the elongation of the embryonic axis and the emergence of the radicle.

We determined the physiological responses of two varieties contrasting in their responses to salinity according to the results of germination tests; Ifrane, being the most tolerant and Siriver, the less tolerant. After exposure to 0 or 200 mM NaCl for 7 days, the seeds developed young seedlings with radicals, hypocotyls and cotyledons. The length and the fresh weight of the young seedlings developed were measured.

All measurements were subjected to a two-way analysis of variance (ANOVA II) with cultivars and salinity treatments as factors, followed by a Student-Newman-Keuls test. A difference was considered statistically significant when p < 0.05. All statistical analyses were performed on Excel.

For the 4 cultivars, the two-way ANOVA revealed a highly significant main effect of both cultivar and salinity regarding final germination percentage and germination rate index (p < 0.001) (Table 1, Table 2). However, the existence of a significant interaction between these two effects (F = 12 for (FGP) and F = 4 for (GRI)) indicated that the populations studied did not respond similarly to the effect of salt at a given concentration of NaCl.

The effects of salt on the final germination percentage (FGP) of alfalfa were shown in Table 3. For all the cultivars studied, Ifrane, Marocaine, Demnate, and Siriver, the highest FGP occurred in the control with 100%, 99%, 96% and 93% respectively. However, increased salinity reduced the final germination percentage of alfalfa seeds compared to the corresponding controls (Table 3). The reduction was more pronounced with increasing NaCl concentrations. For each concentration of NaCl, a significant variation in both seed germination potential and seedling vigor was observed among the four cultivars studied. In addition, all cultivars showed an increase in Mean Germination Time (MGT), indicating that seeds take more time to germinate with increasing salinity (Table 4).

Comparing the mean values of the different genotypes (Table 3), the latter can be divided into two groups with different responses to sodium chloride: The first, composed of the Ifrane and the Marocaine genotypes, showed the highest germination rate of 81% and 75% respectively, while the second group consisted of the Demnate and Siriver genotypes, which had the lowest germination rates of 55% and 50% respectively.

At the control treatment (0 mM) and the lowest stress treatment (50 mM), all the cultivars formed a homogeneous group which showed no significant difference with the control for FGP. At this salinity level, the seeds germinated rapidly and no significant change in germination speed was noticed (MGT not different from the control and between cultivars) (Table 4).

The means of FGP at 100 mM and 150 mM NaCl showed no significant differences, 89% and 84% respectively. However, the results showed considerable variation in the response of seed germination to salinity among the cultivars studied at these two salt concentrations. Thus, at 100 mM NaCl, two groups were distinguished, the first was formed by the cultivar Ifrane, which showed a very high percentage of germination (FGP = 98%) and a high germination index

Df: Degree of freedom, MC: Mean of square, F: Ratio of variance, ***Significant at P = 0.001.

Df: Degree of freedom, MC: Mean of square, F: Ratio of variance, ***Significant at P = 0.001.

In the same row (a, b, c, d, e, f), significant differences (test Student, P < 0.05) if the letters are different. The averages followed by the same letter (A, B, C, D) are not different at the 5% threshold (test Student).

Values followed by the same letters in a row are not significantly different at p < 0.05, (test student).

(Figure 1) (GI = 94.78), and the second formed by the cultivars Marocaine, Demnate and Siriver, with no differences for FGP (93%, 83% and 85% of FGP respectively) and for GI (82.49, 76.5 and 78.57 respectively). For the MGT, there were no significant differences between the four cultivars studied. At 150 mM NaCl, the values of FGP and MGT recorded for each cultivar were significantly different and the means showed no difference with those recorded at 100 mM NaCl. Thus, the Ifrane cultivar showed earlier seed germination MGT (1.25), highest FGP (96%) and the highest seedling vigor (GI = 92.04), whereas the Siriver variety showed the highest MGT (1.84), lowest FGP (70%) and lowest seedling vigor (GI = 51.76).

At 200 mM of NaCl, the results showed considerable variation in the response of seed germination to salinity among the cultivars studied. The Ifrane cultivar showed a high germination percentage of 78% and an earlier seed germination (MGT = 1.56), followed by the Marocaine cultivar with 68% germination percentage and MGT < 2 days. The germination percentage of the Demnate cultivar showed a significant reduction at this salinity level, 53%, the MGT value registered were also under 2 days (1.73). However, the Siriver variety seemed to be severely affected by this concentration, it showed the lowest final germination percentage (FGP = 41%) and the seed germination took longer (MGT = 2.16).

At 250 mM NaCl, the genotypes studied can be divided into two groups according to their germination response to this salinity level. The first group includes the cultivars Ifrane and Marocaine, which showed the highest FGP (73% and 64% respectively), with no difference from those registered at 200 mM NaCl. The second group includes the cultivars Demnate and Siriver, whose germination was strongly affectedby this concentration in the same way, showing the lowest FGP (25% and 13% respectively), with a significant difference with those registered at 200 mM. The MGT values registered indicate that the Ifrane cultivar showed the earliest seed germination (MGT < 2 days) at different salinity levels, while for the Siriver cultivar, germination takes a lot of time to take place (MGT > 2 days); the Marocaine and Demnate cultivars occupied an intermediate position between the tolerant cultivar (Ifrane) and the most sensitive (Siriver). The germination index at this concentration showed a high seedling vigour for the Ifrane and the Marocaine cultivars (GI = 55.19 and GI = 42.43 respectively), whereas the Demnate and the Siriver cultivars showed a lower seedling vigour (GI = 21.84 and GI = 8.71 respectively).

The high salt dose (300 mM of NaCl) caused a significant reduction in seed germination and delayed its initiation in all cultivars. The cultivar Siriver was the most affected by this concentration, almost no germination took place, with a germination percentage not exceeding 2%, and the time to germination gradually lengthened (MGT > 3 days).

The effect of NaCl on the germination behavior of alfalfa results in an increase in the latency time and a decrease in the germination rate (Figure 2). The study of the germination kinetics shows that an increasing salt concentration causes a delay

in the germination. These results show that the Ifrane cultivar is the most tolerant to salt stress, whereas the Siriver cultivar is the most susceptible at the germination stage. Both cultivars with contrasting salinity responses were used for further characterization.

Plant biomass and yield data are used as an agronomic index to define stress tolerance. We analyzed the fresh weight and length of seedlings of two alfalfa genotypes under salt stress (200 mM of NaCl) for 7 days after imbibition. Our results showed that alfalfa growth (fresh weight and length of seedlings) is negatively affected by increasing salinity. Under normal conditions, the Siriver variety had lower fresh weight and seedling length than Ifrane ecotype (Figure 3(a) and Figure 3(b)). However, the fresh weight and length of seedlings of both cultivars were significantly inhibited by 200 mM NaCl (P < 0.01). The reduction was greater in Siriver (262 mg fresh weight and 3.9 mm seedling length) than in Ifrane (314 mg fresh weight and 7.8 mm seedling length) with a significant difference (P < 0.01), indicating that the latter is more tolerant to salt stress than the former.

Soil salinity is one of the major plant stressors, especially in arid and semi-arid regions and can severely limit crop production [21] . The results obtained in this study showed that the salt treatment produced depressive effects at the developmental stages studied. Monitoring of the germination process revealed that salt stress was responsible for both a delay in seed germination and a reduction in seedling establishment in the four alfalfa genotypes studied. This suggests that seed germination takes longer under salt stress. However, a significant difference was noted between the populations studied in their reactions to salt. The Ifrane cultivar had the highest germination percentage and the lowest average germination

time, while the Siriver variety had the lowest germination percentage and the highest average germination time. On the other hand, the Marocaine and Demnate ecotypes showed an intermediate position, with the Marocaine cultivar clearly superior. Thus, the Moroccan alfalfa genotypes used in this study proved to be more tolerant than the variety (Siriver) introduced, testifying to the richness of the local material in terms of tolerance to salinity, and to its usefulness and effectiveness in maximizing the agricultural productivity of alfalfa in areas affected by salinity. The effect of salinity on alfalfa seed germination and seedling establishment has been studied by several researchers, who have shown a wide variation in salt tolerance between alfalfa varieties at different stages of development [12] [22] [23] [24] .

The reduction in seed germination affected by salt stress may be due to the direct effect of sodium chloride on embryo growth [25] . Salinity affects seed germination and reduces it by decreasing the rate of water absorption by osmotic effect and increasing the rate of ion efflux by modifying hormonal and enzymatic activities due to the toxic effects of sodium and chloride ions [12] . The emergence of the radicle is mainly hindered by a reduction in the water potential gradient between the seed and the external environment [14] . Water uptake is necessary to complete an endogenous physiological process during seed germination. Salinity also disrupts the hormonal balance of plants [26] and reduces the use of seed reserves [27] .

A significant variation in germination potential revealed by ANOVA was observed between Moroccan alfalfa populations with respect to the different NaCl concentrations used. This variation is partly a function of their geographical origin. Indeed, the Ifrane and Marocaine cultivars from arid and semi-arid regions (Anti-Atlas), which are representative of the “African” type, are the most tolerant to salinity and showed the highest FGP and strong seedling vigour for high NaCl concentrations. The Demnate cultivar from the High-Atlas mountains, which is related to the ‘Falcata’ type, and the Siriver variety were sensitive to salt stress. The results obtained indicate a certain correlation between the origin of the cultivars and resistance to salinity. This result can be explained by the great diversity of traditional oasis and mountain agro-ecosystems and by the adaptation of populations to the pedoclimatic conditions of their ecosystem of origin, such as drought, cold and salinity.

Similar results have been reported in several studies [15] [28] . Genetic material from Indian and African arid areas excelled in resistance to NaCl during germination, while genetic material related to the European origins ‘Falcata’ was the least resistant [29] .

A reduction in seedling growth resulting from an increase in salinity level has been reported in alfalfa and certain other species [12] [22] [23] [30] [31] . The results obtained in our study showed that salt treatment negatively affects the growth of young seedlings by reducing fresh weight and seedling height, with significant differences between ecotypes. The reduction in fresh weight and seedling length was greater for the Siriver variety than for the Ifrane cultivar. Seedling growth is one of the most important indices of salt tolerance, as shown by various studies [24] [30] [32] . This confirms the concept that plant biomass data are proposed as agronomic parameters to define stress tolerance. The results obtained show the tolerance of the Ifrane ecotype and the sensitivity of the Siriver variety at the germination and young seedling stages.