MATERIALS AND METHODS

Experimental details: The M. arborea seeds were collected directly from the wild from several locations of Mt Hymettus, Athens, Greece (latitude 37°58´ N; longitude 23°48´ E; altitude 350-420 m) in 2003. After collection, immature seeds and those attacked by insects were removed and the healthy seeds were stored at 5°C and 50% RH (relative humidity) until their use.

Three germination experiments were conducted in the Laboratory of Agronomy of the Agricultural University of Athens (AUA) during the summer and autumn of 2003. Both tests were carried out in a completely randomized design under laboratory conditions, in incubators (Conviron T 38/Lb/AP) at constant temperature (30°C) and total darkness.

In the first experiment the effects of the following treatments on seed germination were evaluated: (1) immersion in hot water (100°C) for 1 min; (2) immersion in hot water (100°C) for 2 min; (3) immersion in hot water (100°C) for 4 min and (4) immersion in hot water (100°C) for 8 min.

In the second germination assay, seeds were placed in an aluminum dish and exposed to the desired temperatures in a preheated oven for 5 min: (1) at 50°C; (2) at 100°C; (3) at 150°C and (4) at 200°C. These temperatures were selected because they are likely to be reached at the soil surface or the first few centimetres below ground in fires in arid and semi arid regions (DeBano et al., 1998) while a preliminary experiment showed that prolonged exposure (>5-10 min) at higher temperatures is probably lethal for the M. arborea seeds.

In the third germination experiment the following pretreatments were tested: (1) immersion in water at room temperature (25-30°C) for 20 h; (2) immersion in concentrated (95%) sulphuric acid for 2 min followed by thorough rinsing with running water; (3) immersion in concentrated sulphuric (95%) acid for 6 min and thorough rinsing with running water and (4) immersion in a gibberellic acid (GA₃) solution (100 mg GA₃.L^-1 water) for 30 min.

There were used five replicates (Petri dishes) for each treatment, while untreated seeds were used as control for each experiment. Twenty two seeds were placed between two Whatman No. 1 paper filter disks (Whatman Ltd., Maidstone, England) in each 9 cm Petri dish and 5 mL of distilled water was added. Additionally, distilled water was added whenever there was a need to keep filter papers moist.

Seed germination was recorded every other day and expressed as a % percentage of the total number of tested seeds (Germination percentage, GP). Seeds were considered germinated at the emergence of the radicle (Bewley and Black, 1994). The germination rate index (GRI) was also calculated for each treatment using the following equation:

where G is the germination on each day after placement and 1, 2,...x represents the corresponding day of germination (Esechie, 1994). Corrected germination rate index (CGRI) was obtained by dividing GRI by the Final Germination Percentage or FGP (GP at 20 days after the beginning of seed incubation for our experiments) and multiplying by 100. The number of days lapsed to reach 50% of the final germination percentage (GT₅₀) another widely used index in order to compare relative rate of germination in perennial forage plants, was also calculated (Hsu et al., 1985).

Subsequently, three pot experiments were conducted in a glasshouse of the Agricultural University of Athens (AUA) in 2003. Minimum/maximum air temperature and relative humidity were: 17/40°C and 35/60%, respectively and the plants were subjected to a natural day length ranging between 12-15 h during the experiments.

Twenty five pregerminated seeds of each treatment reaching a radicle of 1-3 cm length were sown at 1 cm depth. Five seeds were planted in each plastic pot (15 cm in diameter) filled with 2.4 l of a slightly calcareous sandy clay loam (SCL) soil. Table 1 shows the physical and chemical properties of this soil. Irrigation was carried out with 200 mL of distilled water in each pot every two days, in order to promote plant emergence. The number of days from sowing to emergence was recorded for all the seedlings.

Statistical analysis: The percentages of germination and emergence (after arcsine transformation) and the rest raw data were subjected to one-way analysis of variance (ANOVA) using the Statgraphics statistical software package (v.5.0, Statistical Graphics Corporation, Englewood Cliffs, NJ, USA). Mean comparison was performed using Fisher’s least significant difference (LSD) method (p<0.05).

Table 1:	The physical and chemical properties of the soil used for seedling emergence of Medicago arborea
SCL = Sandy Clay Loam

RESULTS

In Table 2 it is shown that immersion in hot water for 4 min increased significantly the FGP of M. arborea seeds from 45 (control) to 71% and it was consistently the most effective pretreatment. Furthermore, seed immersion in hot water for 1, 2 and 8 min also resulted to significantly highest germination percentages than untreated seeds (Table 2).

The second germination experiment revealed that dry heat can either promote or hinder seed germination. The temperatures of 100 and 150°C were the significantly most stimulative temperature for the treatment of M. arborea seeds, but still less effective than hot water treatments. In contrast, dry heating of tree medic seeds at 200°C resulted to slightly lower germination percentage even than the untreated seeds (Table 3).

The enhancing or preventing influence of sulphuric acid on seed germination of M. arborea (due to the time of immersion) is shown in the Table 4. Moreover, immersion in water for 20 h has also improved seed germination, while immersion in gibberellic acid for 30 min was rather ineffective on M. arborea seed germination.

Table 2:	Time course for the germination of Medicago arborea seeds in response to different periods of immersion in hot water
Means (germination percentages, %) followed by the same letter(s) within a row are not significantly different at p = 0.05 Fisher’s least significant difference test

Table 3:	Time course for the germination of Medicago arborea seeds in response to several dry heating pretreatments
Means (germination percentages, %) followed by the same letter(s) within a row are not significantly different at p = 0.05 Fisher’s least significant difference test

The germination rate of seeds after an immersion in hot water for 4 min (and secondly for 2 and 8 min) was greatly enhanced, as long as the CGRI values calculated in this method was the highest among all treatments and the 50% of the final germination percentage was obtained after 11 (Table 5). Concerning the dry heat treatments, none of them was really effective on the germination rate, while the temperature of 200°C had clearly negative effects on germination percentage, although it was a quite rapid method. Concerning the germination rate of M. arboreaseeds, immersion in sulphuric acid for 2 min and immersion in water for 20 h were also among the most satisfactory pretreatments. Besides, in Table 5 it is shown that although the emergence percentage and rate of untreated tree medic seeds (control) were relatively satisfactory, some pretreatments, such as immersion in hot water for 2 and 4 min, in sulphuric acid for 2 min and in water for 20 h contributed to the further optimization of seedling emergence.

Table 4:	Time course for the germination of Medicago arborea seeds in response to several pretreatments
Means (germination percentages, %) followed by the same letter(s) within a row are not significantly different at p = 0.05 Fisher’s least significant difference test

Table 5:	Seed germination rate and seedling emergence of Medicago arborea in response to different pretreatments
Means followed by the same letter(s) within a column are not significantly different at p = 0.05 Fisher’s least significant difference test within each experiment. A: First, B: Second and C: Third germination and emergence assay

On the contrary, it has to be noted the negative effect of very high temperatures (e.g., 200°C) and prolonged immersion in sulphuric acid, not only on seed germination, but on seedling emergence, too.

DISCUSSION

Our results indicated that all of the hot water treatments increased significantly the germination and emergence percentage and rate of M. arborea. The beneficial effect of hot water bath on seed germination is common among perennial legumes widespread in arid and semi-arid zones (Clemens et al., 1977; Muhammad and Amusa, 2003). Furthermore, it is well documented that the immersion of the seeds in hot water may lead to the rapture of the coat wall, allowing water to permeate the seed tissues and causing seed germination and further rapid emergence of several other species of arid regions (Agboola and Etejere, 1991; Emongor et al., 2004).

Some of the tested dry heat treatments were also very effective in terms of germination percentage (100 and 150°C for 5 min) likewise in other woody species (Tarrega et al., 1992; Vilela and Ravetta, 2001) while only the temperature of 200°C showed a marked inhibition of germination and emergence, similarly to other woody species (Reyes and Casal, 2006). This positive (or damaging at high temperatures or prolonged exposure) effect of dry heat is well documented in many species, while it underlines the potential effect of fire on seed germination and seedling emergence of Medicago arborea in Mediterranean-type ecosystems (Arianoutsou and Margaris, 1981; Thanos and Goerghiou, 1988; Qaderi and Cavers, 2003).

The beneficial effect of sulphuric acid on seed germination of several Medicago species has already mentioned (Ruiz and Devesa, 1998). Furthermore, concerning the germination promotion of the seeds treated with sulphuric acid and their possible damages caused by prolonged immersion (6 min in our case) or low integumental resistance, there are many reports in several arid-adapted legumes (Sy et al., 2001; Vilela and Ravetta, 2001).

In a contrary way to some-and in accordance to-others hard-seeded legumes native to arid and semi-arid environments, water immersion was also effective on tree medic seed germination and emergence (Sy et al., 2001). Our results confirm previous reports on the positive effect of water immersion on M. arborea seed germination (Huxley, 1992). However, gibberellic acid treatment (immersion for 30 min) did not have significant effects on this species germination and emergence, implying that probably the primary control of germination in M. arborea seeds resides in the seed coats, i.e., the type of seed dormancy is not physiological but mainly physical dormancy, likewise many leguminous plants (Travlos et al., 2006).

In addition, there was a significantly positive correlation between germination percentage (FGP) and germination rate (CGRI and GT₅₀) (r² = 0.88, p<0.05 and r² = -0.6, p<0.05, respectively, suggesting that in our experiments on tree medic, the rapid germination was closely associated with the high germination percentage. Additionally, it seems that M. arborea germination was highly significantly correlated with the emergence of the seedlings (r² = 0.86, p<0.05) and the rate of seedling emergence (r² = -0.79, p<0.05).

Seed dormancy is usually associated with the factors of the protective covering, the seeds coat or the enclosed embryo. Our results indicated that immersion in hot water, in sulphuric acid for 2 min and in water for 20 h can induce significantly seed germination. These results are in agreement with previous reports, providing evidence that Medicago arborea seeds certainly involve physical dormancy (Bass et al., 1998) while there are differences between the several species, ecotypes and specific pretreatments (Uzun and Aydin, 2004). This physical (coat-imposed) dormancy favors the accumulation of persistent seed banks in the soil, spreads germination over time and increases the chance that some seeds will germinate, establish and complete the life cycle successfully (Gutterman, 1993) but is exceptionable for the cultivation of the species.

Medicago arborea is not only included to the national strategies for rangeland rehabilitation in many countries, but it is often recommended as a model legume for revegetation, regeneration and biological reactivation of degraded lands in semi-arid mediterranean areas (González-Andrés et al., 1999; Valdenegro et al., 2001; Amato et al., 2004). Unfortunately, in many cases the cost of the establishment by means of transplanting is relatively high (Bouzid and Papanastasis, 1996) while direct seeding has been shown to be successful in establishing native trees and shrubs in low rainfall environments (Knight et al., 1998). Consequently, there is a need of optimization of seed germination, seedling emergence and first growth of tree medic, in order to exploit the inarguable potential of this species. Therefore, such information can be of considerable practical importance and specific field experiments, monitoring and further laboratory studies must be continued in order to optimize rapid and uniform seed germination and seedling emergence of M. arborea.