Effect of saline water on the vase life of Narcissus poeticus L. flowers

For the cut flower cultivation, plants were grown under field conditions. The own-grown bulbs (Faculty of Horticulture, Mendel University in Brno, Czech Republic) of a cultural form of poeticus L. were originally collected from a local population in Hungary (Pákozd, 4712.9380 N, 1832.7940 E) in 2011 and propagated over the years in the experimental area of the faculty. After lifting and drying in the middle of June 2021, the bulbs were stored at 17°C during summer and planted on 2 October 2021. The soil is brown, with a pH (H₂O) of 7.7, a pH (KCl) of 7.34 and an electrical conductivity (EC) of 330 μS · cm^-1, and the nutrient supply is N-poor, P-very good and K-medium. The soil Na content (measured in March) was 12 mg · kg^-1. After spring sprouting, at 70–100 mm leaf length, a single application of 5 g · m^-2 N was applied, with an irrigation dose of 15 mm. Subsequently, two more applications of 15–18 mm of irrigation water were carried out the last 7 days before flowering.

Flower picking occurred in the early morning (6:30–7:30) of 29 April 2022. The flowers were picked at the pencil stage (scapes standing erect), and after picking, the scapes were cut to a uniform length (190–200 mm).

After cutting, the scapes were placed in distilled water at 4°C for 24 hr. Then, the scapes were rinsed once more with clean distilled water to remove any secreted sap (Armitage and Laushman, 2003; Bittnerová and Martínek, 2007). During the experimental period, transparent 250-mL plastic dishes (‘vases’) were used to store the flowers, with 200 mL of solution in each bottle, at an initial salt concentration corresponding to the treatment. The height of the water column in each vase was 55 mm, into which the 65–70 mm long scapes were diagonally immersed. Laboratory pure NaCl was used to prepare saline solutions. The salt concentrations used in each treatment and the basic chemical properties of the salt solutions are presented in Table 1. Distilled water (0 mM NaCl) was used as the control. The solutions used for the treatments were prepared by diluting the stock solutions of 500 mM concentration prepared on an analytical balance (Kern 770, Kern and Sohn GmgH, Balingen, Germany).

At the start of the experiment, 10 scapes were placed in each vase, and three replicates of each treatment were set up. The vases were kept in a room with an air temperature of 20–24°C. Excluding the monitoring time, no artificial lighting was used, and the plants were protected from direct sunlight. The scapes were kept in the same solution (no change of vase water) throughout the experiment.

The flower opening status recorded every 12 hr for six representative treatments is shown in Figure 1. Graphs for all treatments are summarised in Figure S1 in the Supplementary Material. The flowers remained in the bud stage for a relatively short period of time, with almost all flowers opening within 48 hr of being placed at room temperature (max. 1 bud stage flower per vase). The main differences between treatments were observed in the length of the opening stage and senescence process.

Figure 1.

For the experiment as a whole, a time of 35 hr to 40 hr was needed for half of the plants in the vases to reach the open flower stage (FO+FOfb+FOh). It took 52–60 hr for all flowers to open (Figure 2). The fastest flowers opened in the 10 mM solution, but overall, no correlation was found between salt concentration and opening speed.

Figure 2.

In the 100 mM salt solution, the bouquets were in full flowering less than half as long as in distilled water (44 hr and 92 hr, respectively). From this point of view, at concentrations of 5–10 mM, the flowers appeared similar to the control, and then the time was shortened as the concentration increased. There was no significant difference between the results obtained at concentrations of 15–100 mM. A polynomial trend curve (R² = 0.88) was fitted to the series of results (Figure 3).

Figure 3.

In the solutions with high salt concentrations, the senescence of the flowers started faster and reached the 50% threshold (half of the flowers started to wilt) faster, as well as the stage when all flowers started to wilt (Figure 4). The complete wilting (AF3) stage followed a similar pattern.

Figure 4.

In the control (without any salt), half of the flowers started to wilt after 162 hr from the start of the experiment, whereas in the 90 mM and 100 mM solutions, it took 124–125 hr. Low salt concentrations (<40 mM) were not distinguishable from the 0 mM treatment by statistical analysis. The results for higher salt concentrations (40–100 mM) were also not distinguishable from each other.

The start of senescence of all flowers was observed after 176 hr in the control, whereas it occurred 32–36 hr earlier in the 70–100 mM salt treatment. However, up to a concentration of 40 mM, water quality did not significantly impair the usability of the flowers in this aspect.

The complete senescence of half of the flowers (AF3 stage) occurred 170–177 hr after the start of the experiment in the no-salt and low-salt treatments (0–40 mM), while in the 80–100 mM salt treatments, it occurred 20 hr earlier. The effect of the concentration increase was well described by linear regression (the 10 mM salt concentration increase meant a 2.9-hr reduction in the total vase life).

In terms of total flower senescence, the 0 mM, 5 mM, 10 mM and 40 mM salt treatments were less harmful than the rapid wilting of flowers treated with 100 mM. Over the concentration range studied, the total vase life decreased at a slight slope, as described by a second-order curve.

In the control and low-salinity treatments, where the flowers were in the full flowering stage for a long time, it took only 28 hr from the onset of the first flower senescence to the total wilting of bouquet. In the 70–100 mM treatments (where the opening stage started earlier), this time was 48–52 hr.

Summarising the results presented earlier, the vase life in days is shown in Figure 5. From the vase placement (white bud stage) to the opening of half of the flowers, vase longevity was >6.5 days in the concentration range of 0–15 mM. The longevity was significantly shortened only in the concentration of >30 mM NaCl compared to the control and low-salt treatments. The vase life calculated from the opening of half of the flowers was 5.2–5.3 days in the 0–15 mM concentration range. In the 40–100 mM range, there was no significant difference in the vase life. There is a well-described relationship (R² = 0.94–0.95) between the increase in salinity and the length of the vase lifetime (according to both calculation methods) with a second-order curve.

Figure 5.

The changes in the ornamental value over time are shown in Figure 6. Considering that there were only a few flowers with tepals that were not extended in plane at full flowering, the temporal presence of values close to the maximum ‘10’ score was similar to the parameters associated with 100% full flower in Figures 2 and 3.

Figure 6.

In the initial period, achieving ornamental value scores (OVSs) of 5 and 9 showed no difference in the effect of the salt treatments (Table 3). For the whole experiment, the flower bouquets achieved an average score of 5 after 28 hr from the start and an average score of 9 after 46 hr.

There were significant differences in the decline of the numbers of opened flowers as a result of salt treatments, especially in the high-NaCl concentration treatments. However, there was no significant difference at the time of falling below the 9- and 5-point thresholds in the 0–30 mM concentration range. There were also no significant differences between the 40 and 90 mM salt treatments (especially when the value decreases below a score of 9). However, the regression line fitted to the mean values per treatment showed a strong correlation between the salinity of the solution and the onset of decline in the ornamental value of the flowers. The time of decline below the 5-point threshold (when wilted) occurred after 167–169 hr from the onset in the 0–30 mM treatments and after 30 hr less in the 80–100 mM treatments.

The ornamental value retained 9 or higher scores for 110–113 hr in the 0–15 mM solutions, twice as long as in 100 mM solution. With 5 or higher scores, flowers persisted 30 hr more in the 0–15 mM treatments than in the 80–100 mM treatments (Figure 7). A polynomial trend curve was fitted to the decrease in these durations with good accuracy (R² = 0.90–0.97).

Figure 7.

In the 0–20 mM treatments, the ornamental value decreased from 9 to 5 after 9–13 hr and then from 5 to 0 after an additional 19–21 hr. In contrast, in the 50 mM treatment, the two durations were 19 hr and 35 hr (9 to 5 and 5 to 0 decreases, respectively), and in the 100 mM treatment, the durations were 34 hr and 28 hr (Table 4).

The change in the total ornamental score decreased linearly with the increase in saline concentrations. Starting from the 120 values with the 0–10 mM treatments, the 10 mM concentration increase resulted in a 3-score decrease (Figure 8).

Figure 8.

Colour changes were first observed on the scapes 120 hr after the start of the experiment (Figure 9). At that time, yellowish green scapes were recorded in the 50 mM or higher concentration treatments (at that time, the flowers were still in ‘FO’ condition, except for the 100 mM treatment, Figure 4). Subsequently, in the two treatments with the highest salt concentrations, the scape colour changed very rapidly; more than half of the scapes (the part above the water) discoloured within 12 hr. After an additional 12 hr, completely yellowed scapes were found in these treatments. In the two treatments with the highest salt concentrations, no scapes were green 144 hr after the start of the experiment. 24 hr later (168 hr from the start of experiment), yellowish-green scapes were also found in the 30 mM solution, and in the more-concentrated treatments, the scapes were already drying. However, the colour changes were always only observed above the portion of the scape standing in the solution. In solutions with concentrations >70 mM, all scapes were yellow when the flowers were fully open, and many (57%–87%) had dried. In saline solutions with concentrations >30 mM, the scapes remained green until the flowers were fully open. In higher concentrations (>30 mM) of saline, we also observed broken scapes.

Figure 9.

The cut surface of the scape in saline solutions with a concentration of <30 mM remained intact until the end of the experiment. In the 30 mM and more-concentrated solutions, the length of the damaged part of the scape increased proportionally with increasing concentrations (Figure 10). The coefficient of X of the well-fitting linear regression curve (R² = 0.97) shows that in the concentration >20 mM, a concentration increase of 10 mM increased the scape length decomposition by 3.5 mm at the 108th hr of the experiment and by 6.6 mm at the final senesce.

Figure 10.

No difference was found in the diameter of the flowers at full flowering between the treatments. At full opening, the average petal diameter was 60–62 mm and the crown diameter was 15–16 mm. Likewise, there was no significant difference in the fresh weight of flowers and scapes at the same time (Table 5).

At full flowering, no correlation was found between the salinity of the solution and the dry weight of the plant parts. At the senesced stage, the dry weight changed with the increase in salt concentrations. A polynomial regression curve could be fitted to the mean values with a high degree of accuracy (scapes: R² = 0.90; flowers: R² = 0.81). For the scapes, the lowest values were in the middle salt range, where for the flowers, the highest values were recorded (Figure 11). For salt concentrations >15 mM, the dry weight of the flowers exceeded that of the scapes, with the highest flower-to-scape weight ratio being 1.4–1.5 (40–60 mM NaCl concentration).

Figure 11.

The dry matter content was not clearly affected by the salinity of the soaking solution. At the senesced flower stage, the average dry matter content of the scapes was identical to the previous samples. However, the dry matter content of the flowers changed significantly during the wilting, 1.2–1.6 value increase over the 3 days (Figure 12). In case of the dry matter content of the flowers, the interaction between the salt concentration and sampling time was significant.

Figure 12.

The chlorophyll content of the scapes initially increased with the increase in salt concentrations and then decreased significantly (Figure 13). At the first sampling (full flowering), the chlorophyll a and b contents were the highest in the scapes of the flowers in both the 10 mM NaCl and 15 mM NaCl solutions, and they were the lowest at concentrations of 50 mM and above. At senescence, the chlorophyll a content was the highest in the 5–20 mM solutions, followed by the lowest values for samples from 40 mM and higher concentrations. In the same sampling, the chlorophyll b content was the highest in 10–15–20 mM solutions, followed by a significant decrease from the 40 mM treatment onwards. However, the latter values were similar to those measured in the scapes in the control.

Figure 13.

Comparing the two sampling times, parallel curves with almost identical slopes were obtained for chlorophyll a, with lower values in all treatments at the end of the experiment than in the middle of the experiment. In solutions with salt concentrations of >40 mM, the chlorophyll b content of scapes did not change significantly between the two samplings.

The ratio of the two chlorophyll forms varied with salt concentration. Although the amount of chlorophyll a was higher than that of chlorophyll b in all cases, the ratio was significantly higher in full flowering at concentrations of >40 mM than at the first sampling. The chlorophyll b content decreased more than the chlorophyll a content. At the end of the experiment, the highest ratio was measured in the control, while the concentration effect was not as strong in the solutions containing NaCl as it was in the previous sampling.

Both the chlorophyll a and b contents and their ratios can be fitted with a polynomial function (R² = 0.72–0.96).

The carotenoid content of the scapes was the highest at the first sampling (full flowering) in solutions with 10–30 mM NaCl concentrations and then varied according to a polynomial curve (R² = 0.89). In the 10–20 mM treatments, the carotenoid content decreased significantly upon senescence. However, in the treatments with salt concentrations of >30 mM, the carotenoid content of the scapes increased significantly during flower senescence (Figure 14).

Figure 14.

The plants took up both sodium and chlorine from the solution in the vases (Figures 15 and 16). The regression curves between the solution concentration and the increase in the sodium content of the scapes and flowers were well described by linear and second-order equations. The combined sodium content of scapes and flowers increased in linear proportion to the concentrations of the solutions in the vases (10 mM NaCl concentration increase to 2.25 g · kg^-1 of the sodium content at flowering and 27% more at the flower senesced stage). Among the organs, the increase in the sodium content was higher in the scapes, but sodium was also transferred from the scapes to the flowers. After the main flowering stage, the increase in salinity continued, which was very striking for flowers at high-salt concentrations (in the 100 mM NaCl solution, a threefold increase between 108 hr and 168 hr).

Figure 15.

Figure 16.

Similar to the sodium content, the chlorine content was also higher in the scapes than in the flowers (at the end of the experiment in the 100 mM NaCl solution, the chlorine content in the scapes was 2.7 times higher value than that in the flowers). The chlorine content in the scapes increased steadily with the increase in the solution concentration, and there was a large difference between the two samplings in case of treatments with concentrations of >30 mM, i.e. the chlorine content also increased significantly during senescence. In the flowers, there was no significant difference between the values in the 30–70 mM range at full flowering, but the values measured after senescence increased almost steadily. In the 100 mM solution, however, there was no significant difference between the two samplings; the flowers almost reached the maximum chlorine content at the full flowering stage. Similar to the sodium content, a polynomial regression was well fitted to the curves for the chlorine content.

The potassium content was much higher in the flowers than in the scapes (Figure 17). In the latter, the potassium concentration decreased significantly with the increase in the salt concentration (well characterised by a second-order polynomial curve). The combined potassium content of stems and flowers did not show a close correlation with the salt concentration. However, in the senesced flowers, the value decreased significantly in flowers kept in solutions with high concentrations (>60 mM).

Figure 17.

There was a negative correlation between the sodium and potassium contents of the scapes. As the salinity of the solution increased, the potassium content decreased proportionally (Figure 18). The increase of 10.0 mg · kg^-1 Na content in the scape resulted in a decrease of 1.63 mg · kg^-1 potassium content at flowering, and 1.87 mg · kg^-1 at the end of the experiment. In molecular terms, a 10 mmol · g^-1 increase in the Na content was associated with a 0.96 mmol · kg^-1 decrease in the K content at flowering and a 1.10 mmol · kg^-1 decrease at senesced flowers.

Figure 18.

The pH of the water extracts did not show any correlation with the concentration of the salt treatments (Figure 19). However, it was observed that all scape samples collected at full flowering had lower pH than the initial sample. For the flowers collected at the end of the experiment, pH was higher than that of the initial sample (regardless of the salt treatment). Mean pH values and their corresponding standard deviations (13 treatments) are given as follows:

scapes: open flower stage: 5.09 ± 0.14; wilted stage: 5.61 ± 0.30,

Figure 19.

The EC value of the water extracts increased in both plant parts (at both sampling times) as a function of salt concentration and was well described by both linear and second-order polynomial functions (R² = 0.95–0.98). The results of the two samplings were almost the same. In treatments with a concentration of >15 mM, the EC of the water extract of the scapes was higher than that of the flowers, and the difference between the EC of the extract of the scapes and flowers became higher with the increase in the salinity of the solution in the vases.

In all treatments, the pH level of the saline solutions in the vases was lower than the initial value at the first measurement (54 hr after the start of the experiment), but no effect of the saline solutions was detected. Overall, the mean value for the salt treatments was 5.29 compared to a mean value of 5.53 for the control. The pH measured at the end of the experiment decreased further at lower concentrations (<60 mM), while values above concentrations of 50 mM increased. However, no difference was found in the 60–100 mM range as a result of the change in salt concentration. The mean value of the treatments in the higher concentration range was 5.99 mM, while in the 5–50 mM treatments, it was 5.01 (Figure 20).

Figure 20.

The EC value of the vase solutions was well described by a linear trend with increasing concentrations. The differences in coefficients of the slope (x) of the linear function fitted to the data also indicate that the initial (0 hr) values and those measured during the water usage of plants at higher salt concentrations differed significantly. Comparing the values with the initial data, the EC value of the solution decreased by 1.23 mS · cm^-1 (17%–19%) at the highest concentrations (90–100 mM) compared to the time of the first measurement. However, there was no significant difference between the values of the open flowers and the senesced flowers.

After 168 hr from the start of the experiment, the refractive index (Brix) value of the solutions in the vases and the salt concentrations showed a linear correlation (Figure 21). The correlation was close (R² = 0.996), with a 10 mM concentration increase lifting the Brix value to 0.093. The effect of a 20 mM concentration change was generally clearly distinguishable with the instrument used in the experiment (0.1 precision). All treatments showed higher values at the end of the experiment than at the start. The linear function fitting the two data series was steeper for the data measured near the end of the experiment than that of the initial data series.

Figure 21.